Chess

The Easiest Sicilian
A Black Repertoire with 1 e4 cs 2 tlJf3 tlJc6

GM Atanas Kolev
GM Trajko Nedev
Chess Stars
www.chess-stars.com
Current Theory and Practice Series

The Easiest Sicilian
Translation and editing by Semko Semkov

Cover design by Kaloj an Nachev
Copyright 20 0 8 by Atanas Kolev and Trajko Nedev
Printed in Bulgaria
ISBN: 978-954 8782 66-1
Contents
1 e4 c5 2 t'3 c6
Foreword by Atanas Kolev
Part 1
The Rossolimo Variation 3 bS
Part 2
The Positional Variation 3 d4 cxd4 4 lt:Jxd4 lt:Jf6 S lt:Jc3 eS

6 lt:JdbS d6 7 gs a6 8 lt:Ja 3 bS 9 lt:JdS e7
44
Part 3
10 xf6 xf6 11 c3 0 -0
64
Part 4
12 lt:Jc 2 gs 13 a4 bxa4 14 ruca4 as lS !c 4 !gb8 16 b3 c>h8 17 lt:Jce3 92
Part S
Alternatives to the Main Line after 9 xf6 gxf6 10 lt:JdS fS
llS
Part 6
9 hf6 gxf6 10 lt:Jd S fS 11 exfS MS
129
Part ?
The Main Line 9 xf6 gxf6 10 lt:JdS fS 11 d3 e6
140
Part 8
12 0-0
160
Part 9
6 lt:JdbS d6 7 lt:Jd S
178
Part 10 Unusual Seventh Moves 6 lt:JdbS d6
193
Part 11 Unusual Sixth Moves
20 1
Part 12 Rare Lines 3 c3 ; 3 lt:Jc3 lt:Jf6 4 eS
216
Part 13 The Novosibirsk Variation 9 hf6 gxf6 10 lt:Jd S g7
223
Bibliography
Books
Opening for White According to Anand, vol. 10 by Alexander Khalifman,
Chess Stars 2007
The Complete Sveshnikov Sicilian by Yakovich, Gambit 2005
The Sveshnikov Reloaded by Rogozenko, Quality Chess 2005
The Sicilian Defence. The 5 . . . es System (in Russian) by Sveshnikov,
Fizkultura i Sport 1988
The B bS Sicilian by Richard Palliser, Eve ryman Chess 2005
Periodicals
Informator
New in Chess
Chess Today
Internet resources
Databases
The Week In Chess (chesscenter.com)
10 Days (Chessmix.com)
Inte rnet Chess Club (chessclub.com)
ChessPublishing.com forum
Chesspro . ru
Foreword
About the Title
11 c3 i.gS 12 tlJc2 0 - 0 13 a4 bxa4 14
The book was already finished, but

I still had doubts how to entitle it.
At first I thought about "The Most
Controversial Sicilian. " It seemed a
proper name for a repertoire, based
on the Sveshnikov. Indeed, for only
30 years, this brainchild of sever
al players from Chelyabinsk has
passed through the phases of total
denial, angry attempts of refuting
this defiance to the classical laws of
positional chess, suspicious accept
ance, to be finally adopted by most
leading grandmasters as Kasparov,
Topalov, Kramnik, Leko, Khalif
man, to name a few.
Then a series of internet blitz
games struck me with another char
acteristic feature of the modern
Sveshnikov.
I realised how easy it was to
include it in one's repertoire?
Most of my games reached in
seconds the position on the follow
ing diagram.
1 e4 cS 2 tlJf3 tlJc6 3 d4 cxd4 4
tlJxd4 tlJf6 S tlJc3 eS 6 tlJdbS d6 7 i.gS
a6 8 tlJa3 bS 9 tlJdS i.e7 10 hf6 hf6
!!xa4 as lS i.c4 !!b8 16 b3 <jf h8
We are already in the middle

game, but independent play is still
far ahead. Furthermore, Black's
plan is obvious. He wants to push
f S right away or after . . . g6 in case
White plays 17 tlJe 3 . Strategical
ly, the Sveshnikov is a rather sim
ple opening. You read part 3 and
4, leaf through the "Quick Reper
toire" chapters of the other parts of
the book, and you are ready to test
a whole new Sicilian!
The so-called Positional varia
tion against the Sveshnikov has be
come lately the first choice of White
s
Foreword
players of all levels. A quick check
in my database shows that in
200 6 - 200 8 it occurred two times
more often than the lines with 9
.ixf6 . Apparently fashion, but also
fear of the sharper variations, have
a strong impact on White's prefer
ences. Otherwise it is difficult to ex
plain this fondness of a line which is
too well explored, aspires to a small
positional edge at best, and is of
ten rather boring . Of course Anand
or Shirov may have every reason to
like it, provided it brings them full
points sometimes, but they have su
per technique and deep analyses of
the a rising positions and even end
games . Thelowerthe level, theworse
are White's statistics. Below 2400,
first players scored only about SO
percent in the last two years .
Currently I do not see any
serious theoretical problems
for Black .
I worked hard to neutralise
two fresh ideas of Khalifman and
Anand, and hope that our improve
ments will withstand practical test.
I show that Black's bishop pair is a
fair match to the "magical " control
of the centre, that attracts so many
white players. Most importantly, I
propose an ambitious repertoire,
where White must take considera
ble risks if he wants to aspire even
to the slightest advantage . My aim
was not so much to offer a survival
guide for Black, but rather pick out
variations that lead to rich and dou
ble-edged play, with decent winning
chances for him. I rejected from the
repertoire all the lines where Black
6
would be playing for only two re

sults . For instance, in the Positional
variation I recommend 11. . . 0-0 ,
while 1 1 . . . gS, followed b y 12 ... CiJe 7,
is left for a backup line.
I follow the same approach after
1 e4 cs 2 CiJf3 CiJc6 3 bs. White of
ten tries to ki l any life in the posi
tion, hoping to squeeze us without
any risk thanks to his flexible pawn
formation. I devoted 36 pages to ad
vocate 3 . . . CiJf6 ! in this popular sys
tem. You will find important new
plans, developed by me or Nedev,
which bring about double-edged
unbalanced play. The fine point of
this provocative move is that White
must pick up the gauntlet and push
eS at some moment, or he should
forget about opening advantage. Af
ter eS, however, Black obtains clear
counterplay. In some lines he can
even castle long.
About the Authorship
I have been analysing the Svesh
nikov for years with my friend GM
Vasil Spasov. It is his main reper
toire as Black, while I was more in
terested for the White side. Gradu
ally I discovered that Black was in
perfect shape and I started playing
it for both colours. When I finished
war king on "The Sharpest Sicilian",
I decided to go on with this series
and write about the Sveshnikov.
As a coach of the Bulgarian wom
en's team, I had enough experience
with explaining the most topical
lines of that system. Still, I felt that
I needed an outside critical view on
my analyses. Thus I contacted GM
Foreword
Nedev, who is one of the most de
voted protagonist of the Sveshnik
ov and has ample practical expe
rience. We went together through
all my files to synchronise our as
sessments . During the last year, we
had to repair some variations in the
Rossolimo, (3 bS) analyse the new
ideas, developed by Khalifman, and
fight the sneaky novelty of Anand
against Shirov in Linares 20 0 8 .
W e also dropped some lines o f the
Novosibirsk variation, which turned
to be unfit for playing for win.
The result is a repertoire book
for Black which deals with positions
arising after 1 e4 cS 2 l2Jf3 l2Jc6 . It
covers the Rossolimo (1 e4 cS 2 l2Jf3
l2Jc6 3 bS) and the Sveshnikov, (1
e4 cS 2 l2Jf3 l2Jc6 3 d4 cxd4 4 l2Jxd4
l2Jf6 S l2Jc3 eS) as well as some rare
lines after 1 e4 cS 2 l2Jf3 l2Jc6. The
wide range of other Anti-Sicilians
are beyond the scope of this book.
We have not aimed to offer a his
torical survey or complete study on
the Sveshnikov. We have endeav
oured to provide a sound, yet ag
gressive repertoire, with a focus on
the most topical lines .
About the Structure
This book is above all a practical
guide, so I have arranged the ma
terial in an order of importance.
The Rosso limo variation is a fre
quent guest in tournaments and it
is useful to know it even if you are
not a Sveshnikov fan. Then comes

the Positional variation, which is
the centre of our repertoire. Thus
you'll be able to start playing the
Sveshnikov even before finishing
the book.
The closing Part 13 considers
the Novosibirsk variation. (1 e4 cS
2 l2Jf3 l2Jc6 3 d4 cxd4 4 l2Jxd4 l2Jf6 S
l2Jc3 es 6 ltJdbS d6 7 gs a6 8 l2J a3
bS 9 hf6 gxf6 10 ltJdS g7) It is a
stand-alone system which is out
side our repertoire. We included it
to provide you with a backup line.
You might also want to employ it as
a surprise weapon.
The presentation fallows the
Chess Stars trademark structure,
introduced by "The Safest Sicil
ian" . Every system is examined in a
separate part which contains three
chapters: "Quick Repertoire"; " Step
by Step " ; "Complete Games" .
You start with the "Quick Re
pertoire" . You'll find there all the
vital information that you need to
start playing the variation. These
chapters contain more explanation
and try to extract the essence of
the numerous variations, analysed
branch by branch in the "Step by
Step" chapters.
Finally, the "Complete Games"
sections give practical examples
and sometimes cover backup lines
of the main repertoire.
I suppose that players above Ela
1900 will benefit most of this book.
.
Atanas Kolev
April 2008
Part 1
1 e4 c5 2 lll f 3 lll c 6 3 ibS

QU IC K REPERTO I RE
The most frequent move you are go

ing to face after 2 . . . l!Jc6, is 3 bS.
The lower your opponent's rating
is, the higher the probability of get
ting some Anti-Sicilian with bS.
The so-called Rossolimo Variation is
often seen nowadays even at high
est level. It is a fine choice if White
wants to play "on understanding",
or simply has not done his home
work in the open main lines .
Do not neglect this system in
your preparation as it is deceptively
innocuous. We often defend this po
sition with both colours and we are
well aware of how rich and interest
ing variation the Rossolimo is.
3 ... lllf6
3 . . . g6 is a solid alternative, but
our choice goes for the text due to
several reasons:
1. 3 . . . l!Jf6 allows to build up a
repertoire which is independent of
tricky move orders. For instance,
if White tries 3 l!Jc3, we are happy
to answer it with 3 . . . l!Jf6, not being
afraid of 4 bS . Otherwise 3 l!Jc3
would have been awkward, since
3 . . . g6 could be met by 4 d4 .
2 . By attacking the e4-pawn, we
8
restrict White's choice, because he

has not castled yet so he is unable
to protect it with E'!:el.
3. Should White attempt to slow
torturing us by damaging our pawn
chain with 4 hc6 dxc6 fallowed by
5 d3, we succeed in leading out our
light-squared bishop to g4. This is
an unexplored plan, which leads to
original positions. It has been de
veloped and tested by Kolev and we
are going to arm you with our analy
sis, to ensure you some competitive
advantage over your opponents.
4. The 3 bS adepts usually pre
fer to a void sharp opening lines and
unbalanced positions. That might
make them uncomfortable in the
most challenging lines which in
volve e4-e5.
After our attack on the e4-pawn,
as early as on the fourth move,
White has to settle for a plan.
A. 4 d3
B . 4 hc6
C. 4 e5
D. 4 We2
E. 4 l!Jc3
1 e4 cS 2 l!Jf3 l!Jc6 3 bS
A. 4 d 3
We propose to unbalance abrupt

ly the game by
4 a5+!? S li)c3 li)d4 6 .b4
b5 7 .ib3 li)xb3 8 cxb3 .ib7 9
0-0 d6 1 0 .ig S e6 with totally un
explored play.
B. 4 .ixc6 dxc6
We do not capture with the b
pawn, for White gains the initiative
after c3 and d4. However, such a cap
ture becomes a plausible option at a
later stage of the opening, especially
if White had already played d3 .
5 d 3 .ig4!?
A lot of players refrain from
3 . . . l!Jf 6 in favour of 3 . . . g6 . They be
lieve that the knight is misplaced
on f6 because Black seems unable
to prevent e4-eS, with White's spa
tial advantage. Nedev even made
this system his weapon of choice
as White. We have a fresh idea in
mind, which leads to very interest
ing and complex positions.
6 h3 .ihS!
The key point ! In the overhelm
ing majority of games Black cap
tures on f3 to struggle in a passive
position. Kolev offers another plan:
1. Black wants to play l!Jd7, f6,

eS. I n many cases he castles long:
7 l!Jbd2 l!Jd7 8 l!Jc4 f6 ! ? 9 0-0 es
10 l!Je3 Vlfc7 11 a4 as 12 l!JfS f7+
2 . Should White attempt to pre

vent e7-eS by playing f4, we get
rid of our doubled pawn with . . . c4 ! ,
even at the cost of a pawn in some
lines :
7 f4 c4 !
The idea is to attack the enemy

9
Part 1
pawn-centre with our long-range
pieces, rather than restricting its
mobility by clamping on d4.
8 lLJc3 cxd3 9 cxd3 lLJd7 10 d4 e6
11 e2 lie? 1 2 0-0 0-0 13 adl e8
14 e3 as 15 fel
Superficially, White's _pawn cen

tre should ensure him an edge. On
the other hand, Black has no weak
nesses, and all his pieces are well
placed. The queen has a fine retreat
to a6 , the knight could head for c4
via b6 .
The game Grischuk-Kolev, rapid
Mainz 2 005 followed with 15 . . . ac8
16 /ih 2 , when 16 . . . lLJb6 ! would have
completely levelled the game .
9 .. .f6 10 0-0 d6 11 el 0-0 -0 12

lLJc3 hSt.
White is overextended, with
many weaknesses, while being una
ble to attack anything for his part.
The thematic thrust . . . c4 also
gains in strength after 7 g4 .ig6, for
instance: 8 lLJc3 c4 !
9 gS lLJd7 10 dxc4 ihS ! 11 d3

croo .
Black is enjoying the bishop pair
advantage and very active pieces.
c. 4 es lll dS s o-o lll c7 6
.axc6 dxc6 7 h3 (7 d3 .ig4) its
3. If White advances h is g-pawn,

we get fine counterplay by . . . h7-h5 !
followed up with . . . h4, .ig6-h5:
7 g4 g6 8 eS? ! lLJdS 9 e6
Black solves the opening prob

lems as in the previous line, by lead
ing out the light-squared bishop to
g4 or f5. The c7-knight is longing to
reach d4 via e6 or bS. In case of ex
change on d4 we typically recapture
10
1 e4 cS 2 4J f3 4Jc6 3 ibS
by queen, in order to retain coun
terchances along the d-file with the
breakthrough c5-c4. Note howev
er, that if our bishop were stuck on
c8, we should t ake by pawn in order
to open up the long diagonal - see
game 1 Mortensen-Ermenkov,
Riga 198 1.
In this line Black does not
hurry with castling! His first
task is to activate the c7-knight.
Then he could decide to advance his
kingside pawns or attack in the cen
tre with . . . c5-c4.
5 0-0 4Jc7 6 hc6 dxc6 7 h3 ifS ! ?
8 d 3 h 6 ! 9 l:iJ bd2 e 6 1 0 V!!e2 l:iJ bS !
Now we are going to exam

ine positions where White defends
the e4-pawn by queen and castles
short. That significantly changes
our plans, for we are unable to de
velop the light-squared bishop as
comfortably as in the previous lines.
Accordingly, we must seek another
place for our dark-squared bishop,
too. In all subsequent lines which
we consider, it goes to g7.
Dl. White gains space with 5 eS;
0 2. White builds up a pawn cen
tre with c3 and d4.
0 1 . s e s ds 6 o-o
Occasionally, White attempts
to grab a pawn by 6 c4, but then
6 . . . 4Jcb4 ! is quite awkward for him.
6 ... li:) c 7 ! 7 xc6 dxc6 8 h3 .ig7
9 d3 0-0
11 4Je4 4Jd4 12 4Jxd4 W!xd4 13 4Jg3

ig6 14 @h2 hS ! See for more details
game 2 Movsesian-Chuchelov,
Bundesliga 2005, where Black had
the initiative.
D. 4 'Mfe2 g6
A typical position. Black should

aim for a kingside attack with the
help of .. .f6, gS, h6, intending .. .f5.
He activates the a8-rook with the
manoeuvre . . . b6 , . . . a7-a5, .. . a7.
The game Minasian-Gagunashvili,
Dubai 200 3 saw further:
10 li:) bd 2 li:) e6 1 1 li:) b3 as 1 2 a4
b6 1 3 'Mfe4 ga1 1 4 'Mfh4 f6 ! 1 S g e 1
'Mf d S 1 6 e3 g S 1 7 'Mf g 3
11
Part
Here, instead of 17 . . . hS, Black

should have chosen 17 . . . h6 18 h4 g4
19 llJfd2 fSt . You can see another ex
ample of this plan in game 3 De la
Paz-Handke, Havana 20 0 3 .
0 2 . 5 0 - 0i g7 6 c3 0-0 7 d 1
A fashionable move, which aims
to avoid the old main line 7 d4 dS ! ?
8 e s tlJ e 4 9 .ie3 cxd4 1 0 cxd4 id?
with a fine game for Black.
7. . .e5
I n a n earlier stage of the open

ing, when White was better devel
oped, we used to take with the d
pa wn, in order to keep control of
the centre. In the diagram position,
d4 is not a threat, so we follow the
basic rule to "capture by pawns to
wards the centre".
1 1 i e3 d6 1 2 b4 cxb4 1 3 cxb4
d 5 1 4 ac1 .id7 1 5 .id2 f! e7 1 6
ll:) b 1 ll:) h 5 1 7 a3 ll:) f4 1 8 i xf4 ext4
1 9 f! d 2 g5 with initiative in Tsesh
kovsky-Sveshnikov, Minsk 1976.
E. 4 li:)c3
In the previous lines we have
seen that Black is fine if he achieves
the manoeuvre tlJf6-dS-c7-e6(bS)
d4. Therefore White's most testing
options are connected with limiting
the scope of the f6-knight. The most
fashionable response is 4 . . . Wfc7 aim
ing to prevent e4-eS.
The point ofmy(T N ) reper
toire is not to prevent this ad
vance, but rather to provoke it!
This approach may be risky, but
it does offer more chances to win
the game as the resulting positions
are highly unbalanced strategically.
So we play:
4 g6
White can vary the move or
der, but basically he has two major
plans:
1. White plays .ie3 , Wfd 2, il.h6
and eventually castles long, hoping
for a kingside attack. Black's de
fence is based on . . . es.
2. White restricts o ur knight
with h3, es, g4.
.
8 li:) a 3 ea 9 d 3 a6 10 i xc6
bxc6 !
12
1 e4 cs 2 tlJf3 ttJc6 3 bs
1 . 5 xc6 dxc6 6 h3 g7 7 d 3
0 - 0 8 e 3 b6 9 'Mf d 2 e5 !
1 0 h6
Or 10 0-0 tlJhS ! Black h a s good
counterplay in the centre. He only
has to find the right timing for
cS-c4.
1 o . . . 'Mfd6 1 1 o-o-o as 1 2 xg 7
@ x g7 1 3 h 2 a4 1 4 g 4 g 8 !
Current practice i s favourable

to Black, who stays solidly on the
kingside, while maintaining fair
chances for progress on the other
wing.
g4, the move order of the latter vari

ation is not too precise, since after 8
d3 Black can use the difference with
the main line (where the bishop is
on g7, but the knight is still on g8)
by playing 8 . . . tlJfS ! ?
5 . . . g 7 6 e 5 g8 7 xc6 dxc6
8 d3
8 e2 tlJh6 9 ttJe4 b6 10 d 3 tlJfS
11 gS ttJd4= is analysed in the " Step
by Step" chapter, line E3a.
8 . . . h 6 9 g4
Be sure to meet 9 e3 with
9 . . . aS ! , but not with the com
mon 9 . . . b6? ! which would leave our
strongest piece without prospects.
The key point of our treatment
of these positions is to activate
the queen.
We further examine 10 g4 fS 11
gS tlJf7 12 f4 e6 13 d2 0-0- 0 ! ? see "Step by Step" line E2 .
9 . . 0-0
.
2. 5 h 3
White is following the restrict
ing strategy, started on the previ
ous move. In fact, S eS tlJg4 6 hc6
dxc6 7 h3 tlJh6 often leads to the
same positions. However, if White
does not intend to follow up with
13
Part 1
This is the basic position for line
E. White's primary task is to deprive
the opponent of counterplay. At the
same time he should notforget about
development.
For his part, Black must activate
the h6-knight and find targets in
the enemy camp. He cannot survive
without pushing the f-pawn, but the
question is which move is best, f6 or
fS? Initially we thought that we must
open up the kingside at all cost, so
the answer of that question depend
ed on the placement of White's bish
op: if it went to f4, we would play
.. .f6, while .ie3 would be attacked
by . . .fS . Let us show examples:
a) 1 0 j.f4 f6
b4 ! 14 e3 xb2 with an attack.)

1 1 . . . exf6 12 d2 l2Jf7
Black is well coordinated and

has the bishop pair in an open po
sition. Should White grab a pawn,
we'd get tempi to overrun him on
the queenside: 13 hes e8+ 14 .ie3
bS ! 15 0-0-0 b4! ?
I t remains t o examine White's
third plausible option on move 10:
c) 10 Wfe2
White is unable to keep control

of the es-outpost with 11 e2 , due to
the hit 11 . . . l2Jxg4 ! 12 hxg4 fxeS, Cu
bas-Nedev, Calvia, ol 2004, so he
should prefer 11 d2 fxeS 12 hh6
hh6 13 xh6 f3 14 0-0-0 f8 15
xf8 + @xf8 16 h2 ! The endgame
looks better for White, but the the
matic sacrifice c5-c4 balances the
game, for instance 16 . . . h6 17 el
c4 ! ? 18 dxc4 .ie6 .
b) 1 0 e3 f5 ! 1 1 exf6 (Follow
ing 11 gS? l2Jf7 12 .if4 Black has an
extra tempo for 12 . . . aS ! 13 e2
14
While preparing to castle, White

is waiting for us to reveal our plans.
Thus, he will meet 10 . . .f5 with 11
gS l2Jf7 12 .if4! Although this posi
tion is far from clear, we prefer to
reach better versions of it, for in
stance, with an extra tempo as af
ter 10 .ie3 fS ! 11 gS? l2Jf7 12 .if4. The
real venom of 10 e2 is concealed
in the variation 10 .. .f6 11 .id2 ! ? It
turns out that Black's threat to the
1 e 4 c s 2 tlJf3 tlJ c6 3 bs
eS-pawn can be ignored. White sim
ply develops, counting on his bet
ter pawn structure as in game 5
David-Nedev, ECC 20 07.
However, we can outsmart the
enemy by attacking the b2-pawn
with:
1 0 . . . %Yb6 ! ?
Now, the bishop is pinned to cl

and White is suddenly faced with a
difficult problem: how to disentan
gle his pieces. 11 b3 f 5 12 tlJ a4 (12
gS tlJf7 13 d 2 tlJd8 14 0 -0 -0 tlJe6
15 del tlJd4 16 tlJxd4 cxd4 17 tlJ a4
a6 18 bl bS-+) 12 . . . c7 13 hh6 ,
Wen,Yang-Zhao Jun, Wch U20
Yerevan, 2006, 13 . . . fxg4 ! 14 hg7
gxf3+.
Remember! When White
restrains our knight by g4, we
must look for counterplay by
attacking the queenside with
our queen. Therefore, we play
b6 only against e4.
15
1 e4 c5 2 llif3 lll c6 3 .ibS
Part 1
STE P BY STE P
lll f6
A. 4 d3
B. 4 .bc6
c. 4 es
D. 4 Wie2
E. 4 lZJc3
page
page
page
page
page
17
18
22
23
25
itiative b y opening the centre with

d2-d4. In our case this break would
be with a loss of tempo, so nothing
can stop us fram completing devel
opment.
Besides, we have the daring al
ternative:
4 'Wa5+!? 5 lll c 3 lll d4 6 .ia4
6 ic4 only gives Black the extra
possibility of 6 . . . dS ! ? 7 .bdS lZJxdS
8 exdS ig4.
6 bS 7 i.b3
Or 7 idS lZJxdS 8 exdS ib7 9 0-0
b4.
7 lll x b3 8 cxb3 i.b7
There is no need to hurry with
8 . . . b4, because 9 lZJa4 ! unpleasant
ly targets the cS-pawn. (9 lZJe2 g6!
is OK for Black.)
A. 4 d3
This modest move poses some

move order problems. In case of
4 . . . g6 5 .bc6 dxc6 6 h3 White will
throw us out of the proposed reper
toire. Not that the position is diffi
cult for Black, but it is too static for
our taste . Perhaps 5 . . . bxc6 is more
precise. Usually this recapture is
risky , because White gains the in16
1 e4 cS 2 ltJf3 ltJc6 3 bS
9 0-0
9 gS dS opens up play in Black's
favour due to his bishop pair: 10 0-0
( 10 hf6 dxe4 11 dxe4 gxf6 12 0-0
e6oo) 10 . . .dxe4 11 ltJxe4 ltJxe4 12
dxe4 f6oo .
9 d6 1 0 .igS e6.
This analysis is , of course, just a
starting point for further investiga
tion.
B. 4 xc6 dxc6 5 d3 .ig4!?
Bl. 7 g4
B 2 . 7 llJbd2
B 3 . 7 llJc3
B4. 7 f4
81. 7
A key point in our repertoire.

We catch the chance to lead out
our problem bishop . This possibil
ity is one of the major advantages
of 3 . . . llJf6 over 3 . . . g6. Note howev
er, that our idea is not just to get rid
of our light-squared bishop by trad
ing it for the f3-knight!
6 h3
Hoping for 6 . . . hf3 7 Wxf3 with a
slight edge . In case of 6 ttJbd2 Black
fallows his main idea to clamp on
d4 by 6 . . . llJd7 7 h3 hS 8 g4 g6 9 9
ltJc4 f6 ! when 10 eS? does not work:
10 . . . bS 11 llJcd2 hSt.
6 J.h 5!
We should not part light-heart
edly with the bishop pair.
g4 .ig6
Bla. 8 eS? !
Blb. 8 llJc3
8 ltJeS ltJd7 9 ltJxg6 hxg6 is not
appealing , for the stranded pawn on
h3 is a serious drawback of White's
structure: 10 e3 es 11 llJd 2 d6 12
We2 ltJf8 13 0-0-0 ltJe 6=.
Bla. 8 e5? ! d5 9 e6
This looks like a n overoptimis

tic way of treating the position.
White neglects development, seek17
Part 1
ing structural benefits. His prob
lem is that he is overextended. After
. . . h7-hS he will have to worry about
many weaknesses, while being una
ble to attack anything for his part.
The first interesting option in
the diagram position is 9 . . . hS ! ? 10
ltJeS Wf d6 11 exf7+ hf7 12 l2Jxf7 (12
Wfe2 hxg4 13 l2Jxf7 xf7 14 xg4
Wies+ 15 Wfe4 Wff6t) 12 . . . e6+ 13
Wfe2 Wfxf7= and Black's lead in de
velopment gives him good chances.
We'll consider also:
9 . . .f6 1 0 0 - 0
10 l2Jh4? ! loses a pawn to 10 . . . Wfd6
11 Wf e2 l2Jc7 12 f4 l2Jxe6 13 fS ltJd4+.
10 Wf e2 d6 11 0 - 0 0-0-0 12
l2J c3 (12 l2J a3 hS 13 l2Jc4 c7 14 l2Jh4
e8t) 12 . . . hS 13 l2Jh4 e8 14 l2Jf5
Wf c7t is quite good for Black, but
adventurous players may try also
10 . . . l2Jb4 ! ? with the idea of 11 0-0
(11 a3? hd3 ! ) 1 1 . .. c4 ! . Still we pre
fer to develop our pieces.
10 Wfd6
10 . . . c7 ! ? is another attrac
tive optio n: 11 l2Jh4 0-0-0 12 f4 e8
planning hS. Black's bishops enter
play through the kingside.
11 gel 0 - 0 - 0 12 c3 h5 t.
Blb. 8 c3 c4!
18
9 g5
9 ltJeS cxd3 10 cxd3 ltJd 7 11 l2Jxg6
hxg6 12 e3 eS ! =
9 d7 1 0 dxc4 h5 ! 11 Wfd3
Wfc7oo.
Black is enjoying the bishop pair
advantage and very active pieces.
8 2. 7 ll) bd 2 tl)d7 8 tl) c4

White is trying to prevent . . . es.
8 ltJfl is a plausible alternative. We
can fallow up with 8 . . . es 9 l2J g3 (9
g4 g6 10 l2Jg3 is risky, for it could
be attacked with the thematic sacri
fice 10 . . . hS 11 gS h4 12 ltJfS ih5 13
b3 Wfc7 14 ltJSxh4 0-0-0 with com
pensation) 9 . . . g6 10 0-0 d6 .
(10 . . .f6 11 l2Jh2 f7 12 f4 c4 is un
clear , but Black is somewhat unde
veloped for such a committing ac
tion) We refrain from .. .f6 in order
to discourage 11 ltJf S, which would
stumble into 1 1 . . .hfS 12 exfS f6 .
Thus we win a tempo for the ma
noeuvre l2Jd7-f8-e6. Play could con
tinue with 11 gS (or 11 d2 l2Jf8 12
ltJfS l2Je6 13 c3 f6) 11. . .f6 12 e3
l2Jf8 13 ltJf5 l2Je6 14 ltJ 3h4 f8 15 a4
f7 with a solid position. Black con
trols firmly f4 and will gradually re
pel the enemy knight fram f5.
1 e4 cS 2 llJf3 llJ c6 3 bS
8 g4? ! g6 only gives us more
chances for counterplay:
9 h4 hS 10 gS Wic7 with a possi
ble break f7-f6 .
9 eS e6 10 Wie2 hS 11 gS Wic7 12 b3
0-0-0 13 b2 h4 14 0-0-0 e7 f!-. In
this line Black could aslo try ll . . . h4 ! ?
1 2 llJe4 (12 b3 WiaS 1 3 b2 0-0-0+!)
12 . . . c4 13 dxc4 llJc5 14 llJxcS xcSoo.
9 llJc4 f6 10 0-0 es 11 llJe3 V!ic7
12 a3 Alekseev-Eljanov, 2 004 ,
12 . . . 0-0-0 .
8 . . f6 ! ?
M y (A. K.) treatment of this po
sition is to protect the light-squared
bishop whenever possible. The
game B aklan-Nedev, 2 007 saw in
stead 8 . . . V!ic7 9 d2 eS 10 a4 aS 11 g4
g6 12 llJh4 f6, when 13 llJxg6 hxg6
14 gSt would have been in White's
favour.
9 o-o e5 1 0 e3 Vlf c7 1 1 a4 as
1 2 f5i f7
Black has a satisfactory game,
for instance , 13 b3 g6 14 llJ h6 e6 .
.
83. 7 c3 d7 8 g4
O r 8 e3 e S 9 g 4 g6 1 0 Wid2
e7 11 0-0-0 Wic7 12 llJxeS llJxeS
13 f4 0-0-0 (13 . . . llJf3 ? ! 14 Vlif2 llJ d4
lS fS 0-0-0 16 fxg6 fxg6t) 14 fxeS
V!JxeS= .
8. i
. g 6 9if4
It is very instructive to observe
how Black could take over the ini
tiative if White attempts to display
an activity: 9 es e6 10 gS (10 Wie2
c4 ! Remember this sac ! It proves
especially effective when White has
weakened his centre. 11 dxc4 Wib6
12 a3 hS 13 gS h4+!) 10 . . . V!JaS 11 V!id2

h6 12 h4 c4 !
Killing two birds with one shot.

Said in chess terms, Black opens up
diagonals to both bishops. 13 dxc4
b4 14 0-0-0 llJcS f!- lS hel 0-0 16
e7 llJb3 + 17 cxb3 he7oo . If you
do not believe in the bishop pair's
power, look at the following varia
tion: 18 llJd4 ad8 19 V!ie3 d7 2 0
d2 fd8 2 1 edl b4 ! 2 2 llJ c 2 ( 2 2
a4 cSt) 2 2 . . . xd2 23 xd2 xd2 24
V!Jxd2 hc2 2S i>xc2 V!JxeS+.
9 . . . h5 ! ?
The thematic break 9 . . .f6 1 0
es c4 11 dxc4 i s not so clear, be
cause White controls firmly eS , e.g.
11 ... V!ib6 12 bl ! oo The text is more
consistent.
1 0 g5
Alternatively: 10 llJh4 h7 11
19
Part 1
gxhS (or 11 d2 eS 12 gS e7 13
J,xe7 xe7 14 l!JfS hfS lS gxfS
0-0-0 16 0-0-0 gS=) 11 . . . eS 12 g3
b6 13 d2 g8 recapturing the h
pa wn.
1 o .. h4! 1 1 Ye2 i.h5
This is a fine setup for Black.
.
84. 7 i.f4 c4 !
We see here a typical method of

puting pressure on White's centre.
This approach to solving the open
ing problems has been introduced
in the rapid game Grischuk-Kolev,
M ainz 20 0S. The idea is to attack
the enemy pawn-centre with our
long-range pieces rather than re
stricting its mobility by clamping
on d4.
8 ti) c3
Following 8 0-0 Black could
choose 8 . . . cxd3 as in the main line ,
or the more risky 8 . . . e6 9 l!Jc3 b6 .
I (A. K.) have reached i n m y analysis
a lot of funy positions after 10 E:bl
E:d8 11 d4 b4 12 g4 g6 13 Wffe 2
as 14 gs hS ! ? , for example lS es
hxg4 16 hxg4 l!JdS 17 d8 Wxd8 18
l!JxdS xdS 19 Wg2 fS 20 exf6 gxf6
2 1 E:hl E:g8co. In short, we coun20
terattack all over the board , trying

to unleash the power of our light
sq uared bishop.
8 . . . cxd3
A solid move which offers Black
good chances. Besides , we could
speed up play with 8 . . . Wffb 6? ! , but
we are undeveloped for such ac
tions. White could sacrifice the b 2pawn, as 9 0-0 Wffxb2? (9 . . . e6 i s bet
ter) 10 d2 hf3 11 E:abl Wff a 3 12
E:xb7! would be disastrous for him,
and even 9 E:bl E:d8 10 d4 e6 11 g4
g6 12 Wffe2 b4 13 0-0 as 14 gs
looks good enough .
9 cxd 3 ti) d7
The move order is not too im
portant. Degraeve-Zhao Jun, Paris
2006 saw 9 . . . e6 10 0-0 e7 11 Wff e2
when 1 1. . . 0-0 12 E:adl l!Jd7 would
have transposed to our main line .
1 O d4 e6 1 1 Ye2 i.e7
11. . . f6 is risky. Such a devel
opment of the queen is typical for
some lines of the QGA , but here the
knight is too passive on d7. White
should fallow with 12 e3 hf3 13
gxf3 b4 14 E:gl h6 lS eS ! fS 16
E:xg7 l!JxeS 17 dxeS WffxeS 18 E:g4
J,xc3 + 19 Wfl as when 20 d4
xe2 + 21 Wxe2 E:f8 22 cs E:h8 is
equal, but 20 E:dl ! ? poses concrete
problems.
1 2 0-0 0-0 1 3 ad 1 e8 1 4 Ye3
'Mias 1 5 fe 1
Both sides completed develop
ment so it is time to strike a balance .
Superficially, White's pawn centre
should ensure him an edge. O n the
other hand , Black has no weakness
es , and all his pieces are well placed.
The queen has a fine retreat to a6 ,
1 e4 cs 2 l2Jf3 l2Jc6 3 bs
the knight could head for c4 via b6 .
White's problem i s that his only ac
tive plan is linked with a kingside
pawn storm, but it could easily turn
against him.
hf3 20 gxf3 b4oo) 18 l2Jxa4 V9xa4

19 b3 V9aS.
1 7 . . . .ig6 1 8 es .if6 1 9 b3 c 5=.
(19 . . . e7 ! ? f!)
C. 4 e5 d5
Now I had lS .. . a6 ! ? (control

ling c4) 16 a3 l2Jb6 17 g4 g6 18 ltJeS
(18 h4? l2Jc4 19 V9e2 fS) 18 . . . l2Jc4=,
but the move I have played i s not
bad either.
1 5 . . . ac8 1 6 .ih2
Basically, White i s waiting. Anal
ysis shows that he has no advan
tage, for instance :
16 a3 l2Jb6 (or 16 . . . V9a6 17 g4 g6
18 h4 hS 19 gS l2Jb6 2 0 ltJeS l2Jc4 2 1
V9h3 ltJxeS 2 2 xeS V9c4) 1 7 g4 g6
18 ltJe S V9a6 19 V9g3 l2Jc4 20 h4 ltJxeS
21 hes f6 22 hS f7 23 c7 V9c4 ;
16 V9d3 f6.
1 6...b6!
This simple move would have
completely levelled the game. In
stead I preferred 16 . . . h6? ! 17 a3
g6 (17. . . l2Jb6 ! ? 18 g4 g6 19 ltJeS
h7oo) 18 V9e2 hS 19 V9d3 with a
small edge for White in Grischuk
Kolev, M ainz 200S.
1 7 g4
Or: 17 b3 b4 18 d3 cS ! + ; 17
d3 l2J a4 (17 ... cs 18 dS c4 19 d2
Pushing e4-eS is commonly good

if it repels the knight to a passive
position. In the current situation
it arrives at a central square, when
further advancing by S c4 fails to S . . .
l2Jc7.
Now White has to decide where
to develop the queen's knight so he
chooses, without success :
Cl: S l2Jc3
C2 : S 0-0
C 1 . 5 c3 c7 6 xc 6 dxc6 7
h3 f5 ! ?
This development fits best into
our repertoire. The standard setup
is to fianchetto the bishop with 7 . . .
g6, e.g. 8 d 3 g7 9 e3 b6 10 d2
h6! White would happily trade his
bishop so we should not oblige him.
11 0-0 e6 12 fdl (or 12 l2Je2? dS !
13 d4 .ixf3 14 gxf3 cxd4 lS V9xd4
V9xd4 16 hd4 cS 17 c3 ltJ dS+
21
Part 1
Nanu-Andonov , Belgarde 20 04)
12 .. . d7= preparing ltJdS with a
good position.
8 0-0
We answer 8 d3 with 8 . . . h6 ! in
order to deprive the opponent of 8 . . .
e 6 9 .igS ! .ie7 1 0 he7 V!ixe7 1 1 Wid2
0-0 12 0-0-0 + . Typically for this
line , White's dark-squared bishop
is less useful, than ours.
9 Wie2 l2Je 6 !
Black has less space for manoeu
vering, therefore it is important to
exchange his last short-range piece
through d4. 9 . . . e6 10 ltJe4 ltJbS does
not fulfil that aim due to 11 c3 + .
10 l2Je4 l2Jd4 11 l2Jxd4 Wixd4 12
l2Jg 3 .ie6 13 0-0 c4 ! 14 dxc4 Wixc4 1S
Wixc4 hc4 with a better endgame
for Black in Yu Shaoteng-Zhao Jun,
Wuxi 200 6 . (16 E!dl g6+)
8...e 6 9 d3 .ie7 10 Ye2 lll b 5 !
C 2: 5 0-0 c7 6 .ixc6 dxc6 7

h3
7 d 3 i s seldom seen, probably

because the pin 7 . . . .ig4 is quite an
noying: 8 h3 .ihS (We had already
learnt fram line B to keep the bi22
shop ! ) 9 l2Jc3 l2Je6 10 .ie3 l2Jd4. Now

11 g4 removes the pin, but badly
compromises the king's position,
11. . . .ig6 12 hd4 cxd4 13 l2Je2 hS+t.
7 . ...if5 ! ?
7 . . . g 6 i s also good enough : 8 d3
.ig7 9 .ie3 (9 l2Jc3 b6 10 l2Je4 0-0
1 1 .id2 fS ! 12 exf6 exf6 13 Wicl gS
14 l2Jh2 ifs+ was fine for Black in
game 3 De la Paz-Handke, Ha
vana 20 03) 9 . . . b6 10 Wicl h 6 !
The idea i s not only to preserve

the bishop from exchange, but also
to attack the enemy king with gS
g4. In that scheme we castle long , if
at all ! We offer our analysis of that
novelty:
11 a4 aS 12 E!dl (12 l2Ja3 ltJdS 13
l2Jc4 gS ! ?+t; 12 d4? ! cxd4 13 l2Jxd4 cS
14 l2Jf3 .ib7 lS E!dl Wic8+) 12 . . . ltJdS
13 .id2 (13 c4 l2Jxe3 14 Wixe3 0-0 lS
1 e4 cS 2 l2Jf3 l2Jc6 3 bS
l2Jc3 Wic7 16 d4 cxd4 17 xd4 e6 18
adl ad8=) 13 ... gS ! ? 14 c4 g4 lS
hxg4 l2Jc7 (lS . . .hg4 16 cxdS hf3
17 gxf3 xdS is interesting, but not
quite sound .) 16 c3 hg4 17 Wie3
(17 f4 hS 18 Wie4 Wid7 19 d4 0-0-0)
17 . . . l2Je6 18 l2Jbd2 d7f!.
8 d3
8 l2Jh4? ! e6 9 f4 runs into 9 . . .
gS ! whereas 9 d 3 g6 underlines the
clumsy position of White's knight,
which has deprived of support the
es-outpost.
8 . . h 6 ! 9 li:) bd 2
9 l2Jc3 e 6 10 Wie2 ltJbS ! transpos
es to Cl. S l2Jc3.
9 e 6 1 O YMe2 li:) b S !
.
Commonly, i n the Rossolimo

Black's knight heads for d4 via e6,
but it has another route, too !
1 1 li:) e4 li:) d4 1 2 li:)xd4 YMxd4 1 3
li:) g 3 .ig6 14 h 2 h 5 !
See for more details game 2
Movsesian-Chuchelov , Bundes
liga 2 0 0S, where Black had the in
itiative.
D. 4 Ye2 g6
This is the most challenging ap-
proach. Black does not hinder eS,

on the contrary, he is provoking it.
01 . S eS
0 2 . s 0-0
S c3 transposes to 0 2, while S
l2Jc3 g7 6 eS l2Jg4 is covered in line
E.
0 1 . s e s li:) d S 6 o-o
Occasionally, White attempts to
grab a pawn by6 c4, but H ausrath's
move 6 . . . l2Jcb4 ! is quite awkward :
a) 7 Wib3 a6 8 c4 e6 9 a3 (9 a4 d6
10 0-0 dxeS 11 ltJxeS g7+ Orabke
Hausrath, Bundesliga 2 0 04) 10 . . .
bxc4 11 dxc4 g7 12 axb4 l2Jxb4= ;
b ) 7 a3 a 6 8 a4 b S 9 e4 bxa4 10
axb4 l2Jxb4 11 0-0 , Jens-H ausrath,
Belgium 2003, when best is 11 . . .
dS ! ? 1 2 exd6 fS 13 es f6 14 xcS
Wixd6 lS Wixd6 exd6+.
6 l2Jc3 seems already late. Apart
from 6 . . . l2Jf4 7 e4 l2Je6 8 c4 g7 9
he6 dxe6 10 0-0 0-0 11 el ltJd 4=
Aronin-Shamkovich, Moscow 1961,
Black has 6 ... l2Jc7 ! ? 7 c4 g7 8
l2Je4? ! 0-0 9 ltJxcS d6t.
6 . . . li:)c7! 7 .ixc6 dxc6 8 h 3 .ig7
9 d3 0-0
23
Part 1
A typical position. Black should

aim to push f7-f6. The game Mina
sian-Gagunashvili, Dubai 2003 saw
further:
1 o ll:) bd 2 li:) e 6 1 1 li:) b 3 as 1 2 a4
b6 1 3 '%Ye4 a7 1 4 '%Yh4 f6 ! 1 5 e1
'%Yd5 1 6 ie3 g 5 17 '%Yg3
Here, instead of 17 ... hS, Black
should have chosen 17 . . . h6 18 h4 g4
19 ltJfd2 fSt.
c6 dxc6 8 d3 ltJh6 9 f4 ltJfS ! ? 10

c3 h6f!, which h a s never been tes
ted.
6 . . . 0-0 7 d 1
A fashionable move, which aims
to avoid the old main line 7 d4 dS ! ?
(This i s slightly more precise than
7 . . . cxd4 8 cxd4 dS 9 eS ltJe4 when
White has the option of 10 ltJc3 .
In that line White's dark-squared
bishop goes to gS whereas the text
practically forces it to e3 .) 8 eS (8
exdS xdS 9 dxcS xcS=) 8 . . . ltJe4 9
e3 (9 ttJbd2? ! is dubious du to 9 . . .
cxd4 1 0 cxd4 ttJxd2 1 1 xd2 b6+
or 10 . . . Wb6 ! ? 11 ttJxe4 dxe4 12 xc6
xc6 13 ltJgS fSt Utemov-Smirin,
Podolsk 1990) 9 . . .cxd4 10 cxd4 d7
with a fine game for Black:
0 2. 5 0-0 i g7
11 ltJc3 (ll d3 8:c8 12 ttJbd2 ttJxd2

13 xd2 aS 14 a3 xd2 lS ttJxd2 f6
= Svidler-Leko, Monte Carlo 20 04)
11 . . . ltJxc3 12 bxc3 ltJ aS ! 13 d3 8:c8 =
Svidler-Shirov, Leon 2004.
6 c3
After 6 eS Black chooses between
the promising pawn sac 6 . . . ltJdS 7
c4 ltJc7 8 xc6 (8 xcS b6 9 c4
ttJxeS 10 ttJxeS xeS 11 E'!:el g7+) 8 . . .
dxc6 9 xcS g4 10 ltJd4 d7 with
good compensation, or 6 . . . ltJg4 7
24
7 . . . e5
The point of White's setup is that
7. . . ds 8 es ltJe4?? is no longer possi
ble, so we have to adjust our plan ac
cordingly. The MegaBase shows Tse
shkovsky-Sveshnikov, Minsk 1976
as the source of the text move.
1 e4 cS 2 l2Jf3 l2Jc6 3 bS
8 a3
8 hc6 dxc6 9 d4 (9 ltJxeS e8
10 d4 cxd4 11 cxd4 l2Jxe4 ! favours
Black) 9 . . . exd4 10 cxd4 cxd4 11
l2Jxd4 leaves White's pieces some
what hanging. We can exploit that
by 11 . . . l2Jxe4 !?+ 12 ltJfS MS 13 xd8
axd8+ with overwhelming advan
tage in development.
8 d3 is innocuous . Black achieves
a good game with natural moves:
8 ... c7 ! ? (o r 8 ... e7 9 gS h6 10
hf6 xf6 11 l2Jbd2 a6 12 c4 bS 13
dS b7 14 l2Jfl ab8 1S l2Je3 l2Je7=)
9 l2Jbd2 a6 10 hc6 bxc6 11 l2Jc4 e8
12 gS dS 13 l2Je3 (13 exdS cxdS+)
13 . . . e6+;
Finally, 8 d4? ! exd4 9 cxd4 cxd4
(9 . . . e8 ! ?+) 10 l2Jxd4 l2Jxd4 11 xd4
b6 gives Black the upper hand.
8 .. . e8 9 d 3 a6 1 0 .ixc6 bxc 6 !
Note this capturing. In an earli

er stage of the opening, when White
was better developed, we used to
take with the d-pawn, in order to
keep control of the centre. In the di
agram position d4 is not a threat, so
we follow the basic rule to "capture
by pawns towards the centre" .
1 1 .ie3 d 6 1 2 b4 cxb4 1 3 cxb4
d5
Or 13 . . . l2J g4 ! ? 14 d2 fS = .
1 4 gac1 .i d 7 1 5 .i d 2 V!J e 7 1 6
b1 h 5 1 7 a3 f4 1 8 .ixf4 exf4
1 9 V!i d2 g5 with initiative in Tsesh
kovsky-Sveshnikov, Minsk 1976 .
E. 4 c3 g6
El. S hc6
E2. S h3
E3 . S eS
S 0-0 g7 6 eS ltJg4 7 hc6 trans
poses to El.
E 1 . 5 .ixc6 d x c 6 6 h 3
6 d 3 does not really save a tem
po, for after 6 . . . g7 7 e3 b6 8 d2
l2Jg4 White has to move the bishop
twice: 9 f4 (9 gS f6 10 h4 0-0 11
h3 l2Jh6 12 g4 l2Jf7 13 g3 es 14 l2Jh4
2S
P art 1
e6+ Van Mil-Kuijf, Antwerp 1997)
9 . . . es 10 g3 f6 11 h3 lt:Jh6 12 lt:Jh2
lt:Jf7 13 f4 0-0= Abreu-Hernandez,
Havana 1998.
6 . . g7 7 d3 0-0
.
8i e3
8 f4 should be attacked with
8 . . . lt:JhS! (8 . . . lt:Je8 is too passive and
passes the initiative to the enemy:
9 Vfffd 2 f6 is the most interesting al
ternative: 10 e3 b6 11 h4! g4 12
lt:J h2 ! e6 13 h S.) 9 e3 Vfffd6 10 Vfffd2
eS 11 0-0-0 bS 12 lt:Je2 lt:Jf6 13 h6 aS
with counterplay in Grischuk-Leko,
Dubai 200 2 ;
8 0 - 0 i s less testing. After 8 . . . b6
9 f4 (e3 eS 10 Vfid2 lt:J hS 11 h6
Vfff d6 =) 9 . . . lt:Je8 10 Vfffd 2 f6 1 1 eS Black
can force play with:
11 . . .fxeS ! (Played in Shirov

Kramnik, Wijk aan Zee 2 004. There
26
i s n o alternative t o this move . If

White consolidates, he will main
tain an edge due to his forepost on
eS) 12 hes (12 lt:JxeS Vfffd4 is equal :
13 lt:Je2 V!ffdS 14 c4 Vfffd6 lS lt:Jf7 Vffff6
16 lt:Jh6 + c;t>h8 17 gs Vfffxb2 18 Vfffxb2
hb2 19 he7 hal=)
12 . . . f3 ! ? (This exchange sacri
fice poses problems to White. The
source game saw 12 . . . hh3 13 hg7
lt:Jxg7 14 fel=) 13 hg7 lt:Jxg7 14
gxf3 hh3 lS fel Vffff8 with good
compensation, e.g. 16 Vfffe 3 lt:Je6 17
f4 Vffffs 18 Vfff g 3 f8 19 es Vfffxf4 2 0
Vfffxf4 f o r 1 6 f4 lt:JfS 17 lt:Je4 lt:Jd4
18 lt:JgS h6 19 c3 V!fffS 20 Vfffe 3 hxgS 2 1
cxd4 gxf4- + .
8 . . . b6 9 '%Yd2 e 5 !
Depending on where White cas

tles, the game could take rather dif
ferent courses.
1 0 i h6
The plan with short castling is
linked with the breakthrough f2-f4,
but it could be effectively opposed
by 10 0-0 lt:JhS ! Black has good
counterplay in the centre. He only
has to find the right timing for cS
c4: 11 h6 Vfffd6 12 lt:Je2 f6 13 hg7
(as usual, the interpolation of 13 a3
1 e4 cs 2 lt:Jf3 lt:J c6 3 bs
aS is in Black's favour, 14 lt:Jh2 :a7
lS :adl e6 16 hg7 :xg7 17 lt:Jg4
:d7= , Vachier Lagrave-Lautier, Val
d'Isere 20 04) 13 . . . lt:Jxg7 14 :adl aS
1S lt:Jh2 gS ! ? 16 lt:Jg4 hg4 17 hxg4 hS
18 gxhS lt:JxhS 19 g3 Wffe6 20 @g2 @f7
21 :hl :h8 22 b3 :h7 23 Wffe3 :ah8
24 Wfff3 lt:Jg7= , Vogt-Shirov, rapid,
Mainz 2 0 0S.
Another implementation of the
same idea is 10 a3 aS 11 0-0 a4 12
:ael :e8 13 lt:Jh2 when 13 ... c4 ! ? splits
the enemy pawn chain (13 . . . lt:Jd7 14
f4 exf4 lS :xf4 lt:Jf8 ! 16 :efl :a7 17
@hl lt:Je6 18 :4f2 lt:Jd 4= Pridorozh
ni-Smirnov, Nefteyugansk 20 0 2) 14
dxc4 Wixd2 lS hd2 e6= .
10
...
Wd 6
that he could conduct a success

ful attack without connecting the
rooks . His intention is to wait till
the last moment and even castle
short at an opportunity.
11 ie6 12 Wfe3 ! ?
Anticipating Black's main threat
of cS-c4. This idea of Ponomariov
has not caught up, but the alterna
tives are not any better:
a) 12 lt:Je2 :ad8 13 hg7 (13 lt:Jg3
lt:Je8 14 hg7 lt:Jxg7 1S lt:Jfl f6 16 lt:Je3
bS 17 Wic3 f7 18 gS b4 19 Wid2 fSf!)
13 ... @xg7 14 Wffc3 lt:Jd7 1S lt:JgS h6! 16
lt:Jxe6+ fxe6 ! = , Bologan-Leko, Dort
mund 20 0 3 ;
b ) 1 2 hg7 @xg7 1 3 lt:J h 4 ( o r 13
:gl :ad8 14 0-0-0 as lS lt:Jh4 c4 !
16 f4 cxd3 17 cxd3 Wies 18 @b l c4
19 :g3 bSt Nevednichy-Gladyszev,
La Fere 2003. White should better
come to his senses and develop his
pieces with 13 0-0-0 :ad8 14 :hfl
bS lS lt:JgS h6 16 lt:Jxe6 + Wixe6 17 @bl
c4= , Shirov-Leko, Monaco 2 0 0 2)
13 . . . :ad8 14 0-0-0 c4 lS f4 exf4
16 d4 Wic7 17 gS lt:JhS 18 dS cxdS 19
Wid4+ @g8 20 exdS :fe8+.
12 a5 13 llle2 a4 14 a3 b5
..
The plot is more or less clear

now. We'll witness opposite attacks
where every tempo counts. (White
still could switch to the calmer 11
0-0 lt:JhS ! which was discussed o n
the previous turn.) I n the diagram
position White chooses between 11
g 4 and 11 0-0-0, while 11 hg7 i s not
of independent significance.
Ela. 11 g4
Naturally White hardly believes
Black gained a lot of space on the

queenside which would give him the
27
Part 1
upper hand in an endgame. White's
chances are down the g-file.
15hg7
I n Ponomariov-Leko, Linares
20 0 3 was 15 llJg3 llJe8 ! 16 0-0 f6 17
llJ d2? ! :gas 18 hg7 ttJxg7 19 f4 exf4
2 0 xf4 xf4 2 1 f4 c4 with ini
tiative.
15 . . . xg716 g3 h8
White should think how to keep
the balance.
The key point of Black's
counterplay in this line is to
push c5-c4 at a moment when
the opponent is unable to re
spond with d4 .
Several games at top level show

that the setup with 13 g4, fallowed
by llJc3-e2-g3, is way too slow. Black
can attack in different manners .
Leko prefers the pawn storm with
b and a-pawn, while Ivanchuk fa
vours piece play:
13 g4 a4 14 llJe2 (or 14 llJh4 bS 15
f4 exf4 16 llJf3 llJ d7 17 ttJe2 a3 18 b3
f6 19 d4 cxd4 20 gS e7 21 xd4+
g8 2 2 llJxf4 llJcSf) 14 .. bS 15 llJg3
b4 16 gs :ges 17 llJd2? ! (17 ttJ hs+
llJxhS 18 gxhS b3 19 cxb3 axb3 20 a3
i.a6f!) 17. . . a3t Shirov-Leko, Dort
mund 200 2 ;
1 3 g 4 i.e6 ! ? 1 4 llJgS llJd7 15
ttJxe6 + fxe6 ! t , see game 4 Shirov
Ivanchuk, Edmonton 2005.
Eth. 11 0 - 0 - 0 a5
13 ... a4 14 g4
14 llJ e2 leaves the e4-pawn with
out protection so Black has 14 . . . c4 !
15 llJg4 a3 ! with an initiative.
14 gS ?
The black pawns run faster so

White tries to create threats with
his pieces.
12hg7
Perhaps 12 llJ h2 ! ? at once is
more precise: 12 . . . a4 13 llJg4 llJ hS
14 llJe2 bS ! 15 hg7 xg7 16 h6 +
g8 17 f4 a3 18 b3 f6 19 :gdfl exf4 20
llJxf4 llJxf4 2 1 xf4 xf4+ 2 2 M4
hg4= . After the text move Black
gets the g8-square for his knight.
12 . . . xg713 h2
28
Subsequent play is not forced.

Bot sides have a wide choice on eve
ry turn, but current practice is fa
voura ble to Black, who could stay
solid or try to grab a pawn on the
kingside :
15 llJe2 f6 (Grischuk-Ponoma-
1 e4 cS 2 ttJf3 tlJc6 3 bS
riov, Moscow 2002 saw lS . . . hg4 16
hxg4 e6 17 @bl xg4 when White
has only partial compensation for
the pawn.) 16 @bl a7 17 tlJe3 e6
18 g3 bS 19 f4 d8 20 fxeS fxeS+ was
fine for Black in Fridman-Lanka,
Bochum 20 05.
lS i>bl a3 16 b 3 d4 17 tlJh2 bS 18
tlJf3 d6 19 l2J e2 e6 20 g3 ad 8 2 1
tlJgS c4 2 2 f4 cxd3 23 cxd3 f6 = , Gris
chuk-Leko, Russia (rapid) 20 0 2 .
E2. 5 h3
White i s following the restrict
ing strategy, started on the previ
ous move.
5 .ig7 !
Ever since our first steps i n chess
we have been taught that it is a ter
rible sin to lose tempi in the open
ing, moreover to return a developed
knight to the initial square . On the
other hand, it is a question of con
crete calculations. If White proves
unable to make good use of his tem
porary initiative, we'll complete de
velopment with fair counterchanc
es. Our plan consists of tlJg8-h6, fal
lowed up by f7-fS (or f6) . Notice that
S . . . tlJd4? ! 6 eS tlJxbS 7 tlJxbS tlJdS 8
0-0 g7 9 d4 ! cxd4 10 xd4 is real
ly dangerous for Black.
6 es
White h a s no reason to delay eS
anymore, after all the preparations
he had made . After 6 0-0 0-0 Black's
knight gets the e8-retreat square:
7 es (7 el d6 8 d3 d7=) 7 . . . ttJe 8
8 hc6 dxc6 9 d3 tlJc7 10 tlJe4 (10 e3
tlJe6 11 tlJe4 tlJd4 ! 12 ttJxcS ttJxf3 + 13
xf3 heS=) 10 . . . ttJe6 11 el
. . .
Black has a good game. He can

realise his main plan at once : 11. . .
fS 12 exf6 exf6 1 3 bl b6 1 4 b 3 aS
15 a4 a7 16 b2 e7 17 d2 tlJd4
18 ttJxd4 cxd4=, Adams-Kramnik,
FIDE-Web k.o . g/lS +lO Las Vegas
1999, or simply trade some pieces
first:
11 . . . b6 12 tlJegS (or 12 a4 aS 13
e3 fS 14 exf6 exf6=, planning a8a7-e7 and h6, fS) 12 . . . tlJxgS 13 hgS
h6 14 f4 e6 15 d2 @h7 16 h4
dS=.
6 ... g8 7 .A xc6
It is time to kill the knight or
it will jump to d4: 7 e2 tlJ d4 ! ? 8
ttJxd4 cxd4 9 tlJe4 heS 10 0-0 a6
when White does not get enough
compensation for his central pawn.
7 dxc6 8 d3
8 e2 tlJh6 9 tlJe4 leads by trans
position to a critical position, which
is analysed in line E3a.
. ..
29
Part 1
8 ll.) h 6 9 .ie 3
Another version of this idea is:
9 g4 0-0 10 ie3 . White's idea is
deeper than it seems at first sight.
He is not just trying to win a tem
po for his development. More im
portantly, he hopes to provoke the
move 10 . . . b6? which would deprive
our queen of a pa th to the queenside.
As we will see later, that would con
siderably restrict our counter-chan
ces. Luckyly, we have the nice pawn
sac:
...
10 . . . fS!
11 exf6
Following 11 gS? llJf7 12 if4
Black has an extra tempo for
12 . . .aS! (12 . . . b6! ? is also playable
as in game 7 Iv .Popov-Tregubov,
Krasnoyarsk OS . 0 9 . 20 07) 13 V9e2
V9b4! (or 13 . . . llJdS ! ? 14 0-0-0 bS lS
a3 b4 16 llJbl gbs 17 llJfd2 ie6 lS
'!9e3 idS 19 gh2 llJ e6 20 h4, Menki
novski-N edev, Struga 2 0 0S, when
20 . . . llJd4! 21 hS bxa3 22 llJxa3 gxb2
23 c;t>xb2 gbs + - + would have won
faster) 14 e3 V9xb2 Black has an at
tack, for example, lS c;t>d2 a3 16 h4
gas 17 hS ttJd6 ! - + .
1 1 . . . exf6 1 2 d2
The greedy 12 hcS? unleashes
our bishop pair: 12 . . . ge s+ 13 ie3
(in the blitz game Guseinov-Nedev,
30
2 0 07 was 1 3 cj{fl fS 14 gS llJ f7 1S ie3

cS ! and the bishop takes the other
long diagonal, e .g. 16 h4 b6 17 hS
ib7 l S hxg6 hxg6 19 gh3 d6 ! + 2 0
V9d2 hf3 2 1 gxf3 h2 2 2 c;t>e2 llJxgS
23 gg3 f4 24 gxgS fxe3 -+) 13 .. .fS 14
gS f4 lS gxh6 hh6 16 e2 fxe3 17
llJe4 exf2 + lS c;t>xf2 ifS+ with full
compensation for the pawn.
12 . . . llJf7 13 hes ges+ 14 ie3 bS !
lS 0-0-0 b4! ?
The game Feygin-Nedev, Iz
mir 2 0 04 saw lS . . . VNaS 16 a3 b4 17
llJ e4 gbs lS llJd4 dS 19 axb4 fS !
and eventually I won, but the text is
even better. 16 llJ a4 VNdS 17 llJd4 fS !
lS c4 ! (or lS xb4 f4 19 c4 V9d7 2 0
hf4 xd4 2 1 ie3 f6+) 1 S . . . bxc3
19 llJxc3 d6 20 llJb3 fxg4+.
Having seen this analysis, we
might decide that:
9 g4 0-0 10 if4 is more con
sistent, but then 10 .. .f6 ! offers fair
counterplay due to the hanging
state of White's pieces on the f
file. The point is that 11 V9e2 fails
to 11 . . . llJxg4 ! 12 hxg4 fxeS, Cubas
Nedev, Calvia, ol 2004, so he has to
choose 11 V9d2 fxeS 1 2 hh 6 hh6 13
V9xh6 gxf3 14 0-0-0 fS lS xfS +
c;t>xfS 16 gh2 ! The endgame looks
better for White, but the thematic
sacrifice cS-c4 should balance the
game, for instance 16 . . . h6 17 gel
c4 ! ? lS dxc4 i e6 19 llJe4 hc4 20
ttJcs gas 2 1 ttJxb7 gas 22 b3 gbs 2 3
llJdS ids 24 c4 gbs 2S cxdS gxdS 26
dxc6 gc3+ 27 c;t>b2 gxc6 2S f3 ga3
29 gxeS gxf3 = .
9 Ya 5 !
This novelty i s a result o f my
...
1 e4 cS 2 tlJf3 ttJc6 3 bS
(T.N .) long evolution in understand
ing these structures. (which cost me
a couple of painful losses)
I had played here:
9 . . . b6 10 g4 fS, when 11 exf6?
exf6 12 e2 (Or 12 d2 tlJf7 13 0-0-0
0-0) led to this position in Maciej a
Nedev, Istanbul 20 0 3:
Instead of 12 . . . Wfe7, after the game

I found an improvement: 12 . . . 0-0 ! 13
0-0-0 fS ! and Black takes over the
initiative: 14 gS (14 d4 fxg4 15 dxcS
Wfc7 16 tlJgS gxh3+) 14 . . . ttJf7 15 d4
e8 16 dxcS WfeToo . The threat of 17 . . .
f4 forces White t o move the queen
and we get a tempo to activate the
second bishop on e6 .
Unfortunately, instead of open
ing the centre by 11 exf6?, White
has 11 gS ! ttJf7 12 f4. White has
lost a tempo with this bishop, but
the extra move . . . b6 only hampers
our counterplay on the queenside.
The game Stoj anovic-Majeric, Tuz
la, 2006 saw a similar development
and White had some edge.
Perhaps 10 .. .f6 ! ? would have
been a better option, when 11 Wf e2 ! ?
would b e similar t ogame 5 David
Nedev, Kerner, 05.10 . 2007.
Anyway, Black is not farced yet
to push the f-pawn. It is better to ac
tivate the queen first.
1 0 g4
Alternatively: 10 tlJd 2 tlJfS 11 ttJc4
ttJxe3 12 fxe3 Wfc7+; 10 Wfd2 tlJfS 11
f4 ttJd4=.
1 0 ...f 5 1 1 g 5
It turns out that White i s behind
in development so opening up the
centre is hardly advisable: 11 exf6
exf6 12 Wfd2 (12 tlJd2 ? ! 0-0 13 ttJc4
c7 14 ixcS e8 + 15 e3 bS 16 tlJd 2
fSt) 12 ... ttJf7 13 0-0-0 0-0 14 d4 fS !
15 gS e6 16 a3 bS when our attack
is running very fast.
1 1 . . . li:)f7 1 2 .if4 .ie6
12 . . . Wfb6 ! ? to impede White's
castling is playable, too.
1 3 '%Yd2 0-0-0 ! ?
We are already the active side, so

there is no reason to trade queens :
13 . . . dS 14 ttJxdS Wfxd2 + 15 Wxd2
31
P art 1
cxdS 16 c3 l2Jd 8 17 d4 l2Je6 18 cj{e3
:!%c8 19 :!%acl h6oo.
1 4 a3
Preparing a long castle. 14 h4 c4
15 d4 bS 16 hS b4 17 l2Je 2 dSt or
14 e3? ! dS 15 0-0-0 .ixf3 16 xf3
ltJxeS+ favour Black.
1 4 . . . c4 1 5 d4 c5 1 6 0-0-0 b5t.
E3. 5 e5 g 4
E3a. 6 e 2
E3b. 6 xc6
assess correctly.
White's main positional aim is to
bolster the e5-outpost with f4, :!%el,
but it is not too efficient if we man
age to trade our last knight through
d4. Therefore White tries first to dis
suade us from this idea. The point is
that 9 . . . 4.JfS could be met by 10 c3
b6 11 g4 4.Jh6 12 0-0 0-0 13 d4 cxd4
14 f4 ! ? dxc3 15 bxc3 with an over
whelming position for the pawn.
9 g4 0-0 is considered in E3b.
9 ... b6 1 0 d3
10 l2Jf6 + only helps Black deve
loping: 10 . . . exf6 ! 11 exf6+ cj{f8 12
fxg7+ cj{xg7 13 0-0 :!%e8= .
1 0 ... f5
1 0 . . . 0 - 0 1 1 f4 f6 1 2 0 - 0 ltj f7 13
:!%fel is an example of what Black
should avoid. He is very passive and
has too many minor pieces to acco
modate in a little space.
11 .ig5
E3a. 6 Wfe2 .ig7 7 hc6 dxc6

8 h3 tll h6
9 tlle4! ?
A consistent and logical varia
tion against Black's setup. It com
bines strategical with tactical mo
tives which the engines often fail to
32
Black must make a crucial

choice. He can fulfil his positional
aim by trading the knight, which
could however bring about a rath
er complicated position, or reduce
risk, and winning chances, too !
1 1... d4
1 e4 cS 2 tLlf3 tLlc6 3 ibS

The solid alternative is 11 . . . 0-0 12
c 3 f6 13 if4 fxeS 14 xeS tLlh4 ! The
point. Black gains the bishop pair,
but his pawn structure is cripppled:
15 tLlxh4 xeS 16 0-0 WdS . Black
should be OK here . (A.K.)
12 xd4 xd4 13 f6+ f8
14 c3 dS 15 g4
These moves were farced and
again Black must choose between
the sharper and the calmer options.
The game Tseshkovsky-Nataf, Her
ceg Novi 2005 saw 15 . . . hS 16 tLlh2
Wds 17 tLlf3 fs 18 dl Wxa2 19
0-0 We6 20 fel when Black is un
der bind. 19 .. .f 6 is better, but Kolev
does not like Black's position.
The engines like 15 . . . WdS be
cause it grabs a pawn, but 16 f4
xg4 17 hxg4 d8 (17 . . . h6 18 ih4
gS 19 i.g3 d8 20 0-0 t) 18 0-0- 0 !
h 6 1 9 h4 g S 2 0 if2 Wxa2 2 1 f
is awful for Black who is unable to
connect his heavy pieces. Perhaps
best is :
15 ... f6 16 M4 g5 17 exf6 exf6
18 .ih2 hg4 19 hxg4 d5 2 0
0 - 0 E:e8 2 1 c2 f7 2 2 E:fel=.
8 g4
This is the most topical option.
The alternatives are:
8 d3 g7 (8 . . . tLlfS heading for d4
is playable, too) 9 e3 (9 tLle4 b6 10
f4 tLlfS 11 c 3 ia6) 9 ... WaS and we
have reached the position fram the
main line E2 .
8 tLle4 b6 9 We2 tLlfS 10 c3 aS ! ?
(In Rytshagov-Spasov, Yerevan 1996,
was 10 ... g7 11 g4 tLlh6 12 tLlf6+ when
12 . . . exf6 13 exf6+ cM8 14 fxg7 + xg7
would have been unclear) 11 d3 ia6
12 c4 i.g7 (12 . . . tLld4? ! 13 tLlxd4 Wxd4
14 f4 0-0-0 15 tLld6 + xd6 16 exd6;t
Lilov-Spasov, Plovdiv 19 . 0 3 . 2008)
13 tLlf6 + exf6 14 exf6 + cM8 15 fxg7+
xg7 16 0-0 e8. Black completed
development and has a good game.
8 ...i.g7
8 .. .fS is imprecise. It reveals
Black's plans too early, thus allow
ing the opponent to choose the best
setup : 9 gS tLlf7 10 d3 e6 11 We2
g7 12 h4 ds 13 tLlxdS WxdS 14 c4
We6 15 f4 h6 16 0-0-0;t, Khalif
man-Lautier, Wijk aan Zee 2 0 0 2 .
E3b. 6 hc6 dxc6 7 h3 h6
9 d3
9 We2 0-0 10 d3 Wb6 ! ? transpos
es to the main line.
9 0 -0
33
Part 1
9 . . .f5, intending to block the cen
tre and castle long, is unclear.
1 0 e2
(For 10 ie3 f5 ! or 10 if4 f6 ! see E 2)
This move is flexible to excess !
The reason behind it is seen in the
line 10 .. .f6 11.d2 ! ? It turns out
that Black's threat to the e5-pawn
can be ignored. White simply de
velops, counting on his better pawn
structure. While not ensuring him
a big advantage, such an approach
is quite awkward for Black. You
can see a detailed analysis of the
plan with 10 .. .f6 in the "Complete
Games" section, game 5 David
Nedev, ECC 2 007.
The other major plan of Black
is linked with 10 .. . f5 ! ? 11 g5 ttJ f7 12
f4 Vfff aS ! See game 6 Landa-Mir.
Markovic, Belgrade 1991.
However, we can outsmartthe en
emy by attacking the b2-pawn with :
34
10
Wfb6 ! ?
Now, the bishop i s pinned t o cl

and White is suddenly faced with a
difficult problem: how to disentan
gle his pieces. All plausible choices
do not look encouraging:
a) 11 d2 V9xb2 12 0-0 Vfff a3 13
llJe4 b6 14 if4 f5 15 exf6 exf6 16 d6
8:f7 with a dubious compensation
b) 11 b3 f5 12 llJa4 (12 g5 llJf7 13
d2 llJd8 14 0-0-0 llJe6 15 E!:del llJd4
16 ttJxd4 cxd4 17 llJ a4 Wa6 18 bl
b5) 12 .. . c7 13 hh6 , Wen,Yang
Zhao Jun, Wch U2 0 Yerevan, 2006,
13 . . . fxg4 ! 14 hg7 gxf3+.
c) 11 l (or 11 0-0) We can
take a relieved breath here, as the
long castle would have been a much
more dangerous plan. 11 .. .f5 12 g5
(or 12 exf6 exf6 13 g2 llJf7 14 E!:el f5
15 g5 d7) 12 . . . ttJf7 13 g2 llJd8 14
b3 llJe6 with initiative. You can see
also game 7 Iv .Popov-Tregubov,
Krasnoyarsk 0 8 .0 9. 2 007.
Part 1
1 .e4 c5 2. lllf 3 e6 3.d4 cd4 4.lll d 4

lll c 6 5. lll b 5 d6 6.c4
COM PLETE G AMES
1 . M ortensen-Erm en kov
Riga 1 981
1 e 4 cs 2 li:) f3 li:)c6 3 ibS li:)f6
4 li:)c3 g6 s o-o .ig7 6 es li:) g4 7
ixc6 dxc 6 8 e1 0-0 9 d 3
9 . . . li:) h 6 !
Black's main plan is t o advance
his kingside pawns, but first he
should manoeuvre his knight to d4.
Then he could think about activat
ing the a8-rook with . . . as, . . . b6, . . .
E:a8-a7, and only then turn t o the
kingside. The breakthrough . . .f7-f6
(or fS) will not run away.
1 0 if4 li:)fS 1 1 %Yd2 b6 1 2 li:) e4
li:)d4 1 3 li:) xd4 cxd4
If White's king had castled long,
It would have been better to take on
d4 by queen. With opposite attacks,

it is imperative to have more open
files while the pawn structure falls
into the background. In our case Er
menkov prefers to install his bishop
on the long diagonal, fram where it
will be hitting the enemy king.
1 4 ih6 cs 1 S li:) g 3 ib7 1 6 ixg 7
x g 7 1 7 f4 fS !
Black has realised the main ide

as of this opening and took over the
initiative. The tide is soon going to
turn and it will be Black who will be
attacking.
1 8 exf6+ exf6 1 9 fS %Yd7 20 h4
ae8 21 fx g 6 hxg6 22 h S %Yg4 2 3
'l;Yf2 Y;Yg s !
Mortensen was reluctant t o ad35
Part 1
mit his opening strategy was a fail
ure, and only deteriorated his posi
tio n. Black methodically went on to
build up pressure.
24 xea xea 2S hxg6 xg6 26
ti:) e2 e3 27 ti:) f4+ f7 2a q;h2 f3
29 ti) h 3 \Wes+ 30 h 1 \Wh S 31 g 1
xg2 32 \Wxg2 \Wxh 3
3 2 . . . xh3 ! 33 b7+ g6 34 g2 +
h 6 3 5 fl f5- + was winning, wbile
now White can still resist.
33 \Wxh3 xh3 34 a4 e6 3S as
 d S 36 axb6 axb6 37 a6 q;c6 3a
aa e3 39 f2 q; b s 40 ta e6 41
da q;b4 42 ca es 43 c 6 f s+
44 e2 bS 4S d6 f4 46 c6 c4
4 7 dxc4 bxc4 4a ca h4 49 c3+
b3 SO ba+ q;c2 S1 cxd4 xd4
S2 e3? (52 b4 fS 53 3 d2 54
a4=) S2 .. d3+ S3 q;e4 ts+ S4 @ e s
b 3 ss c a q; d 3 S6 da+ e 3 S7
ca bS+ sa f6 f 4 S9 xc4 f3 60
c3+ f4
0-1
2. M ovsesian-Chuch elov
B u nde s l ig a 200 S
1 e4 c s 2 ti:)f3 ti:)c6 3 bS ti:) f6 4
eS ti:) d S S 0 -0 ti:) c7 6 xc6 dxc6 7
h3 ts a d3 h 6 !
squared bishop fram exchange as

after 8 . . . e6? ! 9 gs e7 10 xe7
xe7 11 ttJbd2 White would have
been slightly better. The point is
that White has not a good place for
his bishop.
9 ti:) bd 2
After 9 ttJc3 e 6 1 0 e2 Black
should not miss the moment for
10 . . . ttJbS ! = since one move later
White would be able to cover the
d4-square: 10 . . . e7 11 ttJe4 ttJbS 12
c3 .
9 . . . e6 1 O '%Ye2 ti) b S !
A key point i n Black's setup !
White was threatening with 10 . . . e7
11 ttJe4 when 11. . . ltJbS would stum
ble into 12 c3 .
1 1 ti:) e4 ti:)d4 1 2 ti:) xd4 \Wxd4
Black has completely equal
ised. Later on he could disturb his
opponent with . . . c5-c4 or . . . h5-h4
while the only active idea of White
is f4-f5.
1 3 ti:) g 3 g 6 1 4 q; h 2 ?
After this move White is deprived
of any counterplay. His chance to
was 14 e3 0-0-0 15 f4, maintain
ing the balance.
1 4 . . . h S ! 1 S f4 h4 1 6 ti) h 1 ?
Thus Black preserves his dark-
A terrible move, which allows
36
1 e4 cS 2 l!Jf3 l!Jc6 3 bS
Black to place a strong blow. Natu
rally, 16 l!Je4 would have been bet
ter. Black would have indeed the
same breakthrough as in the game,
but with White's knight in the cen
tre, it would not be so decisive due
to 17 e3 .
1 6 . . . c4! 1 7 dxc4 hS !
Regaining the pawn with a n ad
vantage in view of the variation
18 d3 xd3 19 cxd3 e2 20 f2
xd3+.
1 8 '%Yf2 '%Yxc4 1 9 fS?
Apparently White cannot be
lieve that he could be worse so early
in the game with White and makes
"active" moves instead of develop
ing . 19 e3 would have made Black
to choose between many appealing
options. He might want to sacrifice
a pawn with 19 . . . e7, when 20 xa7
cS 21 b6 a6 22 c7 g6 23 c3 c6
24 d6 xd6 2S exd6 xd6 26 adl
hS is only slightly better to him.
After the text White should not be
able to level the game any more.
1 9 . . . ext s 20 gs cs ! 21 '%YxfS
The endgame is rather gloom af
ter 2 1 xh4 xh4 2 2 hh4 e2 2 3
fel xh4 2 4 xe2 d8+
21 . . . g6 22 Y;Yf 3 hS 23 f4
o-o-o+ 24 t2?? ts-+ 2s d 3
h S 26 b3 '%Yxc2 2 7 a c 1 '%Yxd 328
'%Yxd 3 xd3 29 xcS g S
0-1
3 . De l a Paz- H a nd ke
H av an a 200 3
1 e4 cs 2 f3 c6 3 b S f6
4 es d S s 0-0 c7 6 xc6 dxc6
7 h 3 g6
We recommend 7 . . .fS , but this
structure is also important for our

repertoire, since it could arise from
other move orders.
8 d3 g7 9 c3 b6 1 0 e4 0-0
1 1 d 2 fS ! ?
Handke conducts the game very

consistently, without subtleties and
fancy move orders. He knows what
he is aiming for, and does not beat
about the bush. Black is set for a
kingside pawn storm.
We prefer first to activate the c7knight with . . . l!Jc7-bS-d4, but the
text is by no means bad. Black has
solved the opening problems.
1 2 exf6
White is unable to blockade the
kingside with 12 l!Jc3 h6 13 h4 l!Je6
14 el due to 14 .. .f4.
1 2 . . . exf6 1 3 '%Yc1 g S !
This i s the right way t o handle
the pawns. Black should bolster up
the gS square before proceeding
with .. .fS.
1 4 h 2 ts 1 S g 3 g6 1 6 f4
fS
This is already too straightfor
ward. Black would have kept the in
itiative with 16 . . . d6t. The text al
lows White to escape into an end
ing with 17 fxgS d4+ 18 hl xb2
37
P art 1
19 1!9xb2 xb2 2 0 ael ltJbS with un
clear play, but White misses this op
portunity.
1 7 ic3?! ixc3 1 8 bxc3 d S 1 9
f3 ?
White cracks under the pres
sure. 19 l2Je2 was more stubborn.
1 9 ... g xf4+ 20 e2 Yd6 21 h4
gae8 22 Yd2 Yf6 23 f3 ge7 24
a e 1 fe8 2S d4 Yd6 26 gf2 ge3
2 7 Yc1 ihS
The rest is clear.
28 e s ixe2 29 xf4 8xeS 30
dxeS Yxe S 31 f2 f4 32 Yd2 ic4
3 3 d 1 Yxc3 34 Yc1 a s 3 S h 2
x h 3 + 36 gxh3 Y g 3+ 3 7 h 1 Yxf2
38 g 1 + f7 39 Yb2 f6
0-1
4. S h i rov- lvanchuk
Edmonton 200 S
1 e4 cs 2 f3 c 6 3 i b S g 6
4 ixc6 d x c 6 S d 3 i g 7 6 h 3 f6
7 c3 0-0 8 ie3 b6 9 Yd2 es 1 0
.i h 6 Yd6 1 1 o-o-o a s 1 2 ixg7
 x g7
1 3 g4
Without any advantage in the
centre, White's attack should not
be lethal. Black's play is even easier.
He entrenches himself with l2Jg8, f6
38
whereas on the queenside his pro

gram includes . . . a4, bS, a3, and
eventually the thematic . . . c5-c4 .
In the diagram position White has
tried to organise play down the h
file, but it has proved quite harm
less ;
13 l2Jh2 a4 14 l2J g4 l2Jg8 15 bl
(or 15 l2Je2 xg4 16 hxg4 Wff e6 17 bl
1!9xg4 18 f3 1!9e6 19 g4 f6 20 f4 a3
21 fS 1!9d7oo, Grischuk-Ponomariov,
Moscow 2002) 15 . . . a3 16 b3 1!9d4 17
l2Jh2 bS 18 l2Jf3 Wffd6 19 l2Je 2 .ie6oo
Grischuk-Leko, Moscow 2002
1 3 . . . ie6
Karjakin-Topalov, Blindfold, Bil
bao 19 . 10 . 2007, saw 13 . . . a4 14 l2Je 2
a7 15 l2Jg3 e7 16 gS l2Je8 17 Wff c3
l2Jc7 18 dfl ltJbS 19 Wffd 2 a3 20 b3
f6+ and Black had a strong pressure
in the centre. Ivanchuk wants to
bind the c3-knigt with the defence
of the a2-pawn. Now, 14 l2Je2 ixa2
15 l2Jg3 fe8 16 Wigs Wffe6 would fa
vour Black, so Shirov has to think
up another attacking plan. 14 l2Jh4
a4 ! 15 W1 gS (or 15 l2Je2 xa2 16 l2Jg3
h8) 15 . . . a3 16 b3 Wffd4 is unappeal
ing, therefore White decides to kill
the awkward bishop .
1 4 g s d 7 1 S xe6+
The attempt of pushing f4 is too
slow: 15 dfl bS 16 f4 b4 17 l2Je2
xa2 18 l2Jf3 f6+. Time and again
we see that without a good centre,
a flank attack has little chances to
succeed.
1 S . . . fxe 6 !
I t took t o Ivanchuk only 1 5
moves t o get the edge with Black!
H i s attack will run very fast while
Shirov will need a lot of tempi to
1 e4 cs 2 l2Jf3 l2Jc6 3 bs
create even the smallest threat.
1 6 gS a4 1 7 h4 '%Yd4 1 8 gdf1 ?
After this passive move White

is lost. He would have had more
chances to resist after 18 l2Je2 ! Wffxf2
19 hS f3 20 hxg6 hxg6 2 1 h6 h8+
(Finkel).
1 8 ... a3! 1 9 hS
Following 19 b 3 b S 20 dl f3
White would be tied up and down.
1 9 . . . c4
Instead of this thematic break,
Black was winning by brute force :
19 . . . axb2 + ! 20 bl a3 21 f4 c3
22 hxg6 hxg6 23 Wff h2 f7 24 Wffh 7 +
e8 25 Wffxg6+ d8 - + . (Finkel)
20 f4 cxd 3 ! 21 gf3 til e s 2 2
hxg6 h x g 6 2 3 Wh2 axb2+ 2 4 @b1
dxc2+? !
Starting fram here, Ivanchuk
gradually begins to lose control of
the game and eventually draws. In
the next few moves he misses sever
al killers, e.g. 24 . . . h8 ! 25 h3 xh3
26 Wffxh3 Wffxc3- + .
2S Wxc2 exf4 ? ! (25 . . . xf4 2 6
fh3 af8 - +) 2 6 Wxb2 g h 8? ! 2 7
g d 1 Wes 2 8 gxf4 ! + g h 3 2 9 tll a4
Wxb2+ 30 tll xb2 gf8 ? ! 31 gxf8
@xf8 32 gd6 gh433 @c2 gh2+ 34
@c1 @e7 3S gxc6 tll xe4 36 gxb6
gh4 37 @c2 gh2+ 38 @d3 tll x g S

3 9 a 4 tll f3 4 0 tll c4 g S 4 1 a s g 4 4 2
@e3 gc2 4 3 tll d 6 tll d 2 4 4 @f4 ga2
4 S a6 ga4+ 46 @ g 3 tll c4 47 tll xc4
g x c4 48 gba ga4 49 g as @f6 so a7
@es S1 @h4 @e4 S2 @ x g4 e s S3
@g3 ga2 S4 @h4 @e3 SS @hS e4
S6 @ g s gas+ S7 @ g 6
%-%
S. Dav id- N ed ev
European C l u b C u p O S .1 0.2007
1 e4 cs 2 tll f 3 tll c6 3 .ibS tll f6 4
tll c3 g6 s es tll g 4 6 .ixc6 d xc6 7 h 3
tll h 6 8 g4 .i g 7 9 d 3 o - o 1 0 We2
1 o . .. t6
During the game I could not
break away from the stereotypi
cal thinking that made me consider
only 10 .. .f6 and 10 . . .f 5. Only after the
game I got the insight to shift my at
tention to the other wing, and try to
punish the opponent for his delay of
development. Then I came up with
the move 1 0 Wb6 ! ? which I ana
lyse in the "Step by Step" chapter.
The text is not that bad, of course,
but David's next move was a sort of
surprise to me.
1 1 .id2 !
I had pleasant experience after
. . .
39
Part 1
11 .if4 lt:Jxg4 ! 12 hxg4 (12 e6 lt:Jh6
13 0-0-0 lt:JfS+) 12 . . .fxeS 13 heS? !
(13 lt:JgS gxf4 14 gxh7!? .if6 ! ? 15
gh6 ! Wes 16 lt:J ce4 hg4 17 f3 .ihS
1 8 0-0-0 Wf8 ! 19 fuhS gxhS 2 0 ggl
Wh6 21 bl gfs+) 13 . . .hg4+, as
in the game Cubas-Nedev, Mallor
ca 2 004, so I expected 11 ie3 . Then
gxg4 gxg4 20 lt:JfgS \Wd7 21 f3 gxgS !

2 2 lt:JxgS \WfS 23 lt:Je4 gfs+) and now
18 . . . hxg6 (or 18 . . . hS 19 .ie3 hf3 2 0
VM fl Wd7 2 1 lt:JxcS \WfS 2 2 ggloo) 19
gd gl fuf4 20 xg4 fug4 21 lt:Jh4
VMc8 22 f3 fuh4 2 3 fuh4 \WfS 24 l&g2
gfg 2 5 gh 3 b6 26 gg3 f7oo is about
balanced. After my positional blun
11 . . . b6 12 0-0-0 lt:Jf7 13 d 4t o r 11. . .

fxeS 1 2 lt:JgS ! b6 1 3 0-0-0oo would fa
vour White . I had in mind 11 . . . lt:Jf7 !
der the game is over as Black has

nothing to oppose to the enemy at
tack on the h-file.
12 hc5 fxeS 13 0-0-0 b6 14 .ie3 cS

15 lt:J gS lt:JxgS 16 hg5 .ib7 17 lt:J e4
h6 18 .ih4 Wd7 with a strong bish
op pair and good prospects for a
queenside attack.
1 8 hxg6 hxg6 1 9 h4 e5 20 &iJfg5

f3 21 Y!if1 gf4 22 Y!ih3 gxe4 23 dxe4
Y!ie7 24 gh8+! ixh8 25 g d8+! 1 -0
1 1 . . . &iJ f7 1 2 0-0-0 Y!ic7 1 3 i> b 1

&iJ d 8 1 4 &iJ e4 &iJ e 6 1 5 h4
6. Land a-M ir. M arkov ic,

Belgrade 1 99 1
1 e 4 c 5 2 &iJc3 &iJ c 6 3 &iJf3 &iJ f6
4 cib5 g6 5 Y!ie2 ig7 6 e5 &iJ g 4 7
ic6 dc 6 8 h3 &iJ h 6 9 g4 0-0 1 0 d 3
f 5 1 1 g5 &iJ f7 1 2 if4 (or 1 2 h 4 f4 ! 13
e6 lt:Jd6 14 hS lt:JfSf) 1 2 . . . Y!ia5 !
Both sides are realising their

plans and now it is evident, that
White started first his attack. That
finally made me take on eS, in or
der to organise counterplay down
the f-file.
1 5 . . .fxe5 1 6 h5 &iJf4 1 7 ixf4
1 7 . . . exf4?
It is difficult to explain this awful
move. I only had to employ the tac
tical motive from my game against
Cubas to get a satisfactory game:
17 . . . hg4 ! 18 hxg6 (18 gd gl fuf4 19
40
This i s another important posi

tion for our repertoire. White has
fixed the kingside in his favour, but
Black has a free hand on the oppo
site wing. His idle minor pieces ca n
easily take a more active role follow
ing the routes .ic8-e6-d5 and lt:Jf7d8-e6. Only the g7-bishop is like-
1 e4 cS 2 l2Jf3 l2Jc6 3 !bS

ly to remain stuck on g7, but it is at
least a good defender.
After the text, long castling is
risky due to 13 . . . bS, (or the more
solid 13 0-0-0 !e6 14 h4 l2Jd 8 lS hS
!xa2 16 hxg6 hxg6 17 l2Jh4 !f7 18
e3 l2Je6) so White's king will re
main in the centre for a while.
Curiously, I had this position
with an extra tempo as White has
played 10 !e3 f5 11 gS l2Jf7 12 !f4
as 13 e2, and following 13 . . . l2Jd 8
14 0-0-0 bS---+ White's resistance did
not last long, Menkinovski-Nedev,
Struga 20 0S.
13 Ye3
I have also analysed 13 a3 !? , pre
paring the castle. Then 13 . . . !e6 14
h4 !dS lS hS ltJd8 (Do not be greedy,
lS . . .!xf3 16 xf3 ltJxeS 17 e3t is
in White's favour.) 16 0-0-0 l2Je6 17
!d2 c7 18 ltJxdS cxdS 19 h4 cj{f7oo
leads to complex play with mutual
chances.
1 3 . . . .ie6 1 4 h4
Still keeping the castling op
tions open. In Kindermann-Hoel
zl, Chalkidiki 20 0 2 White chose 14
cj{fl, when 14 . . . fd8 lS h4 b4 16
bl d4 17 a3 b6 was not too ef
ficient.
It would have been better to acti
vate the knight with 14 . . . l2Jd8 lS h4
!dS (or lS . . . !f7 16 hS gxhS 17 l2Je 2
l2Je6 18 cj{g2 ad8 19 a 3 l2Jd4oo) 16
ltJxdS (16 hS l2Je6) 16 . . . cxdS with a
roughly level game, for example,
17 e6 b6 18 el d4 19 e2 c8
20 ltJeS xe6 21 f3 l2Jc6 22 l2Jxc6
xc6 23 xc6 xc6 24 xe7 f7 2S
e8+ f8 = .
1 4 . . . .idS
1 5 h3
White should consider sacrific
ing the eS-pawn as lS hS .txf3 16
xf3 ltJxeS 17 h3 cj{f7 18 0-0-0oo
or lS ... b4 16 hxg6 hxg6 17 0-0-0
M3 18 xf3 ltJxeSoo would give him
time to complete development.
In my opinion (T.N .) it is bet
ter to maintain the tension with
1S . . . l2Jd 8 , eyeing the a2-pawn. Then
16 a3 (16 hxg6 hxg6 17 a3 l2Je6 18
0-0-0 bS) 16 . . . l2Je6 17 0-0-0 ad8
(to recapture on dS by rook) 18 h4
bS 19 ltJxdS xdS 20 dhl b4 21 hxg6
hxg6 would be interesting to test.
1 5 . . . ad8 ? !
The first critical moment in
this game. Markovich is obvious
ly unable to decide what he wants
to do and hands the initiative to
the opponent. Black has to real
ise that he must free the f7-square
for his king, and that he needs to
bring his knight into play. Both
tasks could be achieved with one
move : 1S . . . l2Jd 8 ! If then White sac
rifices the a2-pawn, he will hardly
have enough compensation after 16
0-0-0 !xa2 17 hS l2Je6 18 d hl bS.
Still, Black is not obliged to take the
gift. Instead he can play in the cen-
41
Part
tre with 16 . . . ttJe6 ! ? 17 bl ad8 18

h5 hf3 19 f3 d4 20 hxg6 hxg6
2 1 hl fd8 2 2 tlJe2 a4 23 tlJc3 b4
24 tlJe 2 c4 25 d4 c5f!.
Let's consider 16 a3 tlJe6 17 0-0-0
(or 17 h5 ixf3 18 W'xf3 tlJd4 19 W'dl
c4t) 17 . . . b5 18 h5 b4 19 tlJbl ab8t.
Black h a s a perfect setup. His pieces
are clearly more dangerous.
1 6 tl) d 2 e6 1 7 tl) f3 d 5 1 8 .tl)d2
'lb4? ! 1 9 0-0-0 'ld4 ? ! 20 ge1
b5? !
Black firstly misplaced his queen,
and now he misses the chance to
trade it. Of course the endgame
would be not too pleasant, but leav
ing the enemy queen alive in this
position is suicidal. But now it is
White's turn to err. He should have
kept the queen with 21 W'g3 ! with
nice attacking prospects.
2 1 .b1 ? ! e6 22 tl) f3? Yxe3
23 gxe3
Now Black is not afraid of being
mated, and his game is preferable .
He must immediately redeploy his
pieces so that the knight reaches e6,
(or g6, should White play h5) for in
stance, 23 . . . d7 24 h5 gxh5 ! 25 tlJe 2
tlJh8 2 6 tlJg3 h4 27 ttJxh4 tlJg6 2 8
tlJ h 5 !h8 .
2 3 . . J;d4? !
Black is obsessed with the d4square. I n a moment when the bat
tle for the key dark squares f4 and g3
is going at full throttle, he presents
the enemy with a clear tempo .
24 tl) e 2 gd7
White should have opened
the kingside with 2 5 h5 gxh5 !
(25 . . . ttJd 8? does not work due to
2 6 hxg6 hxg6 27 tlJh4 f7 2 8 !h 2 !
42
g 8 2 9 tlJf4 !f8 30 tlJf3 +) 26 xh5

tlJd 8 27 !h2 !f7 28 h4 tlJe6 . Black
may be a little better, but White's
pieces are not so useless as in the
game. Instead he decides to defend
passively.
25 g h 2 tl) d 8 26 g3 d 5 2 7
tl) d 2 tl) e 6 28 f4 c4i
This breakthrough is possibly a
bit premature, but we can hardly
blame Markovic for it. It is normal,
thematic move, which fixes a the
clear edge of Black. The only ques
tion is could he press home his ad
vantage.
29 dxc4 xc4 30 tl) xc4 g d 1 + 31
tl) c 1 bxc4 32 g h e 2 gfd 8 33 c3 tl) c 5
34 c 2 tl) e4 3 5 e 1 f8 36 b 3 e 6
3 7 bxc4 a 3 3 8 tl) b3 c 5?
Black's bishop apparently had a
bad day. After having been stuck on
g7 for many moves, it finally broke
free, only to get locked again, this
time on a3. 38 . . . a5+ would have
maintained the edge. The rest of the
game is tragicomic and is irrelevant
for our purposes.
39 d2 g h 1 40 gd3 gxd3 41
xd3 g xh4 42 c2 g h 1 43 e 1
h5 44 g xh6 h 7 45 tl) d 2 tl)xd2 46
1 e4 cs 2 lt:Jf3 lt:Jc6 3 bs
Axd2 g a1 47 b 3 Ac 1 48 g e1 Axd2
4 9 gxa1 Axf4i SO c 2 xh6 (SO . . .
gS ! ) s 1 g h1 + g 7 S2 g b 1 Axes S 3
gb7+ i>f6 S 4 gxa7 g S SS a 4 g 4
S6 a s g 3 S 7 g b 7 g 2 S 8 gb1 f4 S9
@ d3?? (S9 a6 f3 60 a7+-) S9 ... @fS
60 a6 f3 61 e3 \t>g4 62 a 7 Af4+
63 f2 A g3+ 64 e3 f2 6S a8YM
f1 YM 66 YMg8+ h3 67 YMxe6+ h 2
6 8 YM h 6+ g 1 6 9 YMb 6 Ae 1 7 0 YMb2
@h1 71 YMb7 YMf2+ 72 d3 YMd 2+
73 e4 g 1 YM 74 es+ VMgg2 0-1
This game saw a lot of positional
m istakes, but they were quite in
structive, and allowed us to explain
the typical plans in the position with
a closed centre.
7. lv. Po pov-Tre gu bov

K ra snoyarsk 08.0 9.200 7
1 e4 cs 2 f3 c6 3 c3 f6 4
Abs g6 S es g4 6 Axc6 dxc6 7 h 3
h 6 8 g 4 A g 7 9 d 3 o-o 1 0 Ae3? !
ts ! 1 1 g s f7 1 2 At4
Whte has lost a tempo on 10

ie3 ? ! We prefer in such a setup
to develop the queen to as, in or
der to have . . . bS as an option, but
Tregubov's move is also consistent.
It controls d4 and wins a tempo by
attacking b 2 .
1 2 . . .YM b 6 1 3 b3 Ae6
Our recommendation is first to
activate the knight with 13 . . . lt:Jd8e6-d 4, and then to put the bishop to
e6 where it would support ... cS-c4.
1 4 o-o Ads 1 s g e 1 d 8
Black would have more chances
to break through following lS . . . hf3
16 '1Mxf3 '1MaS 17 lt:J a4 lt:Jd8+. Now Po
pov succeeds in entrenching him
self:
1 6 h2
16 lt:JxdS cxdS 17 d4 opens up play
in Black's favour: 17 . . . lt:Je6 18 dxcS
'&xcS 19 ie3 '&as 2 0 lt:Jd4 lt:Jxd4 21
hd4 f4 !
1 6 . . . e 6 1 7 Ad2 YMc7 1 8 f4 h 6
(or 1 8 . . . hS) 1 9 h 4 h x g S 20 h x g S f7
21 xd S cxdS 22 f3 gh8 23 g 2
d 4 2 4 a4 YM c 6 2 S g3 c7
1/2-1/2
8. D ra b ke-Zhi g alko
Kerner 08. 1 0 . 2007
14 ... c4 ! ? 1 S bxc4 A b4+ 1 6 f1

Ac3 1 7 Ae3 cS 1 8 gd 1 YMc6 1 9 g 2
Axes 2 0 g h e 1 f S 2 1 A g s gde8 2 2
gh1 Ac7 23 'tt> g 1 fxg4 24 h4 g h s
2 S YMxg4 e s 2 6 YM g 3 g x h 4 2 7 f4
0 -1
f3+ 28 f2 x g S
43
Part 2
1 e4 cS 2 lll f 3 lll c 6 3 d4 cxd4

4 lll x d4 lllf6 s lllc 3 es 6 lll dbS d6
7 igS a6 8 lll a 3 bS 9 lll d S ie7
We start our survey on the Open Si

cilian with the so-called Positional
variation against the Sveshnikov. In
this part we consider rare continu
ations, which are used as surprise
lines fram White.
A. 1 0 lll x e7 lll x e7 1 1 ixf6
Black meets almost all other
moves with . . . dS. Only 11 \Wf3 re
quires another approach: 11. . . lt:Jd7
12 0-0-0 lt:JcS 13 .ie3 lt:Je6.
1 1 gxf6 1 2 c4
White tries to clamp on dS, but
his pieces are undeveloped. 12 .id3
.ib7 13 \WhS ! ? is a modern attempt
which should be countered with
13 . . . lt:J g6 ! . The knight is eyeing f4,
the h8-rook can occupy the g-file,
while our king would be safe on e7.
1 2 fS!?
1 3 exfS lllx fS 1 4 cxbS 0-0

1 S bxa6
15 .id3 e4 16 he4 V!ie7 gives
Black fantastic attack.
1 S ixa6 1 6 ixa6 as+ 1 7
d2 xa6
Perhaps White can hold the bal

ance by 18 Wie2 , intending to repeat
moves after 18 . . . V!iaS + 19 \Wd2 = .
B. 1 0 ixf6 ixf6
Black already has the initiative.

44
7 ig5 a6 8 a3 b5 9 tlJd5
The main move here is 11 c3. It is
considered in Part 3 and 4.
Lately White discovered that
Black apparently neglects good
preparation against:
1 1 c4
It deprives Black of immediate
counterplay on the queenside and
clamps on d5 "for good ". However,
Black has an active plan, connect
ed with . . . a4, but he must play con
crete chess.
We propose a new idea, con
nected with a pawn sacrifice.
The most topical position aris
es after
1 1 ... b4 1 2 lllc 2 aS 1 3 g 3 0-0
1 4 h4!?
Apart fram restraining Black's
bishop, this move prepares an ex
change of the light-squared bishops
through h3.
The more conventional 14 ig2
ig5 15 0-0 e7 16 ce3 ie6 17 d3
he3 18 xe3 leads to an equal po
sition.
White has more active rooks,

but they can attack only one we ak
ness, o n d6, which is easily defend
ed. A logical continuation would
be: 18 . . . c7 ! ? = . White can open up
the queenside by a3, but then Black
takes on a3 and obtains typical Si

cilian counterplay along the b-file.
Play might become interesting
only in case when White attempts a
kingside attack. However, Black can
then invade White's rear through
the c-file with . . . a4, . . . b3, . . . c6d 4-c2 ! . We analyse this plan in the
"Complete Games" section, see 1 0
Korneev-P. Horvath, Porto San
Giorgio 2007.
After 14 h4, we propose:
1 4... a4!?
The known alternatives 14 . . . g6

and 14 ... ie6 15 ih3 d4 are play
able, but the text is more enterpris
ing. Now 15 ih3 fails to 15 . . . b3, so
White must take the pawn:
1 5 lll c xb4 lll x b4 1 6 lll x b4
YNb6 1 7 a3 .id8
The immediate threat is . . . cs

and . . . ia5, regaining the pawn.
45
Part 2
If White prevented it by castling
quickly, the bishop goes to b6 to un
derline the vulnerability of the ene
my kingside. We would not like to
be in his shoes after 18 i.d3 b7 19
0-0 i. b6 20 tLldS i.d 4 21 bl i.h3 22
el d7. Perhaps he should play 18
d3 i.e6 19 i.h3 cs 2 0 i.xe6 fxe6
2 1 0-0 i.aS 22 adl fd8 with equal
ity.
We can realise the same idea fol
lowing the modern
1 3 Yf3 i.e6 1 4 gd1
tLlxb4 17 tLlxb4 (17 xb4 i.d8 ! = )

1 7 . . . cS !
18 d2 (or 18 e2 i.d8 19 0-0

i.aS 2 0 a 3 0-0) 18 ... 0-0 19 a 3 tb8
20 i.e2 i.d8 21 xd6 xd6 22 xd6
i.e7.
1 6 i.dS 1 7 i.e2 d4 1 8
%Yd3 o-o 1 9 o-o gcs 20 gc1
1 4 a4!?
White hoped to put pressure on
d6 as in the game Nepomniachtchi
Andriasian, Moscow 14.0 2 . 20 0 8 :
1 3 . . . i.e6 14 d l i.e7 l S c S 0 - 0 1 6
i.bS tLl a7 17 i.a4. Kolev's idea rad
ically disturbs his plans.
1 5 cxb4
Now lS cS? ! does not work in
view of lS . . . aS ! .
l S tLlxf6+ does not seem too test
ing either: 1S . . . gxf6 ! ? 16 e2 b3 17
axb3 b8
1 5 Ya5 ! 1 6 a3
Another critical line is 16 c3
46
The fine point of this position

is that the opposite coloured bish
ops work in Black's favour, as after
20 tLlc2 i.xdS 21 exdS i.b6oo. After
the text we continue setting up our
pieces on dark squares :
20 Ya7 2 1 h3 i.a5 22 i.g4
gbs
All our pieces are well placed, the
a4-pawn petrifies White's queenside
and renders his extra pawn useless.
Part 2
1 e4 cS 2 f3 c6 3 d4 cxd4
4 xd4 f6 s c 3 es 6 db S d6
7 i.gS a6 8 a3 bS 9 dS ie7
STEP BY ST EP
I n this part we examine rare lines of

the Positional system:
A. 10 CDxe7
B . 10 M6, fallowed by 11 CDbl or
ll c4.
10 CDxf6? ! + M6 11 e3 loses yet
another tempo and Black should
become even better after 11 . . . e6 12
c4 0-0 ! 13 cxbS (13 e2 CDd4+) 13 . . .
axbS 1 4 xbS (14 CDxbS d S i s excel
lent for Black, for example: 15 exdS
ixdS 16 CDc3 CDd4t) 14 . . . CDd4 15 d3
b6+.
A. 1 0 xe7
White tries either to split the
enemy pawn formation, or obtain
a small strategic edge in case of
10 . . . xe7.
This approach is quite logical,
but it fails the test of practice. It
turns out that in the first case the ri
diculous position of White's knight
on a3 is a more weighty factor than
Black's structural defects, and even
in the latter case, Black retains fair
chances to equalise. Let us shortly
examine 10 .. . xe7 11 c4 0-0 (11. . .
b4? ! proved bad i n Carlsen-Radja-
bov, 20 07 rapid 12 CDc2 0-0 13 f3 h6

14 h4 as 15 e2 e6 16 0-0 fd8
17 <j{hl gS 18 f2 CiJhS 19 el CDf4 2 0
fl Wffc7 21 d2 <j{g7 2 2 acl Wffb7 2 3
CDe3+) 1 2 cxbS CDd4 1 3 d3 axbS 14
CDxbS CDxbS 15 hf6 xf6 16 xbS
b7! (Following 16 . . . Wff g6 17 0-0
Wffxe4 18 xd6 b7 19 f3 Wff e3+ 2 0
<j{hl e 4 21 es ac8, Perez Cande
lario-Illescas, Leon 200 6 , Black
must struggle for a draw after 2 2
a4.)
It is unclear how White could

achieve an advantage here:
a) 17 f3 dS ! ;
b) 1 7 d3 dS (17. . . e6 !?) 1 8 exdS
d6 19 c2 xdSoo;
c) 17 e2 g6 18 f3 dS 19 exdS
hd5 20 0-0 (2 0 a4 fc8 !oo) 2 0 . . . b6 +
21 @hl xa2 2 2 xeS aS = .
47
Part 2
1 o xe7 1 1 ixf6
We'll describe here a number of
unpopular alternatives for White:
...
a) 1 1 9We2 has not been seen in

practice, but it is an interesting at
tempt to use the bishop pair. We
suggest 11 . . . !b7 12 f3 dS 13 0-0-0
Vfic7 14 @bl 0-0 lS exdS ttJexdS 16
c4 bxc4 17 9Wxc4 9We7 ! f ;
b) 1 1 f3 d S As i n most Sicilians,
Black solves the opening problems
if he achieved this break. 12 exdS
ttJfxdS 13 c4 bxc4 14 ttJxc4 f6 lS !d2
0-0 16 !d3 (16 e2 ? ! !e6 17 0-0
ttJb6 18 ttJe3 9Wd4 19 !cl ttJfS ! 20
9Wxd4 ttJxd4 2 1 !dl fd8+ Petrov
Dobrov, Athens 20 03) 16 . . . !fS 17
!as 9Wd7 18 0-0 (18 hfS? ttJxfS 19
0-0 VfibS ! 2 0 9Wb3 fb8 ! + ) 18 . . . VfibS
19 b3 hd3 20 Vfixd3 ttJc6 = ;
c ) 1 1 9Wd3 dS
ll . . . !b7 is a well known old line
which is equal: 12 !xf6 gxf6 13 0-0-0
dS 14 exdS VfixdS lS VfixdS ttJxdS 16
c4 ttJb4! 17 cxbS ttJxa2 + 18 @c2 (18
@bl ttJb4 19 bxa6 !e4+ 20 @al @e7
21 f3 jJS= , Sveshnikov) 18 . . . dS !
19 c4 (19 bxa6 @e7) 19 . . . hc4 20
ttJxc4 axbS 2 1 ttJd6+ @e7 2 2 ttJfS+
( 2 2 ttJxbS? ! hb8 23 ttJ c3 ttJxc3 24
@xc3 a6+) 22 . . . @e6 2 3 t2Jg7+ @e7
24 ttJfS+ @e6= Charbonneau-Nataf,
Montreal 2003 .
1 2 exdS
Or 12 hf6 gxf6 13 0-0-0 d4 14
c3 9Wc7 1S ttJc2 dxc3 16 Vfixc3 9Wxc3 17
bxc3 g8 1 8 ttJe3 (18 g3 !b7; 18 f3
fS ; 1 8 t2Jb4 b7 19 d6 g6) 18 . . . e6
19 @b2 fS = .
48
1 2 exdS ifs 1 3 9Wb3 9WxdS 14

VfixdS ttJ exdS lS c4 bxc4 16 hc4 0-0
17 0-0 ttJ b4 18 fel fe 8 = , Anand
Leko, Linares 200S;
d) 11 !d3 dS 1 2 exdS 9WxdS (after
13 f3 Black equalises with 13 . . . e4 or
13 . . . j,fS) 13 9Wd2 ttJ e4 ! 14 Vfie3 ttJxgS
lS VfixgS VficS 16 0-0-0 0-0 17 hel f6
1 8 9We3 9Wxe3 + 19 xe3 !e6 = .
e ) 1 1 Vfif3 ttJ d 7
Black intends to castle and open

the f-file. White has tested:
1 2 c4 b4 13 ttJc2 aS 14 !d3 0-0 lS
0-0 tlJcS =
1 2 0-0-0 ttJcS 13 !e3 ttJe6 14 @bl
(Or 14 9Wg3 !b7 1S f4? ! he4 16 fxeS
dS+, Sitnikov-Krapivin, Serpukhov
2 0 03 ; 14 c4 Vfic7 lS @bl b4 16 ttJc2
aS 17 h4 !b7 1 8 !d3 Vitolinsh-Lig
terink, 1978 18 .. .fS ! 1 9 Vfih3 f4 2 0 !cl
ttJcS with excellent play) 14 . . . !b7 lS
c4 0-0 (16 cxbS? ! axbS 17 ttJxbS dS ! )
1 6 !d3 ttJc6 ! Vitolinsh-Chekhov,
Daugavpils 1 978 ;
12 b4 f6 1 3 !d2 !b7 14 c4 fS lS
cxbS fxe4 16 Vfig4 0-0 17 bxa6 (17
ttJ c4 !dS ! ) 17 . . .h a6 18 ttJc4 dS ! 19
9We6 + @h8 20 !gS Vfic7! with an ex
cellent game, e.g. 21 he7? hc4 2 2
!xf8 ttJxf8 2 3 Vfih3 !d3 ! -+ .
'
7 gs a6 8 ltJa3 bS 9 ltJdS
11
...
gxf6
1 2 c4
Alternatives:
12 f3 fS 1 3 exfS hf5 14 d3
e6 15 0-0 dS 16 :gadl V!Jc7 17 V!Jf6
0-0- 0 ! shows that Black can castle
long in some lines . The more con
ventional 15 . . . 0-0 16 c4 f5 17 :gfdl
e4 18 V!Je3 bxc4 19 hc4 dS 20 ltJc2
(20 h6 V!Jd6) 20 . . .f4 ! ? oo is another
good option;
12 V!Jd2 b7 13 0-0-0 he4 ! 14
V!Jxd6 V!Jxd6 15 :gxd6 ltJc6 16 f6 (16
f3 e7 17 :gd2 fS+) 16 ... ltJb4 gives
Black a fine game;
12 d3 b7 13 hS ! ? is a modern
attempt. It requires precision from
Black. The other continuations are
less testing:
13 V!Je2 dS 14 exdS (14 c3? ! fS ! )
14 . . . ttJxdS 15 0-0-0 e7 16 e4 ltJf4
17 V!Jf3 he4 18 V!Jxe4 :gc8 19 g3 ltJe6
20 f4 ltJcS 21 V!JfS V!J e6 22 xe6 + fxe6
23 fxeS fxeS= , Zatonskih-Cmilyte,
rapid, Tallinn 20 0 0 ; 13 c4 transpo
ses to 12 .c4 b7 13 .d3 .
13 . . . ltJg6 !
Black does not renounce his
common plan with .. .f5. He only
improves first the position of his
knight.
Nowthe careless 14 g4? ! would

run into 14 .. .fS ! when 15 exfS (or 15
xfS ltJf4 16 g4 V!JaS + ! 17 fl b4
18 :gbl V!Jd4 19 V!Jdl dS ! +) 15 . . . h4
(15 . . . ltJf4 ! ?) 16 h3 hS is in his favour.
Therefore, White should choose ei
ther:
a) 14 0-0-0 ltJf4 15 V!J h6 (15 g4
hS 16 g7? e7!+; 15 V!Jf3 f5 16 bl
V!Jd7=) 15 . . . :gg8 16 g3 (16 xh7? e7!)
16 ... ttJxd3 + 17 cxd3 f5 18 V!Jxh7 (18
:ghel V!JgS + 19 V!Jxg5 :gxgS=) 18 . . . gS+
19 f4 (19 bl fxe4) 19 . . . exf4 2 0 gxf4
V!Jxf4+ 21 bl :gg2 ! or:
b) 14 g3 dS 15 0-0-0 b4 16 ltJbl
V!JaS (16 ... V!Jb6 ! ? 17 ltJd2 0-0 oo) 17
fS ! ? (17 a3 bxa3 18 ttJxa3 0-0 19
f4 :gac8 was good for Black in Arn
grimsson-Deepan, Belfort 2005;
or 17 exdS hdS 18 :ghel 0-0 19
f4 :gfe 8 =) 17 . . . 0-0 18 h 4 (18 exdS
hdS 19 :ghel g7 20 h4 e6 2 1 f3
l2Je7=) 18 . . . dxe4 19 he4 he4 2 0
V!Jxe4 :gac8 with a complex, unbal
anced game.
1 2 fS !?
12 ... b7 is also a very good option. It is better tested, but the effect
of surprise would be lesser as well.
13 d3
It is quite risky to grab the pawn
by 13 cxbS? ! he4 (Practical experi...
49
Part 2
ence confirmed Sveshnikov's evalu
ation of the line 14 a4 dS lS bxa6+
@f8 . He thought that the dynamic
factors are more important, and
they favour Black, for example: 16
b4 g8 ! ? 17 f3 fS 18 g4 c8 ! t) 14
bxa6 (14 . . . dS ! ? 15 bS + @8 is a very
interesting alternative, which also
gives Black a good game : 16 0-0 g8
17 f3 b6 + 18 @hl f5 or 16 d2
g8 17 f3 i.f5 18 0-0 b6 + 19 f2
d4 with fair compensation in both
lines) 14 . . . 0-0 15 e2 l2Jg6 ! ? 16 0-0
ltJf4 17 f3 (17 f3 ixf3 18 xf3 xa6
19 ltJbS dS 20 a4 d4f!) 17 . . .f Sf!
with enough counterchances.
13 . . . g8
14 cxbS
White can hardly delay this cap
ture, since Black is ready to take on
c4 and push . . . dS:
14 0-0? ! bxc4 1S l2Jxc4 dS 16 exdS
xdS 17 f3 (17 e4 xg2 + 18 @hl
f2 ! - + Andriasian-Harika, Yer
evan 2 0 0 6) 17 . . . d8+ Kupreichik
Chekhov, Minsk 1976.
14 gl bxc4 (14 .. .fS ! ? 15 exfS e4
16 c2 ltJxfS 17 d2 dS 18 0-0-0 ! oo
with a very complicated position) 15
l2Jxc4 (15 a4+? ! c6 16 xc4 dS ! +)
lS . . . dS 16 exdS xdS 17 l2Jd6+ (17
a4+ @8 18 0-0-0 was played in
50
Motylev-Shirov, rapid, Bastia 2 0 04,

and here after 18 ... xg2 19 xg2
xg2 20 l2Jd6 dS White's compen
sation seems insufficient) 17 . . . @f8
18 e4 (18 l2Jxb7 xb7 19 b3 c6+)
1 8 . . . aS+ 19 d2 xd2 + 20 @xd2
d8 21 b7 xd6 + 22 @c3 b6 2 3
f3 l2JfS and Black has a tiny edge.
14 . . . xg2 15 bxa6 a6 16 a6
xa6
This position is very unbalanced,
but Black's game is easier, because
his king is well hidden behind the
pawns. Even the endgame could
be in his favour, as the g2-rook be
comes extremely active along the
fourth rank after .. .fS, e.g. 17 d3
b6 ! 18 bS+ xbS 19 ltJxbS @d7
20 a4 fS 21 exfS g4 22 b3 b4 2 3
bl l2Jxf5t.
17 l2Jc4 fS 18 exfS ltJxfS 19 dS
h4 ! 20 xg2 xc4
White's defence is difficult.
12 .. .fS represents another ap
proach . Black plans to castle and
activate the knight. It was recom
mended by Kasparov.
1 3 exf5
In case of 13 cxbS? ! fxe4 White
risks to be overrun by Black's cen
tral pawn cluster. The endgame af
ter 14 l2J c4 (Or 14 bxa6 0-0; 14 a4?
0-0 15 xe4 as + 16 @dl fS+)
14 . . . ltJfS 15 dS e6 16 c6 + @f8
17 dl (17 xe4 axbS 18 l2Je3 l2Jd4+)
17 . . . l2Jd4 18 xd6 + xd6 19 l2Jxd6
d8 is much better for Black.
1 3 tll x fS 1 4 cxb5
Or 14 d3 l2Jd4 15 0-0 b7.
1 4 0-0 1 5 bxa6
lS d3 e4 16 e4 e7 gives Black
fantastic attack: 17 f3 (17 dS e8
...
...
7 gs a6 8 ttJ a3 bS 9 ttJdS
18 f3 e6 19 d2 dS 20 MS MS+
21 @fl f6oo) 17 . . . h4 + ! 18 @e2 (18
@d2 dS 19 MS MSt) 18 . . . e8
Perhaps White can survive here,

but he must defend very precisely:
19 g3? xe4+ 20 @f2 gs 2 1
fxe4 e3 + 2 2 @ g 2 tlJh4+ ! ! 2 3 gxh4
h3 # ;
1 9 V!JdS e6 2 0 Wid2 d S 21 MS
MS+ 2 2 @fl axbSt;
19 d3 dS 20 xdS gS ! 2 1
@elD ( 2 1 adl e6 ; 21 g4? tlJg3 +)
21. .. b7+ 2 2 Wixb7? xg2- + ;
19 el dS (19 . . . xh2 should lead
to a perpetual) 20 xdS e6 21 cs
ac8 22 V!Jb6 xh2 23 @2 tlJh4
24 gl, when 24 . . . cd8 or 24 . . . fS
maintain pressure .
1 5 ixa6
1S . . . V!Jh4 ! ? is also interesting.
1 6 ixa6 Was+ 1 7 Wd2 Wxa6
Black has nice compensation .

White should look for equality by
18 e 2 , intending to repeat moves
after 18 . . . aS + 19 Wid2 = .
The other option i s dubious:
18 h4? ! @h8 (18 .. .f6 ! ? 19 h3
bToo) 19 V!JgS [19 . . . tlJd4 20 V!if6 +
@g8 2 1 gS += (21 0-0-0? ? fc8 +
22 @bl Wixa3 23 bxa3 ab 8 + 24
@al tlJc2 #)] 19 . . . tlJg7 (avoiding the
draw!) 20 h3 f6 !oo.
B. 1 0 xf6 ..txf6
In this part we'll consider only

the rare continuations:
Bl. 11 tlJbl
B 2 . ll c4
Other minor alternatives are:
a) 11 ttJxf6+ Wixf6 (12 e2 b7
13 0-0 0-0 14 c3 fd8 15 tlJc2 (15
f3 ttJe7 16 c4 ttJc6 ! ? 17 cxbS ttJ d4t)
lS . . . ttJe7 16 f3 dS+) 12 c4 0-0 13
cxbS ttJd4 14 tlJc2 (14 bxa6 g6 15
f3 xa6t) and here after 14 . . . dS ! ?
(14 . . . axbS lS ttJxd4 exd4 16 d3 gS
17 0-0=) 15 exdS g4 16 d2 axbS
Black has the initiative;
b) 11 h4? ! ttJe7 (Sveshnikov's re51
Part 2
commendation 11 . . . .ixh4 leads to a
very unclear position after 12 xh4
xh4 13 ttJc7+ e7 14 ttJxa8 xe4+
lS e2 b4 + 16 c3 as 17 0-0-0
.ie6oo) 12 c4 ttJxdS 13 xdS 0-0 ! 14
cxbS (14 xa8? aS+ - +) 14 ... .ie6
lS d2 axbSt.
c) 11 .ie2 This move is seldom
seen because White loses control
over dS after 11. . . 0-0 12 0-0 .igS 13
c3 (13 c4 b4 14 ttJc2 aS lS b3 .ie6 16
a3 bxa3 17 xa3 b8 18 d 3 b7 19
c3 b8 20 g3 .id8= Vladimirov
Panchenko, Tbilisi 1973) 13 . . . ttJe7!
Perhaps at this point White realis
es that his a3-knight is too far fram
the centre. 14 c4 ttJxdS lS xdS .ie6
16 d3 b4 17 ttJc2 b6 18 b3 aS= ,
game 9 Janosevic-Jussupow,
Amsterdam 1978 .
d) ll .id3 .ie6 (11 . . ..i gS is also an
option: 12 0-0 0-0 13 c4 b4 14 ttJc2
aS lS ttJce3 .ixe3 16 fxe3 .ie6 17 hS
ttJb8 18 f3 ttJd7 19 afl a7 2 0 g4
f6+ Black's position is difficult to
attack and White's pawn weakness
es will soon become a problem) 12
c3 .ixdS 13 exdS ttJe7 14 ttJc2 0-0 lS
a4 bxa4 16 xa4 b6 17 b4 cs 18
ttJ e3 .igS+ Bogaerts-Krasenkow, Os
tend 1990 .
8 1 . 1 1 b 1 gb8!
Directed against 12 a4 . If White
persists with it, he will be worse af
ter 12 a4 bxa4 13 xa4 (13 ttJd2? !
xb2 14 ttJc4 b8 lS ttJxf6+ xf6
16 ttJxd6+ e7 17 ttJxc8 + hxc8 18
.ixa6 d8 19 .id3 a8+) 13 ... xb2
14 .ixa6 (14 cl b8 lS .ixa6 .igS
16 dl .id7+) 14 . . . .ig4 ! ? lS cl (lS
S2
f3 d7 16 0-0 ttJb4 17 b4 xb4 18

ttJxb4 b6+ 19 hl xb4+) 1S . . . b8
16 0-0 0-0+. So he must prepare the
wing break with:
1 2 d2
Motylev-Rogozenko, Bucharest
1998 saw the bizarre:
a) 12 b4, which aims to secure
the position of White's knight on c3
and prepare a4. The game went on
12 . . . .ie6 13 ttJ bc3
13 ttJxf6 + ? ! xf6 14 xd6 c8 lS
.id3 offers Black at least two ways
to develop the initiative: lS . . . gS 16
ttJd 2 xg2 17 fl .ih3 18 a4 g6 19
xg6 (19 cs xfl 20 ttJxfl e6 2 1
axbS ttJd4+) 1 9 . . . hxg6 20 axbS xfl
2 1 xfl axbS 2 2 hbS e7 where
White's compensation is clearly not
enough, and 1S . . . e7 ! ? 16 xe7+
xe7 17 a3 ttJd4 18 d2 ttJxc2 19
hc2 hd8 + 20 .id3 .ic4t.
13 a4 hdS 14 exdS e4 lS c3
ttJxb4 16 axbS axbS 17 a3 ttJd3 + 18
hd3 exd3 19 0-0 0-0 20 xd3 b4
is about equal.
Now let us return to 13 ttJbc3 :
Rogozenko chose 13 . . . 0-0 14 a4

ttJxb4 ! lS ttJxb4 (lS axbS ttJxdS 16 exdS
e4 ! ) lS . . . aS with the better game,
for instance, 16 ttJ ca2D (16 xd6?
fd8 17 cs bc8 18 ttJc6 xc6 19
7 igS a 6 8 llJ a 3 b S 9 llJdS

xc6 c8 - +) 16 . . . ha2 17 c3 ie6 18
axbS c7 19 cl (19 c4 axbS 20 ttJ a6
c6 2 1 ttJxb8 xb8 gives Black more
than enough compensation for the
exchange, Rogozenko) 19 . . . b7 (20
llJxa6 bc8 21 e3 id8 !oo) 20 bxa6
xe4+ 21 e3 a8 ! t .
I n the diagram position Black
can also play the logical 13 . . . c8 ! ?
1 4 a4 (14 id3? ttJxb4 lS ttJxb4 xc3+;
14 ttJxf6 +? ! xf6 lS d2 ttJd4 16 a4
0-0+) 14 . . . ttJxb4 ! lS ttJxb4 as with
similar ideas as in the above-men
tioned game, but in an even better
version:
16 bl (Or 16 bl c3 17 axbS
b3 ! 18 cxb3 xb4+ 19 @e2 axbs ;
16 ttJca2? ha2 - +) 16 . . . xc3 17 axbS
(17 xd6? c2 18 axbS? ie7- +)
17 ... ib3 ! 18 id3 (18 xd6 hc2 19
d2 hbl 2 0 xc3 he4+) 18 . . . xb4
19 0-0 axbS 20 b3 xb3 21 cxb3
0-0 and Black is a pawn up.
b) 12 g3 0-0 13 ig2 ie6 14 0-0
hdS lS exdS (lS xdS? ! ttJd4 16
ttJa3 b4 17 ttJc4 c7 ! 18 ttJe3 bS 19
c4 xc4 20 ttJxc4 cS+) lS ... ttJ aS
was equal in Hoffmann-Yakovich ,
Munich 1993 ;
1 2 ... g 5 1 3 a4
White must hurry with this
break, or he will lose the battle in
the centre : 13 id3 ttJe7 14 ttJxe7
xe7+, Stefansson-Schandorff, Co
penhagen 1994.
1 3 . . . bxa4 ! ? 1 4 liJ c4 0-0
lS h4 ih6 16 g4 if4only weak
ens White's kingside: 17 xa4 ie6
18 a6 hdS 19 exdS ttJb4 20 a3
bS and Black seizes the initiative.
1 5 gxa4 liJ e 7 !
White is behind in development

and should think about maintain
ing the balance.
B 2. 1 1 c4 b4 1 2 liJ c2
12 a4? is totally inconsistent:
12 . . . d7 13 ttJxb4 ttJd4 14 dl b8
lS ttJd3 (lS ttJ ac2 ttJxc2 + 16 ttJxc2
xb2+) lS . . .igS 16 b4 fS 17 exfS 0-0
with attack.
1 2...as
We'll examine i n detail:

B2a. 13 ie2
B2b. 13 g3
B2c. 13 f3
White has also tried in practice:
a) 13 cS was played only once
S3
Part 2
and although White gets obvious
positional compensation, it can
hardly be dangerous: 13 . . . dxcS 14
bS ib7 lS ttJce3 0-0 16 .ixc6 ixc6
17 E!cl igS 18 E!xcS E!c8 19 0-0 ixe3=
20 ttJxe3
Following 20 fxe3 Black can
equalise with 20 . . . .ixdS 21 E!xc8 ( 2 1
E!xdS b6) 2 1. . . xc8 2 2 exdS cs
23 d2 (23 d3 f5 24 d6 f4 2S d7
E!d 8=) 23 .. .d8 24 E!cl (24 E!dl f5)
24 . . . d6 2S e4 fS= .
2 0 . . . e7 2 1 E!xaS ( 2 1 cl id7
2 2 E!xc8 E!xc8 23 ltJdS d6 24 d2
fS 2S exfS?? ic6 26 E!cl E!d8 27 f6
xdS- +) 2 1. . . ixe4 2 2 E!el E!fd8 23
g4 c7+ Vachier Lagrave-Ni Hua,
Turin ol . 20 0 6 ;
b ) 1 3 d3 igS 14 E!dl? ! i s asking
for trouble: 14 . . . 0-0 lS ttJde3 ? ! a4 !
16 xd6? b3 ! when 17 xc6 loses to
17 . . .aS+ 18 We2 (18 E!d2 bxc2- +)
18 . . .bxc2 19 ttJxc2 ig4+ 20 f3 ie6+ 2 1 ltJe3 E!ab8 22 E!d2 b6 23 c7
c8.
17 axb3 axb3 18 ltJ a3 is only a
little more stubborn: 18 . . . aS+ 19
d2 cS+;
B2 a. 13 J.e2 is something like

a worse version of line B2b. The bi
shop does not protect e4 and White
lacks the plan with f4. Black pos
sesses two promising setups: the
manoeuvre ltJ c6-b8-d7-cS when the
light-squared bishop goes to b7, and
. . . ie6xdS. Play continues with :
13 . . . 0 - 0 14 0 - 0
14 d3 commonly transposes to
the main line after 14 . . . gS. Black
might also try to lure the opponent
S4
in to a seemingly active position

with 14 . . .e6 lS 0-0-0? ! (lS ltJxf6+?
xf6 16 xd6 E!ac8 17 ltJe3 fd8 1 8
cs ltJ d4 19 xaS ttJxe2 2 0 Wxe2
ixc4+ 21 Wel d2 ! - + Sveshnikov)
lS . . . ixdS 16 xdS b6 17 E!hfl ac8
18 d3 a4 19 ltJe3 gs 20 @bl ixe3
2 1 fxe3 ltJ aS+, Sveshnikov.
14 g5
.
15 Wfd3
The game Frolov-Kramnik, So
c hi 1990 presents a good exam
ple of the ltJb8-d7 manoeuvre: lS
b3 ib7 16 d3 ltJb8 17 E!fdl ltJd7 18
ltJel ltJcS+. Black keeps options to
play on both wings.
15 b7
lS . . . e6 is the favourite plan of
several Chinese GMs. They proved
that Black has sufficient resources:
16 adl ixdS ! 17 cxdS (or 17 xdS
b6 18 bS c7 ! = . White farced
play in Li Shilong-Wang Yue, 20 0 6
and got into serious problems af
ter: 19 cS? ! dxcS 20 xcS E!ac8 2 1
ig4 ltJe7 2 2 xc7 E!xc7 2 3 ltJe3 g6
24 E!d7? ! E!xd7 2S ixd7 E!d8 26 E!dl
ixe3 27 fxe3 ltJc8 ! ) 17 . . . ltJb8 18 ig4
ltJa6 19 g3 ltJcS 2 0 f3 ltJ a4 21 b3
ltJcS 2 2 f3 b6 23 h4 h6 24 ltJe3
ixe3 2S xe3 bS=, Asrian-Ni
7 gs a 6 8 tlJ a3 b S 9 tlJdS
Hua, Taiyuan 200 6 .
1 6 E:adl ll) b 8 1 7 a3
Instead of winning a pawn,
White prefers to deprive the op
ponent of his spatial advantage
on the queenside. 17 ttJde3 tlJa6 18
tlJfS tlJcS 19 xd6 is quite risky in
deed. Black has strong compensa
tion after 19 . . . xd6 20 ttJxd6 c6
or 19 . . . ttJxe4 ! ? 20 xeS f6 21 f4
b8 .
:gd1 fxe4 23 xe4 a7! 24 f3 a4f!

2S g2 axb3 26 axb3 a2 27 :gxd6
xb3 28 :gxh6, draw, Khairullin
Tregubov, Sochi 2 007.
b) lS hS 0-0 16 :gdl (16 h3
xh3 17 fuh3 a4 ! ? f!) 16 . . . a4 !
1 7 bxa3 18 li)xa3 li)a6 1 9

li)b5 li)c5 20 c2 a4, Malakhov Carlsen, Sarajevo 2006, Black has a
satisfactory game.
B2b. 13 g3 0 - 0
You can play 13 . . .gS ! ? if you do
not like the positions after 13 . . . 0-0
14 h4. Then 14 g2 0-0 transposes
to the main line, so we'll consider
here 14 h4 h6 . Now the plan with
f4 is no longer possible, so White
bases his play on the clumsy posi
tion of our bishop on h6. It is practi
cally out of play and the fewer piec
es on the board, the more noticeable
that would be.
a) lS h3 xh3 (lS . . .b7 !? to
avoid exchanges is an interest
ing alternative: 16 hS tlJe7 17 0-0
ltJxdS 18 exdS c8 19 xc8 xc8
20 b3 0-0 21 g2 d2 22 :gadl c3
23 f4 cs 24 fxeS xeS 2S tlJel c3
26 tlJd3 c7+, Berg-Eljanov, Kern
er 2007) 16 :gxh3 tlJe7 looks equal,
which was confirmed by the fallow
ing game: 17 :ghl (17 fl? ! c8 ! )
17. . . 0-0 1 8 fl :gc8 1 9 ltJxe7+ xe7
20 b3 h8 ( 20 .. .fS ! ?) 2 1 d3 fS 2 2
In this system, if Black achieves

to push the a-pawn that far, he usu
ally obtains good counterplay. The
trick is that 17 ltJxb4 is bad due to
ttJxb4 18 ttJxb4 a3 ! , so White must
finally remember about develop
ment:
17 d3 b3 18 axb3 axb3 19 tlJ a3
Nepomniachtchi-Pavlovic, Biel 2007
and here after the natural exchange
sacrifice 19 . . . :gxa3 ! 20 bxa3 as + 2 1
fl h8 Black would have enjoyed
rich compensation and a pleasant
game. Note that 22 tlJf6? ! would be
bad due to 2 2 . . . tlJd4 23 g4 tlJe6 ! + .
14 g2
14 tlJxf6+ xf6 lS g2 ie6 16 b3
(16 xd6? :gac8 17 tlJe3 :gfd8 18 cs
tlJd4+) 16 . . . d8 ! 17 0-0 b6 18 d3
a4= ;
1 4 h 4 has been popular lately.
However, pawns do not move back
wards, and such moves are quite
double-edged. We propose an en
terprising pawn sacrifice which has
SS
Part 2
similar ideas as the novelty in line
B 2c:
14 . . . a4 ! ? lS ltJcxb4 (l S h3 b3)
1S . . . ltJxb4 16 ltJxb4 b6 17 a3 d8
White might be sorry to have

weakened his kingside. We would
not like to be in his shoes after 18
d3 b7 19 0-0 b6 20 ltJdS d4
2 1 bl i.h3 22 el d7. Perhaps he
should play 18 d3 e6 19 h3 cs
2 0 he6 fxe6 21 0-0 as 22 adl
fd8 with equality.
Let us note that Black can also
choose on the 14th move:
a) 14 . . . g6, which occurred in
the game Govorykh-Zhigalko, Ki
rishi 20 07: lS hS (lS h3 b7 16
0-0 ltJ d4f!; lS f3 g7 16 hS fSoo)
lS . . .gS 16 h3 b7 (do not trade
this bishop on h3!) when instead of
17 @fl? ! ltJe7 18 @g2? ltJxdS 19 cxdS
@g7 2 0 ltJel fS+ White should have
played 17 0-0 ltJe7 18 ltJce3 he3 !
with a possible draw after 19 ltJf6 +
@g7 2 0 fxe3 ltJfS 2 1 ixfS (21 exfS
xf6 2 2 g4 gS=) 21. . . xf6 2 2 c8
gs 23 hb7 xg3 + = and :
b) 14 . . . e6 lS i.h3 ltJd4 16 ltJxd4
16 ltJce3 can be attacked by the
surprising move 16 . . . c8 !? , when
17 ltJb6 b7 1 8 ltJxa8 hc4 19 ltJxc4
(19 !fl !e6 ! ) 19 . . . xe4 + 2 0 @d2
S6
ltJ f3 + 21 @cl xc4+ 22 c2 xc2 +

23 cjfxc2 a8 is unclear, as well as
17 he6 fxe6 18 ltJb6 c6 19 ltJxa8
xe4t or 17 ltJxf6+ gxf6 18 g2 a7)
16 . . . exd4 17 he6 !
White wants t o get a good block
ading knight against a poor bishop.
Previously White had tried 17 d3
c8 18 @fl (18 dl cs 19 ltJxf6+
xf6 20 he6 fxe6 21 xd4 xc4 ! )
1 8 . . . cS 19 @g2 hdS 20 exdS c7= ,
Svidler-Elj anov, Bundesliga 2006 .
17 . . . fxe6 18 ltJf4
Here in both known games Black

played 18 . . . e7? ! and was worse af
ter 19 g4 ae8 20 0-0. Instead, we
propose:
18 . . . c8 !
The hit on c4 discoordinates
White's pieces. All defences have
their drawbacks:
19 cl
Or 19 d3 a7 2 0 0-0 eS 21 ltJ dS
g4f!; 19 e2 es ; 19 b3 d3 . After
the text Black free himself using the
hanging state of the cl-rook:
19 . . . c6 20 e2 !
2 0 0-0 ae8 21 d3 (21 el
hh4) 21. . . hh4 = .
2 0 . . . d s 21 exdS
Or 2 1 eS dxc4 22 0-0 d3 23 ltJxd3
(23 xd3 heS=) 23 . . .hh4 with a
7 gs a6 8 ltJ a3 bS 9 ltJdS
drawish position, for example: 24
g4 e7 2S rocc4 (2S xc4 xc4
26 xc4 ac8 27 fcl rocc4 28 xc4
d 8f!) 2s . . . ds 26 c7 f7 27 ltJf4
xeS 28 ltJxe6 e8 29 c4 cs 3 0
xcS xe6 31 xe6 xe6 32 xaS
e2 = .
21. . .exdS 2 2 0-0 dxc4 2 3 xc4
(23 xc4+ xc4 24 rocc4 E!:ac8 2S
fcl xc4 2 6 xc4 xh4=) 23 ... d7!
24 ltJd3 ae8=.
1 4. .igS 15 0 -0 li)e7 16 li)ce3
16 f4 is positionally dubious as
it opens up play in favour of Black's
bishop pair: 16 . . . exf4 17 gxf4 h4
and now:
a) 18 hS ltJxdS 19 xdS (19
cxdS? ! a6+) 19 . . . e6 20 d3 c7
21 b3 a6+;
b) 18 ltJce3 e6 19 d4 (19
eS dxeS 20 ltJxe7+ xe7 21 xa8
cS ! oo with excellent compensa
tion.) 19 . . . c8 20 acl e8 f!;
c) 18 ltJde3 a6 19 d 3 f6 2 0
ltJd4 b6 21 adl a4oo ; Black can ma
noeuvre on the dark squares, main
taining the balance. White should be
constantly watching out for . . . b3. If
he decides to prevent this possibili
ty by playing b3 himself, then Black
will get the a-file in his possession.
White can also try to attack the
d6-pawn:
16 ltJde3 b6 17 d3 (Or 17 hS
h6 18 h4 xe3 19 ltJxe3 e6= ; 18
f4 f6 19 fS gs is risky for White)
17 . . . e6 18 fdl fd8 19 ltJfS, but it
does not achieve the aim since the
d6-pawn is immune and Black is
able to repel the awkward knight
by . . . g6. In a blitz game Kolev con. .
tinued 19 . . .c7 20 b3 g6 21 ltJxe7+

xe7 22 ltJe3 gs 23 ltJdS cs and
Black's chances are already pref
erable in view of his imminent
counterplay down the a-file: 24 h4
h6 2S f3 a4 26 @g2 axb3 27 axb3
hdS 28 cxdS a3 29 abl c3 3 0
e 2 c8 31 Ml a 2 3 2 g4 xd3+
33 xd3 cc2 34 f3 fS 3S exfS e4
0-1.
O ne of the latest game in this
line featured the dubious 16 b3? !
ltJxdS 17 xdS e6 18 d3 b6 19
@hl a4 and Black already had some
initiative in Movsesian-Tregubov,
Dagomys 0 2 . 04 . 200 8 : 2 0 f4 axb3
21 axb3 xal 22 ltJxal exf4 23 gxf4
f6.
16 ... ie6 17 d3
Or 17 f4? ! exf4 18 gxf4 h4 19
d4 (19 hS ltJxdS 2 0 cxdS d7 2 1
e S g 6 2 2 f3 c8 and we prefer the
bishop pair.) 19 d4 c8 20 acl
e8 with good chances.
17...he3 18 li)xe3
This position is about equal, and

quite boring at that. White has more
active rooks , but they can attack
only one weakness, on d6, which is
easily defended. A logical continua
tion of the game would be:
S7
Part 2
18 . . . V9d7 (In fact 18 . . . V9c7 !? looks
more consistent as on the d-file the
queen is X-rayed by the enemy
rook. Anyway, its placement does
not change significantly the char
acter of the position. In contrast,
18 . . . %Vb6 ? ! is weaker on account
of 19 tlJfS!;!;:.) 19 fdl fd8 2 0 acl
ac8= , Balogh-Moiseenko, Moscow
200 6 .
White can open u p the queenside
by a3, but then Black takes on a3 and
obtains typical Sicilian counterplay
along the b-file .
Play might become interesting
only in case when White attempts a
kingside attack. However, Black can
then invade White's rear through the
c-file with . . . a4, . . . b3, . . . tlJ c6-d4-c2 ! .
We analyse this plan in the "Com
plete Games" section, see g ame 1 0
Korneev-P. Horvath, Porto San
Giorgio 2007.
B2c. 13 1Yf3
This is a relatively new idea.
White aims to put his rook on dl
and build pressure on d6 as in the
game N epomniachtchi-Andriasian,
Moscow 14. 0 2 . 200 8 : 13 . . . e6 14
dl e7 lS cs 0-0 16 bs tlJa7 17
.ta4, o r organize a kingside offen
sive: 13 . . .gS 14 h4 h6 lS g4 f6 16
gl 0-0 17 V9g3 Morozevich-Carlsen,
Blitz Moscow 2 2 . 1 1 . 20 07.
We propose a new approach,
connected with a pawn sacrifice:
13 e6 14 gdl a4! ?
Kolev's idea i s t o anticipate the
enemy's activity and radically dis
turb his plans.
S8
15 lll cxb4
Now lS cS? ! does not work in
view of lS . . . WaS ! .
lS tlJxf6 + does not seem too test
ing either: 1S . . . gxf6 ! ? (We believe
that lS . . . V9xf6 16 V9xf6 gxf6 17 xd6
c8 18 f3 @e7 also ensure sufficient
compensation, e.g. 19 d2 hd8co)
16 e2 b3 17 axb3 b8 18 b4 tlJxb4
19 tlJ a3 tlJc6 20 tlJ bS Wi as + 21 d2
@e7 22 V9d3 tlJd4 23 tlJxd4 exd4 24
V9xd4 Wies . Black has full compen
sation for the pawn, for example: 2S
WxcS dxcS 2 6 0-0 b4 27 cl hb8
28 cc2 fS 29 f3 (29 eS f4 !) 29 . . . fxe4
30 fxe4 @f6 31 @f2 @es.
15 1Ya5!
1S . . . tlJxb4 is an inferior option:
16 tlJxb4 %Vb8 ( 1 7 a 3 d8 18 V9d3 0-0
19 e2 as 2 0 V9xd6 V9xd6 21 xd6
tb8co) 17 V9a3 0-0 18 d3 c8 19
0-0 (19 tlJdS hdS 20 cxdS gs 2 1
0 - 0 V9b6co) 1 9 . . . hc4 2 0 hc4 xc4
21 tlJdS;!;:. Computers underestimate
White's advantage in such posi
tions, but practice shows that it is
a very difficult task to defend them.
16 a3
Another critical line is 16 Wffc3
tlJxb4 17 tlJxb4 (17 Wixb4 .td8 ! =)
17. . . WcS !
7 gs a6 8 l2J a3 bS 9 ltJdS
17... d4 18 d3 0 - 0 19 0 - 0
E:c8 2 0 E:cl
17 . . . 0-0 is less precise, as it gives

White a chance to escape from the
pin along the el-a6 diagonal: 18 e2
tb8 19 a3 Wies 20 0-0 d8 ! 21 Wff c2 !
(21 Wffd2 iaS) It is true that even then
Black has probably sufficient com
pensation, e.g. 21 . . . b6 22 ltJd S (22
d3 Wffc7 23 fdl cS=) 2 2 . . . hdS 2 3
xdS Wffc6 = , but we have no reason
to give White extra possibilities.
18 Wffd 2 (or 18 e2 d8 19 0-0
as 2 0 a3 0-0) 18 ... 0-0 19 a 3 tb8
20 e2 d8 21 Wffxd6 Wffxd6 22 xd6
e7.
16 ... i.d8 17 i.e2
Or 17 Wffc3 l2Jd4 18 d3 0-0 19
0-0 Wff c Soo followed by . . . as .
The fine point of this position

is that the opposite coloured bish
ops work in Black's favour, as after
20 l2Jc2 hdS 21 exdS b6oo. After
the text we continue setting up our
pieces on dark square s:
2 0 a7 21 h3 a5
Our play is on the queenside .
21. . .fS 22 exfS hfS 23 Wffdl does not
help us at all.
22 g4 E:b8
All our pieces are well placed, the
a4-pawn petrifies White's queenside
and renders his extra pawn useless.
S9
Part 2
1 e4 cS 2 li)f 3 li)c6 3 d4 cxd4

4 li)xd4 li)f6 s li)c3 es 6 li) d bS d6
7 .igS a6 8 li)a3 bS 9 li)dS .ie7
COMPLETE GAM ES
9. D. J ano sevic - J us supow

Am sterdam 1 978
1 e4 cs 2 ti:) t3 e6 3 d4 cxd4 4
ti) xd4 ti:) t6 S ti:) c 3 ti:) c6 6 ti:) db S d6 7
.it4 es 8 .igS a6 9 ti:) a3 bS 1 o ti:) d S
.i e 7 1 1 .ixt6 .ixt6 1 2 .ie2 0-0 1 3
0-0 .igS 1 4 c3 ti:) e7
We se e a typical scenario when

White simply develops his pieces
without some clear active plan. Now
he must part with his only good
piece, the dS-knight, and Black gets
a free hand on the queenside.
1 S c4 ti:) x d S 1 6 '%Yxd S .ie6 1 7
'%Y d3 b4 1 8 ti) c2 '%Yb6 1 9 b3 as 20 a3
White hurries to undermine b4,
since otherwise Black will play . . . a4
and a3 will be no more possible.
20 ... bxa3 21 xa3 ab8 22 ti:) e 3
60
'%Yes 23 ta 1 .id8 ! 24 .ig4 .ib6 2 S

3a2 g 6
Jussupow could have drawn the
game by trading everything with
25 . . . Wfd4 2 6 dl Wfxd3 27 xd3 he3
28 he6 hf2 + = , but he prefers to
maintain the tension. After the ac
tivation of his dark-squared bish
op, he now embarks on the second
stage of his plan - opening of the
f-file.
26 d 1 bd8 27 ti:) d S
White could have provoked first
. . . hS. That could be achieved by 27
h4 hS 2 8 he6 fxe6, but then h4
would also be weak. Therefore 27 h 3
looks better: 2 7. . . @ g7 28 ad2 hS= .
2 7. . ..ixg4 2 8 ti:)t6+ @g7 2 9 ti:) x g4
t S 30 ext S xt S
Black seized the initiative. Jano
sevic tries to simplify play, but the
bishop is too strong in the end
game.
31 '%YdS dt8 32 '%YxcS .ixcS 33
e1 St7 34 ti:) e 3 hS 3S g3 b8
Now Black's rook penetrates
into White's camp.
36 b 1 tb7 37 xaS xb3 38
xb3 xb3 39 bS d3 40 b7+
@t6 4 1 ti:) d S+ @ts
7 .igS a6 8 l2J a3 bS 9 ltJdS

tll ce3 .ixe3 1 7 tll xe3 e6 1 8 V!f d3
'\Mf d7
White is beyond salvation, be

cause he cannot parry the march of
the Black king. It is very difficult to
point out the exact moment when
White made the decisive mistake.
Jussupow conducted the game con
sistently, realised all typical plans in
this structure, while White seemed
clearly helpless.
42 g2 d2 43 f7+ i> e4 44 h4
d3 4S tll e7 e4 46 f1 d 1 + 47
i>g2 d2 48 f1 xc4 49 tll x g 6
d 4 S O tll f8 d S S1 tll e 6 b 6 S 2
f4 c 3 S3 ts c 2 S4 tll g S d 1
S S tll h 3 d 4 S 6 bS c7 S 7 b 1 +
i>c2 S8 b7 as S9 tll g S d 1 + 60
i> g 2 e1 61 tll e 6 c3 62 c 7 d3 63
tll d4+ b2 64 b7+ a2 6S a7+
i>b1 66 tll b S b2 67 tll x c3 xc3
68 c7+ b3 69 d7 c2 70 c7+
d1 71 g4 hxg4 72 hS i>e2 73 h6
f1 74 e7 xf2+ 7S rbg3 e3 0-1
1 0 . Ko rneev-P. Horvath
Porto San G iorgio 2 6.08.200 7
1 e4 cs 2 tll f3 tll c6 3 d4 cxd4
4 tll xd4 tll f6 s tll c 3 es 6 tll d b S d6
7 gs a6 8 tll a 3 bS 9 tll d S e7 1 0
xf6 xf6 1 1 c4 b4 1 2 tll c 2 a s 1 3
g 3 0-0 1 4 g2 gS 1 S 0-0 tll e7 1 6
Perhaps 18 ... Wffc7 i s slightly more

precise, in order to avoid the oppo
sition of White's rook from dl. In
terestingly, two months earlier Kor
neev faced the same 18 . . . Wffd 7, and
chose 19 adl fd8 2 0 d2 Wffc7. The
game went 2 1 h4 a4 2 2 @h2 l2Jc6 23
f4 f6 24 ltJdS b7 2S fS ixdS 26
WffxdS + @8 27 fdl as, 112-112, Ko
rneev-Nataf, Ourense, lS. 0 6 . 20 07.
The only reason to refrain from c7
could be 19 ltJdS, but then Black
would be fine after capturing by
knight.
In the diagram position White
cannot win by tripling his major
pieces against d6. The game Balogh
Moiseenko, Moscow 2006 saw 19
fdl fd8 20 acl ac8 2 1 ltJdS Wff a7
22 l2Jxe7+ Wfxe7 23 b3 a7 24 c2
Wies 2S cd2 cj/f8 26 f3 @g8 and
a draw was agreed. The only active
plan is a kingside pawn storm, but it
is double-edged.
1 9 f4 f6 20 ts f7 21 g4 tll c 6 22
f3 tll d4 23 g3 a4
Of course Black should not help
the enemy open files on the kingside
by playing . . . h6 .
61
Part 2
24 h 1 b3 25 a3 gac8
White starts attacking first, but

Black has also fixed a target on c4
and might penetrate along the c-file.
Only calculations can help assess
correctly such positions. It seems
that Black's counterplay against
c4 and e4 is just enough to balance
White's pressure along the g-file.
26 g5 fxg 5 27 gxg5 h8 28
gg1
I f we see the game course , we'll
note that two moves later White had
to return his rook to cl. On the other
hand, it is not clear what he should
have played. 28 ih3 , intending f6,
turns out to be not so dangerous.
Black can even ignore the threat by
28 . . . c6 2 9 g3 (29 cl c7 30 g2
Wff a6) 29 . . . c4 30 ttJxc4 xc4 31 f6
xf6 3 2 c8 Wffxc8 with two pawns
for the exchange. 28 g3 has also
drawbacks, as it allows 28 . . . ihS 29
el cs.
28 . . . Wc6
This a good setup, but 28 . . . e8
deserves consideration as well. It
prepares . . . ihS .
29 Wd2 gc7
The first critical moment of
the game. Horvath could have ex62
changed the knights here with

2 9 . . . llJc2 ! ? Then 30 if3 c7 31
xg7? would fail to 31. . . ihS ! where
as 30 ttJxc2 bxc2 31 Wffxc2 c4 3 2
Wffc3 ib3 3 3 g3 c7 34 if3 Wffb7 3 S
Wff h4 b6 3 6 hS ig8 = would create
a fortress on the kingside.
30 g c 1 Wa6
This is an imprecision. Black
should hit e4 in order to prevent
White from manoeuvring the bish
op to d3. So: 30 . . . Wffb7 31 g3 fc8
32 Wffd3 ihS with active pieces.
3 1 gg3 gfc8 32 .if1 Y*lb7 33
Y*f g2 .i g8 34 .id3
34 . . . gfa ?
During the last moves White
has improved his position. Still,
he cannot win without the help of
his knight. Therefore, Black should
have grasped the chance to trade
it by 34 . . . llJc2 ! with unclear com
plications : 3S ttJxc2 bxc2 36 xc2
dS 37 exdS dS ; 3S c2 bxc2 36
xc2 c4oo; 3S llJdS dS 36 exdS
f8oo.
35 gg1 gf6 36 h4 ! ?
Korneev includes his last re
source in the assault. Computers
like noncommittal continuations
like 36 h3 a8 37 gS Wff e8 38 llJ g4
7 !gS a6 8 l2J a3 bS 9 ltJdS

f8, but Black is holding there.
36 .. . h 6 37 h3
37 ... f6?
After this mistake Black is lost.

He should have prevented White's
pawn from reaching hS. The only
move was 37 .. . c8, having in mind
38 hS \We8 ! 39 \Wg4 l2Jc2 ! 40 hc2
bxc2 41 gel !f7 42 xc2 xhS .
38 h 5 h6 39 g 3 f8
It was already late for 39 . . . l2J c2
due to 40 l2Jg4 f8 41 l2Jxh6+.
40 g6 .ih7 41 xd6 d7 42
xd7 '%Yxd 7 43 d5 '%Yf7 44 '%Yg4
d8 45 f1 d6 46 g2 .ig 8 47
'%Yg3 c6 48 c5 d7 49 f6 '%Yxh 5 50
.ib5 .ih7 51 fx g7+
1 -0
63
Part 3
1 e4 cS 2 f3 c6 3 d4 cxd4
4 xd4 f6 S c3 eS 6 db S d6
7 i.g S a6 8 a3 bS 9 dS i.e7
1 0 i.xf6 i.xf6 1 1 c3 0-0
Q U I C K REPERTO I RE
The fallowing two parts are closely

related with the name of the book.
After reading them, you'll be ready
to start playing the system with con
fidence.
Nowadays everybody follows in
the footsteps of the elite, and most
Sveshnikovs reach in seconds the
position after:
1 2 c2 i.gS 1 3 a4 bxa4 1 4
gxa4 a s 1 S i.c4 (lS !bS is rare
ly seen : 1S . . . llJe7 16 llJxe7+ Wixe7 17
llJb4 17 . . . !h3 or 17 . . . !g4) 1 s ...gba
1 6 b3 h8
to defend successfully against direct

attacks. White's targets are obvious
- the pawns on d6 and aS . There is
enough evidence that even if Black
trades the aS-pawn for b3, his posi
tion is not completely immune. De
cisive factor is the piece activity.
All our hopes are connected
with . . . f5.
This move serves multiple pur
poses. First of all, it opens a file
against the enemy king and the f2 pawn. Secondly, the c8-bishop ob
tains a new operating diagonal fram
fS. And finally, with the disappear
ance of the e4-pa wn, Black can hope
to move forward his central pawns.
Should that happen, he can stop
worrying about the aS-pawn, since
his initiative would amply compen
sate it.
We must also be prepared
for negative scenarios.
This is the basic position of the

modern Sveshnikov. It is extreme
ly popular, because White can try
to win it without much risk. He has
no weaknesses and should be able
64
If we fail to develop an initiative

on the kingside, we should switch
to a restraining tactic. In that case
Black usually seeks exchanges, in
order to remove the clamp on dS,
and balances the hit on aS by pres
sure on the b-pawn .
9 ltJdS ie7 10 hf6 ixf6 11 c3 0-0

has two major options : 17 0-0 fS
and 17 ltJce3 g6 . The latter leads to
different pawn structures and is the
subject of the next part of the book.
1 7 0-0 f5 1 8 exf5 i.xf5 1 9
lll c e3 i.g6!
Black is playing "around" White's

pieces . He does not aim to neutralise
them, but rather build his own play
with . . . e4 and . . . ltJeS. In a number of
lines the b3-pawn proves to be weak.
This position has a very good
reputation for Black. In fact, sta
tistically he scores over 50 percent.
That is easy to understand, because
the most natural move:
2 0 d3
leads to mass exchanges, so
White often experiments (unsuc
cessfully! ) with new ideas .
20 . . . hd3 21 xd3 he3 22 fxe3
rucfl+ 23 xfl rucb3 24 c4 bS !
(only move) 25 e4 (or 25 ltJc7 b6
26 ltJdS bS=) 25 . . . cS= .
Another common move is:
2 0 e2?!
It aims t o double the rooks on
the a-file, but this setup encourages
Black's attack with : 20 . . . e4 21 fal
(or 2 1 ibS ltJeS 22 fal ih4 23 g3
cs) 2L . .ih4 !
A typical way to provoke some

holes in the enemy's castling posi
tion.
22 g3 igS 23 ltJg2 (23 i bS ltJeS
24 xa5 c8-+) 23 ... ltJe5 24 ltJel c8
25 meas h3 26 fl h6 . White's
extra pawn is a small consolation
here, Anhchimeg-Rybenko, Ulaan
baatar 20 0 2 .
2 0 b4? ! axb4 2 1 cxb4 i s prema
ture if Black's knight can occupy d4:
21. . . ltJd4.
2 0 f3 prevents ... e4, but leaves
the e3-knight without support. Usu
ally Black uses that to gain control
of dS: 20 . . . if7 21 hl he3 22 ltJxe3
hc4 23 xc4 ltJe7=.
White can get a similar position
with the pawn on b2 if on move 16 he
defended the pawn with his rook:
1 6 a2 h8 1 7 0-0 f5 1 8
exf5 i.xf5 1 9 lll c e3 .ig6
65
Part 3
White's rook is more passive
here as it is charged with the de
fence of the b2-pawn. On the other
hand, White has the possibility of:
20 'Wa4 'Wc8 21 gd1 (21 bS
.ie8)
We follow our general plan with
2 1 e4 22 b3 f7 23 'Wa3
'Wd7 24 lll f 1 lll e5
18 h4 hh4, here the rook i s already

on the second rank and White can
open a passage to the critical h-file
with a tempo (b2-b4 ! ) at the right
moment.
1 8 lll x e3 lll e7
Black has a strong initiative. See

game 16 Socko-Krasenkow,
Plock 20 0 0 .
I n these examples White allowed
.. .f5 and Black obtained counterplay
on the kingside.
Now we are going to examine
the restrictive approach with l2Jc2e3. One small detail will define our
reaction - the position of White's b
p awn. If White plays 16 b3, we pre
p are .. .f5 with 16 . . . g6. It is analysed
in the next part. Now let us focus
on
1 6 ga2 @ h 8 1 7 lll c e 3 xe3!
When the rook is o n a2 and
White has not castled, we must
forget about ... g6.
The reason is that 17 . . . g6 18 h4 !
is very unpleasant. In comparison
to the line 16 b3 ci>h8 17 l2Jce3 g6 !
66
Black solved the problem of the

bad bishop on gS, now he only has to
activate his f8-rook by pushing .. .f5.
Basically, the game is balanced. In
the ensuing middlegame White can
choose to stay either with two minor
pieces, or trade knights.
Wh ite keeps 1 m in or piece
Th is position arises after 19 0-0

fS 20 exfS ltJxf5 2 1 l2Jd5 l2Je7 2 2 l2Jxe7
V!ixe7.
Black has sufficient counterplay
in the centre :
a) 23 b3 V!ic7 24 d2 (24 dS
b7) 24 . . . b7 or 24 . . . .id7= ;
b) 23 dS a4 ! ? 24 xa4 ! (24 dl
Wih4 could be dangerous for White:
9 l2Jd5 e7 10 hf6 hf6 11 c3 0-0

25 g3 h3 26 fl? ! f5 27 d2
d7+) 24 .. . xb2 25 b4 c2 = .
You can see another version of
the same typical position in game
15 Carlsen-Van Wely, Wijk aan
Zee 20 06 . It also shows that Black
easily holds the balance with only
heavy pieces left on the board.
This position arises after 19 b3 ! ?

f5 2 0 exf5 l2Jxf5 2 1 l2Jd5 b7.
With his 19th move White de
fended his b-pawn, thus render
ing 21. . . l2Je7 impossible. Still, Black
has sufficient counterplay on the
kingside, see game 14 Karjakin
Topalov, Wijk aan Zee 20 0 6 .
Wh ite keeps 2 m i nor pieces
Practical advice:
In the Positional variation Black
should try to keep all his three mi
nor pieces, in order to retain chan
ces for attack. If that is not possi
ble, he must seek further exchanges
and play for equalising. Commonly,
he has no problems with 1 or 0 mi
nor pieces.
67
Part 3
1 e4 c5 2 tilf3 tilc6 3 d4 cxd4

4 tilxd4 f6 5 tilc3 e5 6 tildb5 d6
7 .ig5 a6 8 tila 3 b5 9 tild5 .ie7
1 0 ,lxt6 .b.f6 1 1 c3 0-0
STEP BY STEP
Note the move order! W e first cas

tle, to follow up with . . . igS . In the
other lines Wh ite enjoys a small,
but persistent advantage. Howev
er, 1 1 . . . gS 12 lt:Jc2 lt:Je7 is a very so
lid continuation where Black has
all the chances to level the game.
We recommend it as a backup line
in case something goes wrong with
our main repertoire. It is aimed at
avoiding the sharpest lines that are
the subject of the next part of the
book. That comes at a price, though.
Black's winning chances are virtual
ly nonexistent. See game 17 Leko
Carlsen, Linares 0 3 . 0 3 .2008 which
provides enough up-to-date theory
on this topic.
1 2 c2
This is the most flexible and con
sistent variation. White bolsters up
the dS-square and delays castling.
Thus he keeps open sharp options
like h4 . Occasionally, White's bi
shop goes to h 3 .
W e c a n better understand the
importance of precise move order
on the example of the natural-look
ing 12 ie2 gS 1 3 lt:Jc2 ie6 14 0-0
68
lt:Je7 and White is unable to hold

dS, for instance, 15 lt:Jcb4 aS! 16
lt:Jxe7+ V!Jxe7 1 7 lt:J dS (17 lt:Jc6? Wb7
18 V!Jxd6 ga6 1 9 xeS f6 2 0 V!JxbS
gb6 2 1 lt:JxaS fubS 22 lt:Jxb7 gxb7+)
17 . . . b7+.
1 2 d 3 is another innocuous
move. Black can choose a typical
Sveshnikov setup : 12 . . . e6 13 0-0
g b8 14 lt:Jc2 as 15 V!Je2 b4 16 c4
gS= , a s in game 11 Ivanchuk-El
janov, Moscow 2 0 05 , where Black
keeps the rook on f8 in order to help
.. .f 5. Or he might prefer more all-Si
cilian methods like pure queenside
play: 12 . . . gS 13 ic2 (13 h4? ! h6
14 g4 f4 15 lt:Jxf4 exf4 16 lt:Jc2 dS !
17 exdS ge8+ 1 8 Ml lt:JeS 19 ie2
ib7 20 lt:Jb4 aS+) 13 . . . gb8 14 d3
e6 15 gd1 d7 16 0-0 gfc8 17 b4
Arnason-Vukic, Bela Crkva 1 98 3 ,
a n d here 17 . . . lt:Je7! 18 lt:Jxe7+ xe7
19 b3 gc6+ would have been excel
lent for Black.
12
.igS
In the 1970s, Sveshnikov played

both the text and 12 . . . gb8 , which is
meant to prevent a2 -a4. In the lat
ter case, however, White can an-
9 tlJ dS e7 10 ixf6 ixf6 11 c3 0-0

swer 13 h4, restricting Black's bish
op. We prefer to have active pieces,
even at a price.
1 3 a4
The most principled move. We'll
also mention:
a) 13 4Jce3 ixe3 14 4Jxe3 4Je7 lS
e2 (lS a4 b7 16 axbS axbS 17 gxa8
ixa8 18 f3 b6 19 d2 fS ! +) 1S . . . b7
16 f3 . Black is able to hold this po
sition with natural moves, but the
temporary pawn sacrifice 16 . . . dS !
seems best: 17 exdS (17 ttJxdS? !
ttJxdS 18 exdS e4 19 e2 gs 2 0 0-0
ixdS - is fine for Black) 17 ... d6 18
g4 (18 b3 fS ! ) 18 . . . ad8 19 d3
d7! 20 0-0-0 gfd8 21 ttJfs (21
d2 ? ! f6 ! 22 e4 ttJxdS 23 ttJxdS
ixdS 24 ixdS ds 2S xdS gxdS
26 gxdS hS 27 gxhS xf2+) 21. . .f6
22 4Jxe7+ xe7= , Gaprindashvili
Timoshchenko, USSR 1977;
b) 13 d3 is inconsistent as
White loses his grip on dS: 13 . . . e6
(13 . . . 4Je7 14 4Jcb4 aS lS 4Jxe7+ xe7
16 ttJds b7 17 hs d8 18 gd1 e6
19 c2 b4 20 0-0 bxc3 21 bxc3 was
played in Gouliev-Shirov, rapid,
Venaco 200S, when 21. . . c8 ! 2 2
h 3 gb8 would have been great for
Black) 14 4Jce3 4Je7= ;
c) 13 e2 i s too humble a nd can

not aspire to the advantage. Black
fightsfor dS with 13 . . . e6 14 0-0 4Je7
lS 4Jcb4 aS 16 4Jxe7+ xe7 17 tlJdS
b7, 13 Dervishi-Krasenkow,
Ohrid 2001 or:
13 . . . ttJe 7 14 ttJc b4 as lS ttJxe 7 +
xe7 16 ttJds b7 17 d3 gb8 18
0-0 e6 19 gfdl gfc8
White has no active plan, see the

model game 12 Almasi-Topalov,
rapid, Monte Carlo 2001;
d ) 13 g 3 This i s a purely defen
sive setup. Black easily gets a com
fortable game by following the same
development scheme as in the pre
vious examples :
13 . . . 4Je7 14 4Jce3 (Or 14 h4 h6 ;
14 4Jcb4? ! e6 lS g2 a s 16 4Jxe7+
xe7 17 tlJdS b7+ Xie Jun-Ga
liamova, Kazan/Shenyang 1999)
14 ... gb8 (14 ... ixe3 lS 4Jxe3 b7 16
g2 fS is an interesting choice for
courageous players: 17 exfS ixg2
18 tlJxg2 ttJxfS 19 dS + cj{h8 20 0-0
b4 ! ? 2 1 cxb4 b6 2 2 a 3 aS 23 bxaS
gxaS 24 e4 tlJd4oo) lS g2 aS 16
0-0 (16 a3 e6 17 4Jxe7+? ! xe7 18
tlJfS d7! 19 h4 f6 20 4Je3 b4 21
cxb4 axb4 2 2 a 4 d8 ! t Bartel-Rad
j abov, FIDE-Web k.o . Tripoli, 2 0 04)
16 . . . ixe3 17 4Jxe3 e6 18 d3 b6
69
Part 3
19 fdl fd8 2 0 d2
In such positions Black expands

on the queenside by 20 . . . b4 21 c4
ttJc6 with a good game .
e) 13 h4 .ih6 14 g4
An agressive plan, successfuly
used lately by GM Andrei Volokitin.
However, Black has no weakness
es on the kingside, so he should not
fear a direct attack.
14 . . .f4 15 f3
15 gS b7! ? 16 f3 tlJe7 17 tlJxf4
exf4 1 8 0-0-0 tlJg6 19 hS tlJeS 2 0 xf4
fS gave Black strong counterplay in
Safar Zadeh-Agamaliev, Fajr 1997.
15 . . .e6
wins a pawn, but White's numer

ous weaknesses assure Black of a
good game, for instance: 18 . . . b4
(18 . . . c7 ! ?) 19 c4 (19 cxb4 b8 2 0
a 3 dS ! ; 1 9 .ie2 bxc3 2 0 bxc3 as 2 1
0 - 0 xc3+, Bierwisch-Siegmund ,
Obertsdorf 2 0 03) 19 . . . aS ! 20 b3
a3 21 hS b2 22 dl f6 ! f! with
rich counterplay for Black.
16 . . . b8 17 gS e8
Black also has other appealing
options , like 17. . . d7.
18 ttJxf4 exf4 19 xf4 b4
White can still maintain the bal
ance here:
2 0 0-0 bxc3 2 1 bxc3 ttJes 2 2
ha6 as 23 .ie2 xc3 24 ttJe3 b2
25 tlJd l d2 = or:
20 cxb4 ttJxb4 21 ttJxb4 b4 22
b3 (22 d 2 b6 23 b3 dSt) 2 2 ... dS= .
1 3 ... bxa4
This capture is a must in the
Positional variation. Black uses the
b-file for counterplay while our a
pawn could be defended by the
standard manoeuvre .id8 . In, con
trast , after 13 . . . b8 the bS-pawn
would have been an easy target.
1 4 gxa4 as
16 .id3
Alternatively: 16 tlJxf4 exf4
(16 . . . f6 17 gS xf4 18 xf4 exf4
19 0-0-0 fd8= leads to an equal
endgame , but the text move is more
ambitious .) 17 xf4 tlJeS 18 ttJe3
70
9 l'i:JdS e7 10 hf6 hf6 11 c 3 0-0

Main branches are:
A. 15 bS
page 71
B. 15 c4
page 73
15 b4 is also seen, but it neglects
development and can only make
us happy: lS . . . e6 and then White
can close the queenside or maintain
tensio n:
a) 16 bS
Now Black has at least three
good options:
16 ... l'i:Je7 is proposed by Svesh
nikov: 17 l'i:Jce3 xe3 18 l'i:Jxe3 V!ic7
19 c4 l'i:Jc8 ! 20 al l'i:Jb6 21 e2 fc8
2 2 Wid2 WicS ! + and Black is fine;
16 . . . xdS ! ? is a concrete way to
use White's lag in development: 17
WixdS (or 17 exdS l'i:Jb8 18 l'i:Jb4 l'i:Jd7
19 l'i:Jc6 b6 and the a-pawn will
soon be marching forward) 17 . . . l'i:Je7
18 V!ib3 (18 Widl dSt, Wittke-Hoyer,
Germany 1988) 18 . . . dS 19 e2 dxe4
20 0-0 V!idS ! 2 l c4 cs 22 el l'i:Jc8 !
with a fine game ;
16 . . . l'i:Jb8 is recommended by
Yakovich and seems very logical.
The knight is relocated to an ex
cellent outpost on cS : 17 l'i:Jcb4 (17
e2 l'i:Jd7) 17 . . . Wic8 ! 18 e2 axb4 !
19 a8 bxc3 . Black had more than
enough compensation in Yudasin
Semeniuk, Saratov 1981.
b) 16 c4 axb4 17 l'i:Jcxb4
17 xa8 runs into nice tactical
blows after 17 . . . Wixa8 18 l'i:Jcxb4?
l'i:Jxb4 19 cxb4 c8 - + ; 18 l'i:Jc7 bxc3 !
19 l'i:Jxa8 (19 xe6 d2 + 20 fl
aS - + ; 19 0-0 b8 2 0 l'i:Jxe6 fxe6
21 he6 + h8+) 19 . . . d2 + 2 0
Wixd2 ( 2 0 fl xc4+ 21 gl xa8+; 20 e2 xc4+ 21 f3 b3 ! 22
g3 l'i:Jd4- +) 2 0 . . . cxd 2 + 21 xd2

xa8+; 18 h4 d8 ! 19 cxb4 l'i:Jxb4 !
20 l'i:Jcxb4 as with an advantage;
18 cxb4 l'i:Jxb4 ! 19 l'i:J cxb4 c8 20 h4
(20 0-0 xc4 21 l'i:Jb6 b7 22 l'i:Jxc4
xc4 23 l'i:JdS hfl 24 xfl+) 2 0 h4
xc4 21 hxgS xb4 22 l'i:Jxb4 V!ixe4 +
23 l xb4+.
17 . . . l'i:Jxb4 18 cxb4 xa4 19 xa4
Wic8 2 0 b3 xdS 2 1 exdS g4 2 2
0 - 0 c8 Black has the more active
pieces.
A. 1 5 .ib5
This move could be explained
only with White's wish to avoid the
line 15 c4 d7, which is however
quite passive, as demonstrated by
the recent game 18 Shirov-Topa
lov, Morelia/Linares, 19.02.2008.
1 5 e7
16 x e7+
White's only hope to gain some
advantage is connected with intro
ducing a knight on dS. However,
this is impossible:
a) 16 l'i:Jce3 xe3 17 l'i:Jxe3 b6
18 d3 b8 19 c4 fS 20 0-0 fxe4
21 xe4 fS ! 22 l'i:JxfS l'i:JxfS 23 c6
l'i:Jd4 24 Wixb6 xb6= ;
71
Part 3
b) 16 l2Jcb4 also does not work in
view of 16 . . . ih3 .
16 . . . id7 is a well known way to
equalise immediately: 17 l2Jxe7 + (17
hd7? ! axb4 18 xa8 xa8 19 0-0
ltJxdS+) 17 ... he7 18 l2Jc6 (18 hd7
axb4 19 ic6 xa4 20 xa4 bxc3 2 1
bxc3 b8 2 2 0 - 0 id8 = , Svidler
Ivanchuk, Polanica Zdroj 2000)
18 . . . e 8 19 ds ie6 2 0 d3 id7= .
The text is more straightforward.
17 l2Jxe7+
17 gxh3 axb4 18 l2Jxb4 xa4 19
ha4 fS ! (Leko) is dubious since the
white king is rather shaky.
17 . . . xe7 and play transposes to
16 l2J e7 e7 17 l2Jb4 ih3 = .
c ) 1 6 0 - 0 ltJxdS 1 7 xdS ie6 18
d3 b6 = prepares counterplay
down the f-file with .. .f5, for in
stance, 19 c4? ! f5 20 l2Je3 fxe4 21
xe4 a7 2 2 ltJdS (22 fal? af7 ! )
2 2 . . . cS. Perhaps White should
prefer 19 l2Je 3 , but it is clear that the
position after 19 . . .he3 2 0 fxe3 can
not be a problem for Black.
d) Finally, 16 c4 d7 17 a2 c8
18 d3 ltJxdS 19 hdS a4= leaves
Black well developed and with good
prospects.
1 6 ...Wfxe7 1 7 b4
After 17 0-0 b7 18 d3 (18
e2 ie6 19 c4 fS is fine for Black)
18 . . . ie6 19 c4 d8 Black successful
ly redeployed his pieces in Smyslov
Sveshnikov, Leningrad 1977.
1 7 ... i.h3
17 . . .g4 leaves Black fewer winning chances: 18 al (18 ltJdS hdl
19 l2Jxe7 + he7 20 @xdl ab8 21
c4 d8 =) 18 . . . b7 (1 8 . . . c7 is also
72
enough to keep the balance) 19 c6

b6 2 0 h3 (20 ha8 bS ! 21 f3 axb4
22 fxg4 d3 23 dl e3 + 24 e2
cl+ leads to a draw by perpetual
check) 20 . . . ab8 (Rogozenko gives
2 0 . . . axb4 2 1 xa8 xc6 22 f8 +
@xf8 23 hxg4 xe4+ 24 Ml b3, but
the text move is more enterprising)
21 xaS ih4 2 2 0-0 (22 hxg4 xf2 +
23 @dl igS 24 ds fc8t looks
dangerous for White) 22 . . . e2 23
g3 hfl 24 @xfl d8oo. We have
reached an unbalanced position,
which needs further analysis. We
have even explored 24 . . . hg3 ! ? 2 5
fxg3 fS 2 6 exfS fS + 27 @e2 f2 +
2 8 @d3 e4+ 29 he4 xg3+ 30 @c2
f2 + 31 @b3 xb4+ 32 cxb4 e3 +
with equality.
1 8 d5
Aft e r 1 8 gxh3 axb4 19 b4 g6
White's king will never find a safe
haven.
18 c6 ac8 19 xaS hg2 2 0
gl ih3 21 hS looks i n White's fa
vour, but 21. . . h4 2 2 ltJdS d8 2 3
h6 g 6 2 4 a6 e6 allows Black t o
consolidate.
1 8 %Yb7 1 9 .ic4
19 gxh3? ! xbS 20 gl id8 !
is better for Black: 21 b3 (21 c4? !
9 ltJdS e7 10 hf6 hf6 11 c3 0-0

xb2+ , Gouliev-Cheparinov, Lina
res 2002) 21. . . b8 22 b4 c>h8 ;
19 d3 ? ! e6 20 0-0 hdS leads
to a typical Sveshnikov position with
opposite coloured bishops. With a
white pawn on dS Black does not
have any problems. In Luj an-Spas
ov, Santa Cruz de la Palma 2005, he
even was better after 21 exdS ab8
22 xaS d8 23 c6 xb2 24 bs
xbS 25 hbS b6+;
19 c4 cuts off White's bishop
from the kingside: 19 . . . e6 20 0-0
fS 21 exfS fS=.
19
J.d7 20 ga2
In this position Black has tried so

far 20 . . . a4 21 0-0 e6 2 2 d3 c>h8
23 ltJb4 ! t and 20 . . . c>h8 21 0-0 fS 2 2
exfS, when 2 2 . . . fS ! ? i s worth consideration. (22 . . .MS 23 ltJe3 he3
24 dS hf2 + 25 c>hl is an intro
duction to a long farced variation
which is slightly better for White)
We propose:
20 ... gfcS! 21 d3 c6
Now White is farced to compro
mise his pawn structure:
22 b3 a4 23 0-0 axb3 24
gxa8 gxa8 25 ixb3 c5=.
a.
1 5 .ic4 gba
We'll focus mostly o n:

Bl . 16 a2
B 2 . 16 b3
16 b4 is a premature advance
which does not disturb Black:
16 . . . d7 17 a3 axb4 18 cxb4 !e6= ;
1 6 al shifts the queen away
fram the centre and the kingside.
Then the thematic counterplay with
16 . . . c>h8 17 0-0 fS is even more ef
ficient: 18 exf5 hf5 19 ttJce3 g6 .
Black is accumulating striking force
against the enemy king by the help
of the standard e5-e4, ltJc6-e5 with
good prospects, for example : 2 0
dl e 4 2 1 fl e8 2 2 d2 !f7 ! and
Black is fine.
8 1 . 1 6 ga2 @h8
It i s clear that Black must open
the f-file by .. .f5. The big question is
whether to prepare it with . . . g6, or
push it right away.
We prefer to refrain from
weakening the king and the
seventh rank with ... g6.
It is better to stake on dynamical
factors and put pressure on f2 . No
tice that . . . g6 is riskier while White
73
Part 3
had not castled, since he could at
tack it with h4 ! ?
Bla . 1 7 lt:J ce3
Blb . 17 0-0
Without ... g6, 17 h4 would only
weaken White's kingside. The game
Asrian-Khalifman, Bled ol. 200 2
went 17 . . . h6 18 lt:Jce3 he3 19
lt:Jxe3 lt:J e7 20 b3 f S 21 exfS ltJxfS 2 2
lt:JxfS hfS 2 3 0 - 0 e4 2 4 ds hdS
2S xdS xh4 26 xaS f6 = .
Bla. 17 ce3 he3 ! 1 8 xe3
e7
Black solved the problem of the

bad bishop on gS, now he only has
to activate his f8-rook. Basical
ly, the game is balanced, but some
small nuances could be able to em
bitter his life.
19 b3
Practice has shown that Black
has an easy game if he exchang
es any of his minor pieces. After 19
0-0 fS 2 0 exfS ltJxfS (20 . . . MS 21 b3
maintains the tension in White's fa
vour) White cannot prevent the ex
change of the knights due to 21 ltJdS
ltJe7 22 ltJxe7 xe7
74
Black has sufficient counterplay

against White's king and in the cen
tre:
a) 23 b3 c7 24 d2 (24 ds
b7) 24 . . .b7 (24 . . .d7 !?) 2S fal
2S . . . c6 (2S . . . dS leads to a drawn
rook endgame: 26 hdS hdS 27
xdS xc3 2 8 xaS b3 2 9 xeS
xeS 30 xeS b2) 26 fl f4 ! , in
tending . . . g4.
b) 2 3 ds a4 ! ? (or 23 . . .b7 24
d2 d7 2s dS hdS 26 xds bs
27 d3 c6 =) 24 xa4 ! (24 dl h4
could be dangerous for White : 2 S
g 3 h3 26 fl? ! fS 2 7 d2 d7+)
24 . . . xb2 2S b4 c2 = .
Another version o f the same
typical position arises after 2 1 ltJxfS
MS . See game 15 Carlsen-Van
W ely, Wijk aan Zee 20 0 6 .
19 . . .f5 20 exf5 xf5 2 1 d5
With his 19th move White de
fended his b-pawn, thus rendering
2 1 . . . lt:Je7 impossible. Still, 21 ... ib7
should ensure Black counterplay,
see game 14 Karjakin-Topalov,
Wijk aan Zee 200 6 .
Note that the placement of
White's pawn on b3 has its draw
backs, too, as it allows the break
. . . a4 at an opportunity.
9 ll:JdS ii.e7 10 hf6 ii.xf6 11 c3 0-0

Blb. 17 0 - 0 f5 18 exf5 .hf5
19 ltJce3
19 e2 d7 20 dl e4 21 ttJde3
ll:JeS ! is good for Black.
19 i.g6
. . .
xe4+) 24 ... axb4 25 ll:Jxb4= .

22 ii.a6 ! (22 cxb4 ii.xe3 23 fxe3
1 + 24 ix:fl W 25 d2 ii.xdS 26
xdS ll:Jxb4 27 bs bs 2 8 xbS
ll:Jc2+) 22 . . . d7 23 ii.bS (23 ll:Jxb4
ii.e8=) 23 . . . ii.e8 24 ll:Jxb4 (24 cxb4
b7) 24 . . . t7 25 ll:JbdS b7= .
21 e4
2 0 a4
20 f3 hinders the plan with 20 . . .
e 4 (in view o f 2 1 f4), but it weakens
the gl-a7 diagonal. Black uses that
immediately by attacking the cen
tre: 20 . . . ll:Je7 21 e2 c8 22 b3 ii.t7
23 dl ii.xe3+ 24 ll:Jxe3 (24 xe3
ii.xdS 25 hdS ll:Jf5, fallowed by
b6 and ll:Je3, is completely equal)
24 . . . b6 25 @hl dS= , Dominguez
Ramirez, Guayaquil 2003.
20 c S 21 gd1
It is easy to understand White's
wish to reinforce his control over dS .
For example, after 21 ii.e2 ii.xe3 ! ? 2 2
ll:Jxe3 (22 fxe3 xfl+ 2 3 hfl ii.bl !
24 a l xb2) 22 . . . f4 23 a3 ii.t7
24 aal dS+ Black's centre becomes
mobile .
Instead, Papadopoulos played
against Kolev in Kavala 2 007 the
novelty 2 1 b4, which leads to a bar
ren position: 21. .. axb4
Or 21. . . ii.xe3 22 fxe3 xfl + 23
ii.xfl ii. t7 24 d2 (24 e4? ! axb4 25
ii. a6 d7 26 ii.bS g4 ! 27 ii.xc6
The essence of Black's plan is to

put the knight to eS, even at the cost
of the aS-pawn. The threats against
the enemy king should compensate
the small material deficit.
22 b3 .if7 23 a3 d7 24 ltJfl
ltJe5
Black has a strong initiative, 16
Socko-Krasenkow, Plock 20 0 0 .
8 2 . 1 6 b 3 h 8
75
Part 3
has two major options: 17 0-0 and
17 l2J ce3 (it is the subject of the next
part of the book) .
They result in different pawn
structures since in the latter case
Black has to prepare . . .fS by . . . g6 .
Minor alternatives are:
a) 17 V9e2
White takes control of e4 and
prepares to expand on the kingside.
Black obtains good play by simple
and logical moves:
17 .. .fS 18 h4 !f6
The fine point of White's 17th
move is that 18 . . . !h6 is bad in view
of 19 exfS hf5 2 0 g4 hc2 2 1 V9xc2
!f4 2 2 V9e4 V9d7 23 !d3+ with a ter
rible battery on the b l-h7 diagonal,
Morais-Rodrigues, Gaia 2004.
19 exfS MS 20 l2Jce3 !d7
21 V9c2
21 !d3 l2J e7 22 l2Jxe7 V9xe7 23
ltJdS V9f7 24 V9e4? ! is purposeful, but
24 . . . g6 25 hS !gS underlines the fact
that White's king is helpless in the
centre . The best White can do is to
play a pa wnless endgame after 26 f4
!fS 27 hxg6 V9xg6 2 8 fxgS V9xgS 2 9
V9h4 V9xh4+ 3 0 hxh4 !xd3. That's
why Polgar preferred 24 xaS !d8
25 a7 V9xdS 26 xd7 xb3 with to76
tal elimination in Polgar-Kramnik,

Wijk aan Zee 20 05.
2 1 . .. e4 ! ? 22 V9xe4 ltJeS 23 al (23
a3 e8oo Rogozenko) 23 ... e8 24
l2Jxf6 V9xf6 2 5 V9d4 (25 0-0? l2Jxc4
26 V9xc4 !bS+) 2S . . . !c6 . Black has
full compensation for the pawn.
Now 26 0-0 leads to a farced draw
after 26 . . . l2Jf3 + 27 gxf3 V9xf3 28 !dS
hdS 29 V9xdS xe3 30 V9xf3 xf3 = .
Charbonneau-Radjabov,
Calvia
ol. 20 04 saw 26 !dS b4 ! 27 V9d2
hdS 28 V9xdS, when 28 . .. xb3 !
would have leveled the game, for in
stance, 29 0-0 xc3 30 xaS l2Jc6=
(Rogozenko).
b) 17 h4 only provides Black
with a lever on the kingside after
17 . . . !h6 .
Now 18 g4 !f4 1 9 V9f3 ! e 6 20
l2Jce3 he3 ! 21 l2Jxe3 (21 fxe3 l2Je7! t)
2 1 . . . !xc4 22 xc4 l2Je7+ favours
Black.
18 l2Jce3 allows Black to get rid
of his bad bishop with 18 . . . he3 19
l2Jxe3 l2Je7 (20 hS h6 21 0-0 fS 2 2
exfS ltJxfS 23 fufS MS 24 V9dS !d7
2S a2 !g4=) 20 0-0 fS 21 exfS ltJxfS
2 2 ltJxfS hfS 23 V9dS !g6= , Onis
chuk-Filippov, Batumi 1999.
Finally, the tricky 18 V9e2 (hop
ing for 18 .. .fS? ! 19 exfS MS 20 g4)
should probably be answered care
fully with 18 . . . !d7 19 al a4 20 b4
(20 bxa4 b2 ! ?) 20 . . . l2Je7. In these
lines the pawn on h4 is only a cause
for concern to White.
1 7 0-0 f5 1 8 exf5
In 2005 Ivanchuk introduced
the manoeuvre 18 el fxe4 19 xe4
!fS 20 e2 with the obvious inten-
9 ltJd5 ie7 10 hf6 hf6 11 c3 0-0

tion to transfer it to a2 . However,
20 . . . ig4 would provoke a weakness
in White' s castling position, which
could be attacked later, for exam
ple, 21 f3 ih5 22 @hl ltJe7 23 ltJxe7
he7 24 Wal M3 .
18 f3 fxe4 19 fxe4 xfl+ 2 0 Wxfl
ltJe7! is completely equal because
the weak pawns of both sides, b3
and e4 versus aS and d6, counter
balance each other. In Inarkiev
Yakovich, Krasnodar 20 0 2 , Black
managed to exchange light-squared
bishops and even had a slightly bet
ter game: 21 Wd3 ltJxd5 22 hd5
ib7 ! .
1 8 xfS
d4: 21 ltJd4 ltJxd4 22 xd4 b7 23

Wel bf7 24 a2 h5? ! 25 ltJ e3 f4
2 6 d5+, Ivanchuk-Carlsen, Mos
cow 20 07.
21 ltJce3 e4 22 ib5, Ivanchuk
Kramnik, rapid, Monte Carlo 2005.
Here 2 2 . . . he3 ! 2 3 fxe3 (23 ltJxe3
ltJe5=) 23 . . . Wf5 ! would have been
good for Black.
1 9 ig 6 !
Black i s playing "around" White's
pieces. He does not aim to neutral
ise them, but rather build his own
play with . . . e4 and . . . ltJe5.
. . .
. . .
1 9 ce3
19 We2 i g6 does not change the
plans of the sides .
20 dl
White's game is not so uncloudy
as it may seem at first sight. In Pol
gar-Topalov, Wijk aan Zee 2003
Black got the initiative after 20 id3
ie8 2 1 a3 ltJe7 22 ie4 ltJxd5 23
hd5 Wb6+.
20 . . . Wc8
We see here an example where
20 . . . e4 turned premature since
White's knight was able to jump to
This position has a very good

reputation for Black. The point is
that the most natural move 2 0
d3 brings about mass exchan
ges : 2 0 . . . hd3 21 Wxd3 he3 2 2
fxe3 (22 ltJxe3? b3 23 Wc4 Wb6+)
22 . . . fl+ 23 Wxfl xb3 24 c4
b5D 25 e4 (or 25 ltJc7 b6 26 ltJ d5
b5=) 25 . . . c5 2 6 xc5 dxcS 27
Wa6 (27 Wb5 Wd6 = ; 27 Wf7 h6=)
27 ... ltJb8 2 8 Wa8 h6= .
White may attempt to retain
his king's rook by 2 0 E:el, but it
weakens the f2-square. Black un
derlines that by 20 . . . b7, intend
ing 21 . . . bf7. Then 21 ltJfl bf7 2 2
77
Part 3
a2 would be too passive as Black
gets time for 2 2 . . . e4 23 ltJg3 ih4 24
ttJe3 f4 2S d2 ttJeS 26 xd6 gs
with good compensation. So Anand
fallowed up with 21 !d3 hd3 2 2
xd3 he3 2 3 fxe3 ( 2 3 ttJxe3? b3
24 c4 b6+) 23 . . . xb3 24 c4 to
draw after 24 . . . b2 ! 2S xc6 gS
26 ttJf4 exf4 27 xf4 fb8 = , Anand
Kramnik, Wijk aan Zee 200S.
2 0 f3 !f7 21 @hl he3 2 2 ttJxe3
hc4 23 xc4 ttJe7 is also equal.
2 0 b4? ! axb4 21 cxb4 is prema
ture if Black's knight can occupy d 4:
2 1 . . . ltJd4 .
20 e2 ? ! aims t o win the aS
pawn, but this setup encourages
Black's attack with : 20 . . . e4 21 fal
(or 2 1 bs ttJ eS 2 2 fal !h4 23 g3
c8-+) 2 1 . . . 4 (loosening White's
castling position) 22 g3 !gS 23
ltJg2 (23 !bS ttJeS 24 xaS c8-+)
23 . . . ttJeS 24 ttJel c8 2S xaS h3
26 fl h6 . White's extra pawn is a
small consolation here, Anhchimeg
Rybenko, Ulaanbaatar 20 0 2 .
Now we are going t o consider yet
another redeployment of White's
pieces:
2 0 e2
The bishop shifts to f3, while the
a4-rook prevents . . . e4. Black must
reconsider his plans for attack in fa-
78
vour of pressure against b3 :

2 0 . . . !f7 ! 2 1 !f3
We know that 2 1 b4? ! should
be dubious because our knight will
land on d4.
More interesting is 21 ttJc4,
when we must take the chance to
play 2 1 . . . e4 ! , depriving ttJdS of sup
port through !f3 . Then:
22 f4 exf3 23 M3 ttJeS 24 ttJxeS
(24 ttJxaS? bS- +) 24 . . . dxeS= ;
2 2 b4 axb4 2 3 cxb4 ttJe7 (23 . . . hdS
24 xdS ttJxb4=) 24 ttJce3 ttJxdS 2S
ttJxdS e8 ! ? 26 bS eSf!;
22 @hl g8 23 a3. The han
ging b3-pawn is restricting White's
options. 23 . . . ttJe7! 24 ttJce3 (24
ttJxe7 he7 2S ttJxaS c7 26 b4 dSoo)
24 . . . ttJxdS 2S ltJxdS fS ! 26 c4 hdS
27 cxdS b6 t Leko-Gelfand, Po
lanica Zdroj 1998 .
2 1 . . . he3 2 2 fxe3 g8=, Stefans

son-Filippov, Chalkidiki 20 0 2 .
Part 3

4 lll x d4 lll f 6 s lll c 3 es 6 lll dbS d6
7 igS a6 8 lll a3 bS 9 lll d S ie7
1 0 hf6 hf6 1 1 c3 0-0
COMPLETE GAMES
1 1 lvanchuk - E ljanov
M oscow 200S
1 e4 cs 2 li:)f3 li:)c6 3 d4 cxd4
4 li:) xd4 li:) f6 s li:)c3 es 6 li:) d bS d6
7 ig S a6 8 li:) a3 bS 9 li:) d S ie7 1 O
ixf6 ixf6 1 1 c3 0-0 1 2 id 3 ie6
1 3 0-0 b8 1 4 li:)c2 a5
White refrained fram a4 and

did not pose any problems to Black
who is able to choose different set
ups. His last move is consistent, but
probably a bit early against id3 ,
for it helps White start play on the
queenside. Instead . . . ig5, followed
by . . . d7 and .. . fc8, is preferable.
1 S YMe2 !
After 1 5 a4 b4 1 6 cxb4 (16
ib5 hd5 17 exd5 ttJ a7 18 c4 e4)
16 . . . hd5 17 b5 ie6 18 bxc6 xb2 19
llJe3 White would be better, but15 . . .

bxa4 ! leads t o better versions o f the
main lines .
1 S . . . b4
We prefer 15 . . . hd5 16 exd5
llJe7=, favourably changing the
pawn structure.
1 6 ic4 igS 1 7 cxb4 axb4
17 . . . llJxb4 18 llJcxb4 axb4 is so
lid, but dull. The worse Black could
face, is a position with opposite col
oured bishops. It is true that it is
generally more pleasant for White,
but Black's dark-squared bishop
can easily be transferred to b6 to
hold the draw. Compare such posi
tion with game 17 Leko-Carlsen,
where Black's bishop is idle on h6,
and still he was not lost at all .
1 8 fd 1 YMaS? !
Elj anov follows a wrong strate
gy, playing on the queenside, where
White is stronger. He should have
remembered that in the Sveshni
kov Black's play is connected with
. . .f5. Instead of the queen's sortie,
18 . . . cj/h8 19 ib5 hd5 20 xd5 llJe7
21 ddl f5 seems fine.
1 9 a3! bx a3 20 b4 YMd8 21 bS
li:)aS 2 2 ia2
79
Part 3
1 2 Alm a s i - To p alov
M o nte C a rlo, ra p id, 2001
1 e4 c5 2 tll t3 tll c 6 3 d4 cxd4
4 tll x d4 tll t6 5 tll c3 e5 6 tll db5 d6
7 i g5 a6 8 tll a 3 b5 9 tll d 5 ie7 1 O
ixt6 ixt6 1 1 c3 ig5 1 2 tll c2 0-0
1 3 ie2 tll e 7 1 4 tll cb4 a5 1 5 tll x e7+
VNxe7 1 6 tll d 5 '!Nb7
2 2 ... @h8
White has considerably im
proved his position during the last
few moves. He has made a passer
and needs only 2-3 tempi to con
solidate and rearange his minor
pieces . Black realised that and de
cided to switch to the tested plan
with . . .fS. Perhaps he could main
tain the balance attacking the ex
tended White pawn, for example,
22 . . . \Wd7 23 llJxa3 fc8 24 b6 !d8
25 tt:Jbs c>h8 26 llJbc7 !xdS 27 llJxdS
hb6 28 llJxb6 xb6 29 ht/ \Wxf7
3 0 xaS= , or 2 2 . . . Wc8 2 3 llJxa3 g4
24 f3 d7 25 c>hl \Wes 26 b6 d8 ,
but here 2 7 dcl \Wxa3 2 8 llJf6 + gxf6
2 9 hf?+ f7 30 xa3 might turn
in White's favour.
23 tll x a3 t5 24 ext5 Axt5
Now 25 llJbl ! \Wd7 26 llJbc3 d8
would be pleasant for White in view
of the clumsy position of the aS
knigh t. I nstead Ivanchuk thrusts
his passed pawn forward. . . to lose it
in few moves.
25 b6? ! tll c6 2 6 tll b5 tll e7 27
tll bc7 tll c 8 28 b1 tll x b6 29 Axt5
tll x d5 30 g xd5 VNxc7
%-%
The extra pawn is worthless.
80
This is a model position for Black

in the cases when White refrains
from a4. The dS-knight looks glori
ous, but in fact it is rather useless
as it has no targets. White's bishop
is not any better. Black has active
plans on the queenside, connect
ed with . . . b4, or in the centre. ( . . .fS)
They ensure him good counterplay.
1 1 VNd3 gb8
17 . . .b4 is a fair alternative, but it
allows White to close the centre with
18 c4. (18 cxb4 axb4 leaves Black
more chances. In practice Black of
ten emerged with some initiative,
for example : 19 \Wb3 e6 20 c4
ac8+ Anand-Kramnik, Dortmund
1997 or 19 \Wg3 h6 20 0-0 c>h8 2 1
b3 b8 2 2 adl fSf Almasi-Shi
rov, Cannes 2005) 18 . . . e6 19 dl
a4 20 0-0 aS= .
1 8 0 - 0 e6 1 9 gtd 1
I n a later game Almasi tested
9 llJdS e7 10 hf6 hf6 11 c3 0-0

adl, but soon discovered, that the
best setup of his rooks is on cl and
dl.
1 9 . . . fc8 20 a3
White must be careful not to give
up the c-file as in the game Tivia
kov-Van Wely, Leeuwarden 2003
which went 20 g3 h6 2 1 b3 cs 2 2
c 4 hdS 23 rocdS xdS 2 4 cxdS c8
2s d3 cs+.
20 . . . h6 21 g3 .id8
This move anticipates possi

ble b4 in case Black played . . . cs,
when Black should not capture,
but retreat to c8 or c6. It also re
locates the dark-squared bishop to
its best place, b6. Almasi regularly
plays this position, although with
out great success. Against Peter
Heine Nielsen in 20 04, he preferred
22 h4 to restrict Black' bishop . We
think that the same 2 2 . . . a7, as in
the current game, would be the best
answer, for instance, 23 d2 b6=.
22 g d 2 V!fa7 23 g 2 gc5
Now Black seizes the initiative,
because b4 is impossible and the
positional threat of 24 . . . hdS forc
es White to retreat the knight to e3
under a pin fram gS.
24 e 3 .ig5 2 5 h4 .ixe3 26
V!fxe3 V!f c7 27 g ad1 gc6 28 h 2

.i b 3 29 g g 1 .ic4 3 0 g g d 1 .ixe2 3 1
V!fxe2 b 4
Finally we can talk about a slight
edge for Black, because he has two
target pawns against only one in his
camp .
32 axb4 axb4 33 gd3
White would have had more
chances to survive after 33 cxb4.
Now his rooks are too passive.
33 . . . bxc3 34 bxc3 gb3 35 V!f d2
V!fb6 36 @ g 2 gb2 37 V!f e3 V!f xe3 38
gxe3 @ta 39 ged3 @e7 40 g9 3
gc2 41 gdd3
Ironically, in Linares 2008 To
palov lost the same pawn struc
ture, but this time he had the pas
sive rooks . Obviously, humans face
great difficulties defending 4 rooks
endings with passive pieces.
41 . . . h5 42 @f3 g6 43 @ g 2 e 6
4 4 @f 1 gc4 45 e 1 f5
46 f3 g g 2 4 7 exf5+ gxf5 48 f4
e4 49 gd2 g g 1 + 50 @f2 g c 1 5 1 ga2
g4 xc3 52 gxc3 gxc3 53 gas d 5 54
gea+ d 6 5 5 g95 d4 56 gxf5 gc2+
57 e 1 e 3 58 g9 5 g g 2 59 @f1 gxg3
60 @e2 g g 2+ 6 1 @f3 gd2 62 gea
d 5 63 g95+ c4 64 f 5 gf2+ 65
@g3 d3
0-1
81
Part 3
1 3 D erv i sh i - Krasen kow
E U - ch Ohrid 2 0 0 1
1 e 4 c s 2 f3 c6 3 d 4 cxd4
4 xd4 f6 S c3 e s 6 d b S d6
7 ig S a 6 S a3 b S 9 dS ie7 1 0
ixf6 ixf6 1 1 c 3 0-0 1 2 c2 i gS
1 3 ie2 ie6 1 4 0-0 e7 1 S cb4
as 1 6 xe7+ \Wxe7 1 7 dS \Wb7 1 S
\Wd3 ixd S 1 9 \Wxd S \Wxd S 20 exdS
g abs
In general, this structure is i n

Black's favour. His plan i s simple:
to neutralise possible White's at
tempts on the queenside, (meeting
a4 with . . . b4) and centralising the
king. The usually bad dark-square
bishop, here is clearly superior to
its enemy counterpart.
21 g3 fS 22 a3 g6 23 g 2 gfcS
24 gfd 1 g7 2S h4 if6 26 g d 2 e4
27 f3 exf3 + 2S xf3 hS 29 g2
ieS 30 gf1 f6 3 1 gf3 e7 32 if1
d 7 33 ie2 c7
Krasenkow decided to use the
king as a defender of the bS-pawn.
At the same time it is close to the
centre and can enter into play
through cS.
34 b3 b6 3S c4 bxc4 36 ixc4
a4 37 gc2 a xb 3 3S gxb3+ as 39
gxbS gxbS 40 ie2 gb3 41 if3
82
White is lost, because his bishop is

tragically passive and g3 is an eter
nal weakness. It is not so important
whether Black can win against best
defence, or not. In practice, White
can withstand 20 or 30 moves, but
eventually he is likely to miss some
tactic and lose. For his part, Black
could be squeezing as long as he
likes, waiting for a mistake.
41 . . .g xa3 42 ge2 b4 43 gc2
g d 3 44 f2 b3 4S ge2 gc3 46
g2 c4 47 ga2 d3 4S f2
id4+ 49 g 2 ie3 SO ge2 d4 S 1
g a 2 e s S 2 g b 2 g c 1 S 3 ga2 g g 1 +
S4 h 2 f4 S S gxf4+ xf4 S6 ig2
g d 1 S7 g a4+ id4 SS ga3 ieS S9
h3 t S 60 gf3 + if4 6 1 ga3 t6
62 gb3 g d 4 63 if3 gci2 64 ig2
i eS 6S g f3+ g7 66 ge3 h7 67
ga3 if6 6 S i e 4 g d 4 69 ge3 g b4
70 g 3 g7 7 1 h 3 h 6 72 g 3
i d 4 73 ge2 i e S + 7 4 g 2 if6 7 S
g 3 g b 3 + 7 6 if3 ixh4+ 7 7 x h 4
gxf3 7 S ge6 g f4+ 7 9 h3 g s SO
gxd6 g d 4 s1 g d s h4 s2 d6 g d 3 +
S3 h 2 g4 S 4 d 7 g S SS g2 h3+
S6 t2 h2 S7 ghs gd2+ ss e 1
g x d 7 S9 g x h2 g3 90 ga2 gf7 9 1
g a s g 4 9 2 g a 3 + h2 93 g a 2 + h 3
94 g a s g 3 9 S g h s+ g 2 96 e 2
g 1 97 g a s g h 7
0-1
9 ld dS e7 10 hf6 hf6 11 c3 0-0

1 4 Karjakin - Topalov
W ij k aan Zee 24. 0 1 . 200 6
4 tll xd4 tll f6 s tll c3 es 6 tll db S d6
7 .igS a6 8 tll a 3 bS 9 tll d S e7 1 O
.ixf6 xf6 1 1 c3 .igS 1 2 tll c2 00 1 3 a4 bxa4 14 gxa4 as 1 S c4
gba 1 6 ga2 q; h 8 1 7 tll ce3 i.xe3 1 8
tll xe 3 tll e 7 1 9 b3 ! ? fS 20 exfS tll xfS
2 1 tll d S
2 1 . . .b7
This is our model game how to
treat the position when each side
has two minor pieces. We would like
to exchange one of them, but that
would have cost a pawn. ( 21 . . . lde7
22 ldxe7 Wixe7 2 3 E:xaS) So we should
switch to kingside play where the
b7-bishop would be well placed on
the main diagonal, eyeing g2 .
22 o-o gca 23 'Mfd3 tll h4
Topalov sets a nice trap - 24 b4?
axb4 2 S cxb4 ldxg2 26 cj/xg2 e4 27
Wixe4 Wigs + 2 8 cj/hl E:xc4, but his
move is not best. Black takes his
knight awayfram the centre too ear
ly. 23 . . . E:cS ! is better. Then, if White
continues as in the game with 24
E:dl, Black answers 24 ... WigS ! and
White cannot repel the queen with
2S Wffg3 , whereas 2S b4 axb4 26 cxb4
E:cc8 27 E: a7 E:b8 i s also fine, since

f ram fS the knight goes to the terrif
ic outpost d4. The same happens af
ter 24 b4 axb4 2 S cxb4 E:c8.
24 gd1 h6
Another strange, waiting move
of Topalov. He hardly counted on
2S E:ad2? ! a4 ! 26 bxa4 Wigs. More
likely, he decided that a luft would
be useful when White finally push
es b4. However, 24 . . . h6 also weak
ens the light squares around Black's
king. Later in the game Topalov had
to resort to tactical tricks to cover
the critical diagonal bl-h7. 24 . . . E:cS
seems more consistent. Then after
2S Wig3 ldfS 26 Wig4 Black can return
to his initial plan to seek exchanges
with 26 . . . lde7 27 ldxe7 Wixe7 28 Wffe6
Wixe6=. Or 2S b4 axb4 26 cxb4 E:c8
27 b3 ldfS or 27 . . . WigS .
2s Yg3 CiJts 26 Yg4 gcs 27
g ad 2
27 b4 axb4 2 8 cxb4 E:c8 29 E:a7
E:b8 is slightly better for White,
mostly due to the weakened light
squares around Black's king. Still,
30 d3 hdS 31 MS Wigs 32 WixgS
hxgS 33 E:xdS E:xfS 34 bS E:f4 is a
draw.
27 . . . ca 28 Ye4 .ib7
Topalov underlines the fact that
White lacks an active plan.
29 h3 tll h4 30 .id3 gfs ! 3 1
.ib1 ?
White stays seemingly well, but
Black's pieces are constantly attack
ing something, thus hindering the
enemy to reset his forces in the cen
tre. Now 31 c4= would have been
equal, but Karjakin overestimates
his position.
83
Part 3
1 5 C arlsen - Van Wely
Schagen 0 1 .05 . 20 0 6
1 e4 c 5 2 tll f3 tll c 6 3 d 4 cxd4
4 tll xd4 tll f6 5 tll c 3 e5 6 tll d b 5 d6
7 ig5 a6 8 tll a3 b5 9 tll d 5 ie7 1 O
ixf6 ixf6 1 1 c3 ig5 1 2 tll c2 0-0
1 3 a4 b x a4 14 g x a4 a5 1 5 ic4
gb8 1 6 ga2 i> h 8 1 7 tll ce3 ixe3 1 8
tll xe 3 tll e 7 1 9 b 3 f5 20 exf5 tll xf5
2 1 tll xf5 ixf5
3 1 . . Jxc 3 ! 32 YMg4
All Black pieces are hanging, but
at the same time they dominate the
board. The queen has no retreat
square. Even the relatively best 33
Wffa4D c6 34 liJxc3 ha4 3S MS
tlJxfS 36 bxa4+ would favour Black.
32 . . . h5 33 Y;Y e2 Y;Y g 5-+ 34 f4
34 e4 would cover the criti
cal square g2 for only one move:
34 .. cf3 ! 3S M3 M3 36 g3 xb3.
34 . . Jxf4 35 i> h 1
0 r 3 S ttJxf4 tlJf3 .
35 . . . tll x g 2
Only the b7-bishop is n o t direct
ly hitting White's king, but in fact
it makes possible all the nice varia
tions that remain behind the scene.
36 YMxg 2 gg3 37 tll xf4 ixg 2+
38 tll x g 2 gxh3+ 39 i> g 1 g g 3 40
gf2 <it> g 8
I n such positions with a bare
king, the queen generally triumphs
over a rook and two pieces.
41 gxd6 h4 43 if5
Or 43 d6 e4 44 d4 h3.
43 . . . gxg 2+ ! 44 g xg2 YMxf5 45
gcg6 YMt7 46 g sg4 YMf6 4 7 i>h2 i>t7
48 i> h 3 e4 49 gg5 e3 50 i> xh4 g 6
0-1
.
84
This is one of the most boring

positions. Black has no problems,
neither with bishops, nor without
them. The mutual weaknesses bal
ance themselves - the aS and d6pa wns versus b3, c3 and f2 .
22 0-0 ie4 23 id5 ixd5 24
YMxd5 YMc7 25 gxa5 (or 2S cl Wffb 6=)
25 ... gxb3 26 c4 h6 27 gas gb6 28
gfa1 YMb8 29 h3 gxa6 30 gxa6 YMb2
9 ltJdS :lle7 10 hf6 hf6 11 c3 0-0

A typical situation for this line.
Black does not protect his pawns,
but instead attacks the enemy's
ones.
31 f3 'Mfc 1 + 32 @t2 gba 33 gas
'Mfc2+ 34 @g3 'Mfg 6+ 3S @h2 gxa8
36 'Mfxa8+ @ h 7 37 'Mf dS 'Mff6 38 @ g 3
'Mf g6+ 3 9 @f2 'Mfc2+ 40 @ g 3 % - %
1 6 B . Socko - Krasen kow

P lock 03.0S .2000
1 e4 cs 2 f3 c6 3 d4 cxd4
4 xd4 f6 s c 3 es 6 d bS d6
7 .i g S a6 8 a3 bS 9 dS .ie7 1 O
.ixf6 .ixf6 1 1 c3 0-0 1 2 c2 .ig S
1 3 a4 bxa4 1 4 gxa4 as 1 S .ic4 gba
16 ga2 h8 17 0-0 f S 18 exfS .ixfS
1 9 ce 3 .ig6 20 'Mf a4 'Mfc8 2 1 g d 1
e4
This position best illustrates the

aims of both sides in the Positional
variation. Black stakes on dynami
cal factors like piece activity, while
White trusts the classical principles,
according to which he should be
better in view of the weaknesses on
aS and d6 . Modern understanding
of chess tends to take into consider
ation all possible nuances. Practice
shows that both sides must find the
right balance between attack and

defence. For instance, Black cannot
simply give up his pawns and thrust
all his forces against the enemy
king. White has no weaknesses, his
pieces are well coordinated in the
centre, so he should be able to with
stand a direct assault. Therefore,
Black must try first to break this co
ordination by exerting pressure in
the centre. Only when White dis
connects his knights, can he think
about sacrificing the aS-pawn.
22 b3 .if7 ! ?
Both knights are under attack
and White must be constantly con
sidering possible exchanges on e3
or dS.
23 'Mf a3 'Mf d7 24 f1
White is unable to improve his
position without redeploying the
knights. For example, 24 b4? would
have failed to 24 . . . axb4 25 cxb4
ltJxb4 26 ltJxb4 he3 27 hf7 :Iles .
Computers like 24 h3, which
radically prevents . . . \Wg4. Howev
er, this is a permanent weakening
and White will probably regret it at
some moment. Black could main
tain the tension with 24 . . . :/ld8, re
suming the idea of . . . ltJeS. The text
intends ltJd5-e3 and requires fram
Black crucial decisions.
24 . . . e S ! 2S 'Mfxa S 'Mf g4 26
de 3 .ixe3 27 xe 3 f3+ 28 h 1
'Mff4
It is White's turn to make a dif
ficult choice. 29 g3? ! is obviously
out of question. After 29 . . . \Wh6 3 0
ltJfl flhS ! the bishop will replace the
knight on f3 with a devastating ef
fect.
85
Part 3
2 9 tlJfl looks safe, but innocu

ous. Black can simply capture on b3
(29 .. . xb3=) or maintain the initia
tive with 29 . . ..ixc4!? 30 bxc4 tlJeS.
Socko grabs the knight, butthattums
to be at least impractical. White's de
fence is difficult and requires a lot of
calculations. No wonder Socko end
ed up in a time trouble.
29 g xf3 \Wxf3+ 30 @ g 1 ixc4 3 1
bxc4 gf6
Albeit being a piece up and no
mate on the horizon, White's de
fence proves to be amazingly diffi
cult. His problem is not only that
Black can win the h2-pawn and ad
vance his own passer. Black has im
minent threats on the f2-square.
Let us consider:
a) 32 dS. Then Black wins by
force with 3 2 . . . g6 + 33 fl f8 34
d4 h3 + 3S e2 hS+ 36 el
gl + 37 tlJfl h3 38 e2 f3 + 3 9
cj{el e3 ! - + .
Apparently White must pro
tect with his queen the e3 and f2squares:
b) 32 a7 bf8 33 c2 hS ! Not
so much to advance a future passer,
but rather to make a retreat square
to the king. 34 d4 h4 3S c5 dxcS 36
xcS cj{h7! !
86
Most surprisingly, White i s i n a

some sort of zugzwang.
37 d6 weakens the first rank:
37 . . . b8 38 dl h3 39 d4 bf8 and
next Black captures the h 2 pawn. 37
e7 loses to 37 . . . g6+ 3 8 fl hl +
3 9 e 2 f2 + 40 xf2 f3 + 41 cj{el
xe3 + ; 37 d4 g6 + 3 8 fl hl +
3 9 e2 xh2 4 0 fl f4 i s also hap
less .
Nevertheless White can still save
the game by building an interesting
fortress. Krasenkow shows the cor
rect variation:
c) 32 gS ! g6 33 xg6 hxg6 . In
such positions the knight is a very
good defender. The same idea was
possible in the game:
32 g e1 ggs+ 33 t1 gfa 34 gd2
h 6 35 Wf d 5 gf4 36 \Wa8+?
In the time trouble White miss
es 36 cS ! dxcS 37 xcS h7 3 8 es
gs 39 xgS hxgS 40 cj{gl+.
36 . . . h 7 37 \Wd8 \Wh3+ 38 @e2
\Wh5+ 39 @f1 Wfxh 2 40 d1 \Wh 1 +
4 1 @e2 \Wf3+ 42 @f1 gfg4
0-1
The exemplary attack of Krasen
kow displays the rich attacking pos
sibilities of Black when his pawn
reaches e4. The fine point is when to
drop the aS-pawn. We should await
some discoordination of White's
pieces .
.
9 ltJdS e7 10 hf6 hf6 11 c3 0-0

1 7 Leko - C arlsen
L in a re s 0 3.03. 2008
Comments b y Kolev
1 e4 cs 2 f3 c6 3 d4 cxd4
4 xd4 f6 s c3 es 6 d b S d 6
7 .ig S a 6 8 a3 bS 9 d S .ie 7 1 0
.ixf6 ixf6 1 1 c3 .ig S 1 2 c2 e 7
Black uses this move order to

evade the much sharper variations
that arise after 11. . . 0-0. Its only
drawback is that instead oftrying to
overtake the initiative and attack on
the kingside, Black aims first of all
to neutralise the enemy pieces. It is
evident, that without knights Black
cannot hope for active play. On the
other hand, play is easy and clear,
and the importance of home prepa
ration is not so considerable.
1 3 h4
Two days earlier Anand did not
obtain even the slightest edge with
13 a4 bxa4 14 ltJcb4 0-0 lS Wxa4
lt:JxdS 16 ltJxdS d7 17 Wa2 aS 18 d3
ic6 19 0-0 Wb8 20 c4 @h8 21 b3
t5= Anand-Carlsen, Linares 200 8 .
1 3 h 4 aims t o displace the bishop to
h6, f ram where it would be difficult
to activate it .
1 3 . . ..ih6 1 4 a4 bxa4 1 S cb4
0-0 1 6 \Wxa4 xdS 1 7 xdS a s 1 8
.ibS .ie6
18 . . . @h8 ? ! loses a pawn to 19 b4
fS 20 c6 E:a 7 21 exfS MS 22 bxaS
d3 23 bs hbS 24 WxbS+ Karj a
kin-Shirov, Wijk aan Zee 2007 while
18 .. .fS? ! 19 exfS MS stumbles into
20 g4 e6 21 c4 g6 22 gS g7 2 3
ltJf6 + E:xf6 24 gxf6 Wxf6 2 S Wc6+.
1 9 .ic6 gb8 20 b4!
This i s the only way to fight for

an opening advantage. Khalifman
recommends 2 0 b3 @h8 21 E:a2
but after the simple 21. .. g6 (21. . .fS
2 2 exfS E:xfS 23 0- 0t) 2 2 b4 (22 hS
WgS ! ; 2 2 g3 fS) 2 2 . . . axb4 23 cxb4 fS
24 h5 fxe4 2 S Wxe4 E:f4 Black was
fine in Kolev-Gladyszev, Villa de
Navalmoral 20 07.
20 . . . ixd S 21 ixd S axb4 22
cxb4 \Wb6 23 g b 1
Only ten years of global internet
and computerisation were sufficient
to turn modern chess theory into an
information swamp which threat
ens to suck dry any creativity in the
opening stage. The engines help
players avoid obvious blunders in
their preparation so most novelties
present some little improvements
which rarely change significantly
the previous assessments. The cur87
Part 3
rent game is a typical example of
such approach . \Ve have reached
move 23, but I' m sure that both op
ponents looked thoroughly at this
position at home. Perhaps Leko
had discovered that the position
was still not completely exhaust
ed and tries to test his young rival,
without running any risks himself.
Let us note that Dominguez-Jako
venko, Faros 2 0 07 had seen 23 0-0
xb4, when even the ingenious 24
d7 would have given \Vhite just a
tiny edge after 24 . . . hS 25 g3.
23 . . . h S ! ?
I n the recent game Felgaer
\Vang Hao, Gibraltar 200S was
23 . . . d4 24 0-0 d2 25 :gfdl :gxb4,
when \Vhite could have got an an
noying initiative by 26 d7 ! , for ex
ample: 26 . . . g6 ( 26 . . . b6 27 :gal h6
2s :gas :gxaS 29 has g6 30 as ;
26 . . . bl 27 :gxbl) 27 :gal g7 2S
g3 ! ? (2S :ga7 f4 ! 29 :gfl g3 30
hf7 xa7 3 1 xa7 :gxfroo) 2 S . . . :gb2
29 :ga7 b6 30 :gb7! d4 31 :gc7!
b6 32 hf7 hS (32 . . . h6? 33
e7 ! ; 3 2 ... xc7 33 xc7 f7 34
xd6 c3 3 5 :ga3 :gb1 + 36 g2 d4
37 :gd2+) 33 :gc6 b7 34 xb7 :gxb7
3 5 hg6 with a difficult ending for
Black. Most probably Leko has no
ticed this option, but Carlsen de
viates first, anticipating any play
against f7.
24 o-o ts 25 VNa5
In my opinion, with queens
\Vhite has more chances: 25 bS fxe4
26 he4 dS 27 g3 dS 2 S :gbdl d4 29
c4 with a small, but lasting edge;
25 exfS ? ! :gxfSf! would only help
Black, for example: 26 c6 (26 g3?
SS
:gbfS 27 c2 f2 !) 26 . . .xc6 27
hc6 d2 2S bS aS=.
2 5 ... fxe4 26 VNxb6 xb 6 27 gb3
gca 28 ga1 g 6
2 S . . . g 5 i s also possible, intend
ing to meet 29 hS with 29 . . . g4.
It is incredible that a top grand

master could lose this position, but
Black still has a few obstacles on his
way to the draw. On the other hand,
Black should not play the Sveshni
kov at all, if he is afraid of this end
game.
29 gas
\Vith four rooks Black could ac
tivate his pawn centre, for example:
29 he4 :gcb S ! (29 . . . :gc4? ! 30 :gas +
g7 31 :ga7 + hS 32 ds :gcxb4 33
:gc3) 30 :ga bl :gcs 31 bS :gc4f! .
29 .. . xa8 30 ixa8 if8? !
I t looks like Carlsen thought that
the worse part is over and he needs
only to put the bishop on b6 to fix
the draw. It might be so, but still it is
always better to stay active. At this
moment 30 . . . d2 ! 31 bS e3 32 fxe3
el ! (hitting h4) would have level
led the game.
31 b5 ie7 32 g 3 id8 33 ixe4
d5?
Perhaps Black panicked here.
9 ttJd5 e7 10 xf6 ixf6 11 c3 0-0

The sacrifice would have been good
enough , but the bad pawn structure
on the kingside aggravates Black's
defence. 33 . . . 8:b8 with the typical
placing of the bishop on the gl-a7
diagonal would have been rather
drawish.
34 .ixd S d6 3S ic6 ib6 36
gb2 d3 37 @g2 @g7 38 .ie4
ga3 (38 . . . d4 ! ) 39 g4! id4 40 c2
gb 3 4 1 c7+ @h8 42 c8+ @ g 7 43
gc1+ @h8 44 b7 b 2 4S h S !
Black managed t o stop the b

pawn, but now the attack on the
other side is decisive. Later on both
opponents made some mistakes,
but the final result is quite logical.
4S . . . xt2+ 46 @ g3
46 @ h 3 i s even stronger, e . g.
46 . . . E!f4 47 d5 E!:f8 ( 47 . . . gxh5 48
g5 ! 8:f8 49 @h4 + -) 48 g5.
46 . . . t4 47 h 6 ta 48 c7 it2+
49 @ g 2 ie3 SO gS?
A mistake, which questions the
victory. The best way was: 50 8:c6 !
and White's passed pawn is ready
for a triumphant march.
so . . . b8
It i s highly probable that Leko
missed this move, when playing
50 .g5? Now Black grabs the g5-
pawn, while stopping the passer.

S1 c3 id4 S2 c6 @ g8?
52 . . . e3 would have been enough
for a lucky draw: 53 @f3 xg5 54
d5 (54 b6 d8) 54 . . . hh6 55 b6
f8 ; 53 8:e6 ixg5 54 b6 hh6 55 b7
f8 and Black seems to hold on.
S3 idS+ @ta S4 ic4+- @e7
SS c7+ @d6 S6 xh7 e4 S7 g 7 !
@ c s s a c7+ @d 6 S 9 c6+ @ e s 6 0
x g 6 @ts 61 d 6 .i e 3 62 h 7 1 -0
1 8 S h i rov - To palov
More l i a 1 9 .02.2008
Comments by Kolev
1 e4 cs 2 li:) t3 li:) c 6 3 d4 cxd4
4 li:) xd4 li:) t6 s li:) c3 e s 6 li:) d b S d6
7 igS a 6 8 li:) a3 bS 9 li:)dS ie7 1 O
ixt6 ixt6 1 1 c3 igS 1 2 li:) c2 0-0
1 3 a4 b x a4 14 xa4 as 1 S ic4
id7
We recommend 15 . . . 8:b8 .
1 6 0-0 li:) e 7 1 7 a3 li:) xd S 1 8
ixd S b8 1 9 b4
Khalifman advocates 19 8:a2
with the idea of sacrificing the ex
change : 19 . . . a4 20 ttJb4 g6 2 1 8:xa4
ixa4 22 xa4oo.
1 9 . . . axb4 20 li:) xb4
Although White hasnot created a
passed b-pawn, the other positional
factors ensure him a lasting edge.
The excellent control of d5 and the
a-file make possible the occupation
of the seventh rank.
20 . . . '%Yb6 21 %Ye2 ibS
Topalov made this move quick
ly and he was obviously confident
about his position.
22 ic4 tc8 23 ixb S %Yx b S 24
%YxbS xbS 2S d1
89
Part 3
I think that Black should avoid

such positions. This game, and the
previous one, clearly show that
when Black is passive in the Svesh
nikov, his prognosis is not positive.
25 . . . g6 26 g3 @ g 7 27 til d 5
O f course White should not re
lease his grip with 27 E!xd6? ibe7 2 8
E!dS ibxb4 2 9 cxb4 E!xb4 30 E!xeS
f6 31 E!aaS E!cc4 = .
27 . . . c4
27 . . . E!b2 with the intention of
bringing gS to d2 is met by the
logical 28 E!d3 . Play might continue
28 . . . E!bl + ( 2 8 . . . E!e2? ! 29 fl E!xe4?
30 f3 E!ec4 31 l2Jb6) 29 g2 E!el 30
h 4 i.h6 31 E!a7 E!xe4 3 2 E! f3 E!f8 33
E!f6 E!e2 34 E!xd6 E!d2 35 E!dd7 and
White's pieces dominate on the
board.
28 ga7 id8
Topalov tries to solve the prob
lem of his bishop . 28 . . . E!xe4 29 CiJc7
E!b8 30 E!xd6t would only under
line the different energetic level of
the two armies.
29 g d 7 !
Shirov prepares t o double his
rooks on the sevent rank.
2 9 . . . iaS
2 9 . . .ibb6 3 0 E!d3 E!xe4 3 1 CiJxb6
90
E!xb6 3 2 E!f3+ illustrates the poten

tial of White's rooks .
30 g e 1 !
The sharp lines would let Black
escape: 30 E!d3 E!xe4 31 E!f3 E!xdS 3 2
E!fxf7+ h6 3 3 g 2 E!d3 34 h 4 hS
35 f3 h6 36 fxe4 g4oo 37 fl (37
E!f6 E!xg3+ 38 fl gS 39 E!xh6 gxh4
40 E!g7+ f3 41 E!xg3 + hxg3 42 E!xd6
xe4 43 c4 b4 44 E!dS ibc3 45 E!d8
ibd4 46 g2 f2 =) 37 . . . gS 38 e2
E!xc3 39 E!xd6 E!xg3 40 E!fS E!g2 + 4 1
fl E!c2 = .
3 0 . . . ib6 3 1 lll xb6 g xb6 32
g93
We have seen a textbook case of

transformation of the advantage.
White gave up his knight, but he
forces Black into a terrible bind.
32 . . . gca 33 gf3 gfa 34 t1
The march of the king will fur
hermore improve White's position,
so Black must try to break-through
on the kingside.
34 . . . g S
Perhaps the setup with 34 . . . hS
gave more chances for survival. It
would have retained the thematic
Sveshnikov break f7-f5, for exam
ple: 3 5 e2 E!b2 + 36 d3 E!bl 37
h4 E!el 3 8 E!c7 E!dl+ 3 9 c4 g8 40
9 4J d5 fJ.e7 10 ii.xf6 ii.xf6 11 c3 0-0

b5 f5 41 exf5 e4f! and suddenly
Black is breathing again.
3S h4 g4? !
This weakness will prove to be
fatal for Topalov. 3 5 . . . gxh4 36 gxh4
g6 37 !!f5 !!c6 was somewhat bet
ter.
36 ts h6 37 e2 c6 38 d 2
g 6 3 9 h S + g7 4 0 d3 gb6 4 1
gc7 ! g b 1 4 2 c4 g d 1 43 b S
g8?
Only this move irreversibly ru
ins Black's game . It is unclear how
White's king could cross the c-file
after 43 . . . !!cl, for example: 44 b6
!!b8 + 45 a7 !!f8 46 b7 !!c2 47 c4

!!c3 48 !!c6 !!d3 49 c7 !!d4.
44 gf6 !
The rest is agony.
44 . . Jd2 4S c6 g 7 (45 . . . !!d3
46 !!e7 !!xc3 + 47 xd6 c2 48
xe5 +-) 46 g g 6+ h7 47 gxg4
gxf2 48 xd6 gea 49 c4 gd2+ so
c6 gfa s1 cs gd4 s2 gb7 ha S3
b S gd1 S4 c6 gc1 SS b 6 gca S6
c7 gea S7 ga7! (57 !!b8? !!bl + 58
a7 !!al+ 59 b7 !!bl +=) S7 .. gb1 +
sa c s gc1 + S9 d S gc2 60 g a s
h 7 61 g c s g d 2 + 6 2 c s g a s
1 -0
53gh4
91
Part 4
1 e4 cs 2 f3 c6 3 d4 cxd4
4 xd4 f6 5 c3 es 6 dbS d6
7 .igS a6 8 a3 bS 9 dS .ie7 1 O
i.xf6 .bf6 1 1 c 3 0-0 1 2 c2 .igS
1 3 a4 bxa4 1 4 gxa4 aS 1 S .ic4
gb8 1 6 b3 @ h8 1 7 ce3 g6
QUICK REPERTO IRE
We examine this position in a

separate part, for it offers sharp
Pay
special
attention
to
this variation! It is a frequent
play of a quite different character

in comparison with the previous
part of the book. The knight on e3
guest in Internet and club-level
forces Black to prepare .. .fS with g6.

This little pawn move significantly
Alternatively:
a) 2 0 'We2 ! ? is a new idea ofKarj a
kin. (You can see a detailed anal
ysis of game 19 Karjakin-Shi
rov, Khanty-Mansiysk 1 1 .12 . 2 0 0 7
i n the "Co mplete Games" sectio n.)
In short, avoid Shirov's 20 .. .fS ? ! We
prefer the thematic 20 . . . l'i.Je7 2 1 f4
l'i.JxdS 2 2 l'i.JxdS exf4 2 3 gxf4 f6
changes the pawn structure with all

the ensuing long-term consequenc
es. A lot of positions that were as
sessed as drawish in the previous
part, might turn quite unpleasant to
defend , due to the weakness of the
seventh rank and Black's castling
position as a whole. Another particu
larity is White's option to push h4,
intending to further compromise
Black's kingside. You should also
have in mind that two great players
and theoreticians, Anand and Kha
lifman, have recently chosen this
system as a main weapon against
the Sveshnikov. We shall present
clear recommendations (and novel
ties) against their approaches.
A. 1 8 h4
This variation is critical for our
repertoire. Its current status is fa
vourable for Black, but you must
watch out for new discoveries.
92
games.
18
i.xh4 1 9 g3 .ig S 20 f4
Black has a reliable position.

b) 20 ga2 is too sophisticated.
We can simplytake on e3 : 2 0 . . . xe3
21 l'i.Jxe3 e6 = .
2 o. ef4 2 1 gxf4 ih4+ 22 @d2

.
On fl the king is X-rayed by the
12 l2Jc2 0-0 13 a4 bxa4 14 xa4 as 15 c4 b8 16 b3 h8 17 l2Jce3 g6

f 8-rook. The variation 22 Ml f5 23
a2 fxe4 24 ah2 g S 25 l2Jg2 b7
demonstrates two important defen
sive resources of Black: . . . b7 pro
tects h7, while . . . gS often cements
our bishop on h4.
22 e7!
Black repels the initial attack

and the game enters the stage of
balanced manoeuvring. In the next
few moves Black seeks to destroy
both white knights, while the enemy
tries to swing the a4-rook onto the
kingside. The latter is not so easy,
because the king on d2 is barring the
second rank. Therefore White often
moves it to cl. Note that 23 xh4? !
ltJxdS 24 hl l2Jf6 25 ltJdS hS ! is
pleasant for Black, so we'll focus on:
23 c1
23 c2 leads to similar posi
tions: 23 ... ltJxdS 24 ltJxdS e6 25 al
f6 26 h2 g7 27 d2 a4 ! ?
Here White should b e able to

make a draw with ingenious play,
but we clearly do not risk much.
The fine point is that xh7 is not
such a dangerous threat as long as
we control the h8-square . It is only
one check!
23 @c1 xd5 24 xd5 .ie6
From the comment on the pre
vious move we know that Black is
safe if he rules over the main diag
onal (25 a2 xdS 26 xdS f6) so
White chooses:
25 Y9d4+ g8
We'll soo n reach a position with

bishops of opposite colour. Our king
will be safer behind the pawns, but
White compensates that with more
active rooks. The remedy is to sac
rifice a pawn to open files, for in
stance:
26 a2 xdS 27 xdS f6 2 8
d2 g3 29 fl dS ! ? ( 2 9 . . . hSoo)
93
Part 4
30 xdS would leave the fl-rook
hanging to 30 . . . gS, while 30 exdS
fe8 hints that the tide is turning
and Black is already the active side.
The latest top level game in this
variation s aw:
26 <t!?b l hdS 27 xdS f6
We would say that the game is
level. See 2 0 Jakovenko-Shi
rov, Faros 20 07 in the "Complete
Games" section.
B. 1 8 0-0 f5
Our king may be weakened, but
we see already the first benefits of
. . . g6 - 19 exfS gxfS would give us
a mobile pawn centre. Then 20 f4
would block it indeed, but at the
cost of providing an outpost on eS
for our knight and opening play in
favour of our bishop pair: 2 0 . . . exf4
2 1 tlJc2 2 1 tlJc2 ttJeS. (or 2 1 . . . d7 !?)
1 9 Y;7d3
This has been recommended
by Khalifman as a main repertoire
against the Sveshnikov in the Chess
Stars book "Opening for White Ac
cording to Anand 1.e4", vol. 10.
W e propose a new arrangement of
Black's pieces:
1 9 .ie6 20 gd1 Y9d7 21 f3
Y9g7 !
94
In this structure White would

be happy to exchange pieces. That
would only underline Black's nu
merous weaknesses and especially
his poorly protected king. Kolev has
however completely different inten
tions ! He wants to launch a kingside
pawn storm with the f and g-pawns.
For that, Black needs his queen on
the right wing.
It is also very important to
set up correctly the bishop pair.
The best places are d7 and d8.
Then we can maintain the ten
sion in the centre, or close it and ad
vance the g-pawn. Let us see some
examples:
a) 2 2 ttJfl d7 23 a2 d8
24 ttJde3 b6, when it would be

risky for White to take on d6 and
open the file to our rook: 25 xd6
fd 8 .
b) If White chooses to stay with
his knights on dS and e3: 2 2 <t!?h l
d7 23 a2 , we can follow up with
23 . . . d8 or 2 3 . . . h6, but 23 . . .f4 ! ?
i s more interesting: 24 tlJc2 d8 2 5
tlJ a3 g S 2 6 tlJbS h6oo
We think that such a position,
with a clear-cut attack against the
enemy king, should appeal to any
Sveshnikov fan.
12 lOC2 0-0 13 a4 bxa4 14 xa4 as 15 c4 b8 16 b3 h8 17 ll:Jce3 g6
Play might continue with 27

ll:Jdc7 f6 28 ad2 g4oo .
C. 1 8 1Mfe2 !?
This was a novelty in game 21
Anand-Shirov, Linares 20 0 8 .
1 8 ...fS 1 9 h 4 ixe3 2 0 1Mfxe3
f4!
And the last move is our im
provement. Shirov opened the cen
tre, but his king proved to be vul
nerable.
After 20 .. .f4 White's future plan

is not clear. He had already com
mitted himself with 19 h4, so short
castling will hardly be good. White
will have to leave his king in the
centre, but then he will be unable to
advance the b-pawn. For his part,
Black will fallow up with 21 . . . h5 to
fix the weakness on h4 . The game
might continue with 21 Wf d3 h5oo or
21 Wfd2 h5 22 f3 h7 23 Wfff2 g8 24
e 2 g7 25 hal e6. We are eager
to see more tests of Anand's idea.
95
Part 4
1 e4 c5 2 f3 c6 3 d4 cxd4
4 xd4 f6 5 c3 es 6 db5 d6
7 .ig5 a6 8 a 3 b5 9 d5 .ie7 1 O
ixt6 ixf6 1 1 c3 0-0 1 2 c2 .ig5
1 3 a4 bxa4 1 4 gxa4 a 5 1 5 .ic4
gb8 1 6 b3 <it>h8 1 7 ce3 g6
STEP BY STE P
page 96
A . 18 h4
page 102
B . 18 0-0
page 105
C. 18 We2
Khalifman also mentions as
risky 18 Wffd3 f5 19 f3 i.h4+ 20 @e2
f4 21 tlJfl Wd7 22 @dl Wa7 with un
clear consequences.
Line A. 1 8 h4 considers the shar
pest attempt of White to refute
Black's setup . There is enough prac
tical evidence that Black should be
able to survive the first attacking
wave. Then commonly arise posi
tions with bishops of opposite col
our where Black seeks to open files
for his rooks by sacrificing a pawn
with either . . . a4 or . . . e4. In no way
should he defend passively as in
such positions even without a pawn
White would have winning chances.
At top level Black achieves good re
sults and during the last year the
popularity of 18 h4 has waned con
siderably.
18 0-0 has long been assessed as
innocuous, but Khaifman advocat
ed it in"Opening for White Accord
ing to Anand 1.e4", vol . 10. He con
nects it with the new idea of hold
ing firmly the centre by Wffd3 . At our
turn, we also propose a new setup for
96
Black, which i s fairly aggressive.

18 We2 earned Anand a point in
Linares 20 0 8 , but it is still early to
panic. We recommend a new ap
proach which needs further tests .
A. 1 8 h4
This aggressive continuation is
critical for o ur repertoire. It brings
about sharp forced variation which
might refute the whole setup with
early castling. Fortunately, the cur
rent status of this sacrifice is favour
able for Black, but you must watch
out for new discoveries.
1 8 . . . .ixh4 1 9 g3 .ig5
20 El:a2
A2 . 2 0 f4
A3 . 2 0 We2 ! ?
Al.
12 ttJc2 o-o 13 a4 bxa4 14 roca4 as lS !c4 b8 16 b3 wh8 17 ttJce3 g6

A 1 . 20 a2 xe3 2 1 xe 3 ie6
From a theoretical standpoint,
this is the main move, as it gives
Black a comfortable game. On the
other hand, 21 . . .fS is significantly
more hazy, but that could be a prac
tical plus if one is well prepared:
22 d2
22 f4 exf4 23 ah 2 (23 gxf4
b6oo) displays the reason behind
White's 20th move, but Black holds
firmly after 23 . . . b7 24 gxf4 g7 ! f! .
2 2 . . .f4 2 3 gxf4 exf4 2 4 xd6 c7
2S ltJg2
The alternatives are: 2S dS?
fxe3 26 xc6 g7 27 d4 xd4 2 8
cxd4 exf2 + 2 9 W fl a4 ! ; 2S ttJc2 !g4
26 f3 bd8 27 xd8 xd8 28 ttJd4
!hS 29 !dS ltJeS = ; 2S ttJfS gxfS 26
dh6 b7 27 xh7+ xh7 2 8 xh7+
xh7 29 d6 ttJe7 30 es+ g7 31
xf4 ltJg8= .
2S . . . !g4 26 xc6 (26 f3 bd8)
26 ... xc6 27 xg4 xe4+ 2 8 @d2
(28 Wfl bl+ 29 ttJel be8 30 xh7 +
wxh7 31 d7 + @h6 32 h3 + Wg7 33
d7+ @f6 34 c6 +=) 28 . . . bd8 +
29 Wcl a4 ! with a double-edged po
sition according to Rogozenko.
22 g d 2
I n Sandipan-Spasov, Turin oL
20 06 , White chose the tricky 2 2

d2 , intending t o meet 2 2 . . .hc4?
with 2 3 ttJfS ! gS 24 dl . Spasov
answered 2 2 . . . g8 ! 23 al hc4
24 ttJxc4, when he should have tak
en the pawn: 24 . . . xb3 ! 2S dS (2S
ttJxd6? ! f6 26 dS gb8 ! 27 xc6
3b6 28 c7 xd6 29 xf7 bl+ 3 0
We 2D lb 2 + 31 W f3 hS+) 2S . . . ttJe7
26 xf7 (26 xd6? ! xd6 27 ttJxd6
Wg7+) 26 . . . g7 27 e6 (27 f3 ttJc6 !
2 8 ttJd 2 bS+) 27 e6 c7! ( 2 7. . . c8
28 xc8 + ttJxc8 29 @d2 f7=) 2 8
0-0 (28 c l ltJ g 8 29 ttJxeS xc3 3 0
@ d 2 xcl 31 xcl e7+) 2 8 . . . c8+.
22 . . . '%Ye7 ! 23 gxd6 d4 24 g d s
2 4 xd4? i s insufficient: 2 4 . . .
exd4 2 S xd4+ f6 2 6 ttJdS hdS 27
hdS fd8+.
24 . . . c6 2 5 gd2 ixc4 26 xc4
26 bxc4 cs is also roughly
equal : 27 d7 fd8 28 0-0 Wg7 29
dS a3 30 f3 e7= .
26 . . . gfd 8
Black can maintain the balance
with other moves as well: 26 . . . 1!9b7!?
27 b2 (27 dS xb3 28 xb3 xb3
29 Wd2 c8 30 cs bb8 31 al
Wg7=) 27 . . . fd8 28 e2 d7! 29
d2 (29 0-0 1!9d3=) 29 . . . b7= .
21 gds=
97
Part 4
We would have stopped here, as
suming that the position is clearly
equal, hadn't Rogozenko claimed
"some advantage" for White. More
likely, there is none.
27 . . . @ga 28 o-o Yb7 2 9 xe5
(29 d6 bS) 29 .. Jxd5 30 Yxd 5
x e5 3 1 Yxe5 Yxb 3 3 2 Yxa5 gca
33 gc1 Yb2 34 Yg5 g ea 35 c4
gxe4 36 c5 g es=.
A2. 20 f4 exf4 2 1 gxf4 .ih4+
llJe3 g6 30 ds tt:Jes 3 1 e2 . Here

Kolev likes the move 3 1 . . . tt:Jd3 32 fS
hS+ ! 3 3 d2 f3 34 lh2 tt:JeSf!
with a good position for Black.
23 . . .hfS 24 llJxfS fufS 2S g4
gs 26 d3
26 g2 ? fails to 26 . . . hf4 27
h4 (27 tt:Jxf4 gs 28 llJxg6 + g7
0-1, Delchev-Kotanjian, Kusadasi
2 006) 27 . . . gS + 28 xgS xgS+ 29
f3 es- + . After the text, the game
ends up with mass elimination:
26 d3 xdS 27 xg6 f6 28
xh7+ g 8 29 f7+ xh7 30 xdS
tt:Je7 31 e4+ g7 32 xgS+ xgS
33 fxgS xb3 34 xaS xc3 = .
2 2 . . . e7
Now the main line branches to:
A2 a. 23 c2
A2b . 23 cl
A2c. 23 gl ! ?
Black's defence is based on . . .fS,

followed by b7. In some lines White
proves unable even to shift the bish
op fram h4, if we get the chance to
support it with . . . gs.
22 i>d2
O n fl the king i s X-rayed by the
f8-rook:
22 fl f S 23 exfS
23 a2 fxe4 24 ah2 looks very
purposeful, but 24 . . . gS 2S tt:Jg2 b7
26 llJxh4 gxh4 27 xh4 g7 28 hS
fS defends everything. Stangl-Kindermann, Altensteig 1987 saw fur
ther 29 h6 , when Black counter
attacked with 29 . . . a4 ! ? 30 bxa4
b8 31 bS tt:Je7= . Hracek-Mis
ta, Czechia 2006 improved with 29
98
2 3 h4? ! allows Black t o re

pel White's attack with 23 . . . ttJxdS
24 hl tt:Jf6 , for instance, 2S c2
b7t or 2S llJdS hS ! 26 tt:Jxf6 xf6
27 xaS g7+. (Rogozenko)
A2a. 23 c2
N owthe c3-pawn is well defend
ed by the king, but the a4-rook can
not reach the h-file via the second
rank. Anyway, we'll see later that
even if White takes on h7, he is still
far from winning .
23 lllxd5 24 lllxd5 e6 25 E:al
2S d4+ g8 26 aal hdS 27
xdS b6 is unclear.
25 f6 26 E:h2 ig7 27 Wfd2
a4! ?
12 4Jc2 0-0 13 a4 bxa4 14 xa4 aS lS c4 b8 16 b3 h8 17 4Jce3 g6
Remember this thematic break!

It is the only way to open files and
obtain counterplay. The game Ko
rneev-Solak, Vrsac 200S shows that
White still has a draw:
28 E:ahl (28 xa4 a8f!) 28
axb3+ 29 hb3 hd5
29 . . . hS also leads to a draw: 30 fS
xdS 31 xdS \Wb6 3 2 \Wcl bc8 33
f6 ! hf6 34 xhS+ gxhS 3S xhS+
g8 36 gS + hgS 37 \WxgS+ h7
3 8 \WhS + g7 39 \WgS +=.
30 E:xh7+ <;f;>g8 31 hd5 f6
32 %Yd4= \Wxd4 33 cxd4 xd4 34 eS
dxeS 3S hf7 + f7 36 h8 + g7 37
lh7+ f6 38 xf7+ xf7 39 xb8
exf4, draw.
A2b. 23 @cl xd5 24 xd5

.ie6
In this variation Black main

tains the balance by counterattack
ing c3. Here is a typical example: 2S
a2 xdS 2 6 xdS \Wf6 27 \W f3 fc8
28 h3 hS 29 xaS el 3 0 c4 xb3
31 axhS+ gxhS 32 \WxhS+ g7 =.
That's why i n practice White choos
es:
25 %Yd4+ <;f;>g8 26 <;f;>bl
We chose this as a main line be
cause it occurred in the latest top
level game, Jakovenko-Shirov, Fa
ros 20 07. Alternatives are :
a) 26 c2 is rather inconsistent.
Balck is able to choose whether to
play with opposite coloured bishops
after 26 . . .xdS 27 xdS f6 28 \We3
g7oo or maintain the tension with
26 . . . c8 ! ? .
b ) 2 6 fS xdS 27 xdS ( 2 7 \WxdS
\WgS + 28 bl \We3 ! ) 27 . . . \Wf6 2 8
fxg6 hxg6= .
c ) 2 6 a2 After this move Black
gets good play on the dark squares,
thanks to the position of the white
king and the pawns c3 and f4 . It
would have been better for the
white king to be on a light square,
thus preventing possible checks.
26 . . . xdS 27 \WxdS
27 xdS? ! f6 28 \We3 \Wc7 fa
vours Black, for instance 2 9 ah2
a4 ! ; 29 c4 fc8 30 ah2 dS ! ; 29
h3 fc8 30 c4 \Wes . (o r 30 . . . a4 ! ?
3 1 xa4 a8)
27 . . . \Wf6 (27 . . .g3 28 \Wd2 \Wf6
transposes to 27 . . . \Wf6) 2 8 \Wd2 g3
29 fl dS ! ?
Topalov-Leko, Linares 200S,
went on with 29 ... hS and the game
was very tangled, but White grad
ually outplayed his opponent. The
99
Part 4
text was proposed by Rogozenko
and passed the test in Korneev- De
vereaux, Port Erin 2006.
34 b4 (34 E:a6? E:xc3 ! ) 34 . . . E:fc8 35

E:a3 h5 (Rogozenko) would have
been quite promising for Black.
26 hd5 27 11Mxd5 (27 hd5
f6 28 d3 c7) 27 f6
Black needs open files for his

rooks . We had seen before the sac
rifice . . . a4 with similar aims .
Now, 30 e5? ! Wf5 31 hd5 g5 !
would be slightly better for Black.
Perhaps the most testing answer is
30 exd5 E:fe8 (Rogozenko suggests
30 . . . h5, but after 31 :gxa5 E:a8 32
E:xa8 E:xa8 3 3 @b2 Black's compen
sation is not too clear, e.g. 33 . . . Wd6
34 b4 ih4oo) 31 E:f3 (or 31 E:xa5 E:el +
32 @c2 E:xfl 33 hfl hf4) 3 1 . . .'Wh4
32 d6 Wg4 (32 . . . E:e4 33 E:a4 @g7
34 e2;t) 3 3 E:fl (33 e2? ! E:e4 34
E:a4 :gxa4 35 bxa4 g5 ! ) 33 . . . E:e4 34
d7 E:d8 35 Wd3 E:xf4 36 ib5 E:xfl+
37 Wxfl e5oo . In these lines White
has a strong passed d-pawn, which
is balanced by constant threats by
Black. In the source game Korneev
preferred to keep the e-file closed :
30 xd5 g5 31 @b2 hf4 32
E:xa5
Now 32 . . .hd2 33 E:xf6 E:fc8
leads to a drawish endgame, e.g. 34
hfl+ @g7 35 E:f3 g4 36 E:d3 hc3 +
37 E:xc3 @xf7 Devereaux retained
the tension with:
32 . . . E:bc8 33 Wd3 , when 33 . . . E:c7
=.
100
Black has neutralized the direct

threats and his king is well protect
ed, which is an important factor in
positions with bishops of opposite
colour. On the other hand, White's
rooks are more active and restrict
Black's play. We would say that the
game is level. See 2 0 Jakovenko
Shirov, Faros 20 0 7 in the "Com
plete Games" section.
A2c. 23 VMgl! ?
I n the previous lines we have
seen White trying to find the per
fect balance between attack and de
fence . Now we'll examine the most
straightforward approach, which is
frequently met in my (Kolev) ICC
blitz games.
23 xd5 24 xd5 e6
24 . . . h5 25 a7 (25 d4+ @h7
26 E:aal e6oo) gives White a slight
ly better ending, thanks to his cen
tralised king: 25 . . . ie6 26 ifMxa5 (26
12 ltJc2 0-0 13 a4 bxa4 14 xa4 aS 15 c4 b8 16 b3 @h8 17 ltJce3 g6

xaS f2 27 rfxf2 xaS 28 rfd4 + f6
29 ltJxf6 rfb6 ! ) 26 . . . g3 ( 26 . . . @g7
27 rfxd8 hd8 28 a6t) 27 xd8
fxd8 28 fSt.
25 h2
25 rfd4 + @g8 26 @c2 hdS
brings about already familiar posi
tions : 27 xdS g3 ! ? (27 . . . Wb6 2 8
rfd 2 f6 29 rfh2 h S 30 rfgl rfxgl 3 1
xgl @g7 3 2 xaSt) 2 8 fS (28 fl
f6) 2 8 . . .eS+! 29 fxg6 (29 d3
@g7) 2 9 . . . hxg6 30 d2 rff6 ! 31
xaS a8= or 27 hdS f6 28 rfd3
b6 +!;
25 @c2 f6 26 fS (26 h 2 hS)
26 . . . hdS 27 hdS e7 28 fxg6
fxg6 29 xg6 looks dangerous, but
29 . . .eS= reminds that White's king
is not safe, too.
25 g5
example: 3 1 . . .hc4 32 ltJf5+ @g6 33

bxc4 rfb6 34 gS + ! fxgS 35 ltJe7+
@g7 36 rff5 dS 37 ltJxdS with at
tack.
26 rbc2
26 @d3? fS ! earned me the point
in a ICC blitz game . More hazy is 2 6
gl ! ? f6 27 rfd4 b7oo.
26 f6 ! ?oo
With this move we prepare a de
fence along the seventh rank with
b7. 26 . . . h6 is also playable and
needs tests.
A 3. 20 yge2 ! ?
Black's defence hangs by a thread,

but it is a very strong one ! White
had saved a tempo by leaving the
king on d2, and now we are able to
survive thanks to the check: 26 fxgS
hgS+ 27 @c2 h6 28 aal g8 !?oo,
intending g7. Note that the setup
with 28 . . . @g7 29 agl h8 offers
more chances to White: 30 f2 f6
(30 . . . hdS 31 hdSoo) 31 ltJe3 ! , for
This is a new idea of Karjakin

which he used against Shirov in the
World Cup. The white queen is eye
ing the h 2-square while freeing dl
for the king. The source game saw
2 0 . . .fS? ! and Black got a winning po
sition . I (Kolev) was also impressed
by Shirov's play and fallowed in
his footsteps in a later game, but a
more thorough analysis convinced
me that Black should look for bet
ter options. See my detailed analy
sis of game 19 Karjakin-Shirov,
Khanty-Mansiysk 11 .12.2007 in the
101
Part 4
"Complete Games" section. My con
clusion is that White can continue
with 21 exf5 ! hf5 22 l2Jxf5 gxf5 23
f4 exf4 24 h2 b7 25 gxf4 f6 26
@dl, fallowed by a2, with a very
unpleasant attack. Therefore, I pro
pose to refrain from 20 .. .f5 in fa
vour of:
20 . . . tll e 7
O u r knight i s not very efficient
on c6 and it is better to trade it
for the l2Jd5. Thus we will reduce
White's attacking potential.
21 f4
Or 21 l2Jxe7 he7 22 f4 exf4
2 3 gxf4 (23 h2 h5 24 gxf4 f6)
2 3 . . . f6 24 l2Jd5 e8f!.
2 1 . . . tll x d S 22 tll xd5 exf4 23
gxf4 if6
Black has a reliable position.

Now 24 h 2 h5 25 a2 e8 or 24
f3 e6 25 a2 @g7 26 ah2 h8
leave White struggling to prove that
he has enough compensation for
the missing pawn and bare king.
B. 1 8 0-0 f5
B l . 19 exf5
B 2 . 19 d3
10 2
exf5 gxf5 20 f4
White can delay f4, but that
would only help us improve our
piece s: 2 0 h5 d7 ! ( 20 .. .f4? ! 2 1
d3 b7 2 2 l2Jc4) and now:
21 f4? ! exf4 22 l2Jxf4 l2Je5 23 a2
a4 ! ;
2 1 hl e4 ! ? (21. . .e8 2 2 h3 f4
2 3 d3 d7! 24 xd7 hd7 25 l2Jc4
b3 26 l2Jxa5 l2Je7 27 aal bb8 2 8
l2Jxe7 he7=) 2 2 f4 exf3 23 gxf3 ( 2 3
xf3? l2Je5 24 h3 h6 - +) 23 . . . l2Je5
24 a2 he3 ! 25 l2Jxe3 f4+.
21 fal e4 2 2 dl (22 ltJfl g8 !
23 e2 g6 24 @hl f8 25 4a3
h6 26 gl g7 27 l2Jg3 e5t) 22 . . .
f4 2 3 l2J c 2 e8 24 xa5 ( 2 4 l2Jd4
l2Je5 25 xa5 f3 26 g3 e3 27 5a2 e2
28 l2Jxe2 l2Jxc4 29 bxc4 fxe2 30 xe2
g6+) 24 . . . l2Jxa5 25 xa5 f3 26 g3 .
Khalifman claims that "White has a
sufficient compensation for the ex
change, but not more than that . "
81 . 1 9
2 0 ... exf4 2 1 tll c2

21 ltJxf4? loses due to the pin
2 1 . . . b6 22 f3 (22 el e8 23
ltJfd5 a7 24 hl he3 25 f3 f4 26
f4 a6 - +) 2 2 . . . l2Je5 23 h3 ( 2 3
f2 M4 24 xf4 l2Jg4 25 el 8- +)
12 tlJc2 0-0 13 a4 bxa4 14 a4 as 15 ic4 b8 16 b3 h8 17 ttJce3 g6

2 3 ... id7 24 aal tlJg4- + .
2 1 c2 e s
2 1 . . . id7 ! ? 2 2 a3 ttJ e 5 also de

serves attention: 23 ttJxf4 ttJxc4 24
bxc4 b2 = . (Khalifman)
22 \Wd4
White's bishop on c4 is unsta
ble and should be supported. 2 2
ttJd4 id7 23 a2 runs into 2 3 . . . a4 !
24 ttJxf4 c8+ while 22 ttJxf4 id7
23 al b6 ++ is also in Black's fa
vour. After the text White's prob
lems come from the other wing:
22 . . . f3 ! ? 23 gxf3 gga 24 @ h 1
g g 7 2 5 f4 ( 2 5 ttJce3 f4 ! 2 6 ttJxf4
ttJc6t) 2 5 . . . xc4 26 gxc4 (26 xc4
if6 !) 26 . . . ib7.
B2. 1 9 \Wd3
This has been recommended
by Khalifman as a main repertoire
against the Sveshnikov in the Chess
Stars book "Opening for White Ac
cording to Anand 1.e4", vol. 10 . His
main line runs as: 19 .. .f4 20 tlJc2 f3
21 g3 d7 (21. . . h5 22 fal h4 23 tlJel
hxg3 24 hxg3 ig4 25 ib5i) 22 fal
h3 23 ttJde3i. We propose a new
arrangement of Black's pieces :
1 9 . . . ie6 20 gd 1
White might start with 2 0 f3 . We

propose to fallow up with the same
setup as in the main line: 20 . . . d7 ! ?
2 1 hl f4 ! ( 21. . . g7 i s inferior: 2 2
exf5 gxf5 23 f4) 2 2 tlJ c 2 id8 23 fal
g5 24 b4 g4 with counterplay.
20 gd 1 \Wd7 2 1 f3 \Wg 7 !
Khalifman considers only 21 . . . a7

2 2 hl he3 23 ttJxe3 hc4 24 c4
with a slight advantage. Of course,
in this structure White would be
happy to exchange pieces. That
would only underline Black's nu
merous weaknesses and especially
his poorly protected king. Kolev has
however completely different inten
tions ! He wants to launch a kingside
pawn storm with the f and g-pawns.
For that, Black needs his queen on
the right wing. It is also very impor
tant to set up correctly the bishop
pair. The best places are on d7 (to
avoid tactics based on the hanging
state of the bishop on e6) and d8.
We shall examine:
B2a. 22 tlJfl
B2b. 2 2 hl
These moves keep hold of the
centre and are in the spirit of Kha10 3
Part 4
lifman's strategical approach.
The computers also like total
ly inconsistent moves like 22 exfS
gxfS 2 3 hl, but there is no reason
to pay them much attention. Black
can get a strong initiative with 23 . . .
e4 ! 2 4 fxe4 fxe4 ( 2 4. . . ltJeS 2S d4
fxe4 26 E'!:xaS is likely to be drawn
after 2 6 . . . fuc4 27 xg7+ xg7 28
l2Jxc4 h6 ! intending ... e3) 2S xe4
8:be 8 .
in view of ltJxdS 29 exdS e4oo.

25 exf5
2S xd6 is risky: 2S . . . 8:fd8 2 6
hl fxe4 27 fxe4 he3 2 8 l2Jxe3 g4
2 9 l2Jxg4 (29 xc6 hdl=) 29 . . . 8:xd6
30 8:xd6 e7 31 8:xc6 d7 32 dS
xg4 33 h3 d l+ 34 h2 8:f8 .
25 ... hfS ! 26 Yd5
26 d2 would let through 26 . . .
e4 ! 2 7 dS ltJeS 2 8 fxe4 d'Too.
26 ... .id7 27 Wfxd6 (or 27 hl
cS) gbd8
B2a. 2 2 fl i.d7 23 ga2 .id8
All Black pieces went to their op

timal places . It is not easy for White
to create threats.
24 de3
24 l2Jfe3 goes halfway to Black's
plan: 24 . . .f4 2S l2Jc2 gS 26 l2J a3 g4t .
24 ... i.b6
The dark-squared bishop is very
active and that allows Black to sacri
fice the d6-pawn in order to open the
d-file. However, the position is dy
namically balanced and both sides
have different options of rough
ly equal worth. For instance, now
24 . . . c7 ! ? is a good alternative : 2S
dS b6 26 hl l2J e7 27 l2J c4 ic7oo,
when 2 8 ltJxaS? ! is hardly advisable
104
White is unable to prevent 27 . . .

e4, which will activate the "fian
chettoed" queen. Our analysis indi
cates that the game is still balanced.
We'll show its main line:
2 8 i.d5
Or 28 hl e4 29 dS xe3 30
hc6 f4 3 1 cs hc6 3 2 8:xd8
8:xd8 33 xc6 E'!:dl 34 g3 E'!:xf1 + 3S
g2 8:xf3 = .
2 8 . . .e4! 29 hc6 i.c8
29 . . . ifS ! ? also appears to be
equal: 30 d7 he3 + 3 1 l2Jxe3 8:xd7
3 2 cs E'!:xdl+ 33 ltJxdl exf3 34 gxf3
e6 35 E'!:xaS hb3 36 l2Je3 8:xf3 = .
30 .id7 gf6 31 Ye5 e xf3 3 2
gd6 gxd6 3 3 Yxd6 he3+ 3 4
xe3 gxd7= .
The game is level. White can
12 CZJc2 0-0 13 a4 bxa4 14 xa4 as 15 c4 b8 16 b3 cjfh8 17 l2Jce3 g6

force a draw with 35 f4 f7 36 c4
c7 37 d4 xc3 38 d8 + g8 39
f6 + g7.
30 . . . ttJxdS 31 exdS fSt.

Our recommendation however
goes to:
23 . . .f4! ? 24 c2 d8 25 a3
g5 26 b5 lh6a:>
B2b. 22 @ht
This prophylaxis seems best.
Black proceeds with the same setup
as in the previous line:
22 i.d7 23 E:a2
..
Now the familiar 23 . . . d 8 ,

maintaining the tension, i s playable
and leads to a double-edged game:
24 exf5 gxf5 25 f4 exf4 26 ttJxf4 b6
27 ttJedS (27 e2 cs ; 27 e6 xe3
28 xe3 fe8 29 xd6 xb3 30 gl
xe6 3 1 l2Jxe6 f6 32 c6 xe6 33
c8 + g7 34 xaS xc3 =) 27 . . . CZJe S
2 8 fl d 8oo .
The diagram position is very
rich and needs practical tests . We'll
show another interesting option for
Black:
23 . . . h6 24 CZJfl d8 (you can
also try 24 . . .fxe4 25 xe4 fS 2 6
e2 e 4 27 fxe4 dToo) 25 ttJde3 c7
26 ds (26 l2Jd2 !?) 26 . . . ttJe7 27 l2Jd2
gSoo 28 CZJ dc4 (28 CZJec4 bS) 28 . . .
f4 2 9 l2J c 2 (29 ltJfl CZJxdS 30 xdS
bS) 29 . . . f6 with counterplay.
30 CZJxaS? ! would be risky due to
We think that such a position,

with a clear-cut attack against the
enemy king, should appeal to any
Sveshnikov fan. Play might conti
nue with :
27 dc7 E:f6 28 E:ad2 g4oo.
C. 1 8 Y!! e 2 !?
This innovation of Anand is the

latest hit against Black's setup . At
first it seems like a simple transpo
sition. We answer "thematically":
1 8 ...fS 1 9 h4
105
Part 4
Now we realise that in case of
19 . . . !xh4? ! 2 o exfs Jbns 21 g3 gs
22 lt:JxfS we are out of our proposed
repertoire, and even worse, play did
transpose, but to a variation which
is known to be dubious for Black. (18
h4 h4 19 g3 gs 20 V9e2 fS? !) So it
turns out, that we have to deal with
a completely new system where we
cannot rely on the usual bishop pair
to plug up the gaps in our castling
position .
19
...
.ixe3 20 xe3 f4!
It was not easy to reach this de-
106
cision. We spent a lot o f time ana

lysing the course of the stem game
20 . . . fxe4 21 hS! gS 22 V9xe4. Finally
we decided that White retains some
ad vantage. You can see more details
in the "Complete Games" section
game 21 Anand-Shirov, Linares
20 0 8 .
Commonly, i n this line Black
aims to take on e4 in order to open
the f-file and organise some play
on the kingside. In the current sit
uation, however, White had already
committed himself with 19 h4, so
short castling will hardly be good.
White will have to leave his king in
the centre, but then he will lack an
active plan such as advancing the b
pawn. We like Black's position after
20 . . . f4 . He will follow up with 21 . . . hS
to fix the weakness on h4. The game
might continue with 21 V9d3 hSoo or
21 V9d2 hS 22 f3 cj{h7 23 V9f2 g8 24
cj{e2 g7 2S hal e6 . We are eager
to see more tests ofAnand's idea. Or
was it a one-game novelty? !
-
Part 4
1 e4 cS 2 f3 c6 3 d4 cxd4
4 xd4 f6 s c3 es 6 d bS d6
7 .igS a6 8 a3 bS 9 dS .ie7 1 O
.bf& .bf6 1 1 c3 0-0 1 2 c2 .igS
1 3 a4 bxa4 1 4 gx a4 as 1 S .ic4
gb8 1 6 b3 @hS 1 7 ce3 g6
COMPLETE GAM ES
1 9 K arjak in - S h i rov
Kha nty-M a ns iysk 1 1 . 1 2. 2007
Comments by Kolev
1 e4 cs 2 liJ f3 liJ c6 3 d4 cxd4
4 liJ xd4 liJ f6 s liJ c3 es 6 liJ d bS d 6
7 .i g S a 6 8 liJ a 3 b S 9 liJ d S .ie7 1 O
.ixf6 .ixf6 1 1 c3 .igS 1 2 liJ c2 0-0
13 a4 b x a4 14 gxa4 as 1 S .ic4 gba
16 b3 @ h 8 1 7 liJ ce3 g6 18 h4 .ixh4
1 9 g3 .i g S 20 YMe2
20 . . . fS ? !
We recommend 2 0 . . . li.J e 7 2 1 f4
li.JxdS 2 2 li.JxdS exf4 2 3 gxf4 f6 .
I must confess that Shirov's play
in this game had impressed me and
I fallowed in his steps in a game of
mine. Analysing it, however, made
me change my mind . The prob-
lem is that 2 1 exfS ! is rather awk

ward. Then 2 1 . . . gxfS 2 2 \WhS b7 23
f4 exf4 24 gxf4 f6 2 5 @d2, plan
ning a4-al-gl, would give White a
clear edge since the coordination of
the black pieces is rather poor. Re
mains:
2 1 . ..ixf5 2 2 li.Jxf5
2 2 . . .gxfS
Or 2 2 . . . xfS 23 d3 xb3 (23 . . .
e 4 24 xe4; 2 3 . . . f7 2 4 xg6 g7 2 5
e4 + - xb3 2 6 xh7! ; 23 . . . \Wg8 2 4
MS gxfS 25 \Wc4+-) 2 4 hfS gxfS
25 \Wc2 \Wb8 2 6 0-0 +- .
Now 2 3 f4 i s already good and
gives White an advantage: 23 . . . exf4
24 \Wh2 b7 25 gxf4 f6 2 6 @dl, fol
lowed by a2, with a strong attack.
In the game White chooses a
wrong move order:
107
Part 4
2 1 t4 ext4 22 gxt4 .ih4+ 23 d1
E!b7
A typical defence o f the seventh
rank and particularly the sensitive
h7-square.
24 h2
It is already late for 24 exfS due
to 24 . . . ixfS 25 tt.Jxfs gxfS 2 6 gxh4
1Mixh4 27 1Mie8+ g7 28 1Mixc6 1Mihl+
29 c2 1Mig2+ and the best White
can hope for is a draw after 3 0
d3 , because 3 0 bl? would fail to
30 . . . ghs ! 31 1Mixb7+ h6 32 1Mixh7+
xh7 3 3 tt.Jf6 + h6+.
24 . . . g S 2S .i a6
2 5 exfS, as in the game Chirli
an-Kolev, 2 00 8 , is more testing:
25 . . . his 26 tt.Jxf5 f5 27 ga2 tt.Je7
I was sure this move was best,
but playing a last round I tried some
thing more complicated: 27 . . . gg7? !
2 8 1Mih3 gft7 29 ge2 tt.J e7 30 tt.Jxe7
gxe7 31 gxe7 gxe7 32 1Mif5 gg7 33
fxgS and White has good compen
sation. The game eventually ended
in a draw.
28 tt.Jxe7 (28 gd2 tt.JxdS 29 .ixdS
gc7 30 fxgS .ixgS 31 il.e4 gf4+)
2 8 . . . gxe7 2 9 fxgS .ixgS with a lev
el game: 3 0 gf2 gf6 = . 30 gg2 a4!
is also equal after the correct 31
Wfh3 ! gfeS 32 ggh2 h6 33 1Mixh6+
.ixh6 34 gxh6+ ! g7 35 gh7+ 6
36 g7h6+ = with a pretty perpetual
check.
2S . . . E!xb3 26 c2 .ixa 6 1
I n such totally unbalanced posi
tions only piece activity matters.
27 xb3 txe4 28 txg S
Or 28 gxe4? ! il.d3 29 ga4 Wfd7
and Black is ahead with his attack.
28 . . . b8+
10 8
First critical moment in the

game. White still has a draw, but he
had to demonstrate nerves of steel
and play 29 tt.Jb4 ! , for example:
29 ... axb4 30 1Mixh4 bxc3+ 31 xc3.
It turns out that Black has noth
ing decisive: 3 1 . . . gf7 32 g6 tt.J aS 33
gbl Wxbl 34 g7 + ! with perpetual, or
3 1 . . . 1Mia7 32 Wfxh7+ Wfxh7 33 xh7+
xh7 34 gxa6 gcs 35 g a4 = .
Instead, Karjakin panicked and
landed in dire straits after:
30 YMxh4 bxc3+ 31 xc3 YMa7-+
32 YM xe4 tiles 33 g 6 .id3 34 E! x h7+
YMxh 7 3S YMxeS+ dxeS 36 g xh7
ixh7
Incredibly, Shirov failed t o win
this . . . .
3 7 ttl g4 E!c8+ 3 8 b 4 e 4 3 9
ttl g t6 .its 4 0 lil h s .ig4 4 1 ttl g 3 it3
(41 . . . .ie6 42 tt.Je3 gel- +) 42 lil ts
E!c1 43 ttl c 3 h 7 44 c4 E!c2??
Second a nd last critical moment.
44 . . . g6 was easily winning. Now,
despite his big material advantage,
Black is unable to break the block
ade of the enemy's knights. It seems
to be a funny positional fortress in
the middle of the board.
4 S d4 g6 46 ttle3 E!d2+ 47
es E!d3 48 ttl ed S
12 l2Jc2 0-0 13 a4 bxa4 14 xa4 aS 15 c4 b8 16 b3 h8 17 l2Jce3 g6

1 6 b3 h8 1 7 llJ ce3 g6 18 h4 xh4
1 9 g3 g s 20 f4 exf4 21 gxf4 h4+
22 d2 llJ e7 23 c1
48 . . Jd 2 49 llJ xe4 ga2 SO liJ f4+

h6 S1 llJe6 gas+ S2 t4 d 1 S3
liJ d4 g6 S4 llJc3 h S ss llJe4 gd s
S6 e3 gda S7 llJc6 gea sa d4
gas S9 llJ es+ @ g7 60 liJd6 t6 61
llJ e4+ e 6 62 llJ cS+ @ts 63 llJ c6
gea 64 liJ b4 gda+ 6S e3 ea 66
liJ bd 3 bs 67 liJ b4 gha 68 liJ bd 3
g h 3 + 69 d 4 g h 4 + 7 0 e 3 gc4
71 liJ b3 ge4+ 72 d2 gh4 73 c3
 e4 74 liJd2+ d S 7S liJ b4+ d6
76 llJc2 gh 3+ 77 b2 dS 78 llJ a 3
a4 79 llJ c 2 @ c s 8 0 c1 g d 3 8 1
llJ e 1 gc3+ 82 b 2 b4 8 3 liJ b 1
gb3+ 8 4 c 1 g h 3 8S liJ c2+ c4 8 6
llJ e 1 g h 1 8 7 d 2 gh2+ 8 8 e3 d 1
89 liJd2+ d S 90 llJ ef3 ge2+ 91 f4
gea 92 g3 gfa 93 g 2 xf3+ 94
llJxf3 e4 9S liJ d 2+ d3 96 liJf1
gf6 97 liJ g 3 gf4 98 liJ h s gfa 99 liJ g 3
 e 3 1 00 liJf1 + e 2 1 0 1 llJ g 3+ e1
1 02 llJ e4 gf7 1 0 3 liJ g 3
%-%
20 J akoven ko- S h i rov

Foro s 27.06. 2007
1 e4 cs 2 liJf3 llJc6 3 d4 cxd4
4 llJxd4 liJ f6 s llJc3 es 6 liJ d bS d6
7 gs a6 8 llJ a3 bS 9 liJdS e7 1 0
xf6 xf6 1 1 c3 g s 1 2 llJ c2 0-0
1 3 a4 bxa4 1 4 gxa4 a S 1 S c4 gba
White's last move clears t he se

cond rank for the manoeuvre a4a2-h2 . Meanwhile the rook stays
on a4 in order to prevent a possible
sacrifice . . . a5-a4, which would dis
tract White fram the king side and
open the a-file for invasion.
23 . . . llJ xd S 24 llJ xd S e6 2S
\Wd4+ g8 26 b1
26 a2 looks more critical, but
current practice is favourable to
Black. Jakovenko intends to make
all the useful prophylactic moves
first, before committing himself to a
concrete plan. However, this game
shows that Black is not deprived of
counterplay and is also able to gene
rate threats.
26 . . .xd S 27 \Wxd S f6
27 . . . f6 ! ? is more active. Then
28 d2 g3 29 fl hS would over
take the initiative, so White should
try 28 es xf4 29 exd6 e3 ! with
double-edged play. For instance, 30
b2 ? ! el ! is better for Black. The
text is more restrained.
28 c2
In Lahno-Voiska, Turin oL 2006
10 9
Part 4
White chose 2 8 E:xaS which is rather
inconsistent. White suddenly allows
play to o pen, in case of 28 . . . .ixc3 ! 29
E: a 6 ( 2 9 E: a 7 Wf6) 2 9 . . . E:a8 ! 3 0 E:xd6
Wb8 . In the game Voiska preferred
28 . . . Wc7 and subsequently lost.
28 . . .b6 29 fif1
29 E:xaS would have given Black
a ple asant choice between (29 . . . e3
30 Wd3 Wxf4 31 Wh3 hS and
29 . . .Wf2 + 30 Wd2 Wf3 31 E:el E:a8.
29 . . . '5e3 30 '5d3
gds '\Wc7
Black i s sticking t o h i s war of
nerves. If now White repeated
with 36 E:fS, Black might deviate
by 36 . . . g7 37 eS dxeS 38 fxeS ih4
39 Wg4 f6 with sharp play. Still, his
king looks slightly safer in this line .
White, however, seems lulled by Shi
rovs repetitions and he soon misses
the oppo rtunity to force play.
36 '5h3 g a7 37 '5e3 gba 38

'5d3
I n the time trouble White mi ss
es 38 E:xd6 \Wxd6 39 Wxa7 xf4 40
xf7+ h8 4 1 ie2 ! with a likely
draw.
38 . . . gda 39 '5d2 @g7 40 gd3 h5
It is time for recapitulation.

Black won the theoretical dispute.
Now he must trade queens, but
his chances for converting the ex
tra pawn would be minimal. Shirov
steps back, probably to underline
that he is the moral winner of the
battle so far. Of course, he cannot
avoid the endgame after 30 . . . WcS 3 1
dS, but his psycho trick sudden
ly "succeed s" . The truth is that with
queens White has enough threats to
We see a typical position for

this line. Black still has an ex
tra pawn, but his rooks cannot en
tactical skills . The result of the game

proves that he was right.
ter play. At this moment, however,

the 40-moves control has passed,
and Jakovenko takes an important
(and wrong ! ) decision. He willing
ly opens up a file . . . Instead, a wait
ing game like 41 E:dS (but not 41 E:al
a4 ! ) would have offered a stubborn
defence.
30 . . . \WcS 31 g d 1 gb6 32 '5g3

gas 33 g d s '\Wc7 34 gfs '5d8 35
41 e5? dxe5 42 fxe5 ie7 43

'\Wf4 gxd 31 44 '\Wxf7+ @ h 6
keep the balance, but Shirov has al

ways been very confident of his own
110
12 '2Jc2 0-0 13 a4 bxa4 14 roca4 as 15 ic4 b8 16 b3 @h8 17 '2Jce3 g6

1 8 . . .fS 1 9 h4 .ixe3 20 Y:Yxe3
White's position amazinglycrum

bles down in a flash . 45 @xd3 1!9xe5
is rather hopeless. Perhaps Jako
venko simply missed the following
queen "sacrifice". It often happens
after a time trouble, when the con
trol is over and the player feels re
lieved. The mind takes a little nap
and stops functioning for a while.
Now White is beyond salvation.
Even the ingenious attempt 45 1!9f4+
gS 46 1!9f5 does not help, in view of
46 . . . xc3 + ! 47 @xc3 1!9d7! 48 xd7
ib4 + - + .
4 S xd3 Y:Yxc3+! 4 6 <t!? xc3 b4+
47 <t!? d4 xf7 48 e6 g g7 49 ga1 gS
SO gf1 g4 S1 <t!?e4 e7 S2 .ie2 h4
0-1
S3 ts g3
21 Anand - S h i rov
L i n are s 28.02.2008
Comments by Kolev
4 tll xd4 tll f6 s tll c 3 es 6 tll d b S d6
7 .ig S a 6 8 tll a 3 b S 9 tll d S .i e7 1 0
xf6 xf6 1 1 c3 gs 1 2 tll c2 o-o
1 3 a4 b xa4 14 gxa4 as 1 S .ic4 gb8
1 6 b3 <t!?h8 1 7 tll ce3 g 6 1 8 Y:Ye2
This is a novelty, which leads to
totally unexplored positions.
20 . . .fxe4? !
We recommend 2 0 . . .f4 ! keeping
the centre closed and eyeing the h4pa wn.
In the following commentaries
I'll try to shed more light on Shi
rov's approach. I watched the game
online and I must admit that at first
I admired his decision to open play
and stakes on attack.
21 hS!
The point o f Anand's idea. Now
most exchanges are in White's fa
vour as all endings are much bet
ter for him. Positions with only ma
jor pieces are also difficult for Black.
Thus his only hope is to get some
how to the enemy's king.
21 . . . gS 22 Y:Yxe4 .ib7
2 2 . . . ttJe 7 is a strategical mistake:
2 3 '2Jxe7 1!9xe7 24 0-0 (24 xaS gf4)
24 . . . ib7 25 d3 f4 and now sim
plest is 26 idS ! + xa4 27 bxa4.
Black's castling positions is hope
lessly loosened.
23 Y:Ye3 e4!
Played in typical Sveshnikov fa
sh ion - Black seeks maximum piece
activity on the kingside.
24 o-o til e s
111
Part 4
At the surface Black's position

looks appealing. He needs only one
move ( . . . Ve8) to hit the hS-pawn.
Should that pawn fail, Black will
have a strong attack, connected
with . . . g4 and . . . ct:Jf3, or a rook lift to
the h-file. Let us consider:
a) 2S 'M/xe4? ! puts the queen un
der pin and encourages 2S . . . ic6 26
ga3
Or 26 gaal (26 ga2 a4) 26 . . . a4!
(26 . . . gf4 27 'M/e3 gh4 2 8 f4 f4 2 9
ct:Jxf4) 27 'M/d4 axb3 2 8 gfb1 e8
with counterplay, e.g. 29
b3
gxb3 30 .ixb3 'M/xhS 31 ct:Jb4 hg2
(31 . . . gf4 ! ? 32 'M/xd6? ct:Jg4- +) 32
xg2 'M/f3+ 33 gl gf4;
26 . . . gf4 27 e 3 (27 'M/ e 2 gh4oo
28 f4? gxf4 29 ct:Jxf4 'M/b6 + - +)
27 . . . gh4 2 8 f4 (28 gfal a4 2 9 ct:Jb4
id7) 28 . . . gxf4 29 ct:Jxf4 ct:Jxc4 3 0
bxc4 gb2 3 1 ct:J g 6 + hxg6 3 2 Vh 6 +
g8 3 3 xg6+ h8 3 4 'M/h6 + = .
I suppose that Anand has not
even considered 2S 'M/xe4 ?! serious
ly. At the same time, the move he ac
tually played without much think
ing, is hardly very good, too! I had
to abandon Shirov's idea in view of
another p ossibility:
b) 2S 'M/d4!
112
I was not able to fi nd acceptable

ideas for Black. White does not ac
tually need the e4-pawn. He symply
puts his pieces on the best places,
having in mind something like gfal,
ct:Je3, possibly ie2 . The p in along
the main dark-squared diagonal is
extremely awkward for Black. Main
options now are :
2S . . . g4 2 6 ct:J e3 and Black has no
time for capturing on hS as his centre quickly falls appart: 26 . . . 'M/gS 27
gxaS gf6 2 8 ie2, while 2 6 ... gf6 27
ie2 ! ic6 28 ga3 leaves him disco
ordinated;
2S . . . e3 26 fxe3 fl+ 27 hfl ic6
28 ga3 hdS 29 'M/xdS 'M/f6 30 'M/d2
g4 31 'M/f2 with an edge;
2S . . . ic6
At first this move inspired some
hopes as it brings about interest
ing tactical possibilities after 2 6
ga2? ! I n some critical variations
the rook is hanging there! 26 . . . g4
27 ct:J e3 gf6 28 gel (28 ids hdS 2 9
'M/xdS gh6) 2 8 . . . a4 ! ( 2 8 . . . 'M/e8 29
idS hdS 30 ct:JxdS i s also possible,
but inclusion of the 28 . . . a4-break is
better: 29 bxa4 (29 b4 'M/e8 30 idS
ibS 31 c4 id7 32 bS 'M/xhS 33 .ixe4
gbf8 34 ct:Jfl gh6 3S g3 g8 36 gxa4
ct:Jf3 + 37 ixf3 gxf3 38 'M/dS+ 'M/xdS
39 cxdS .ixbS=) 29 . . . 'Ml e8 30 ie2
'M/g8 ! oo and the queen not only de
fends g4, but is also hitting the a2rook. This tempo proves to be vi
tal. Without it, Black cannot organ
ise decent counterplay. Therefore,
White should retreat to a3 :
2 6 ga3! g4 (26 . . . g7, intending
h6 , is too slow due to 27 ct:Je3) 27
ct:Je3 and I lack good advice about
what to do with Black.
12 l2Jc2 0-0 13 a4 bxa4 14 xa4 as lS c4 b8 16 b3 h8 17 l2Jce3 g6
27 . . . f6
28 e2 binds Black to
the defence of the g4-pawn, while
27 . . . a4 (this is necessary, to prevent
xaSxeS ! ) 28 bxa4 gs 29 b3 is
clearly better for White, for instance
29 . . . xb3 30 hb3 f6 31 bl g7 32
ds or 31. . . g3 32 fxg3 xg3 33 fl.
25 gfa 1 ? !
Perhaps Anand correctly as
sessed the position in White's fa
vour, and thought it was time to
collect the fruits of his opening sur
prise. His impatience, however, al
lows Shirov to fulfil his dream and
swing the queen to the kingside.
25 .. .'%Yea 26 gxa5 Wxh 5 27
Wxe4 gbe8 28 .ie2
28 ... Wh4? !
An incredible move! Being short
of time, Shirov willingly trades
queens, using the doubled h-pawn

as a ram to shatter White's pawn
shelter. This is an original and in
teresting idea, but the obvious 28 . . .
g4 ! would have been the better
choice. Then 29 d4 gS ! , threat
ening . . . e6 and g3, would be rather
unpleasant: 30 l2Je3 (30 g3? ! e6 3 1
fl ef6 ! -+ ; 30 fl e6 3 1 a7 c6
32 a6 h6 ! -+ ; 30 b4 e6t) 30 . . .
g 3 ( 3 0 . . . f4 3 1 xd6 f2 i s about
equal : 32 xf2 f4 + 3 3 f3 g3 + 34
e2 hf3+ 3S gxf3 xf3+ 36 d2
W'f2 + 37 dl W'xe3 38 d4 xd4 +
39 cxd4 g2=) 31 f3 f4. Perhaps
this position is balanced or at least
I could not find anything decisive
for either side :
32 W'b6 (32 xd6 f6 33 W'd4

W'h6 34 ltJfl l2Jc6 3S W'd7 xe2 3 6
W'xb7 ltJxaS=) 32 . . . h4 ( 3 2 . . . .ixf3 ! ?
leads t o perpetual check: 3 3 hf3
W'h6 34 ltJfl f3 3S gxf3 l2Jxf3 + 36
g2 l2Jh4+ 37 gl l2Jf3 + 38 g2 =)
33 ltJfl h6 (3 3 . . . c6 ! ?oo) 34 W'xb7
W'h4 3S l2Jxg3 xg3 36 a8 W'h2 +
with perpetual check: 37 fl W'hl +
38 2 l2Jd 3 + 39 hd3 W'h4 + 40 l
W'hl + 41 f2= .
29 Wx h4 g xh4 30 e 3 h 3 3 1
gxh3 f3+ 32 xf3 gxf3 33 g h 5
g g a + 34 t1 ggfa
113
Part 4
Black has a very strong initia
tive, but Anand plays up to the end
like a machine . . . in the good sense.
Finally it turns out that White's ex
tra pawn survives to bring him a full
point.
35 tl) d 1 g d 3 36 gh4 if3 37 gd4
g xd4 38 cxd4 gf4 3 9 tl) e 3 gxd4 40
g a4 gd3 41 gf4
114
41 . . . .ih5?
Perhaps on c6 the bishop would
have been more useful. After the
text White succeeds in trading the
powerful bishop.
42 b4 d5 43 g2 .ig6 44 tl) f5
@gs
This loses easily, but 44 . . . MS
45 f5 b3 46 f4 g7 47 h4+ is
also difficult. We know about the
rook endgames being drawish, but
. . . except the lost ones. Black's prob
lem is that his king is cut offfar from
the b-passer.
45 tl) e7+ @g7 46 tl)xg6 x g6
47 gf3 gd1 48 gb3 d4 49 f3 d 3 50
e3 g h 1 51 b5 g xh3+ 52 f3 g h 1 53
b6 ge1 + 54 xd3 gea 55 b7 gba
56 e4 h 5 57 f4
1 -0
Part 5
1 e4 c5 2 Eilf3 Eilc6 3 d4 cxd4

4 Eilxd4 Eilf6 5 Eilc3 es 6 Eildb5 d6
7 ig 5 a6
QU IC K R EPERTO I RE
In this part we deal with tricky al

ternatives to the Main line. At least,
they were tricky in the early days of
the Sveshnikov, and used to take a
heavy toll of points. In modern com
puter times, tactical gambling does
not war k, but one has to know some
basic lines.
8 Eila3
8 hf6 gxf6 9 l2Ja3 b5 ! transpo
ses to the main line.
8 b5 9 ixf6 gxf6 1 0 Eild5 .
f5
12 . . . g7 ! 13 l2Jf6 + hf6 14 xc6 +

d7 15 xd6 e7 16 0-0-0 xd6.
Black has full compensation, Mu
ratov-Timoscenko, Beltsy 1977.
b) 11 g3 g7 12 g2 fxe4 13 he4
e6 14 l!Jf6 +? (14 h5 is wiser, but
then Black gets a fine game with
simple methods: 14 .. . El!c8 15 0-0
l2Je7= .) 14 ... hf6 1 5 hc6 + @e7 16
ha8 xa8
1 1 xb5
White has also tried to use the
weakening of the a8-hl diagonal. In
these cases we should not be afraid
to part with some material for a
strong initiative :
a) 11 d3 ? ! fxe4 12 xe4
115
Part s
The raving bishops are ready
to tear White's position apart. Ad
ditional resources are the h and b
pa wn s:
17 f3 h5 ! 18 'M'e2 h4 19 0-0-0
Wc6 ! + ;
1 7 g l e4 ! ? 1 8 c 3 b4+
c) 11 ttJxb5 axb5 12 hb5 !b7 13
exf5
13 . . . !g7 14 f6 ! ? hf6 15 f3 !e7!

16 ttJb4
White regains the piece remain
ing a pawn up, but in return Black
gets a very active rook:
16 . . . c8 17 ttJxc6 Wb6 18 ttJ a7+
xb5 19 ttJxb5 hf3 20 gxf3 xc2oo.
d) 11 exf5 hf5 12 d3 e6 13
e4 !g7 14 'M'h5 c8 15 dl ttJe7
and Black shakes off the blockade
on d5.
1 1 axbS 1 2 xb 5 ga4!
...
116
You can stop here as the result

ing positions are thoroughly cha
otic and it is impossible to memo
rize everything. It is important how
ever to understand the principles of
Black's play:
1. We grab everything along the
fourth line:
13 ttJbc7+ @d7 14 0-0 xe4, or
13 b4 (intending to open files on
the queenside in case Black attempts
to hide his king there) 13 . . .xb4 14
ttJbc7+ @d7 15 c4 xc4.
2 . We hide our king on the
queenside if the b-file is closed : 13
ttJbc7+ @d7 14 0-0 xe4 15 'M'h5
ttJd4 16 c3 ttJe 2 + 17 @hl cj[c6 18 g3
@b7. Conversely, after:
13 b4 b4 14 ttJbc7+ @d7 15 0-0
g8 !
Black leaves the king in the cen

tre and adopts the principle that at
tack is the best defence:
16 ttJx b4 ttJx b4 17 ttJd5 ttJxd5 18
xd5 @e7+.
You can find a detailed analy
sis of this position in the "Com
plete Games" chapter, 22 Cres
po-San Segundo, San Sebastian
0 1 . 04.2007.
Part 5
1 e4 c5 2 tll f3 tll c 6 3 d4 cxd4

4 tll x d4 tllf 6 5 tll c 3 es 6 tll d b 5 d6
7 ig5 a6
STE P BY STEP
8 tll a 3
8 hf6 gxf6 9 lt:J a3 b5 ! transpo
ses to the main line. The move or
der with 8 hf6 gives Black the ex
tra option of playing 9 .. .f 5, which
is highly praised by many authors,
but there are certain positions we
do not like :
10 lt:J c4 ! ?
1 0 !d3 i s another good choice,
when it would be wiser to return to
familiar waters by 10 . . . b5 (10 .. Jg8
is considered attractive, but the po
sition after 11 g3 lt:Jd4 12 lt:Jd5 fxe4
13 he4 !g4 14 d3 f5 15 !g2 !h6
16 f4 does not seem satisfactory for
Black) 11 lt:Jd5 !e6 .
1 0 . . . b5 11 lt:J e 3 b4
Or 11. . . fxe4 12 lt:Jxe4 !e6 13 a4 !
lt:Jd4 14 c3 lt:Jb3 15 bl bxa4 (15 . . . d5?
16 lt:Jxd5) 16 !c4t.
12 lt:J cd5 fxe4 13 a3 ! bxa3 14 xa3
!g7 15 !b5 !;t and Black has to de
fend a slightly worse position with
no winning chances whatsoever.
This is certainly not our intention
when choosing the Sveshnikov. For
such cases they invented the Rus
sian game !
8
b5 9 .bf6
We examined 9 lt:Jd5 in the pre

vious parts. There is a third option
though, which we'll mention here:
9 lt:J abl? !
A time consuming manoeuvre,
which does not pose any problems
to Black.
9 . . . !e7 10 hf6
Or 10 a4? ! b4 11 hf6 bxc3 ! 12
he7 cxb2 13 a2 lt:Jxe7 (13 . . . xe7
14 xb2 0-0+) 14 xb2 !e6+ (Ka
linitschew)
10 . . .hf6 11 a4 b4 12 lt:Jd5 !g5
13 !c4
The setup with 13 lt:Jd2 0-0 14
lt:Jc4 (14 lt:Jf3 !h6 15 !c4 h8 in
tending f7-f5) 14 . . . !e6 15 !e2 fa
vours Black: 15 . . . lt:Jd4+.
13 . . . 0-0 14 lt:Jd2 h8 15 0-0 f5
16 f3
Perhaps White should seek equ117
Part s
ality with 16 exf5 MS 17 f3, as 17
ltJb3 aS 18 V9e2 ltJe7! gives Black the
initiative.
16 . . . ltJe7! 17 hl as 18 ltJxe7
V!Jxe7 19 ds b8 2 0 ltJc4 V9c7 21
exf5 MS 22 b3, Fillipenko-Gore
lov, 198 3 , when 2 2 . . . bd8 would
have been better for Black due to
his bishop pair.
9 gxf6 1 0 lll d 5
With this move White prepares
c3, ltJ a3-c2 . He can also relocate the
a3-knight via bl-d2 , but this ma
noeuvre does not really activate the
knight, as it is rather useless on d2 .
On the other hand, Black gets the
d4-square:
10 ltJ abl
Rogozenko called it "the move for
Sunday morning" as he had to face
it at 9 a . m. in a Bundesliga game.
But then, every move is good/bad
enough at this time of the day.
10 . . . fS! 11 ltJd 2
Alternatively:
11 a4 b4 12 ltJ dS g7 13 ltJd2 0-0
14 c4 h8 15 V!ff h S (15 0-0 fxe4 16
ltJxe4 fS+) 15 ... ltJd4 16 0-0-0 fxe4 17
ltJxe4 fS with initiative, Motylev
Gre bionkin, Internet 2 0 04;
11 g3 g7 (11 . .. ltJd4 ! ? is also ap
pealing) 12 g2 ltJe7 13 0-0 (13 exf5
MS ! 14 0-0 c8+) 13 . . . 0-0 14 exfS
(14 V!ffh S? ! b4 15 ltJdS ltJxdS 16 exdS
e4 17 c3 b8+) 14 . . . MS ! ? 15 ha8
V9xa8oo.
11. .. g7 12 V!ff h S (12 g3 b7 13
g2 ltJd4 14 0-0 0-0 15 ltJb3 e8 16
ltJdS hdS 17 exdS ltJxb3 18 axb3 e4+
Gufeld-Sveshnikov, Moscow 1973)
12 . . . fxe4 13 ltJdxe4 ltJd4 and Black
has the initiative: 14 d3 (14 0 -0-0

dS 15 ltJg3 0-0t) 14 . . . ds 15 ltJgS a7
16 0-0-0 b4 17 ltJe 2 h6 18 ltJh3 (18
ltJxd4 hxgS 19 V9e2 V9f6 20 hel
e7+) 18 . . . ltJxe2 + 19 he2 V9f6+.
10
...
fS
...
11 8
In this chapter we shall examine

some rare moves, while 11 exfS and
11 d3 will be a subject of the next
parts.
A. 11 V9d3 ? ! page 119
B . 1 1 ltJxbS page 119
C. ll hbS p age 120
D. 11 g3
page 122
E. 11 exfS
page 123
F . 1 1 c3
page 125
We should also mention 11 g4? !
This move looks more like a "mouse
slip ". It attempts to take control over
the central light squares, but in fact
it only achieves to weaken White's
positio n: 11 . . . fxe4 (11 . . . fxg4 is good
enough, too: 12 c3 g7 13 e2 hS 14
h3 gxh3 15 ltJc2 ltJ e7 16 ltJxe7 V!Jxe7
17 ltJe3 e6 18 xhS was played in
Velimirovic-Vukic, Cetinje 1990,
when 18 ... V!JgS+ would have earned
Black a small edge.) 12 g2 hS ! ?
(Sveshnikov's suggestion 1 2 . . . g8
6 ltJdbS d6 7 gS a6
is not so clear due to 13 l2Je3 ! oo) 13
h3 (13 gxhS fSt) 13 . . . hxg4 14 he4
(14 hxg4 xhl + 15 xhl fS 16 gxfS
h4 17 g2 a7!t) 14 . . . bB 15 hxg4
xhl + 16 hhl h4+.
A. 1 1 d 3?! fxe4 1 2 xe4
1 3 .J.xf6 1 4 xc6+ d7 1 5
xd6 e7 1 6 0-0-0
16 xe7 + @xe'Too is at least equal .
(16 . . . he7! ?oo is also appealing.
Here is just one, though not oblig
atory, line: 17 d3 fS 1 8 f3 0-0 19 00-0 c6 2 0 hfl ac8 21 del? ! e4
2 2 fxe4 fxe4 23 f8 + f8 24 he4
h4 ! 25 g3 gS+ 26 \t>dl d8 + 2 7
d3 f3 ++)
1 6 xd6 1 7 gxd6 e7
(17 . . . @e7! ?oo) 1 8 gds f6 1 9 gd2
.ie6.
Black has fu ll compensation ,
Muratov-Timoscenko, Beltsy 1977.
B. 1 1 xb5 axb5 1 2 xb 5
ib7 1 3 exf5
Due to the threat of 13 l2J f6 +

White wins a pawn, but i t turns out
that Black's bishop pair and superi
or development more than com pen
sate for it.
1 2 .ig7
12 . . . d7 is less energetic, but
playable : 13 f4 (13 g4? ! c8 14 h4
g7 15 0-0-0 l2Je7! 16 l2Jxe7 xe7
17 h3 c6 18 fS 0-0+ Muratov Deev, 1978) 13 . . .fS ! ? 14 f3 g7 15
hS+ <M"S 16 0-0-0 e6f 17 fxeS? !
dxeS 18 l2Jf4 f6 19 l2Jxe6+ xe6t.
1 3 f6+
13 l2Je 3 dS is a well known pawn
sacrifice which gives Black enough
compensation. (13 . . .b7! ? also seems
satisfactory, because of the delicate
position of the white queen: 14 ltJf5
f8 15 0-0-0 d7 and Black is fine.)
14 xdS (14 ltJxdS fS 15 f3 l2Jd4 16
hS + \t>f8 with initiative, according
to Sveshnikov.
White has grabbed 3 pawns for

the piece, occupied the light squares,
and now he hopes to promote his
queenside pawns in an endgame.
However, Black's piece activity
proves to be a weightier factor.
1 3 .ig7
13 . . . aS is a very natural con
tinuation and White needs precise
play to hold the balance : 14 d3 (14
a4? ! xbS 15 axbS l2Jd4 gave Black
a strong initiative in a blitz game
119
Part s
Chiburdanidze-Sveshnikov, 16 l!Je3
h6 17 0-0 gs lS l!Jg4 f4 ! -+ ; 14
c4? is warse, due to 14 . . .xbS lS
cxbS l!J d4 16 l!Je3 hg2 ! 17 l!Jxg2
\Was + lS cj/fl WxbS+ 19 cj/el Wc6 2 0
f3 l!Jc2 + 2 1 cj/f2 l!Jxal- +) 1 4 . . . g7 1S
b4 (An only move, as lS Wc4? ! cj/fS !
16 b4 l!Jd4 17 bxaS \WxaS+ lS cj/fl
WaS ! 19 l!Je3 l!JxbS favours Black)
1S . . . e4 16 Wxe4+ MS 17 bxaS \WxaS+
lS cj/d l (lS cj/fl? hal 19 hc6 hc6
2 0 We7+ cj/gS 2 1 WgS+ g7 2 2 l!Je7+
cj/fS 23 l!Jxc6 \WbS+-+) lS . . .hal 19
\We7+ cj/gS 20 WgS+ cj/f8= leads to a
curious draw.
1 4 f6!?
14 0-0? ! is too slow a nd White's
compensation seems inadequate
after 14 . . . 0-0 lS c4 (lS a4 l!Jd4+)
lS . . . WgS 16 a4 cj/hS 17 a3 WxfS
(17 . . . e4 ! ?) l S l!Jb6 (lS l!Je3 was bad
for White in Sulskis-V an Wely, Mos
cow 20 04) lS . . . adS 19 as l!Je7 2 0
a 6 c6 2 1 a 7 dSt ;
The tricky 14 Wf3? ! i s tactically
refuted by 14 . . . cj/fS ! 1S f6 (lS c3 l!Jd4
16 cxd4 \Was + - +) 1S . . . l!J d4 16 fxg7+
cj/xg7 17 Wg4+ Ms lS c4 l!Jxc2 + 19
cj/e 2 l!Jxal+.
1 4 .bt6 1 5 Wt3 ire 7! 1 6
b4
16 l!Jxe7? loses to \Was+ . After
the text White regains the piece re
maining a pawn up, but in return
Black gets a very active rook:
1 6 gc8 1 7 xc6
17 hc6+? c6 1S l!Jxc6 Wb6+.
1 7 Wb6 1 8 a7+
Alternatively:
lS l!JxeS +? WxbS 19 Wxf7+ cj/dS ;
lS l!Jxe7+ cj/xeroo . (lS . . . \WxbS??
19 l!JxcS hf3 20 l!Jxd6 ++-)
120
1 8 Wxb5 1 9 xb5 i.xf3 20

gxf3 gxc2iii
To be fair, White has all the
chances to draw the game.
C. 1 1 ixb5 axb5 1 2 xb 5
ga4
In the early days of this varia

tion Black used to play 13 . . . a7, but
later the focus of interest shifted to
the text move. It not only evades the
knight fark, but also attacks e4.
In the diagram position majo r
continuations are:
Cl. 13 llJ bc7
C2 . 13 b4
13 c4? is much weaker: 13 . . . xc4
(13 . . . WaS + ! ? 14 b4 b4 is not too
clear: lS 0-0 xbS 16 cxbS l!Jd4 17
WhS e6oo or lS l!Jf6 + cj/dS 16 0-0
bs 17 cxbS l!Jd4-+) 14 0-0 g7 lS
l!Je3 (1S l!Jf6 + MS ! 16 \Wxd6 + \Wxd6
17 l!Jxd6 d4 lS l!JxcS hf6 19 fcl
l!Jb4-+) 1S . . . d4 16 \Wc2 l!Je7 17
l!Jxd4 exd4 1S l!JxfS l!JxfS 19 exfS 0-0
2 0 a4 (2 0 acl WgS ! +) 2 0 ... WgS 21
a S .bfS+.
C 1 . 1 3 tll bc7+ <j;>d7 1 4 0-0
White has also tried:
6 tLJdbS d6 7 gs a6
a) 14 b4 xb4, when lS 0-0 trans
poses to 13 b4, whereas lS WhS?
loses to 1S . . . xe4+ 16 cj/fl Wh4 17
Wxf7+ (17 WxfS+ cj/d8 18 Wxf7 e7)
17 . . . e7- + ;
b ) 1 4 WhS? xe4+ l S cj/ fl lLJe7 16
Wxf7 cj/c6+;
c) 14 c4? ! xc4 lS 0-0 lLJ d4 ! 16
lLJb6+ ( 16 WhS xc7 17 Wxf7+ e7+) 16 . . . cj/xc7 17 lLJxc4 b7 18 cl
cj/b8 19 c3 g8 - + ;
d ) 14 exfS? ! lLJe7 l S 0 - 0 ? d4- + .
1 4 . . Jxe4 1 5 \Wh 5
1 5 . . . tl) d4
lS . . . lLJe7 16 Wxf7 cj/c6 17 c4 Wd7
18 lLJa8 ! lLJg6 19 lLJb4+ cj/b7 20 WdS+
cj/b8 21 liJc6 + ! cj/xa8 22 Wbs Wb7 2 3
Was+ Wa6 2 4 Wc7 Wb7 2S WaS += is
a well known farced drawing line.
1 6 c3
16 Wxf7 + is a consistent alterna
tive, which could be answered by:
16 . . .e7 !?
16 . . . cj/c6 17 lLJb4+ cj/b7 18 lLJbS +
Wd7 19 WdS + cj/b6 2 0 a4 ( 20 lLJxd4
xd4 21 Wb3 cj/a7-+) 20 . . . lLJxbS 2 1
axbS b4 2 2 c4 leads t o a highly un
balanced position, where 22 . . . h6
appears to be in Black's favour, but
the game remains messy.
17 liJbS
Or: 17 f3 e2 18 c3 f8 19 Wxh7
llJe6 2 0 lLJxe6 cj/xe6 21 adl b7+;
17 c3? ! Wf8 18 WhS (18 lLJf6 + cj/xc7
19 Wc4+ cj/b8 20 lLJxe4 lLJ f3 + ! 2 1
gxf3 fxe4-+) 1 8 . . . h4 ! 1 9 Wdl Wh6
20 h3 xh3 ! - + .
17 . . . Wf8 1 8 lLJf6 + ( 1 8 liJb6 + cj/d8
19 WdS lLJxbS 20 WxbS We8 ! 21 Was
Wc6+) 18 . . . cj/d8 19 Wxf8 + f8 20
lLJxe4 lLJxbSt or 2 0 . . . fxe4!? t with
Black's advantage in the sharp end
game.
1 6 . . . tl) e 2+ 1 7 h 1 c6 1 8 g 3
18 Wxf7 Wd7 1 9 Wh S lLJf4+; 1 8
ael? ! lLJf4 - + .
1 8 ... b7 1 9 gae1 gc4 20 tl) a 6 !
2 0 Wxe2 xc7 2 1 WbS+ cj/ a 8 2 2
Was + ( 2 2 lLJxc7+ Wxc7+) 2 2 . . . cj/b8
23 Wxc7+ Wxc7 24 lLJxc7 cj/xc7 is
rather grim for White.
20 . . . ie6 !
I n the game Mastrovasilis-Illes
cas, Calvia 20 04, Black went on to
win after 20 . . . cj/a8 21 b3 xc3 2 2
Wxe2 b7 2 3 cj/gl xdS 24 cl c8
(24 . . .xcl 2S xcl Was 26 b4 ! ) 2S
Wbs b7 26 Wa4 Wb6 , but White
missed his chance to draw with
27 lLJb4+ Wa7 (27 . . . cj/b8 28 xc8 +
hc8 29 cl) 28 :!xc8 + hc8 29
Wc6+ Wb7 3 0 Wa4+= .
2 1 tl) ab4 g c s 22 \Wxe2
Now Black eliminates to a slight
ly better ending, but other moves are
worse: 2 2 Wf3 e4 23 Wxe2 xdS 24
lLJxdS xdS 2S f3 e3 ! 26 Wxe3 hS+;
22 lLJe3 f4 23 Wxe2 fxe3 24 Wa6+
cj/b8 2 S liJc6 + xc6 26 Wxc6 d7+.
22 . . . ixd S + 23 tl)xd5 gxd5 24
\Wf3
Perhaps White should try 24
Wc4 ! ? d 2 2S Wxf7+ where Black
121
Part s
faces serious technical problems.
24 . . . e4 25 !8xe4 fxe4 26 Wxe4
Wb6 27 Wxd 5+ Wc6i.
Black has so me winning chances
in this endgame.
C 2 . 1 3 b4
White wants to open files on the
queenside in case Black attempts to
hide his king there .
1 3 . . . gxb4 1 4 tl) bc7+ d7 1 5 0-0
Or 15 c4 xc4 16 0-0 ltJd4 ! 17
ttJb6+ xc7 18 ttJxc4 !b7 19 cl (19
d2 g8 ! 20 as + d7 21 a7 (21
ttJb6+ e6) 2 1 . .. c7- +) 19 . . . d7!
(19 . . . b8 20 bl ! oo) 20 f3 (20 a4+
c6 2 1 a7+ e6 22 exfS+ f6 - +)
20 . . . g8+.
1 5 . . . gg a !
1 6 tl) xb4 ! ?
16 g 3 was practically refuted in
the game Luther-Leko, Essen 20 0 2:
16 . . . b7 1 7 hS gs 1 8 xf7+ e7
19 ttJxe7 ttJxe7 2 0 ttJe6 g6 2 1 llJf8 +
xf8 2 2 xf8 fxe4 23 tbl (23 a4
c7 24 as a6- +) 23 . . . c7 24 a4 e3 !
25 fxe3 e4 0- 1.
1 6 ... tl) xb4 1 7 tl) d 5
1 7 c 3 xc7 18 cxb4 b7 19 a4+
e7+.
1 7 . . . tl) x d 5 1 8 Wxd 5 e7+
122
You can find a detailed analy

sis of this position in the "Com
plete Games" chapter, 22 Cres
po-San Segundo, San Sebastian
0 1 . 0 4 . 20 07.
D. 1 1 g3 .ig7
White hurries to take a firm grip
over the light squares, but leaves his
a3-knight out of play for a long time.
Black must remember to not clutch
onto the e4-pawn with 11 . . . fxe4 and
12 . . . !fS, but continue developing.
We shall see that White's threat of
winning the exchange turns against
him.
1 2 .ig2
12 exfS hfS 13 g2 e6 usual
ly transposes to other lines. For in
stance, 14 llJf6 + ? ! xf6 1S hc6 + e7
16 xa8 xa8 is the game Solomon
Spasov, Novi Sad (ol) 1990, (see the
sub-line to move 14) where Black's
compensation for the exchange is
very strong, or 14 0-0 0-0 15 c3 b8
16 llJc2 as, which is the main line 11
c3. Finally, 14 hS c8 15 0-0 ttJe7
16 adl ttJxdS 17 xdS 0-0= leads to
the same position as in the current
main line with 12 g2 .
1 2 fxe4 1 3 xe4 .ie6
6 tlJdbS d6 7 gS a6
1 4 YMh5
Here is the first critical moment.
It is important to examine 14 tlJf6+?
f6 15 hc6 + e7 16 xa8 xa8
tlJc2 dS 20 f5 gS) 18 tlJbl h6 19 c3

aS 20 cxb4 axb4 21 a3 bxa3 22 ttJxa3
d7 23 d2 0-0+;
16 fdl ttJxdS 17 hdS 0-0 18 c3
cs 19 f3 d7! 20 d2 h6 !+, Za
pata-Illescas Cordoba, Linares 1994
1997;
16 gS? ! ttJxdS 17 xg7 f6 ! 18
xf6 ttJxf6 19 b7 c7 20 h a6 b4
21 ttJbS xc2 22 ttJxd6+ e7+.
17 gl
17 0-0 is obviously bad after 17 . . .
b4 18 tlJbl h3 1 9 f3 hfl+;
17 f3 looks playable, but 17 . . . hS !
proves the opposite: 18 e2 h4 19
0-0-0 (Or 19 2 dS ! 2 0 ael a7+
21 g2 h3 + 22 fl g7 !? 23 tlJbl
fS 24 tlJd2 d4+, intending . . .dS .)
19 . . . c6 ! 20 e3 ( 20 d2? ! hxg3 21
hxg3 xhl 22 xhl xf3+ a nd Black
went on to win in Solomon-Spasov,
Novi Sad (ol) 1990) 2 0 . . . c8 2 1 d2
b4 ! 22 tlJbl ha2+.
17 ... e4 ! ? 18 c 3 b4 The raving
bishops tear White's position apart.
19 cxb4 hb2 20 bl c3 + 21 fl
h3 + 22 g2 e3 23 f3 dS 24 tlJc2
d4+.
1 6 xdS
This exchange allows Black to
castle. 16 . . . cS? ! is premature and
hands the initiative to White after 17
ttJxe7 ! ? xe7 18 b4 c7 (or 18 . . . c3
19 tlJbl c7 20 f4-+) 19 c4 (19 d2
dS ! 20 hdS xb4=) 19 . . . bxc4 2 0
tlJ c 2 with compensation.
1 7 i.x d5 0-0 1 8 c3 gcs 1 9
i. b 3 YMd7=.
1 4 gca 1 5 o-o
15 dl ttJe7 is similar to the main
line .
1 5 e7 1 6 gad 1
White is struggling to maintain
the game level. Worse alternatives
are:
16 ttJxe7 xe7 17 adl b4 (17 . . .
h6 ! ? i s very interesting: 1 8 c3 cs 19
E. 1 1 exf5 ixf5 1 2 i.d 3

12 c3 g7 is the subject of the
next part.
12 f3? ttJd4 13 ttJ c7+ xc7 14
xa8 + e7 is yet another exam
ple where White wins the exchange
only to discover that it was a Greek
gift.
123
Part 5
cj{f6 21 g4 was the only way to pro
long the agony: 2 1 . . . i.e3 22 fxe3
i.d3 + 23 hd3 VNxd3 + 24 cj{g2 VNe2 +
25 cj{g3 VNxe3+ 2 6 cj{g2 VN e2 + 27 cj{gl
e4 ! -+ ) 20 . . . id3 + 21 hd3 VNxd3 +
22 cj{ gl !Lle2 + 23 ci>fl !Llg3+ 24 cj{gl
i.f4 ! !
15 c3
Alternatively:
a) 15 i.d3 VNa5 + 16 ci>fl hd3 + 17
cxd3 VNd2 18 VNe4 fS 19 VNe3 VNxb2 20
g e l cj{f7 ! + ;
b) 1 5 gdl !Llxc2 + 1 6 !Llxc2 hc2
17 VNd5 (17 gd2 i.h6 ; 17 gd5 VNa5+
18 cj(e2 VNxa2+) 17 . . . hdl 18 VNxdl
i.h6 - +
15 . . . b4 ! 16 cxb4 VNb6 ( 1 6 . . .i.h6 !?
i s an interesting option as well: 17
VNxa6 gb8 ! 18 i.c4 i.e4 ! t with a
strong initiative, e.g. 19 VN a5? VNxa5
20 bxa5 hg2 21 ggl gxb 2-+) 17
ha6 (17 i.c4 VNxb4+ 18 cj{fl VNxb2
19 gel VNxa3 2 0 VNa7+ cj(d8 21 VNb8 +
i.c8 2 2 VNb6 + cj{ d 7 2 3 VNa7+ cj(c6 24
VNa 8 + i.b7 25 i.d5+ ( 2 5 VNe8 + cj(b6)
25 . . . cj(c5 ! (threatening mate VNd 3)
26 VNa7+ cj(xd5 27 VNxb7+ cj(e6 2 8
VNc8 + cj{f6 - +)
17 . . . VNxb4+ 18 cj{fl (18 cj(d l VNa4 +)
18 ... VN d 2 ! t . Amazingly, this position
is still ocurring in tournaments, so
we shall give more details. To be
franc, we could not resist the temp
tation to show the exquisite mate on
the next diagram:
a) 19 gel?? id3 + 20 hd3 VNxd3+
21 cj{gl !Lle 2 + 2 2 ci>fl !Llg3+ 23 cj{gl
VNfl+ 24 fufl !Lle2 # ;
b ) 1 9 h 3 i.h6 2 0 VNxh8 ( 2 0 VNb7+
124
White has n o defence against the

mating threat of 25 . . . VNfl ! !
c) 19 h4 ! ih6 20 VNxh8 (20 VNb7+
cj{f6 21 g3 gc8 - + , intending . . . gc6
and i.e4, for instance, 22 cj{ gl gc6)
20 . . . VNxb2 21 gdl VNxa3 22 ic4 VNa4 !
23 gxd4 (23 ib3 VNb5+ - +) 23 . . . exd4
24 ib3 VNb4 . Black has full compen
sation for the exchange.
12
...
ie6 1 3 ie4
13 VNf3 counts on the "trap"

13 . . .i.g7 14 !Llf4, (14 !Llf6 + ?! i.xf6
15 VNxc6 + cj{e7+) but it turns out to
be in Black's favour after 14 . . . exf4
15 VNxc6 + ci>f8 ! 16 0-0-0 gc8t. Per
haps simpler is 13 . . . hd5 ! ? 14 VNxd5
!Lle7 15 VNb7 (15 VNf3 d5 16 VNf6 ? ! gg8
17 !Llxb5 i.g7 18 !Lld6+ cj{d7 19 VNxf7
cj(xd6+) 15 . . . i.g7 16 0-0 d5 with a
considerable space advantage: 17
gadl VNb 8 18 VNxb8 + gxb8 19 c3 b4
20 cxb4 gxb4 21 b3 gb6+.
13
...
ig7
6 llJdbS d6 7 gS a6
1 6 gcs 1 7 lll x e7
Alternatives:
17 tLJb4? ! b6 18 gs f8 19
e3 fS with an edge since White's
bishop cannot retreat to g2 ;
17 gS? ! llJxdS 18 xg7 f6 19
xf6 llJxf6. The ending favours
Black, for 2 0 b7? ! e7 21 ha6
loses material after 22 . . . b8 22 d3
b6 23 b4 xa6 24 bxcS xa3 - + .
1 7 xe7 18 lll c 2
After 18 0-0 dS 19 ifs hfS 20
xfS e6 21 hS 0-0 2 2 d3 fS 2 3
fdl d4 ! ? 24 cxd4 exd4+ the only
use of the a3-knight is to protect the
c2- square.
1 8 d S 1 9 i.f5 d4!?
Black has successfully passed the
fifth rank with his pawn and does
not risk to be cramped any longer.
20 0-0 dxc3 21 bxc3 i.xf5
22 xf5 0-0 23 lll e 3 e4=.
1 4 h5
More th an once we observed
White struggling with the extra ex
change : 14 llJf6 + M6 1S hc6 + e7
16 ha8 (or 16 c3 c8 17 ids b4 ! ? 18
cxb4 b6 19 0-0 xb4+) 16 ... xa8
17 0-0 g8 18 f3 (18 g3 b4 19 llJbl
h3 20 f3 hfl 2 1 xfl e4t) 18 . . . ih3
19 f2 a7 20 g3 igS+.
14 llJe3 is more reasonable, but
still Black obtains active play, for ex
ample, 14 . . . d7 (14 . . . c8 15 c3 llJe7
16 ib7 b8 17 ha6 d7 !oo is also
worth considering) 15 0-0 0-0 16
dS ad8 17 c3 llJe7 18 llJ ac2 fSf .
14 gc8 1 5 gd1
15 c3 runs into a forced line: 15 . . .
b4 ! 16 llJc2 ( 1 6 cxb4 llJxb4 ! 17 llJxb4
as 18 llJ ac2 xc2 19 hc2 xb4+
20 fl xb2 21 dl 0-0 2 2 a4 c8+)
16 . . .bxc3 17 bxc3 as 18 llJce3 llJe7
19 d l llJxdS 20 llJxdS c4 ! 21 e2
(21 f3 fS ! 22 MS f8 23 g4 e4+) 2 1 . . . xdS 22 hdS xc3+ 23 fl
d4+.
1 5 lll e7 1 6 c3
Similar is 16 llJxe7 xe7 17 0-0
( 17 c3 b4 ! +) 17 ... cS ! , when 18 b4
c 3 19 llJbl c4 20 llJd2 xb4 21
ic6 + f8 2 2 llJe4 loses to 2 2 ... dS
23 hdS hdS 24 dS xe4.
F. 1 1 c3 i.g7
White commonly uses this move

order to avoid 11 exfS MS 12 c3 ie6
13 llJc2 h6 , which is (albeit playa
ble) outside our repertoire anyway.
1 2 .id3
In most games you will face here
125
Part s
12 exfS which we analyse in the next
part of the book.
12 l!Jc2? ! is an ambitious attempt
to seize control over the kingside
light squares after 12 . . . fxe4 13 l!Jce3
e6 14 g4. Black should refrain from
castling and cut across the enemy
plan with 14 . . . l!Je7 15 ig2 l!JxdS 16
l!JxdS hS with the better game, Ya
kovich.
12 l!JxbS? is a dubious sacri
fice: 12 . . . axbS 13 .txbs b7 14 l!Jb4
(14 exfS f8 15 0-0 l!J e7+) 14 . . . d7
126
1 5 exfS f8 ! 16 0-0 g8 17 g3 as
18 hc6 hc6 19 l!Jxc6 xc6+ A.
Sokolov-Lautier, Val d'Isere 2 0 04.
12 exfS transposes to the next
part of the book.
1 2 .ie6
After this move play transposes
to line A of Part 7.
Black can also choose as a back
up line 12 . . . l!Je7, which leads to the
Novosibirsk variation. It is the sub
ject of Part 13.
...
Part S

4 lll x d4 lll f 6 s lll c 3 es 6 lll dbS d6
7 .igS a6
COMP LETE GAM ES
22 C re spo - S an Seg u ndo

San Seba stian 0 1 .04.2007
1 e4 cs 2 li:)f3 li:)c6 3 d4 cxd4
4 li:)xd4 li:) f6 s li:) c3 e s 6 li:) d b S d6
7 .igS a6 8 li:) a3 bS 9 ixf6 g xf6
1 0 li:) d S fS 1 1 .ixbS axb S 1 2 li:) xb S
g a4 ! 1 3 b4 g x b4 14 li:) bc7+ d 7
1 s o-o g ga !
Black seeks t o weaken the enemy

castling position. In the future that
might be important, as in Luther
Leko, Essen 200 2: 16 g3 b7 17 hS
gS 18 xf7+ ie7 19 Cfjxe7 ljj xe7 2 0
ljj e 6 (Following 2 0 ljj dS Black has
a pleasant choice between 20 . . . d8
or 20 . . . g7 with a nearly winning
position.) 20 . . . g6 2 1 ljj f8 + :B:xf8 2 2
xf8 fxe4 2 3 tbl ( 2 3 a 4 c7 2 4 as
ia6 - +) 23 . . . c7 24 a4 e3 ! 25 fxe3
e4 0-1 Luther-Leko, Essen 200 2 .

1 6 li:) x b4 li:) x b 4 1 7 li:) d S
O r 1 7 c 3 xc7 1 8 cxb4 b7 1 9
a4+ e7+ .
1 7 ... li:) xd S 1 8 'Wxd S e7
1 9 g a b1
White hopes t o generate some
threats using the open b-file, but
the course of the game shows that
Black manages to consolidate and
his advantage soon becomes deci
sive. Let us examine:
19 a4 f4
Like in the game, Black stakes
on the attack. 19 . . . c7!?, threaten
ing to put the bishop on the long
diagonal, is another appealing op
tion . Then 20 aS? would lose to
20 . . . ib7- + , so White should con-
127
Part s
tinue with 2 0 E:fbl (or 2 0 E:abl fxe4
2 1 as h6 ! 22 xe4 E:g4 23 xh7
c6+) 2 0 . . . fxe4 ! 21 as (2 1 xe4 E:g4
22 a8 d7+) 21 . . . e3 22 a6 a7!+
a nd Black arrives just in time to
stop the passer.
20 hl (20 E:a3 e6 21 b7+
f6 2 2 aS dS-+) 2 0 . . . d7 21 f3 (21
as E:xg2 ! ) 21. .. a7!
The queen blockades the a-pawn

and restricts the fl-rook in view of
the threat f2, followed by h3 . It
is important to keep this possibili
ty because the line 2 1 . . .a6 22 E:fbl
a7 23 c6 f2 24 E:gl e2 25 as
M3 26 c7+ 6 27 d8 + = leads
only to a draw.
22 as f6 23 a6 (23 b3 a6
24 E:gl e7 25 b6 xb6 26 axb6
b7 27 E:a7 E:b8 28 E:bl e6-+)
23 . . .e7
It seems that Black is unable to
make any further progress . . .
2 4 c6 e6 2 5 b7, ( 2 5 E:abl
E:c8 2 6 b7 E:c7- +) but h e possess
es a study-like solution:
2S . . . e3 ! 2 6 E:gl
26 a7 loses again to 26 . . . e2 27
E:gl h3 .
128
26 . . . f2 ! (threatening h3) 2 7
bS (intending fl) 2 7 . . . E:g6 ! 2 8 a 7
g3 ! - + .
A very amusing positio n ! Black

abandoned the whole queenside,
but his rook is enough to finish the
game. The conclusion of this anal
ysis is that Black has the edge after
his move 18.
1 9 . . . f4
19 . . f6 ! ? favours Black as well:
20 exfS c7 21 f4 h6+.
20 gb3 f6
20 . . . e6 21 b7+ f6 22 E:c3
aS-+ was also posible.
2 1 gc3 ie6 22 c6 ie7 23 a4
as-+
The a-pawn is stopped and White
is helpless . The rest of the game is
irrelevant.
24 gb3 .ix b3 2S cxb3 cs 2 6
gc1 xc6 27 gxc6 g b 8 2 8 g c 3 g b4
29 f3 id8 30 f2 ib6+ 31 f1 id4
32 gd3 e 6 33 g 3 fS 34 g2 fxg 3
3 S x g 3 fxe4 36 fxe4 d S 37 exdS+
xd S 38 g 2 gb7 39 f1 e4 40
e2 gg7 41 gh3 gg2+ 42 f1 gf2+
43 e1 gb2 44 gxh7 d3 4S gh3+
ie3 46 f 1 g b 1 +
0-1
Part 6
1 e4 c5 2 f3 tll c6 3 d4 cxd4
4 tll x d4 f6 5 tll c3 e5 6 db5 d6
7 i.g5 a6 8 a3 b5 9 ixf6 gxf6
1 0 d5 f5 1 1 exf5 i.xf5
QUICK REPERTO I RE
This part is devoted to one of the

three most important pawn struc
tures in the Sveshnikov. The moves
obtains the d4-square and his ini

tiative shapes up in a direct attack:
1 2 c3 .ig7 1 3.tll c 2 0-0 1 4

tll c e3 i.e6
lead to an open position with a
mobile pawn centre and a bishop
pair for Black.
19 . . .fS ! 2 0 gxfS xdS 21 llJxdS
White hopes to seize control of

the light squares and put pressure
on d6. However, the tour of his
knight around the board cost six ( ! )
tempi, and even one more undevel
oping move sets up the ground for
tactical blows:
15 g4 b4 !
Remember this pawn thrust !
It is a thematic way of shattering
White's position. In case of 16 cxb4
gb8 17 a3 as 18 bS llJ d4 19 a4 Black
gxfSt, Rodriguez-Jussupow, Am
sterdam 1978 .
15 a4? ! also encounters 15 . . . b4 !
16 cxb4 fS, whilst 15 g3 provides a
lever on the kingside, which can be
used by 15 .. .fS 16 ig2 f4 !? 17 llJc2
ifS ! ?
These examples suggest that
White must develop without allow
ing any pawn weakness :
1 5 .id3 f5 1 6 0-0
16 WfhS, intending long castling
or g4, is certainly appealing, but
again the lack of development
tells :
16 . . . e4 17 ic2 llJe7!
129
Part 6
White's queen deprived the dS

knight of support. Accordingly, we
seize the chance to break the block
ade and set our central pawn cluster
moving. The b4-break is an impor
tant additional resource:
18 ib3 f4 19 ltJxe7+ W!xe7 2 0
ltJfS 'Wf6 2 1 g 4 h8 2 2 ltJxg7 hb3
23 axb3 'Wxg7 24 0-0-0 b4 !+, Ljubo
jevic-Shirov, Monte Carlo blindfold,
20 0 3 .
1 8 ltJf4 i f7 1 9 ib3 dS, Anand
Kramnik, Frankfurt 2000,
1 8 dl b4 ! 19 ltJxb4 a S 2 0 ltJbdS
ltJxdS 21 ltJxdS b8 22 ib3 a4
Black has sufficient counterplay,
see a detailed analysis of the stem
game 23 Fressinet-Gelfand, En
ghien les Bains 2 003 in the "Com
plete Games" chapter.
1 6 <at?hS
130
It seems that Black's pawns will

shortly overrun the enemy army,
but it is not so simple to achieve
that. 16 . . .f4 stumbles into 17 'WhS
and 16 . . . e4 into 17 ltJf4.
After the text move, White has to
redeploy his pieces in order to op
pose the enemy threats.
17 f4? ! ltJe7 18 ic2 ltJxdS 19 ltJxdS
c8 20 ib3 aS only helped Black
developing an initiative in K. Geor
giev-Van Wely, Bled ol. 20 0 2 ;
1 7 ic2 i s well met with 1 7. . . ltJe7 18
ib3 ltJg6 ! 19 f4 exf4 20 ltJc2 ieS = .
Most popular is:
1 7 h5 e4 1 8 i.c2 e7! 1 9
gad 1 i.f7 !? 2 0 h3 xd 5 2 1
xd 5 g 5 !
Black's active pieces assure him
of a good game, 24 Olsson-Spa
sov, EU-chT Gothenburg 200S.
Part 6
1 e4 c5 2 lll f 3 lll c 6 3 d4 cxd4

4 lll x d4 lll f6 5 lll c 3 es 6 lll db5 d6
7 .ig 5 a6 8 lll a 3 b5 9 ixt6 gxf6
1 O lll d 5 f5 1 1 exf5 ixf5
STE P BY STEP
1 2 c3 ig7 1 3 lll c 2
Alternatives:
13 f3? ! e6 ;
13 d3 e4, followed up by . . . 0-0
a nd ... b4;
13 g3? ! e4 ! (14 g4 xdS 15
Wfxg7 f8 16 gl b4 17 ltJc4 bxc3 18
bxc3 c7 19 Wff6 ie6+) 14 f3 xdS
15 WfxdS ltJe7 16 Wfd2 dS 17 ltJc2 0-0
18 i g 2 Wfb6 ! + .
1 3 0-0
13 . . .e6 could be a good backup
line. This move anticipates 14 ltJce3
e6 15 id3, which is a major possi
bility after 13 . . . 0-0 . The fine point
is that 13 ... i.e6 14 ce3 does not
simply transpose to our repertoire
following 14 . . . 0-0, since Black has
the interesting option 14 . . . ltJe7.
Now 15 g3 is not so good, as

White's knight is already commit
ted to e3. This enhances the effect
of . . .f5-f4.
In practice White has also tried:
15 a4 ltJxdS 16 ltJxdS 0-0 17 ie2
bxa4 18 xa4 as 19 0-0 b8 20 b4
axb4 21 cxb4 e4 22 bS Wf gS ! which
is satisfactory for Black;
15 ltJxe7 xe7 16 d3 dS 17 0-0
0-0 18 WfhS, when Black occupies
the centre and succeeds in defend
ing it by tactical means: 18 . . . e4 19
c2 fS 2 0 b3 ad8 2 1 adl Wies 2 2
d2 @h 8 23 fdl f4 2 4 ltJxdS e3 !+.
White can also attack 13 . . . e6
with 14 a4. This is a minor op
tion, which has been occasional
ly tried by Kasparov, Anand, Topa
lov, Leko. The best answer is:
14 . . . 0-0 ! 15 axbS (Or 15 ltJce3
ltJe7! 16 g3 fS ! 17 g2 f4 18 ltJxe7+
xe7 19 ltJdS f7 20 gxf4 exf4 2 1
f3 h3 ! 2 2 axbS ae8+ 2 3 Wd2
axbS 24 as est, Ding Yixin-Zhao
Jun, Shandong 20 0 7.)
lS ... axbS 16 xa8 Wfxa8
131
Part 6
17 ltJce3
The tactical background of Black's
idea is seen in the variation 17
hbS? ltJd4 18 ltJe7+ h8 19 cxd4
xg2 2 0 c6 e4 2 1 :gfl gs . A Domi
nation theme. The e7-knight is trap
ped. White can attempt a counter
strike with 22 ltJdS hdS 23 h4, but
Black's queen completes his full
tour around the board, to arrive vic
toriously in the centre: 23 . . . d8 24
hdS as+ 2S d2 xdS+. 17 ltJc7
a2 18 xbS is not satisfactory ei
ther: 18 . . . ltJ a7! ? 19 ltJxe6 fxe6 2 0
d3 xb2 2 1 hS ( 2 1 0 - 0 xc3+)
2 1 . . .xc3 + 22 e2 h6. White's at
tack is not impressive, for example,
23 g6 f6 24 e 8+ f8 2S d7! e4
26 he4 c4 + 27 d3 g4+ 28 fl
f7 2 9 e8 f4+.
17 . . . ltJd4! 18 ltJc7
Or 18 ltJe7+ h8 19 cxd4 exd4
20 ltJ3fS e4+ 2 1 e2 MS 22 lLJxfS
xfS = .
18 . . . a2 1 9 ltJxe6 fxe6 !
This important novelty brings

about a sharp endgame with fine
compensation for Black. The game
Azarov-Nedev, Turin (ol) 2006
saw further 2 0 cxd4 xb2 ! 2 1 d2
(White must seek to trade queens
before the f8-rook came into play, as
in the variation 21 ltJc2 e4 ! 22 d2
b4 23 ltJxb4 bl+ 24 e2 b8 -+)
2 1 . .. xd 2+ 2 2 xd2 f2 + 2 3 el
a2 24 dxeS (24 hbS exd4 2S c4
al+ 26 ltJdl dS 27 e2 eSco; or 24
d S h6 !) 24 . . . heS 2S e2 al+co.
The only drawback of the
move order with 13 ... i.e6 is
that it enables the fianchetto
14 g3 , which leads to a balanced
positional game.
By lea ving the bishop on fS, Black
prevents 14 g3 in view of 14 . . . e4.
1 4 lll c e3
14 a4 is seldom seen: 14 ... ltJe7
(14 . . . e6 is also known to lead to
equality) lS d3 (lS ltJxe7+ ? ! xe7
16 ttJb4 e4 ! 17 f3 b7 18 axbS axbS
19 xa8 xa8 2 0 hbS dS ! +) 1S . . . e6
16 e4 (16 hS fS 17 ltJxe7 + xe7
18 axbS e4 19 e2 f4 was better for
Black in Santo Roman-San Segun
do, Moscow ol. 1994) 16 . . . c8= ,
Anand-Khalifman, Belgrade 1999;
14 .ie6 1 5 .id 3
The alternatives are clearly infenor:
a) lS g3 is quite popular, but
White's knight on e3 is begging to
be attacked with lS .. .fS 16 g2
Or 16 h3 b4 17 0-0 (17 cxb4
b8 18 a3 ltJd4 assures Black of the
initiative: 19 0-0 e4 or 19 . . . aS ! ? 20
bxaS? ! xb2) 17 0-0 bxc3 18 bxc3
as 19 bl h8+.
13 2
7 gs a6 8 ttJ a3 bS 9 hf6 gxf6 10 ltJdS fS 11 exfS hfS

16 . . .f4 ! ? 17 ltJc2 fS ! ?
This pawn sacrifice was intro

duced in the game Dragiev-Chepa
rinov, Sofia 2003. After 18 ltJxf4
exf4 19 hc6 e7+ 20 cj{fl h3 + 2 1
g2 (or 2 1 @gl ac8 2 2 !f3 cs 2 3
ltJd4 gSoo, Hunt-Berkvens, Esbjerg
2 003) 2 1 . . . e4 22 f3 Black should
have opted for 22 . . . c4+ 23 @f2
e6 24 el cs + 2S ltJd4 es 26 a3
aSoo with good control of the board.
b) lS a4? ! encounters 1S . . . b4 ! 16
cxb4 fS 17 cl (17 c4 is hardly bet
ter: 17 .. .f4 18 ltJc2 18 ltJb6 xb6 19
he6 + @h8 20 ltJdS b7+) 17 . . . ltJd4
18 ltJc7 fl
Afunnyposition. White's king got

stuck in the centre and Black's mo
bile pawns will soon sweep away his
last defenders . Look at the instruc
tive game Franchini-Timoshenko ,
Thessalo niki 26 . 0 8 . 2 007:
19 ltJedS (19 ltJxa8 xa8 2 0 c4

hc4 2 1 fuc4 f4) 19 . . . @h8 20 g3 e4
21 e2 a7 22 0-0 f4 23 hS hdS
24 ltJxdS f3 2S c4 ltJe2+ 26 @hl
gs 27 g4 h4 2 8 xe4 !eS 2 9 xeS
dxeS 30 d2 g7 3 1 ltJe3 d8 32 c2
ltJf4 33 d l dg8 34 ltJfl xhS 0-1.
c) lS g4 White is undeveloped for
such ambitious moves. Of course,
Black must organize an attack very
quickly before the enemy consoli
dated :
lS . . . b4 ! 16 g2
16 cxb4 b8 17 a3 aS occured
in the game Rodriguez-Jussupow,
Amsterdam 1978 : 18 bS ltJd4 19 a4
fS 2 0 gxfS hdS 21 ltJxdS xfSt.
16 ... bxc3 17 bxc3 c8 18 d3
e4 ! ? 19 he4 ltJeS, J. Polgar-Gel
fand, Pacs 2003, when 20 xa6
would have been the only move .
Even then, Black has strong com
pensation after 20 . . . ltJxg4 21 ltJxg4
hg4oo. (Rogozenko)
15
...
fS
Our main line branches here to :

A. 16 a4
page 134
B. 16 hS
page 134
C. 16 c2
page 13S
page 136
D. 16 0-0
133
Part 6
A. 1 6 a4
White is unnecessaryly provok
ing the following sacrifice:
1 6 ... b4! 1 7 0-0
17 l!Jxb4? is very bad due to
17 . . . l!Jxb4 18 cxb4 e4 19 c4 hc4
20 l!Jxc4 dS+. 17 cxb4? ! e4 is not any
better either.
1 7 ... @ h8!? 1 8 ic2
Or 18 cxb4 e4 19 l!Jf4 (or 19 !c2
hb2 19 !c4 f4t) 19 . . .d7 20 !c4
hb2 intending . . .es with unclear
play.
1 8 ... bxc3 1 9 bxc3 lll e7
Black has a fine game, Barua
Sermek, Calcutta 2 0 0 2 . It went
2 0 bl l!JxdS 21 l!JxdS c8 22 b7
cs 23 !b3 Wa8 24 llJc7 l&xb7 2S
l!Jxe6 xc3 2 6 dS (26 l!Jxf8? l&xb3
27 Wxd6 Wa3 2 8 Wxa3 xa3 2 9
l!Je6 xa4 3 0 bl ctt g 8 3 1 h 3 !f6+)
2 6 . . . 1We7 2 7 l!Jxf8 Wxf8 2 8 l&hS cS
29 dl e4+.
B. 1 6 V9h5
A logical attempt to highlight
the weaknesses in Black's castling
position. An imminent threat is g4.
Black has no choice, but to strike
first.
1 6 ... e4 1 7 ic2 lll e7! 1 8 E!d1
Other continuations do not pose
any problems to Black:
1 8 l!Jxe7? ! + l&xe7 1 9 b3 f4 ! 2 0
l!Jds Wb7 2 1 0-0-0 as ;
1 8 !b3 f4 19 l!Jxe7+ Wxe7 2 0 l!JfS
(2 0 l!JdS l&b7!) 2 0 . . . Wf6 2 1 g4 ctt h 8
22 he6 (22 l!Jxg7 hb3 23 axb3
l&xg7 24 0-0-0 b4! 2S c4 aS+, Ljubo
jevic-Shirov, Monte Carlo blind
fold, 200 3 . ) 22 . . . Wxe6 23 0-0-0 b4 !
1 34
24 l!Jxg7 Wxa2 ! 2S WdSD WxdS 2 6

xdS cttx g7 2 7 xd6 bxc3 2 8 bxc3= ;
1 8 l!Jf4 f7 1 9 b3 dS 2 0 gS
l!Jg6 ! 21 Wxd8 axd8 22 l!JexdS
l!Jxf4 23 l!Jxf4 b4 24 l!Je6 he6 2 S
he6 + , Anand-Kramnik, Frankfurt
2 0 0 0 , 2 S . . . ctt h 8 = .
1 8. . . b4!
We often meet this thematic

thrust in the Sveshnikov. It allows
Black to escape the bind by opening
files on the queenside.
1 9 lll x b4
In the first game where White
encountered 18 . . . b4, he chose 19
cxb4? ! l!JxdS 2 0 l!JxdS hb2 21 0-0
es 22 ctt h l ctt h 8 23 f3 c8t , Go
loshchapov-Volzhin, Moscow 1999.
1 9 0-0 bxc3 20 bxc3 ctt h8 is level
since the weakness of c3 balances
the defects of Black's pawn forma
tio n: 2 1 f3 l!JxdS 22 l!JxdS hdS 2 3
xdS Wb6 + 2 4 ctt h l Wb2 2S b3 e3
26 xd6 e2 27 el ae8= .
1 9 ... a s 2 0 lll bd5 lll x d5 2 1
lll x d5 E! b 8 2 2 i b 3 a 4 2 3 ixa4
gxb2 24 ib3 @h8!
Black has sufficient counterplay,
7 igS a6 8 l2Ja3 bS 9 hf6 gxf6 10 llJdS f5 11 exf5 hfS

see a detailed analysis of the stem
game 23 Fressinet-Gelfand, En
ghien les Bains 2003 in the "Com
plete Games" chapter.
C. 1 6 i.c2
White intends to redeploy the
bishop to the a2-g8 diagonal. We
should energetically oppose this
or we risk to end up with a slight
ly worse position, as in the case of
16 . . . h8 17 hS e4 18 l2Jf4 f6 19
0-0 l2Je7 20 ib3 .
1 6 ...f4
This should lead to a more or
less farced draw. Black can deviate
at his own risk by 16 . . . E!a7 ! ? 17 hS
E!af7
18 g4
18 0-0 l2Je7 19 ib3 l2Jg6f! is OK
for Black.
18 . . .hdS 19 llJxdS e4 20 0-0-0 !
The most persistent continua
tion. 20 l2Je3 b4 ! ; 20 l2Jf4 aSoo or
2 0 f4 b4 2 1 ib3 bxc3 22 l2Jxc3 d5 ! 23
0-0-0 (23 llJxdS aS+---+) 23 . . .ixc3
24 bxc3 l2Je7+ are in Black's favour,
but the text is a sterner test . The
only sensible answer is:
20 . . . b4oo with a very complicat
ed position:
21 E!hgl bxc3 (21. .. llJeS 22 l2Jxb4)
22 gxfS. Black's king looks more

vulnerable.
16 .. . h8 is another way to es
cape the draw. Then 17 hS e4 18
E!dl would transpose to game 24
Olsson-Spasov, EU-chT Gothen
burg 2 0 05 after 18 . . . l2Je7 19 0-0 if7
2 0 h3 llJxdS 2 1 llJxdS gS, but 18
l2Jf4 ! ? f6 19 0-0 l2Je7 20 ib3 is
more unpleasant.
1 7 V;YhS
17 llJfl? ! is too timid and hands
the initiative to Black: 17 . . . l2J e7 18
ie4 (18 ib3 h8+) 18 . . . E!c8 19 d3
E!cSt. (or 19 . . . h8 !?)
1 7 gf7
1 8 bh7+
Alternatively:
a) 18 l2Jg4? hdS 19 xh7 + f8
20 ig6 (2 0 l2Jh6 E!c7 2 1 llJfS f6 - +)
20 . . . E!c7- + ;
b) 18 xh7+ f8 19 g6 (19 ifS
e8 ; 19 ig6 fxe3 2 0 hf7 exf2 + 2 1
xf2 hf7 2 2 E!hfl hdS 2 3 gl+
if7 24 E!xf7+ xf7 25 E!fl+ e6 26
xg7 l2Je7- +) 19 . . . e8 20 ie4 fxe3
21 fxe3 (21 l2Jxe3 E!c8+) 2 1 . . . b4 ! ? 2 2
c4 ( 2 2 h 4 bxc3 2 3 bxc3 b8 24 hS
c8 25 h6 ih8 26 h7 l2Je7 27 h6 +
e 8 2 8 xe6 xe6 29 l2J c7+ d7
30 l2Jxe6 xe6+) 22 . . . E!b8 23 h4 (23
135
Part 6
0-0-0 b3 24 a3 ttJ aS !) 23 . . . d7 24
hS ttJe7 2S ttJxe7 cj{xe7 26 h6 if6+.
1a
ta 1 9 .tts
Thus White rescues his hanging

knights. The other moves lose:
19 g6? e8 20 ttJ b6 ( 20 ttJg4
hdS 21 xd6+ ttJ e7- +) 20 . . . fxe3
21 ttJx a8 exf2 + 22 d2 e7 ! - + ;
1 9 ltJg4? hdS 20 ltJ h 6 f6 2 1
ttJg4 if7-+ .
1 7 YN h S
Black's last move drew the sting
of 17 f4? ! in view of 17 . . . ttJe7 18
ic2 (18 ttJxe7 xe7 19 fxeS? a7 ! )
1 8 . . . ttJxdS 1 9 ltJxdS c8 2 0 ib3 aS
21 a3? ! e8 22 ttJe3 exf4 23 ttJdS
22 ttJb6 dB 23 ttJfS (23 ttJedS

e4 24 0-0-0? ttJeS) 23 . . . e6 24 ttJ e3
h6 = is similar to the main line,
but Black might also try 22 . . . fxe3
fl+, K. Georgiev-Van Wely, Bled

ol . 2 0 0 2 ;
1 7 ic2 i s also well met by
17 . . . ttJe7 18 ib3 ltJg6 ! 19 f4 exf4 2 0
ttJc2 ieS= .
1 7a 4 b4 i s already familiar. Black
has good co mpensation following 18
cxb4 e4 19 ttJf4 ig8 20 ha6 ttJxb4 !
21 ic4 hc4 2 2 ttJxc4 dS 23 ttJe3 d4
24 ttJe6 b6 2S ttJxf8 dxe3t .
23 ttJxa8 exf2 + 24 e2 dSoo .

Now the tactical clash ends up
with a repetition:
1 7 e4 1 8 .tc2 e 7 ! 1 9
gad 1 .tf7 !?
1 9 YN eS 2 0 .txe6 YNxe6 2 1
YNg4 YN h6 2 2 ts

22
YN e6 23 fe3 YN h6=.
D. 1 6 0-0 h8
Black commonly equalises here

with 19 . . . cB, followed up by a dou
ble exchange on dS, but we prefer
active play on the kingside. Accor
digly, we want to place the queen's
rook on e8. The text is meant to en
able . . . gs .
20
3 xd5 21 xd5 YN gS!
Black's active pieces assure h im

of a good game . You can find a de
tailed analysis of this position in the
"Complete Games" section, 24 Ols
son-Spasov, EU-chT Gothenburg
2 0 0S .
136
Part 6
1 e4 cS 2 tllf3 c6 3 d4 cxd4
4 xd4 f6 s c3 es 6 dbS d6
7 ctgs a6 8 lll a 3 bS 9 i.xf6 gxf6
1 0 d S fS 1 1 exfS i.xfS
COMPLETE GAMES
23 Fress inet - G elfand ,

E n g h ien les B a in s 20 0 3
1 e4 c s 2 tll f3 tll c 6 3 d 4 cxd4 4
tll x d4 tll f6 s tll c3 es 6 tll d b S d6 7
.ig S a6 8 tll a 3 bS 9 xf6 gxf6 1 0
tll d S fS 1 1 c3 ig7 1 2 exfS xfS 1 3
tLl c2 0-0 1 4 tLl ce3 e6 1 S d3 ts
1 6 YMhS
Lately White prefers a more re
strained approach as 16 0-0 , but
obviously Fressinet was aiming for
a big advantage in the opening.
Thus the idea of castling long and
crash the opponent by a direct at
tack seems the most logical choice.
1 6 . . . e4 1 7 c2 tll e7 1 8 gd1
b4 ! 19 tll x b4 as 20 tll bd S tll xd S
2 1 tll x dS gba 22 b 3 a4 23 xa4
gxb2 24 ib3 h8
Only this move was a novelty.

24 . . . j,xdS practically farced a draw:
25 hdS+ @h8 26 0-0 hc3 27 ie6
Wie8 28 Wixe8 8:xe8 29 MS E'!:eS 30
E'!:cl E!:xfS 3 1 8:xc3 8:xa2 3 2 f3 exf3 33
8:cxf3 112- 112 Van de r Wiel-Cmilyte,
Wijk aan Zee 200 3 .
2 S 0-0
It was high time to castle. The
attempt to defend c3 with 25 Wffh 3
would be punished with 25 . . . e3 ! 26
fxe3 f4 ! 27 Wixe6 Wih4 + .
2 S f4 2 6 gfe 1 ?
This is the critical moment of
the game. White's move is a deci
sive mistake. We shall examine the
other options:
26 l2Jxf4? hb3 27 l2Jg6 + @g8 2 8
l2Jxf8 hdl - + ;
2 6 c4 ! ? 8: f5 ( 26 . . . ieS? 27 cS) 2 7
Wffg4 ic8 ! ? (27 . . . e 3 i s only equal
after 28 ltJxf4 E'!:xf4 29 Wffxf4 e2 3 0
Wixd6 exdl Wi 3 1 E'!:xdl=) 2 8 ltJ b4 ( 2 8
l2Jxf4? Wfff8 2 9 g 3 ih6) 2 8 . . . Wfff8
with compensation;
Perhaps best is 26 l2Jb4, when
26 . . .j,xb3 27 axb3 8:xb3 (27 . . .f3 2 8
g3) 2 8 WidS i s i n White's favour, so
Black should continue with 26 . . . E!:fS
27 Wffg4 c8 with an interesting
double-edged game.
. . .
137
Part 6
26 . . JU5 27 1M/g4 e 3 !-+
Prolonging the second rank up
to the sensitive point g2 . The greedy
27 . . . eS? would have let White back
into the game : 2 8 xf4 xb3 2 9
axb3 hdS 3 0 c4 !b7 3 1 cS ! f! .
28 gxe3
Fressinet tries a last trick since
the rest was hopeless:
28 fxe3 gs 29 f3 !g4 ;
2 8 @hl f2 29 ltJxe3 fxe3 30
he6 Sf4;
28 e2 b3 29 ltJxf4 ds 30
xg7+ @xg7 3 1 ltJxe6+ @ g 8 32 del
f6 3 3 axb3 d2
2 8 ...fxe 3 29 xe3
H oping for 29 ... hb3 30 ltJxfS .
2 9 . . . h5! 30 '%Yg 6 gf6 31 1Mfxh5+
gh6 32 1M/f3 1Mf h4
0-1
24 Ols son-S p asov

E U -chT Gothenburg 200 5
1 e4 c5 2 f3 c 6 3 d 4 c xd4 4
xd4 f6 5 c 3 e5 6 d b5 d6 7
ig 5 a6 8 a 3 b5 9 xf6 g xf6 1 O
d 5 f5 1 1 c3 ig7 1 2 exf5 ixf5 1 3
c2 0-0 1 4 ce 3 ie6 1 5 id 3 f5
1 6 ic2 h 8
We consider 1 6 . . .f4 more relia
ble, but it is known to lead to a draw.
Spasov wants to keep on fighting,
but his move has positional draw
backs.
1 7 1Mfh5 e4 1 8 g d 1
After this move the game trans
poses into our main line D, which is
double-edged. 18 ltJf4 ! ? f6 19 0-0
ltJe 7 20 !b3 is more unpleasant as
White retains his grip on the centre
while keeping the queen on the ac
tive hS-square.
138
1 8 ... e 7 1 9 0-0 if7 20 1Mfh 3

xd 5 21 xd5 1M/ g5
22 h 1
The diagram position is crucial
for line D so we shall examine it in
depth.
a) 22 f4 only opens play in
Black's favour: 22 . . . exf3 23 MS
(23 f3 hdS 24 xdS cl+ 2S !dl
xb2+; 23 xf3? !hS+) 23 . . . xg2 +
24 xg2 fxg2 2S xg2 ae8 with an
initiative thanks to the bishop pair,
for instance : 26 f2 !hS 27 dd2 (27
dfl? !e2 - +) 27 . . . es+ (27 . . . f7 ! ?) ,
o r 26 !g4 es+;
b) 22 e3 is an attempt to orga
nize a piece blockade on f4, but it
proves inefficient:
22 . . . h4 ! 23 g3
Or 23 f3 exf3 24 f3 (24 xf3?
JihS) 24 . . . hdS 2S xdS ae8 26
d3 e l+ 27 fl !eS ! t .
23 . . . h3 24 f4 ( 2 4 f3 hdS 2 S
xdS f4 ! t) 2 4 . . . exf3 2S f3 ae8
(2S . . . !hS? 26 ltJf4) 26 f2 hdS
27 xdS f4 28 gxf4 (28 f4 f4
2 9 gxf4 g4+ 3 0 @hl e2 31 !dl
f2 32 hg4 !h6=) 28 . . . g4+ 29
@fl (29 g3 el+ 30 @f2 e6f!)
2 9 . . . !eS with counterplay.
c) 22 ltJe3 ha2 23 xd6
7 gs a6 8 ltJ a3 bS 9 xf6 gxf6 10 ltJdS fS 11 exfS MS

23 b3? ! ac8 24 ltJdS (24 al
hb3 2 S hb3 xc3 26 dS f4 27
he4 h6 +) 24 ... cS favours Black.
23 . . . g8 24 f4 (24 fdl aSf!)
24 . . .exf3 2S '!!fxf3 (2S xf3 ae8f!)
2S . . .es 26 ddl (26 d2 ae 8 ; 2 6
c6 ae8 2 7 ltJxfS '!!fd2oo) 2 6 . . .b 4 2 7
ltJxfS bxc3 2 8 bxc3 ac8 = .
2 2 . . . ae8 !
Bringing the last piece into play.
2 2 . . . hS? ! is premature: 23 f4 exf3
24 gxf3t
23 t4
23 b3 hS 24 f4 exf3 2S gxf3
e2 26 gl Wh6t, for example one
amusing line : 27 g2 xg2 28 cj{xg2
es 29 cj{f2 (29 ltJc7 dS ! 30 ttJxdS
i.xf3+ 31 Wxf3 Wxh 2+ 32 cj{fl g8+) 29 . . . '!!fgS 30 ltJc7 f6 31 ltJe6
'!!fg4 ! ! t
Black's pieces are better coordi-
nated and are able to create various

tactical threats.
White cannot contest the e-file,
since 23 fel hS 24 f4 (24 al?
f4 2S he4 xe4 26 xe4 WxdS+)
24 . . . exf3 2S gxf3 (2S xe8 is rath
er hapless after both 2S . . . xe8 ! ?
2 6 gxf3 e2 27 MS xb2 2 8 e4
Wh6 29 Wc8+ f8 30 h3 :gxa2+ or
2S . . . '!!fx g2 + 26 Wxg2 fxg2 + 27 cj{xg2
xe8 28 d2 28 . . . es 29 ttJ b4 h6
30 xd6 e 2+ 31 cj{g3 f4 + 32 cj{h3
gS---+ ) 2S . . . xel+ 26 xel Wd2t en
sures Black the initiative.
23 ext3 24 gxt3 ge2 2S ixt S
'%Yh 6 26 Wxh 6 ixh 6
With the bishop pair and a rook
on the second rank, Black is clear
ly on top .
27 g g 1 g xb2 28 g g 2 g x g 2 2 9
xg2 i g7 3 0 ie4 i e S 3 1 a 4 bxa4
32 ga1 ie8 33 ga3 ibS 34 c4
Axc4
The intermezzo 34 . . . g8+ was
more precise : 3S cj{hl d7- + .
3 S gxa4 i b S 3 6 g a 3 g g 8 + 37
h3 Ad4 38 t4 gg1 39 Ad3 Acs
40 gb3 Axd 3
From practical point o f view,
40 . . . c6, keeping the bishop p air,
was better.
41 gxd3 as 42 li:) c3 g e 1 ?
The decision to trade rooks is
wrong. 42 . . . al should be winning.
43 li:) a4 g93+ 44 gxe3 Axe3 4S
g4 g7 46 ts t7 47 e4 Acs
48 ts e7 49 h 3 it2 so li:) c3 d7
S 1 e4 c6 S 2 t S i. h 4 S3 li:)dS ig S
S4 d4 bS SS t6 a4 S6 li:) c3+
b4 S7 li:) xa4 xa4 S8 d S ixt6
S9 xd 6 b4 60 e6 Ad8 61 ts
c3 62 g4 d 2 63 h S
%-%
139
Part 7
1 e4 cS 2 lll f 3 lll c6 3 d4 cxd4

4 lll x d4 lllf6 s lll c 3 es 6 tll d bS d6
7 i.g S a6 8 lll a 3 bS 9 i.xf6 gxf6
1 O lll d S fS 1 1 i.d3 i.e6
Q U I C K REPERTO I RE
This is one of the most aggressive

systems against the Sveshnikov. We
could call it ''The Optimistic Ap
proach". White should deeply be
lieve that Black's setup is outright
dubious and could be crushed by
a direct assault. Otherwise it is dif
ficult to understand why he would
willingly accept to play against a
powerful pawn cluster in the centre
rather than use the gaping hole on
dS for putting pressure on the back
ward d6-pawn. In fact, we face here
a different approach. White concen
trates on the split kingside pawn
structure. It offers him only tempo
rary ad vantages as Black needs two
three tempi to castle and consoli
date, after which he would become
himself the active side on that very
wing. Therefore, White must strive
to obtain immediate benefits and
he is ready to even shed in a piece.
No wonder that the main lines of
this part are critical for the ex
istence of the Sveshnikov. There
were times when line B. 12 hS
seemed to be a possible refutation
of it, and lately 12 c3 ig7 13 l!JxbS
proves a stern test of Black's open140
ing strategy. We must however re

assure our readers, that Black is in
good theoretical shape in these cri
tical lines, provided that our origi
nal analysis of the piece sac on bS
holds true.
Let us start with it:
A. 1 2 c3 i.g 7 1 3 lll x b S
Capturing is not obligatory and

in blitz, or at lower level, 13 . . . ixdS ! ?
14 exdS l!Je7 15 l!J a3 e4 16 ic2 0-0
looks quite appealing for Black. In
practice he achieves good results
even though his compensation for
the pawn is not that clear.
The most principal answer is
undoubtedly:
1 3 axbS 1 4 i.xbS i.d7 1 S
exfS 0-0
7 gs a6 8 ltJ a3 bS 9 hf6 gxf6 10 ltJdS fS 11 d3 e6

that Black had tried it before, but
it got negative commentaries and
failed to gain popularity.
1 7 a4 gea
This position has only recently

become topical and is rather unex
plored. We had really hard time find
ing a decent plan for Black. Having
analysed the available games, we
have reached to the conclusion that
Black needs major improvements of
his play. We shall present a new, ac
tive approach, in order to rehabili
tate this line.
1 6 0-0
White may try to blockade the
e-pawn from e4 by 16 g4, but the
rash sortie of the enemy queen al
lows us to activate the f8-rook fram
g8 instead of the usual place e 8 :
1 6 . . . @ h 8 17 0-0 (17 e4? ! as 1 8 a4
ltJe7 !) 17 . . . g8 1 8 e4 h6 .
1 6 e4!
Only this move gives counter
play. Practice has seen Black strug
gling after 16 . . . b8 ? ! 17 a4 e8 18
g4 @h8 19 e4.
The whole idea of contest
ing the d5-square by trading
pieces is wrong.
Instead, we must seek ways for
activating our pieces . Firstly we
make room for our f8-rook on eS.
The other one could be brought into
play from g8 .
Of course, 16 ... e4 is so logical,
1 8 Wg4 (18 el es) h8 1 9

gad 1
19 fel es 20 ltJe3 f6 2 1 adl
g8 22 hS h6 demonstrates our
best setup .
Black's pieces are so active and

close to the enemy king, that we can
part with some material, but still re
tain a strong attack:
23 ltJg4 xg4 24 xg4 hfS 2 S
g3 ltJe7 26 c 4 e 3 2 7 fxe3 e4-+ .
19 g e s ! 20 e 3 Wf6 2 1
We2
Or 21 hS a7! 22 ltJg4 xfS 2 3
xfS rocfS 2 4 xd6 cS ! when Black
even has some initiative.
21 gga! 22 c4 Wxf5 2 3
xe5 xe5
141
Part 7
14 exfS hdS 15 f6 h6 ! ? 16 fxg7
@xg7 17 l!Jc2 e6oo, or
14 0-0 fxe4 ! 15 he4 fS 16 l!Jf4
exf4 17 hc6 c8 18 e2 es 19
i!Mf3
Black has good chances. We pro

vide further analysis in the " Step by
Step" chapter.
In such a complex positions it is
impossible to foresee everything, but
the essence of Black's play is clear:
He attacks the fS-pawn with . . . es
hoping to place the other rook on
g8. Meanwhile he protects the d6pa wn with . . . f6 . In many cases the
knight fork on g4 is not very dan
gerous as White's knight is the most
unpleasant enemy piece which is se
verely cramping Black's position.
Besides the piece sacrifice, on
move 13 White often chooses 13
hS 0-0
19 . . . b4 !
This thematic break prolongs
the diagonal to our dark-squared
bishop and balances the game. See
game 27 Kramnik-Van Wely,
Wijk aan Zee 2005.
B. 1 2 h5
This version o f the queen's lunge
is significantly more venomous .
White hopes to see 12 . . .g7? ! when
13 0-0 would give him some stable
edge, not to mention that it would
throw us out of the proposed rep
ertoire which is based on 12 0-0
hdS !
1 2 g9s 1
The combination of moves c3

and hS is rather inconsistent
and looks like mixing two different
plans . Black has no particular prob
lems, for instance:
142
7 gs a6 8 l2J a3 bS 9 xf6 gxf6 10 ltJ dS f5 11 d3 e6

This counter-strike saves the
day! Black remains the active side
in otherwise extremely chaotic po
sition. His king is stuck in the cen
tre, but at least it is well hidden be
hind pawns. In contrast, White's
king "escapes" with a long castle,
only to discover that it is not a safe
haven at all.
White has tried so many op
tions in answer to 12 .. . :Bg8 ! , that it
would be unreasonable to remem
ber them. The keypoint is to take on
g2 and then play in the centre:
13 c3 xg2 14 f3 g4
15 exf5 xdS 16 xdS l2Je7 17

b7 c8 18 xc8 + .
The endgame i s preferable for

Black.
Usually White prefers to take
care of the g2-pawn:
1 3 g3 d4 1 4 c3 fxe4 1 5
xe4 g4 1 6 xh 7 gg7 1 7
h6 f3 +
White i s o n the defensive. H e

has a narrow path to equality. See
game 28 Mastrovasilis-Johan
nessen, Athens 200 3
Typically for the Sveshnikov,
Black has strong counterplay in the
sharpest lines. Perhaps even more
important is that play is rich and in
teresting, and both sides have room
for creativity.
143
1 e4 cS 2 lll f3 lll c 6 3 d4 cxd4

4 lll x d4 lllf6 s lll c 3 es 6 lll d bS d6
7 igS a6 8 lll a 3 bS 9 ixf6 gxf6
1 O lll d S fS 1 1 id3 ie6
Part 7
STEP BY STE P

chooses between 12 .c3 , 12 h5 or
12 0-0.
12 c4? ! is clearly weaker due to
12 .. . a5 + ! 13 @fl (13 d2 xd2 +
14 @xd2 h6 + t) 1 3 . . . fxe4 14 he4
c8 . Black's pieces are much more
active: 15 ltJf6 + @d8 16 cxb5 ltJd4
17 bxa6 e7! 18 ltJh5 xa6 + 19 d3
b7 ! + or 15 cxb5 ltJd4 16 :gel :gxcl 17
xcl axb5 18 b4 a4 ! 19 g3 e7+.
A. 12 c3
B. 12 h5
page 144
page 150
We'll examine the most popu

lar continuation 12 0-0 in the next
chapter.
A. 1 2 c3 ig7
Black can also reach this posi

tion via the move order 11 c3 g7
12 d3 e6 .
Al. 13 ltJxb5
A2 . 13 h5
Other options transpose to line
A2 or to variations that are covered
in the next part of the book:
13 0-0 hd5 (13 . . . 0-0) 14 exd5
ltJe7 15 h5 e4;
13 lDc2 0-0 14 0-0
14 lDcb4? ltJxb4 15 lDxb4 fxe4 ! 16
he4 a5 ! overtakes the initiative, as
17 ha8 xa8 18 lDc2 xg2 19 :gfl
h6+ would be bad for White;
14 lDce3 fxe4 15 he4 f5 16 c2
f4 17 h5 transposes to lines with
early ltJce3 .
14 . . . hd5 (14 . . . fxe4 15 he4 f5
16 lDf4 ! exf4 17 hc6t) 15 exd5 ltJe7
16 :gel d7.
A 1 . 1 3 lll xb5 ! ?
The fine point of this sacrifice is
that Black's bishop is already com
mitted to e6 and cannot defend the
c6-knight from b7. We are uncer
tain what answer to recommend.
144
7 gS a6 8 l!J a3 bS 9 hf6 gxf6 10 l!J dS f5 11 d3 e6

Capturing the piece leads to po
sitions where White has enough
compensation and significant space
advantage. Black is able to trade
all minor pieces and stop the ene
my queenside pawns, but then he
would not be able to display any ac
tivity, as White's passers would be
too advanced. Still, White's game
is easier and his mistakes usually
have no fatal consequences. That's
why we propose another plan, con
nected with counterplay in the cen
tre and on the kingside.
On the other hand, 13 . . . hdS ! ?
14 exd5 l!Je7 15 l!J a3 e4 16 c2 0-0 17
0-0 looks quite appealing for Black.
In practice he achieves good results
even though his compensation for
the pawn is not that clear.
Aggressive players will certainly

like Black's side. Perhaps 17 . . . b6 ! ?
1 8 bl (18 b3 aS) 1 8 . . . cS 1 9 b3
(19 d2 l!JxdS 20 fdl l!Jxc3 2 1 bxc3
xa3 2 2 b3=) 19 . . . aS is the most
interesting approach. Unfortunate
ly, the only game that featured this
line, Lupulescu-Stamenkov, EU
chT, Gothenburg 2 005, lasted only
one more move and the opponents
signed a draw.
1 3 . . .axb5 14 ixb5 d7 1 5 exf5 0-0
This position has only recently

become topical and is rather unex
plored . We shall present a new, ac
tive approach, in order to rehabili
tate this line.
1 6 0-0
White may try to blockade the
e-pawn fram e4, but he is undevel
oped for shuch moves. After 16 g4
in the game Vallejo Pons-Topal
ov, Linares 20 04, Black equalised
with 16 . . . l!Jb8 17 0-0 hS 18 f3 , and
here 18 . . .hbS 19 f6 c>h7 leads to a
draw: 2 0 fS+ (20 xhS+ h6 2 1
fS+ ci>h8 2 2 hS ci>h7=) 20 . . . ci>h6
21 f4 (21 h4 g80 22 gS+ ci>h7 2 3
fS + ci>h6=) 2 1 . . . g80 2 2 gS +
c>h7 23 xhS+ h6 24 xf7+ c>h8
(25 l!Je7 hfl 26 fl l!Jc6 27 l!J g6 +
xg6 2 8 xg6 f8 29 f3 a7) 25
hS c>h7= . He might, however, pre
fer 16 . . . c>h8. Usually he plays this
move after . . . e8 first, but here he
can use the premature arrival of the
queen at g4 and activate the rook
from g8, as in the case of 17 0-0
(17 e4? ! aS 18 a4 l!Je7! or 18 c4
gS) 17 . . . g8 18 e4 h6 .
1 6 . . . e4 !
Only this move gives counter-
145
Part 7
play. Practice has seen Black strug
gling after 16 . . . b8? ! (16 . . .Ee8 17
g4 @h8 18 e4 f6 19 a4 is similar)
17 a4 e8 18 g4 @h8 19 e4
- Rogozenko. Alas, the simple 32 a6

wins immediately, 3 2 . . .xb6 33 a7
a6 34 g3 ! + - .
The moral o f this analysis i s that
Black should not wait passively.
We can also see that the rook is
more useful on a8, from where it
stops the a-pawn.
1 7 a4
17 el 8:e8 18 a4 E'!:eS transposes
to the sub-line on move 18.
1 7 . ea
.
Black's idea is to gradually break

the clamp on dS and include the
dark-squared bishop into play. It is
too slow and gives the enemy time
to advance the a and b-pawns: 19 . . .
f6 ( 1 9 . . . l!J e7? 2 0 f6 + -) 2 0 b 4 l!Je7
(20 . . . c8, hoping to get a lever on
the kingside after 21 g4? ! , is well
countered by 2 1 c4 ! with a clear
edge) 21 hd7 xd7 22 l!Jxe7 xe7
23 aS ! ec8 24 tbl ! Almasi-Wang
Yue, Paks 20 06 .
Here, instead of 24 . . . d7, Rogo

zenko suggests 24 . . . f7, but it does
not help either: 25 bS dS 26 d3
f8 27 b6 cs 28 bS ! d4 (28 . . . e4
29 bl ! ) 29 c4 e4 ! 30 xe4 xc4
31 bbl b4 "This position is very
likely to end in some sort of draw"
146
1 8 YMg4
18 el es 19 l!Je3 l!Je7 brings
about an equal endgame: 20 hd7
xd7 21 f6 hf6 22 g4+ xg4 23
l!Jxg4 e6 24 l!Jxf6+ f6 2 5 xe4
@f8 = .
1 8 . . . h8 1 9 gad 1
19 fel es 2 0 l!Je3 f6 2 1 E'!:ad l
g8 22 hS h6 demonstrates
Black's best setup .
7 !gS a6 8 ltJ a3 bS 9 hf6 gxf6 10 ltJdS fS 11 !d3 e6

His pieces are so active and close
to the enemy king, that we can part
with some material, but still retain
a strong attack:
23 ltJg4 xg4 24 xg4 MS 2 S
g3 ltJe7 2 6 c 4 e 3 2 7 fxe3 e4-+ .
1 9 .. J e S !
Aiming to reach the position o f
the last diagram. 19 . . . ltJb8 ? ! 20 fel
es 21 ltJe3;t; is rather gloom.
20 ll) e3 Yf 6 21 Y e2
21 hS a7! is another critical
line. The reason behind our last move
is seen in the variation 22 ltJg4 xfS
23 xfS S 24 xd6 cS ! when
Black even has some initiative: 2S
ltJe3 !es 2 6 ds ds 27 ttJxdS ltJe7
28 hd7 ttJxdS 29 !c6 ltJf4.
White might try 22 fel.
Then 2 2 ... hfS? is bad d ue to
23 ltJdS xdS 24 xdS ltJe7 2 S fS
xfS 26 xfS ltJxfS 27 e4 !es 2 8
f4 !f6 2 9 g4 ltJh4 30 M2 d S 3 1 e2
h6 3 2 d2 ttJg6 33 ds llJxf4 34 fs
!gS 3S b4 ltJe6 36 cj/e2 e7+ .
Black should prefer 22 . . . ltJe7
23 hd7 xd7 with sufficient
counterplay: 24 b4 (24 llJg4 xfS 2 S
xfs fs 2 6 e4 aS=) 24 . . . d s 2 S
b S ( 2 S g4 c 6 26 b S c8oo) 2S . . . a7
26 bl b6oo .
Let u s return to 2 1 e2 :
It seems that White is better in

view of the threat 21 ltJg4, but now
we complete the redeployment star
ted with 19 . . . eS :
2 1 .. _ g ga ! 22 tl) c4 Yxf5 23 tl) x e 5
.ixe5
Black has good chances. We shall
provide further analysis, in order to
help you in practice:
24 g3 Yg6 2 5 f4
White cannot survive without
this move: 2S e3 fS 26 f4 exf3 27
xf3 f4-+ .
2 5 . . . exf3 2 6 gxf3 f5
27 gdf1
The other way to take control of
f4 is worse:
27 d2 hS-+ . The text prevents
this option in view of the threat 2 8
hc6 .
27 c4 c8 28 d3 !e6oo is dou
ble-edged.
27 ...f4 28 gxf 4 .ixf 4 29 gxf 4
The weakness of White's first
rank and light squares balances the
game .
29 ... g ea 30 'ld2 .ih3 3 1 q;t2
Or 31 if1 hfl 32 xf1 ltJeS 33
d4 (33 b4 e4 34 d4 xd4+ 3S
cxd4 ltJc6=) 33 . . . cj/g7 34 aS f8 = .
3 1 . . .d S ! 3 2 b 4 'lb1 3 3 g 4 ge6
147
Part 7
34 g5 1Mfh 1 35 g 3 1Mfg 1 + 36 xh3
g e3+ 37 h4 tiles 38 i e8 g94
39 ih5 tll c4 40 g x e4 tll xd2 41
g e a+=.
Draw b y repetition. In such
complex positions it is impossible
to fore see everything, but the idea
is clear: Black attacks the f5-pawn
with . . . 8:e5 hoping to place the other
rook on g8. Meanwhile he protects
the d6-pawn with . . . Wfff6. In many
cases the knight fork on g4 is not
very dangerous as White's knight
is the most unpleasant enemy piece
which is severely cramping Black's
position.
A2 . 1 3 1Mfh5 0-0
A2 a. 14 exf5? !
A2b. 14 0-0
White also tries occasion ally:
a) 14 l2J e3 f4 15 l2Jf5 Wfff6 16 g4. The
knight stays nicely on f5, but the to
tal lack of coordination ruins White:
16 . . . b4 ! 17 l2Jc4 bxc3 18 bxc3 8:fd8
19 E'!:bl h6 2 0 8:b6 ti:Je7 2 1 h4 l2Jxf5
2 2 exf5 e4 ! - +(Movsisian-Smirnov,
Yerevan 2004 saw 22 . . .hc4? 23 g5
Wfe7 24 hc4 Wffc7, when 25 gxh 6 ! +
148
would have favoured White ;

b) 14 g4? fxg4 15 h3 g3 ! 16 E'!:gl !
(Trying to open up files. 1 6 fxg3
f5 is grim for White: 17 0-0 f7 18
Wff d l ixd5 19 exd5 Wffg5 ! + or 17 00-0 f7 ! 18 Wff e 2 hd5 19 exd5 l2Je7+)
16 . . . gxf2 + 17 @xf2 8:a7! and the tide
turns desively: 18 ti:Je 3 f5 ! 19 exf5
@h8 ! 2 0 e4 d5 2 1 E'!:adl if7 2 2 Wffg4
ih6 - + ;
c) 14 l2Jc2? ! fxe4 15 he4 f5 16 ti:Jf4
exf4 17 hc6 8:c8 18 Wfff3 d5 19 l2Jb4
aS 20 hd5 Wffd 6 21 ti:Jc6 @h8 ! - + ;
d ) 1 4 0-0-0? fxe4 1 5 he4 f5
16 ic2 b4 ! 17 l2Jc4 bxc3 18 bxc3
l2Jd4 ! + ;
e ) 14 l2Jc7 Wffxc7 15 exf5 d 5 16 f6
h6 17 fxg7 @xg7 18 ti:Jc2 b4 19 c4 e4
20 cxd5 exd3 2 1 l2Je3 d2 + !+.
A2a. 14 exf5?! hd5 15 f6 h6

This move allows Black to play
for win. 15 . . . e4 should lead to a
draw in a rather forced way, see 25
Naiditsch-Chuchelov, Belgium
20 0 3 .
16 fxg7
16 Wfff5 hands Black the initia
tive: 16 . . . e4 17 fxg7 8:e8 18 ie2
18 hb5 e6 19 Wfff4 looks ap
pealing, but we can take the piece
by 19 . . . axb5 ! 20 Wffxh6 f5 and White
cannot prove even equality: 2 1 l2Jxb5
(21 0-0 l2Je5 22 f4 ti:Jg4+; 21 h4 l2Je5)
2 1 . . . l2Je5 22 l2Jd4 Wffd7+.
18 . . . 8:e5 19 Wfff4 (19 Wff h3 Wffg5+)
19 . . . Wffg5 20 Wffxg5 8:xg5
20 . . . hxg5 ! ? is a good alterna
tive: 2 1 l2Jc2 @xg7 (22 h4 ie6 23 E'!:dl
8:d8=) 2 2 l2Je3 e6 23 E'!:dl 8:d8 = .
2 1 l2Jc2 ie6 2 2 l2Je3 b4 ! ? Break-
7 gs a6 8 ltJ a3 bS 9 hf6 gxf6 10 ltJdS fS 11 d3 e6

ing White's pawn chain. 23 f4 exf3
24 hf3 bxc3 2S bxc3 (2S hc6
cxb2 26 bl b8 27 @d2 ha2+)
2S . . . c8 = .
1 6 xg7 17 c2
17 fS? ! is inefficient, as Black
can ignore the "threat" by 17 . . . gS !
when 18 h7 +? @f6 would favour
Black. Remains 18 xgS+ hxgS 19
ltJc2 e6 20 a4 bxa4 2 1 xa4 as 2 2
@d2 fS 23 hal ab8+, Tokmachev
Samojlov, Serpukhov 1999 (39) ;
17 h4 ! ? is more consistent since
h3 would be decisive. 17 . . . e6 18
g4 b4 ! and again we see White un
developed for a killing blow: 19 gS
hxgS 20 hxgS h8 2 1 h7 @f8 .
Perhaps he should prefer 19 ltJ c4
bxc3 2 0 bxc3 ltJe7 21 gS (21 ltJe3
ltJg6 ! +) 2 1 . .. h8 22 f3 ltJdS 23
gxh6+ @xh6 24 e4 ltJf4 2S ha8
xc4 with complications.
17 .ie6
.
23 h3 g6 24 xg6+ fxg6 2S ltJ e3

d4+ 26 ltJc2 f4 27 d2 hS 2 8 b3
@g7! (28 . . . dS 112- 112 Anand-Ivan
chuk, Linares 2002) 29 @dl aS 30
@el b4 3 1 cxd4 exd4 32 @dl ltJeS 33
ltJel @f6 (33 . . . c8 34 ltJd3 ltJxd3 3S
hxd3 xh4 36 xd4 hl + 37 rJ/e2=)
34 rJ/e 2 c8+ /+ e . g . 3S ltJd3 ? e4+
36 @dl c3 with a big advantage.
A2b . 14 0 - 0 fxe4!
This farced variation solves the
opening problems. The older line
14 .. .f4 leads to very complex play
with mutual chances . We cover it
in considerable detail in the "Com
plete Games" section - see game
26 Svidler-Van Wely, Wijk aan
Zee 2004.
15 he4 f5 16 f4 exf4 17
hc6 gc8
Now:
18 a4 fS ! ?oo (18 . . . bxa4? ! 19 xa4
aS 20 ltJe3t) 19 axbS axbS 20 xa8
xa8 21 0-0 b7oo ;
18 e3 dS 19 h4 (19 fS gS)
19 . . . f6oo 2 0 fS MS 21 fufS+
@h7 2 2 dl ( 2 2 h3 g6) 22 . . . ad8
18 e2
It is risky to capture the a6pawn: 18 b7 cS ! = or 18 f3 dS 19
b7 cS 20 ha6 b6 2 1 .bbS xbS
22 ltJxbS xbS.
18 .te5 19 f3
19 b7 is still dubious, but due to
different reasons than on the previ
ous turn. This time Black aims for
149
Part 7
a direct kingside attack, using the
retreat of White's queen: 19 .. .:gc7!
(19 .. .:gb8 20 ha6 + ; 19 .. . :gcs 20 f3 !
h4 2 1 E!fdl h8 2 2 lDc2 E!c7 23 ltJd4
c4 24 Wffd2 E!g7 25 b3 g8 26 Wffd3
Wff6 27 a4+) 2 0 ha6 Wffa8 ( 20 . . . b4
21 cxb4 f3 22 Wffxf3 Wffh4 23 h3 Wffxb4
24 d3 hb2 25 ltJb5) 2 1 hb5 f3 ! 2 2
gxf3 ( 2 2 Wffxf3 dS 2 3 Wffh 3 hg2 2 4
Wffxg 2 + E!g7+) 2 2 . . . E!f6-+ ;
19 f3 Wffh4 2 0 E!fdl E!f6 also
looks like fun for Black.
19 b4!
Black must try to prolong the dia
gonal of his dark-squared bishop.
Consolidation would mean advan
tage to White, who has a much bet
ter pawn formation, for example,
19 . . . h8 2 0 dS d7 21 ltJc2 Wffh4
22 E!adl ie8 23 Wff d3 E!c7 24 ltJd4i.
2 0 cxb4
White would be worse if he ac
cepted a weakness on c3 : 20 ltJc2
bxc3 21 bxc3 b6 22 dS hdS 23
WffxdS+ h8+.
2 0 . . .i.xb2
is the a3-knight, which h a s not any

stable square. d5 would be a terrific
outpost for it, but both approaches
to it, c3 and e3, are firmly control
led. See game 27 Kramnik-Van
Wely, Wijk aan Zee 2005.
a.
1 2 Yhs ggs
The game is balanced. Black's

pawns are split and look easy to
collect, but in fact White cannot at
tack them efficiently. His problem
1 50
Bl. 13 c3
B 2 . 13 f4
B3. 13 g3
Minor alternatives are:
a) 13 0-0? ! f4 14 h3 (Or: 14 c3
g4 15 Wffxh7 E!g6+; 14 c4 b4 15 ltJc2
g4 16 Wffxh7 E!g6 17 f3 e6 18 hl
E!h6 19 Wff g8 f5- + ; 14 Wffdl h3 ; 14
Wffxh7 E!g6 15 Wffh 5 E!h6 16 Wffd l Wff h4+) 14 . . . E!g6 15 c3 ltJe7f! with excel
lent prospects for Black;
b) 13 0-0-0 E!xg2 ! 14 Wfff3 (14 f4
ltJd4 ! transposes to line B - 13.f4
E!xg2 14 0-0-0) 14 . . . E!g4 15 exf5 (15
h3 E!h4) 15 . . .hd5 16 Wffxd5 ltJb4
17 Wffb7 (17 Wffb3 E!f4 18 c4 bxc4
19 Wffxb4 Wffc7t ; 17 Wfff3 E!f4 18 Wff g2
ltJxd3 + 19 E!xd3 E!c8 20 ltJbl Wffc7+)
17 . . . d5+ ;
c) 13 c4? ! hd5 (14 cxd5 ltJb4
15 e2 fxe4 16 Wffxh7 E!g6+) 14 exd5
7 gS a6 8 l!J a3 bS 9 hf6 gxf6 10 l!J dS fS 11 d3 e6

as + (14 . . . l!Jd4 ! ? 15 o-o g4 ! ) 15
w fl l!Jd4t;
d) 13 Wfl h6 14 h3 hdS lS exdS
l!J e7+.
8 1 . 1 3 c3 gxg2 1 4 \Wf3
Queenside play with a4 is insuf
ficient: 14 l!Jc2 f4 1S a4 bxa4 16 xa4
g7 17 l!Jcb4 l!Jxb4 (18 bS+ axbS 19
xa8 l!Jd3 + 20 Wfl xa8 2 1 l!Jc7+
Wd8 22 l!Jxa8 wins Black's queen,
but it will be retaliated by the raving
rook on the second rank: 22 . . . f2 +
23 Wgl xb2) 18 l!Jxb4 aS 19 l!JdS
b8 2 0 b4 axb4 2 1 cxb4 c8+.
1 4 \Wf3 gg4
17 f3 is passive: 17 ... f4 18
e2 c7 (18 ... b4 ! ? 19 cxb4 b6f)
19 l!Jc2 b7! ? (19 ... dS 20 l!Je3) 2 0
f3 h6+.
1 7 . . . \Wc8 1 8 Yxc8+
The endgame is preferable for
Black, but keeping queens on is
even worse: 18 b6? l!JdS 19 as (19
xbS+ axbS 20 xbS+ We7 21 xdS
xa3 ! - +) 19 . . . c6 ! ?+ (19 . . . l!Jf4 2 0
e4 dS+) 2 0 0-0-0 l!Jb4 2 1 c4 ( 2 1
c2 l!Jxc2 2 2 l!Jxc2 a4- + ; 2 1 bl
h6+ - +) 2 1 . . .h6+ 22 W bl xc4
23 cxb4 d4+.
1 8 . . . &fJ x c8 1 9 &iJ c2 &iJ b 6i.
82 . 1 3 f4 g x g 2 1 4 0-0-0
Typically for the Sveshnikov,

Black has strong counterplay in the
sharpest lines.
1 5 exf5
15 h3 is of little use: lS . . . g8
It is also interesting to keep the
rook on the 4th rank: 1S . . . h4 ! ? 16
exfS xdS 17 xdS l!Je7 18 g2 (18
b7 c8 !) 18 ... dSf, Boudre-Kouat
ly, Ales 1984.
16 exfS xdS 17 xdS l!Je7 18
e4 g7 19 0-0-0 dS 20 h4 b6 !
2 1 w bl f6 ! 22 h6 0-0-0 f!.
1 5 . . . ixd 5 1 6 \Wxd 5 &fJe7 1 7
\Wb7
This attempt to crush Black

in the centre fails because of the
stranded knight on a3 . Without it
White lacks resources to complete
the attack. The possibility of the c6knight to jump to d4 tips the bal
ance in Black's favour.
Note that the double attack 14
l!Je3 is bad due to 16 . . . as+ 15 Wfl
d2 ! 16 l!Jbl (16 l!JxfS b4 ! + ; 16
exfS b4 ! 17 f3 d7 18 bl 0-0-0
19 l!J dS cs 20 b4 d4+) 16 ... xd3
17 cxd3 b4 18 exfS (18 e2 fxe4
151
Part 7
19 dxe4 xe4 - +) 18 . . .hfS 19 WffxfS
Wffxb2+.
1 4 ... d4! 1 5 e 3
Aft e r lS c3 xdS 1 6 exdS, the
thematic 16 . . . b4 ! + comes at rescue;
Other options: lS hgl fxe4 16 xg2
exd3+ ; lS bl xdS 16 exdS g4 !
threatening to trap the white queen
with h4.
1 5 . . JU2 1 6 exf5
According to Kramnik, Black
is on top after 16 hfl fl 17 xfl
c8 .
1 6 . . . .ixa2 1 7 fxe5 dxe5
17 . . . c8 18 xbS+ ttJxbS 19 ttJxbS
axbS 20 exd6 is unclear.
1 8 xb50 .ih 6 ! !
This surprising move underlines
how vulnerable White's king is. The
stem game Brodsky-Kramnik, Her
son 1991 saw further 19 hel (19
xh6 xc2 + ! mating) 19 hel axbS!
2 0 xbS+ e7 21 h4 + f6 2 2 xf2
.if7 with an overwhelming attack.
d2 (16 c3 gs 17 Wff h4 ttJf3 - +) 16 . . .

fxe4 1 7 xe4 b 4 ! 1 8 ttJxb4 gs 19
dl dS+;
b) 14 xh7 g6 lS hS fxe4 16
xe4 ttJd4 ! t ;
c ) 1 4 c 3 g6 ! l S Wfff3 (lS tlJ c 2 fxe4
16 he4 .ig4 17 Wffh4 Wffx h4 18 gxh4
fS 19 hS g7! 20 h3 xh3 2 1 .if3 .ig4
22 tlJf6 + d8+) lS . . .xdS 16 exdS
e4 17 WffxfS gs 18 Wfff4 exd3 19 dxc6
es+ 20 fl b6 21 Wfff3= .
14 c3
14 0-0-0? ! c8 l S f4 (lS c 3 h6 ! ;
lS bl h6 !) lS . . . g6f!;
14 ttJe3? gs lS xh7 fxe4 16
xe4 dS 17 .ig2 as++;
14 h3? xdS lS exdS e4 16 0-0-0
f6 ! + .
1 4 . . . fxe4 1 5 .ixe4 .ig4 1 6 1Mfxh 7
gg7 1 7 \W h 6 f3+
83. 1 3 g3
Black checked first and that puts

White on the defensive. He has a
narrow path to equality.
B3a. 18 fl? !
B3b. 18 e2
1 3 . . . d4
13 . . . c8 is a decent alternative.
Main lines are:
a) 14 0-0-0? ttJd4 lS bl h6 16
1S2
Practice has also seen 18 dl,

but after 18 . . . tlJgS+ 19 f3 g6 20
Wffh4 ttJxe4 21 fxg4 l!Jf2 + Wffc8 ! 2 2 gS
7 gs a6 8 ttJ a3 bS 9 hf6 gxf6 10 ttJ dS fS 11 d3 e6

ttJxgS 23 ttJc2 WfffS Black is slightly
better.
B3a. 18 @fl? ! E:g5!

Threatening 19 . . . ttJd2 . 18 . . . g6
19 Wffcl + is worse.
19 f6+
Or 19 Wffh7? ! ttJd2 + 20 el g7+.
19 e7 2 0 h8
After 2 0 ttJdS+ d7 2 1 ttJf6 + c7+
the king slips to the queenside.
2 0 d2+ 21 g2 xe4 22
xe4 E:g6 23 h4+
23 hel Wffd7 ! allows Black to con
solidate : 24 f3? hf3 + 2S xf3 h6
26 Wff g8 Wff fS+ 27 e3 xh2 28 Wigs+
WffxgS+ 29 tiJxgS h6- + ; 24 Wffh4+ f6
2S Wffh7+ g7 2 6 Wff h6 f3 + ! 27 xf3
g3 + 2 8 hxg3 hh6 29 adl dS 3 0
@ g 2 Wffc6- + ; 2 4 ttJc2 Wff b7t.
23 h4+ e6 ! 24 1l;Yxd8 E:xd8
b4! (28 . . . h6 29 ttJ d3 as 30 hel;t)

29 E:hel bbl 3 0 E:xbl bxc3 31
bxc3 (The source game A. Sokolov
Skripchenko, France (ch) 20 03
saw 31 ttJd3? c2+) 31 E:xc3 32
E:b6+ .id6 33 E:e2 (33 xa6 c2)
33 E:a3+.
B3b. 18 e2 g5+ 19 fJ xe4

2 0 fxg4 c8 !
21 1l;Ye3
Or 2 1 h3? xg4 2 2 Wffh7 Wies 2 3
hxg4 Wfff2 + - + ; 2 1 d3 Wffxg4 2 2 ael
ttJf2 + 2 3 d2 Wfff3 t ; 21 ttJe3? g6 22
Wffh 3 dS+.
21 1l;Yxg4+ 22 VffJ xt'3+ 23
xfJ f5 24 c2
24 ttJe3 f7 2S ttJxfS? fails to 2S . . .
dS ! + .
24 @f7
Practice has shown that the end
game is balanced, but Black's game
is a bit easier. See game 28 Mas
trovasilis-Johannessen, Athens
2003
Black's central pawns threaten

to overrun the enemy:
25 fJ (2S ltJc2? fS+) 25 .tf5
26 E:adl d5 27 b l gc8 28 f2
1S3
Part 7
1 e4 c5 2 lll f3 lll c 6 3 d4 cxd4

4 lll x d4 lll f6 5 lll c 3 es 6 lll d b5 d6
7 ig5 a6 8 lll a 3 b5 9 ixf6 gxf6
1 O lll d 5 f5 1 1 .id 3 .ie6
COMPLETE GAMES
2S N a id itsch - C h uch elov

B el g iu m 20 0 3
1 e 4 c s 2 li:) f3 li:) c 6 3 d4 c xd4 4
li:) xd4 li:) f6 s li:) c 3 es 6 li:) d b S d6 7
ig S a6 8 li:) a 3 bS 9 ixf6 gxf6 1 0
li:) d S ts 1 1 id3 ie6 1 2 c3 ig 7 1 3
YhS 0-0 1 4 exfS ixd S 1 S f6 e4
We suggest 1S . . . h6 ! ? , which leads
to complicate play. The text is known
as a reliable equaliser.
1 6 fx g7 gea 1 7 ie2 ges 1 8 Yh6
b4!
d l cs 24 ids e8) 23 ... cs 24

ids e8 ! (Chuchelov assesses the
position after 24 . . . xc3 + 2S b2
cs 26 he4 b8 + 27 al ltJb4 2 8
d2 a s slightly better for White) 2 S
ib3 ( 2 S hel ltJb4 2 6 he4 ltJxa 2 +
2 7 d 2 ec8 =) 2 S . . . ltJ aS 2 6 b2
b8 27 hel dS with some initia
tive for Black.
22 Yxd 6 g d 8 ! 23 Yxc6 ge6 ! ?
23 . . . cs also leads t o a draw: 2 4
xa6 d2 + 2S fl fS 26 hf7+ !
f7 27 e2 xc3 (27 . . . e3 2 8 xd2
xd2 29 f3=) 28 dl xdl + 29 xdl
e3 30 d8 + with perpetual check.
24 ixe 6
White cannot avoid the perpe
tual check, since 24 c7 d2 + 2 S
l e 3 o r 2 4 b7 d2 + 2S l e3
are hopeless for him.
24 . . . Yd 2+ 2S @f1 'ld3+ %-%
1 9 li:) c4
After 19 lDc2 bxc3 2 0 bxc3 ie6 2 1
0 - 0 Black equalises like i n the game:
21. . . gS 22 xgS xgS 23 fdl cS=
24 xd6 xc3 2S ltJe3 ltJb4 ! = .
1 9 . . ..ixc4 20 ixc4 bxc3 2 1
bxc3 Yg S ! 2 2 Yxd 6
Or 2 2 xgS xgS 23 0-0-0 (23
2 6 Sv idl er - Van Wel y

W ij k aan Zee 2004
Comments by Nedev
1 e4 cs 2 ll:) f3 ll:) c 6 3 ll:) c3 ll:) f6
4 d4 cxd4 s li:) xd4 es 6 li:) d bS d 6
7 i g S a6 8 li:) a3 bS 9 ixf6 gxf6 1 0
li:) d S ts 1 1 id3 ie6 1 2 c3 ig7 1 3
YhS 0-0 1 4 0-0 f4
1S4
7 .igS a6 8 lt:J a3 bS 9 .ixf6 gxf6 10 lt:JdS fS 11 .id3 .ie6
Our
main
recommendation
is 1S . . . fxe4 ! because it is a forced

equaliser. However, the text is a fair
alternative which offeres both sides
a lot of interesting possibilities.
1 5 gad 1
a) lS g 3 fS 16 gxf4 h 8 It is al
ways risky to open up play against
a bishop pair. For the pawn Black
seizes the initiative and White must
quickly return the material: 17 lt:Jc2
.if7 18 %Vh3 exf4 19 lt:Jxf4 gs+ 2 0
lt:Jg2 f4 2 1 %VfS %VxfS 2 2 exfS lt:JeS
23 gfdl .ihS 24 .ie4 ixdl 2S gxdl
gae 8 f! ;
b ) lS gfdl I t is not evident which
rook should go to dl. Still, the
game focus is on the king side and
White might need a rook on fl later.
1S . . . h8 ! ? 16 lt:Jc2 fS 17 a4 (17 exfS ? !
ixdS 18 f6 .ih6 ! 1 9 xh6 ga7t In
arkiev-Filippov, Sochi 20 0S) 17 . . . .if7
18 h3 .ixdS 19 exdS lt:Je7oo ;
c) l S g4 White closes the kingside,
but in an open Sicilian this is not an
nov, Sochi 2 0 0 6 , or regroup first on

the kingside: lS. . . h8 16 hl gg8
17 gS lt:Je7! 18 f3 .if8 19 gadl gg7 2 0
.ic2 gc8 2 1 ggl gcs 2 2 .ib3 lt:JxdS 2 3
ixdS ixdS 2 4 gxdS gxdS 2 S exdS
a8+ Gonzalez de l a Torre-San Se
gundo, Lugo 20 07;
d) lS lt:Jc2 fS ! 16 lt:Jcb4 lt:Jxb4 17
lt:Jxb4 aS 18 exfS .if7 19 %Vh3 f6 ! 2 0
lt:Jc2 Anand-McShane, Bundesliga
2 0 0 3 and now 2 0 . . . b4! 21 cxb4 dSoo
would have been fine for Black.
1 5. . . @ h S
Black anticipates g 3 and shifts
his king from the open (in future)
file. Topalov preferred 1S . . . ga7 ! ?
when 16 lt:J c2 h 8 1 7 lt:Jcb4 lt:Jxb4 1 8
lt:Jxb4 d7 1 9 .ie2 fS 2 0 exfS fS
was double-edged in Svidler-Topa
lov, Cap d'Agde rapid 20 0 3 . More
testing is 16 g3 fS 17 gxf4 exf4 1 8
lt:Jxf4 .if7 1 9 h3 %VgS + 2 0 g3 ( 2 0
lt:Jg2 f4 2 1 e s .ihS ! ) 2 0 . . . h6 ! with
compensation.
1 6 g3
If White refrains fram under
mining the f4-pawn, Black activates
his rook: 16 lt:Jc2 gg8 17 Whl .if8 18
.ie2 ggs+.
1 6 ... ggs 1 7 @h1 ts 1 8 e2
g g 5 1 9 '1Mf3 (19 h4 gc8) 1 9 . . . f5
achievement as Black retains his

typical queenside play. He can start
with lS .. . gb8 ! ? 16 lt:Jc2 as 17 a3 lt:Je7
18 gadl d 7 19 h3? ! ixdS 20 exdS
fS 21 f3 gf6t, as in Amonatov-SmirlSS
Part 7
20 exf 5
Anand-Topalov, Monte Car
lo rapid 20 0 1 saw 2 0 gxf4 exf4 2 1
ttJxf4 a2
ficult for both sides since it requires

a lot of calculation without clear cri
teria for evaluation.
21 . . . ext4 22 Wf xc6 gca
22 exfS xfS 23 e4 f7 24
ihS? ! (24 ttJe6 e6 2 5 xe6 e7=)
24 . . .f6 2 5 c4? ! tlJeSt a nd Black's
pieces showed better coordination.
should probably try to keep the
main diagonal closed by playing
immediately :
2 2 c4 tlJ eS 23 e3 g8 24 cxbS
(24 exfS ih6f) 24 . . . i.h6 25 exfS,
but 25 . . . axbS is level. (Rogozenko
proposes the sharper line 25 . . . h4
26 d4 .bf4 27 xf4 dS+ 28 f3
gs 29 gl h6 30 d4 ac8 ! 3 1
xg 8 + xg8 3 2 bxa6 h3 33 f2
h6 34 d4 h3=)
20 ... ixfS 21 li:) xf4
2 1 gxf4 exf4 ( 2 1 . . . e4 2 2 e3 g6
23 gl t) 22 gl is also balanced.
(22 ttJxf4 is extremely risky, for in
stance, 22 . . . e8 23 e3 tlJeS 24
tlJdS g6oo when 25 ttJc7? would
lose to 25 . . . c8 26 ttJxa6 ie4+ 27
f3 ib7 28 ttJ b4 ih6 29 f2 g8 ; 2 2
xf4? ! ih6) 2 2 . . . ih6 23 xgS xgS
24 gl d8 25 hS f8 26 f3 c8
(26 . . . g6 27 h3 tlJeS 28 ttJe7!) 27
ig4 g4 28 xg4 e8oo. Play is dif-
2 3 Wff3
Of course White cannot capture
on a6 due to 23 xa6 e4+ 24 f3
a8- + , but 2 3 b7 is possible, al
though White's queen risks to be
trapped in some variations:
a) 23 . . . b8? ! 24 f7 (24 xa6
e4+ 25 f3 a8) 24 . . . ie4 + 25 gl
f5 26 a7 is probably in his fa
vour;
b) 23 . . . g7 24 xa6 (24 f3 ig4;
24 g 2 e8 ! 25 if3 d7) 24 ... ie4 +
25 gl b7 2 6 ttJxbS a8 27 ttJxd6
xa6 28 ttJf7+ xf7 29 xd8 af6t;
c) 2 3 . . . dS ! ? 24 id3 (O r 24 gxf4
g7 25 xa6 g6 ! 2 6 xbS h6 - + ;
2 4 g 4 d7 2 5 tlJc2 c6+) 2 4 . . . g7 !?
(24 . . . ixa3 ! ? 2 5 .bf"S fS 26 bxa3
xc3 27 fel fxg3 28 fxg3 f8 2 9
g l c2t) 2 5 xa6 ig4 26 f3 ih3
27 gl b4 28 cxb4 fxg3 29 g3 h4
30 dgl xg3 31 hxg3 hS+ .
23 . . .fxg 3
Van Welly could have main
tained the tension with 23 . . . e7 ! ?
(threatening . . . ig4) White can take
up the gauntlet by 24 d4 (24 id3?
156
7 !gS a6 8 ttJ a3 bS 9 xf6 gxf6 10 tlJdS fS 11 !d3 !e6

fxg3 2S fxg3 !g4 ; 24 gxf4 !e4 2 S
fxgS cs 2 6 gl xgS+) 24 . . . !g4 2 S
xg4 xg4 26 !xg4, when 26 . . . b8 !
ensures counterplay. Black's idea is
to meet 27 f4 by 27 . . . b4 28 cxb4
!h6 2 9 c4 dS 30 c2 d4 with a tan
gled game.
24 fx g 3 d S !
This tactical trick allows Black
to level the game. The point is 2S
xdS? !e4 !
2S Wfg2
Or 2S e3 ! ? !e4 + (2S . .. e7 26
xe7 he7 27 fS MS 2 8 !g4 cf8
29 fixfs fs 30 g2 !cs 31 d2 + )
26 !f3 !cs 27 1!Me2 !xa3 28 bxa3
e7 29 d4 eSoo.
2S . . . .ixa3 26 bxa3
26 dS ! ? should be met by
26 . . . !d7! 27 fu:gS xgS 28 bxa3 fu:c3
29 m cs 30 d2 !c6 31 !xc6 (31
!g2 i.xg2 + 3 2 xg2 c6= )31 . . .xc6+
32 gl 1!McS+ 33 f2 f3= .
2 6 . . . Wf e7 27 .if3 .ie4 28 Wf d2
Wfg7
28 . . . fixf3 + ! 29 M3 e4 30
1!Md4 + e s leads to a n equal rook
ending: 31 gl (31 1!Mxe4 dxe4 3 2
e 3 c4) 3 1 . . . 1!Mxd4+ 3 2 d4 el+
33 M2 cl= .
29 .ixe4 dxe4 3 0 Wfe3
Black's strong passed pawn in
the centre balances the game.
30 .. . es 31 @ g 1 (31 d6 ! ? f8 ! )
gf8 ! 3 2 gxf8+ Wfxf8 33 Wfd4 Wfe7
33 . . . cS ! was more precise: 34 M2
g8 3S e3 dS 36 xcS xcS= .
3 4 @ f2 Wff6+ 3 S @e2 Wff3+ 3 6 @d2
e3+ 37 @c1 Wff6 38 Wfd8+ Wfxd8 39
gxd8+ @g7 40 @d1 gfs 41 gas gf2
42 h4 gd2+ 43 @e1 gxa2 44 gxa6
gc2 4S gc6 hS 46 gcs @g6
%-%
27 K ra m n i k - Va n We ly
W ij k aan Zee 2 S. 0 1 .200S
1 e4 cs 2 f3 c6 3 d4 c xd4 4
xd4 f6 s c 3 es 6 d b S d6 7
.igS a6 8 a 3 bS 9 .ixf6 gxf6 1 0
d S ts 1 1 .id3 .ie6 1 2 c3 .ig7 1 3
Wf h S 0-0 1 4 0-0 fxe4 1 S i.xe4 fS
1 6 f4 exf4 1 7 .ixc6 gca 1 8 Wf e2
i.eS 1 9 Wff3
1 9 . . . b4!
Perhaps this is the most impor
tant tactical resource in the Svesh
nikov. Sometimes it is used to gain
control of d4, or open the b-file.
Here it prolongs the diagonal to
Black's dark-squared bishop and
deprives White's knight of a central
outpost.
20 cxb4 .ixb2 21 .id S
2 1 adl ! ? counts on 2 1 . . . !xa3 2 2
ids hdS 2 3 xdS hb4 24 b 3 re
gaining the piece. Perhaps simplest
is 2 1 . . . !es 22 ids 1!Mf6 23 d3 !xdS
24 xdS + 1!Mf7= .
2 1 . . . Wff6
White retains some initiative
after 2 1 . . .hdS 22 1!MxdS+ h8 2 3
adl ha3 24 b3 f3 2S 1!Mxa3 (2S
g 3 f4 26 xa3 1!Md7 27 b2 + g8 28
dSt) 2 S ... fxg2 2 6 fel.
2 2 gad1 gc3 !
1S7
Part 7
22 . . .hd5 23 xd5 Wffc3 24 d3
Wffx b4 25 b3 Wffd4 26 b7 clearly fa
vours White, but 2 2 . . . h8 was play
able .
2 3 gd3
23 fe l !? is an interesting op
tio n: 23 . . . xf3 24 xe6 . However,
Black has 24 . . . Wffc3 25 ltJbl Wffd 3 26
xd6+ h 8 27 hf3 Wffb5 and h e is
not worse, at least.
23 ....ixd S 24 Yxd S+ Yf7 2S 'lf3
gfc8
The activity of Black's pieces and

the clumsy position of ltJ a3 level the
game.
26 b 1 gc1 27 d2 gxf1 +
It was worth retaining the ten
sion by 27 . . . ie5 ! ? with the following variations:
28 a3 Wff a7 29 Wffd 3 Wffb 7=;
2 8 a3 fl+ 2 9 l!Jxfl Wffc4oo ;
2 8 ltJb3 xfl+ 29 xfl Wffc4f!.
28 xf1 .ieS
The bishop cements very well
the split pawns, building a strong
Black cluster in the centre.
2 9 ga3
Or 2 9 g3 Wff c4 30 a3 Wff e4 3 1 <j)g2
fxg3 32 hxg3 f7= .
29 ... Yc4 30 'lhS Yf7 !
Black'skinghas notapawnshield,
158
so it is good to trade queens. On the

other hand, Black should keep rooks
on. In short, the good combination
is rook+bishop vs . rook+knight, in
stead of a queen +bishop vs . queen +
knight The latter might arise after
30 . . . Wffxb4 31 Wffxf5 Wffxa3 32 Wffxc8 +
g7 33 Wie6 with some edge.
31 'lf3 Yc4 32 'lb7 gc7 33 Ya8+
gc8 34 Yxa6 Yxa6 3S g x a6 gc2=
White is unable to keep both
passed pawns o n the queenside.
36 ga3 gb2 37 gb3 gxa2 38 g3
fxg 3 39 h x g 3 .id4 40 e3 @f7 41
bS .ib6 42 @f1 @e6 43 c4 .ixf2
44 b6 dS 4S b7 .ia7 46 gb6+ @d7
47 gd6+ @c7
%-%
28 M astrovas il i s - Johannes sen

Ath ens 20 0 3
1 e 4 cs 2 f3 c6 3 d 4 c x d4 4
xd4 f6 s c3 es 6 d b S d6 7
.igS a6 8 a3 bS 9 .ixf6 g xf6 1 0
d s ts 1 1 .id3 .ie6 1 2 'lhS gg8
1 3 g 3 d4 14 c3 fxe4 1 S .ixe4 .ig4
1 6 Yx h7 gg7 1 7 'lh6 f3+ 1 8 @e2
g s+ 1 9 f3 xe4 20 fxg4 Yc8 2 1
Ye3 Yxg4+ 22 'lf3 Yxf3+ 23 @xf3
ts 24 c2 @f7
2S a4
7 !gS a6 8 ltJa3 bS 9 xf6 gxf6 10 ltJdS f5 11 !d3 !e6

The previous moves are com
mented in the "Step by Step" chap
ter. White has also tried 25 ltJce3
setting the trap 25 .. . i>e6? 26 ltJxfS ! ,
but 2 5 . . . ltJf6 ! is enough for equal
ity. If the other knight goes to e3
- 25 ltJde3, then 25 . . . ltJgS+ avoids
the trap. White's attempt to break
through the queenside is logical.
2S . . . e6 26 ce 3 gba
The mobile pawn centre offers
Black many possibilities for devel
oping the initiative, for instance :
26 . . . ltJgS+ ! 27 cj/g2 b8 28 ltJ b4
(28 axbS xbS) 28 . . . aS 29 ltJc6 b6
30 axbS xbS 31 a2 cs 32 ltJd8+
cj/f6 33 fl f4 34 gxf4 ltJe6+ 35 cj/ hl
ltJxf4. Johannessen prefers to keep
the knight in the centre, but now
White had 27 ltJxfS ! and White is
at least not worse, e.g. 27 . . . cj/xfS 2 8
ltJ e 3 + cj/e6 29 cj/xe4 bxa4 3 0 hfl,
maintaining the clamp on dS.
27 b4? ! cs 28 as?
Mastrovasilis commits a terri
ble positional mistake. Instead of
seeking to reduce the material with
28 ltJc6 c8 29 axbS axbS 30 ltJ a7,
he seals the queenside. That leaves
him without any counterplay.
28 . . JU7 29 e d S .ih6 30 g h e 1
gga 3 1 g ad 1 f4 32 g 4
3 2 . . ..ig S !
O f course, Black should not de
stroy his pawn chain with 32 . . . e4+?
33 xe4+ ltJxe4 34 el ! After the
text White is unable to prevent a
killing check fram the h-file.
3 3 c6 gha 34 b4 a4?
34 . . . ltJb3 would have finished the
game. Now White is kicking again.
3S d4+ xdS 36 fS+ c6 37
gxd6+ b7 38 gg6 gfh7 39 h4?
The final mistake. 39 xgS h3 +
40 cj/g2 xh 2 + 41 cj/f3 2h 3+ 42 cj/g2
ltJxc3 43 hS 3xh5 44 gxhS xhS
45 xeS looks close to the draw.
39 . . . ixh4 40 gxeS .id8 4 1
d 6+ @ as 4 2 e a g h 3+ 4 3 xf4
gfa+ 44 e4 xc3+ 4 S d4 gf4+
4 6 @cs gc4+ 47 d 6 gd3+ 48 e 6
g c 6 + 49 t 7 gd7+
0-1
159
1 e4 c5 2 lll f 3 lll c 6 3 d4 cxd4

4 lll x d4 lll f6 5 lll c 3 es 6 lll d b5 d6
7 i.g 5 a6 8 lll a 3 b5 9 xf6 gxf6
1 O lll d 5 f5 1 1 i.d3 i.e6 1 2 0-0
Part 8
For a quarter of a century White

had been linking his hopes to refute
the Sveshnikov with this very line.
Practice experience has found that
12 . . . i.g7 13 hS favours White, so
1 2 xdS!
is the only decent choice here.
Lately all the top guys prefer this
capture.
1 3 exd5 lll e7
When the players from Chelya
binsk elaborated the whole system
in the 1970s, they linked it with a
direct attack on the kingside where
the open g-file seems to offer Black
good tactical prospects. He push
es f4 in order to gain space advan
tage on the flank, and tries to bring
his heavy pieces closer to the ene
my king . This unsophisticated tactic
is easy to follow and it often reaps
good results, but objectively it is
hardly the right way to treat the po
sition. Look at the game
M azi Nedev
Antalya, E U - c h . 2004
14 c3 .tg7 lS l!Jc2 0-0 16 el f4
17 a4 bxa4 18 xa4 as 19 l!Ja3 c8
20 l!JbS cs 21 c4 fS 22 d2 l!Jg6 23
f3 l!Jh4
160
24 xaS? e4 2S b4 gs 26 fl
bS 27 cxbS .td4+ and Nedev
soon won.
If it were so simple to cr.ush
White's army, everyone would
have played nothing other. In fact,
White lost because he neglected
his defence . One more prophylac
tic move in the diagram position,
24 hl ! , a nd the tide could turn
against Black. In blitz such ap
proach could be rewarding, it even
proved good at an European cham
pionship, but still it counts on poor
defence.
We will advocate another ap
proach , which is more reliable.
Instead of gaining space on the
kingside, we should expand in the
centre with . . . e4:
7 g5 a6 8 llJ a3 b5 9 xf6 gxf6 10 llJdS f5 11 d3 e6 12 0-0

any weaknesses. For his part, Black
Accordingly, our target will not

be the g 2-pawn, but that on d5 . It
could be assaulted by :ga8-c8-c5.
When we capture on d5, our f5pawn is likely to perish, so we must
be sure to protect in advance our
outpost on e4. Therefore, we put the
other rook on e8, having in mind the
manoeuvre 1. .. li:Jxd5 2 VNxfS :ges. In
general, our setup should be the fol
lowing:
1 4 :ge1
After 14 V!fh5 e4 15 e2 g7 16 c3
0-0 17 llJc2 f4 White must take into
account the threat of . . .f3 and par
ry it with 18 f3, when 18 .. .fS main
tains tension.
is eyeing the d5-pawn: 18 llJc2 llJxdS

19 V!fxf5 :gest or 18 :gadl :gc8 19 :ge3
:gcs with a balanced game .
The proposed setup is good
enough when Wh ite keeps his bish
op on the fl-a6 diagonal, planning
to open up the queenside. However,
in the 1990 s White invented a veno
mous plan. He retreated the bishop
to c2 and broke open the kingside
with f3 or g4 after several prepara
tory moves:
1 4 c3 (be sure to meet 14 llJxbS
by 14 . . .g7! with compensation)
1 4 ... g7 1 5 h5 e4 1 6 .ic2 0-0

1 7 gae1 ca
The change of White' s plan re
quires modifications of our ac
tions, too. We must stay passive
ly in the centre and aim for quick
counterplay with b4.
Note that the move order is
vital here !
1 8 c;t> h1
1 4 ... Ag7 1 5 c3 o-o 1 6 h5

e4 1 7 f1 gea
Both sides have defined their

plans . White counts on his better
pawn formation which is without
This position may have been

critical for Black until 20 0 2 , but it
is fun to play it nowadays. Perhaps
the following novelty of Leko, (in
troduced in Dortmund 200 2) was
the last major discovery in this line.
White failed to produce any serious
16 1
Part 8
idea ever since. That game ran:
1 8 gb8!
This sneaky move waits for f3,
while preparing . . . b4. The fi ne point
is that immediate 18 . . . b4 loses due
to the possibility of a rook lift along
the empty third rank: 18 . . . b4? 19
cxb4 hb2 2 0 e3 !
1 9 f3 b4 2 0 ltJbl? ! bxc3 2 1 bxc3
hc3 2 2 l2Jxc3 (22 e2 es 23 fxe4
f4oo) 2 2 . . .9*Vxc3 23 fxe4 f4 ! , Shirov
Leko, Dortmund 200 2 . Black easi
ly repels the attack.
White attempted improvements,
but in vain. Black is holding firmly,
for instance :
20 fxe4 bxa3 21 exf5 g6!
Thus Leko drew the sting of the

most dangerous plan in White's possess1on.
We would also like to draw your
attention to White's attempt to gain
an advantage with 1 4 c4
White had castled already, and
now he opens up the queenside
where he has a pawn majority. On
.
16 2
the other wing, he intends to restrain

Black's counterattack with f3 .
14 ..i g 7 1 5 gb1 e 4 16 i.e2

bxc4 1 7 xc4 0-0 1 8 f3
We see a typical position for this

pawn structure. White can also pre
vent 18 .. .f4 by 18 f4, but it would
give Black a free hand in the centre
where the dS-pawn is a juicy tar
get . After f3 the weakness of the dS
pa wn also ensures full equality:
1 8 gb8 1 9 @h1 gbs 20 Y!!c2
xd5 21 fxe4 b4 22 Y!! a4
fxe4=.
See game 3 0 Efimenko-Moi
seenko, Zlatibor 20 0 6 .
Part 8
1 e4 cs 2 f3 c6 3 d4 cxd4
4 xd4 f 6 s c3 es 6 dbS d6
7 .igS a6 8 a3 bS 9 xf6 gxf6
1 0 d S fS 1 1 .id3 cie6 1 2 0-0
STEP BY STE P
12
i.xd S! 1 3 exd S e7
1 6 .ic4
White aims to cover his queen
side by the bishop while leaving the
dl-hS diagonal open for the queen.
16 e2 0-0 17 d2 tlJg6 gives Black
ample compensation, for example:
18 E!:fel ieS 19 fl h8 with ... 8:g8
and ... f6 to fallow.
1 6 g6 1 7 hS
17 d2 would retain White's
pawn formation flexible, but at the
cost of letting the strongest black
piece come to a striking position:
17 ... 0-0 18 E!:ael (o r 18 E!:fel es
with a standard kingside attack)
18 . . . h4 19 g3 (19 b3? ! h6 20
e2 ttJeS-+ ; 1 9 tlJ d l f4 20 8:xe4 fS
2 1 E!:eel f3-+) 19 . . . h3 2 0 f4 exf3 2 1
8:xf3 ttJeS when White has t o sac
rifice the exchange in search for
salvatio n: 22 E!:xeS (22 E!:f4 h6+)
2 2 . . . xeS+.
1 7 xc3 1 8 bxc3 f6 1 9
h6
It seems sensible to deprive
Black of castling.
Alternatively:
a) 19 f4 0-0 2 0 b3
Or 2 0 E!:ael E!:fc8 21 b3 (21
e2 xc3 2 2 b3 as 2 3 a4 8:ab8t)
A. 14 ttJxbS
B . 14 hS
C. 14 c4
D . 14 E!:el
E. 14 c3
page 163
page 164
page 166
page 167
page 170
A. 1 4 xbS .ig7 1 S c3 e4
163
Part 8
2 1. . . xc3 22 cj{hl cj{h8 ! ? 2 3 gs (23
g4 e3 ! 24 WffxfS WfxfS 2S gxfS l2Jh4
26 f6 gs+) 23 Wigs cj{g7 24 h3 (24
g4 Wffx gS 2S fxgS ltJh4 2 6 gxfS ltJf3t)
24 . . . g8 2S cj{h2 h6 26 Wff h S l2Jh4+
M anion- Shaked, USA 199 2 .
2 0 . . . fc8 2 1 g4 xc3 ! 2 2 gxf5
l2Jh4 2 3 Wff g5 + Wffx g5+ 24 fxg5 l2Jf3+
25 cj{f2 f6+ ;
b) 19 ael g8 20 f4 Ms 2 1 e3
(21 Wffh 6 + g7 22 e3 c8 23 ixa6
xc3 24 xc3 Wffxc3 25 bl e3 + 26
cj{ h l Wffxf4+) 2 1 . . . c8 2 2 b3 (22 e2
aS; 22 ixa6 xc3 2 3 xc3 Wffx c3 24
bl cj{g7+) 2 2 ... xc3 23 xc3 xc3
24 g3 (24 xf5 l2Jh4) 24 . . . a5 ! 25 a4
Wff6 2 6 cj{hl cj{g7 27 e2 c8 28 dl
Wffc3 29 fl h5 ! + Ehlvest-Van Welly,
Moscow 2004.
1 9 . g9a
b5+ cj{e7 23 Wffxg7 xg7 24 a4 l2Jh4

White would be already worse. 2 0
ael i s a better option, intending to
undermine the centre by f3. Then it
would be interesting to try 2 1 . . . g7
22 f3 cj{f8 23 fxe4 e8 24 cj{hl xe4 .
It seems that the game would be
balanced, e.g. 25 xe4 fxe4 26 Wffg5
f6 27 Wffxf6 + Wffxf6 28 M6 + f7 2 9
xf7 + cj{xf7 3 0 cj{ g l cj{f6 3 1 cj{f2 cj{e5
32 cj{e3 l2Jh4 33 g3 l2Jf5 + = .
B . 1 4 '1M h 5 e 4 1 5 e2 g7 1 6
c3 0-0 1 7 lll c2 f4
..
Black has compensation for the

pawn thanks to his more active
pieces . However, we should not un
derestimate White's possibilities.
As poor as his bishop may look, it
is a long range piece, and Black still
has to find a safe haven for his king.
Therefore, trading queens is usual
ly in his favour, because his king can
be quickly centralised. For instance,
after 20 abl Wffxc3 21 ixa6 g7! 2 2
164
The mere fact that Black is able

to play this move without any sacri
fices shows that he solved the open
ing problems.
1 8 f3
White might want to capture the
saucy pawn by 18 Wffg5 f5 19 Wffxf4,
but then 19 . . . l2Jxd5 2 0 Wffd2 l2Jb6 2 1
adl d 5 levels the game accord
ing to Gorelov, as 22 l2Jb4 Wffd6 ! 23
l2Jxd5 ad8 (23 . . . fd8 24 c4 l2Jxd5
25 cxd5 ac8oo is also possible) 24
c4 l2Jxd5 2S cxdS c8 gives him suf
ficient compensation.
In Smirnov-Nijboer, Istanbul
2003, instead of taking on f4, White
7 gs a6 8 ttJa3 bS 9 xf6 gxf6 10 ttJdS fS 11 d3 e6 12 0-0

chose 19 fdl, when 19 .. .f3 leads to
a balanced position: (the alternative
19 . . . ttJg6 2 0 \Wxd8 fxd8 2 1 a4t is in
ferior, but 19 . . . @h 8 ! ? 20 xf4 ttJg6
deserves consideration) 20 gxf3
to gain an advantage, but 18 . . .fS ! 19

fel (19 axbS f3 2 0 c4 axbS 2 l hbS
b8 2 2 c6 fxg2 23 @xg2 f6 with
an attack) 19 . . .f3 2 0 gxf3 ttJxdS 2 1
fxe4 ttJf4 2 2 \Wf3 fxe4 2 3 \Wg3 @h8 +
leaves the White king unprotected.
1 8 ...fS 1 9 fxe4
Black intends to play . . . es fal
lowed up by . . . f6 so White has no
time for moves like 19 @hl. On the
other hand, 19 a4? ! is dubious due
to 19 . . . b6 + .
1 9 . . .fxe4 20 .ig4 V!fc7
2 0 . . . @h8
20 . . . f6 ! ? is another good option: 21 W hl g6 22 e3 h6 23
d4 (23 f4 f8 ! 24 f3 f6 2S fxe4
\Wh4 26 fl fxe4 27 xe4 f8f! Ro
gozenko) 23 . . .f4f! 24 gl f8 2S
xg6+ ttJxg6f!, Nijboer-Avrukh,
Plovdiv 200 3 .
2 l @hl eS ! ?
Grischuk-Krasenkow, Bundesli
ga 2003 saw 2 1 . . .g8 22 e3 ttJg6 23
ttJd4 ! es 24 ttJxfS (24 ttJc6 h4 2S
ttJxeS ttJxeS 26 f4 ttJg6 27 gl h6 ! f!)
24 . . . exf3 with compensation.
22 f4
Or 2 2 ttJd4 g8 23 h4 f8 24
ttJe6 ttJg6 2S hS f6f!.
2 2 ... g8 (22 ... f6 23 \WhSt) 23
h4 g7 24 hS
Or 24 ttJd4? xd4 2S d4 ttJg6+;
24 ttJe3 ! ? ttJg6 2S \Wxd8 axd8 26
ttJxfS gf8 ! 27 ttJxg7 xg7 28 a4
4 29 fl df8 30 axbS axbS 31 f3
exf3 32 xbS ttJeSf!.
24. . . \Wf8 2S ttJe3 f6 26 h3 h6
27 ttJxfS ttJxfS=.
18 a4? i s another logical attempt
21 gad 1
This is the latest attempt of
Shabalov to shake the assessment
of the diagram position as pleasant
for Black. White prepares ttJd4 . In
stead, 21 ael c4 22 e6 + @h8 23
a3 b3 is double-edged.
21 ... <at?h8 22 lll d 4
It is arguable that the fl-rook
would stand better on el: 22 fel ! ?
cs+ 23 @hl f3 ! ? 2 4 gxf3 (24 xe4?
f2 2 S gl xc2 26 xe7? fxg2 + 27
xg2 fl + 2 8 gl f2 - +) 24 . . . \Wf2
2S ttJd4 hd4 26 cxd4 exf3 2 7 M3
fS 2 8 e2 (28 g4 g8 2 9 e2 \Wxe2
30 \Wxg8 + @xg8 31 xe2 ttJxdS 32
g4 ttJe3=) 28 . . . xe2 29 xe2 xhS
16S
Part 8
30 xhS c8 = .
2 2. . .%Yc5 23 h 1 'Wxd 5 24
llJts 'Wes
25 l2Jxe7 Wixe7 26 fS h6 ! is
also level: 27 he4 (27 d4 'WeS)
27 .. .'Wxe4 28 Wixh6 f3= .
2 5 llJxd6 'Wx h 5 2 6 i.xh 5 e3
27 g3 i.e5 28 e4 fxg3 29
hxg3 gf5 30 i.e2 gaf8!=.
The last finesse. In the stem game
Krasenkow played 30 . . . g8, which
eventually also led to a draw.
C. 1 4 c4 i.g7
Perhaps this is the most consis

tent attempt to gain an advantage.
White had castled already, and now
he opens up the queenside where he
has a pawn majority. On the other
wing, he intends to restrain Black's
counterattack with f3 .
1 5 gb1
a) 1 5 cxbS? ! e 4 16 e2 hb2 17
l2J c4 hal 18 'Wxal 0-0 19 b6 llJxdS
20 dl llJf4 2 1 fl dS+;
b) 15 l2Jc2? ! e4 ! (15 . . . 0-0 16 cxbS
e4 17 e2 axbS 18 hbS hb2 19 bl
eS) 16 e2 bxc4 17 hc4 0-0+;
c) 15 Wff b 3 hides more venom. It
brought White success in Vescovi166
M atsuura, Sao Paulo 2006 after

15 . . . 0-0 16 cxbS e4 17 e2 f4 18 f3 e3
19 acl l2Jf5 20 c4 l2Jd4 21 Wffd 3 'Wf6
2 2 b6 . We think that Black should
take up the gauntlet by:
15 . . . bxc4 ! ? 16 'Wa4+
16 l2Jxc4 has no much sense
as the queen on b3 will only help
us activate the rook, for instance:
16 ... 0-0 17 f3 b8 18 'Wa3 llJxdS 19
hfS llJf4+.
16 . . . Wffd7 ! ?
1 6 . . . M8 contradicts the principles of quick and harmonious de
velopment which Black typically
embraces in the Sveshnikov. Still,
play is unclear after 17 l2Jxc4 e4 18
c2 Wffc7 19 b3 d4 20 @hl cs.
17 c2 c3 18 l2Jc4 d8oo . Now
both 19 Wixd7+ @xd7 or 19 Wixa6 0-0
would be fi ne for Black.
d) 15 Wid2 e4 16 e2 has prac
tically disappeared due to the the
matic positional sacrifice 16 . . . b4 ! ?
Black keeps the enemy bish

op passive and gains time to cas
tle and start his kingside offensive.
That will enable him to control the
game. As to White's extra pawn, Tal
had put it nicely long ago : Black has
an extra open file in exchange !
17 Wfxb4 b8
7 gs a6 8 liJa3 bS 9 xf6 gxf6 10 liJdS fS 11 d3 e6 12 0-0

17 ... 0-0 ! ? needs more tests : 18
d 2 (18 abl f4 19 cS ! ? is worth
considering) 18 . . . liJg6 19 abl (19
cS dxcS 20 liJc4 f4 21 cj{hl d4 2 2
d 6 f6 i s good for Black) 1 9 . . .f4 2 0
f3 (or 20 liJc2 f3 2 1 gxf3 1!Mh4--+) 2 0 . . .
e 3 2 1 W!el W! gS--+ gave Black a tre
mendous position in Koch-Hor
vath, Bischwiller 1999.
18 W!a4+ Ms 19 abl hb2 .
Black has good chances for a
kingside attack, see game 29 Kot
ronias-Shirov, Calvia, ol. 2 004.
1 5 e4 1 6 i.e2 bxc4
This is much more topical then
16 . . . 0-0 . White often fails to defend
his king.
1 7 ti:)xc4 0-0
W!d2 bS ! ? 20 liJ e3 W!b6 ! , when even

the best answer 2 1 hbS would not
have saved White from trouble, e.g.
2 1 . .. d4 2 2 fel liJxdS 2 3 cj{hl xe3
24 e2 dS 2S b3 aSoo or 2S . . . cS ! ? 2 6
b 4 d6 2 7 g 3 d4oo.
19 . . . bs 20 liJe3 xb2 21 xb2
hb2 22 d2 g7 23 bl W!c7 24
ha6 b8 2S cl W!a7 26 fl b2 27
e l xa 2+ Ivanchuk-Lautier, Odes
sa rapid, 2006 ;
b ) 1 8 W!d2 b8 !
Threatening t o grab the dS
pa wn. 18 . . . liJg6 is less concrete and
leaves White a tiny edge follow
ing 19 f4 exf3 2 0 xf3 f4 2 1 b4 liJeS
22 h3 liJxc4 2 3 hc4 es 24 cj{hl
Areshchenko-Wang Yue, Lausanne
200 6 .
1 9 b4 bs 2 0 liJe 3 f4 2 1 xbS fxe3
22 W!xe3 axbS 23 W!xe4 liJg6 ! Black
has the better game, since White's
pawns are weak, for instance, 24
bcl a8. Instead, Dworakowska
Aksiuczyc, Brzeg Dolny 1996 saw
23 .. .fS? which only compromises
Black's position.
1 8 gb& 1 9 h1 E!b5 20 c2
Or 2 0 liJe3 b4.
20 ti:)xd S 21 fxe4 ti:) b4 22
a4 fxe4=.
This position was reached in
game 3 0 Efimenko-Moiseenko,
Zlatibor 200 6 . It is rather equal.
1 8 f3
White must prevent 18 .. .f4. He
can also do that by:
a) 18 f4. This move does put a
radical stop to Black's kingside ac
tivities, but now we get a free hand
in the centre. The dS-pawn is a juicy
target.
18 . . . b8 19 cj{hl
One can appreciate such prophy
laxis after seeing the game Stefans
son-Krasenkow, Gausdal 1991: 19
D. 1 4 E!e1
With this move White starts a
typical middlegame redeploying of
his forces. It enables the d3 to re
treat to fl which significantly hin-
167
Part 8
ders Black's counterplay. The fine
point is that . . .f4-f3 will be a strike
at thin air without the bishop being
o n e 2 . Although we can find a game
dated of 1994, this plan attracted at
tention after the convincing victory
of Yurtaev:
14 E!el ig7 15 E!bl 0-0 16 c4 bxc4
17 llJxc4 aS 18 'WhS e4 19 i.fl E!c8 20
llJ e3 f4 21 llJfS e3 22 id3 llJg6 23
fxe3 E!e8 24 E!fl fxe3 25 llJxg7 e2 26
he2 cj{xg7 2 7 E!xf7+ cj{xf7 2 8 '\Wxh7+
cj{f6 29 E!fl + cj{es 30 V!ixg6 V!ib6 + 3 1
cj{ h l V!ie 3 3 2 i.bS 1-0 , Yurtaev-Gre
bionkin, Samara 2 0 0 2 .
Lately 14 E!el tends t o displace 14
c3 as most popular line.
14
ig7
1 8 cxb5
In
Lutz-Moiseenko,
Plovdiv
2 0 0 3 , White preferred 18 b4, but
it does not affect Black's plans:
18 . . . i.eS . (18 . . . \WgS 19 \&cl ! ?) Here
Lutz went wrong with 19 V!ihS? !
V!if6 2 0 E!e3 h6 2 1 llJc2 ? ! llJf4 2 2
'Wdl cj{h7+. Rogozenko suggests 19
g3 V!igS 20 cj{h l with double-edged
play.
18 'WhS? ! only loses time :
18 . . . V!if6 19 cxbS axbS 20 hbS llJf4
2 1 V!idl V!ig6 t .
18 c x b 5 axb5 1 9 i.xb 5
The most testing continuation.
Now 19 . . . hb2 20 llJc4 (20 E!xb2
E!xa3 =) 20 . . . i.eS 21 a4 cj{h8 22 g3
E!g8 23 i.c6 E!a7 24 E!b3 , Czarnota
Dl. 15 E!bl
0 2 . 15 c3
01 . 1 5 gb1
White intends t o advance his
queenside pawns and create a pas
ser.
1 5 . . 0-0 1 6 c4 e4 1 7 f1 g6
.
Like i n most Sveshnikov lines

with opposite coloured bishops, the
initiative, especially on the kingside,
is more important than the pawns.
16 8
Felgaer, 2 0 07, is equal, but Black

should have went on with 24 . . . 'Wf6 !
25 cj{hl i.d4 26 V!id2 (26 llJxd6 hf2
2 7 E!fl V!ixd6 2 8 f2 f6 2 9 V!id4
E!c7 30 E!c2 llJeS 31 E!bc3 E!b 8 ! f!)
26 . . . icS 27 aS llJeS = . Another de
cent option is 19 . . . i.eS 20 i.c6 E!a7
21 ctJc4 E!xa2 22 V!ihS V!if6 2 3 i.d7 f4
24 E!xe4 hb2 = , Oleksienko-Kulj as
evic, Pardubice 20 07. However, we
recommend :
1 9 . . .'%Yg5
7 igS a6 8 ttJ a3 bS 9 ixf6 gxf6 10 ttJdS fS 11 id3 ie6 12 0-0
Black's attack is very danger

ous and even the exchange sacrifice
slows it down only tern porary:
2 0 E:e3 ieS 21 ic6 a7 22 ttJc4
ttJh4 23 g3 hg3 24 hxg3 ltJg6 2S
a4 f4 26 ttJxd6 e7t, Murariu-Gen
gler, Crete GRE 29.10. 20 07.
White can also trade queens,
with some drawing chances:
2 0 cl xcl 21 excl ixb2 2 2
xb2 a3 2 3 cc2 ttJeS+, Kalash
nikov-Oleksienko, Moscow 20 07.
0 2 . 1 5 c3 0-0 1 6 \Wh 5
White can open up the a-file by
16 ttJc2 e8 (16 . . . d7! ?) 17 a4, but
that does not ensure him an edge:
17 . . . b6 18 axbS axbS 19 ttJb4 xal
20 xal e4 21 ifl cs 22 ttJa6 b6 ,
Negi-Alekseev, Kirishi 200 S
1 6 . . . e4 1 7 f1 g e 8
1 8 g ad 1
The dS-pawn needs protection.
18 ttJc2 ttJxdS 19 xfS es gives
Black an initiative: 2 0 g4 (No one
wished to copy White's play from
the game Inarkiev-Babula, Pardu
bice 20 03 which saw 20 h3 ttJf4 2 1
g4 ttJ e 6 2 2 f4 exf3 2 3 xf3 ltJgS 24
b7 ttJ e4 2S ttJ b4 e7 2 6 c6 dSf!)
20 . . . hS 21 dl gs 22 a4 ae8 (Or
22 . . . b4 23 cxb4 e3 24 fxe3 ttJxe3 2S
ttJxe3 xe3 26 xd6 ae8.) 2 3 axbS
axbS 24 ttJe3 ttJf4f! planning b4 and
dS with a satisfactory game.
1 8 . . . gc8
18 ... ttJg6 is a popular, but not
very convincing gambit: 19 xfS
es 2 0 g4 fS 2 1 g3 ! ? f4 2 2 g4
f6 23 ttJc2 ae8 24 ttJd4 (24 f3 !?)
24 . . . :gxdS 2S a4t, Zude-Srienz, Dres
den 07. 04 . 2007
19 g93
Apart from this move, White
tried 19 ttJc2, but the freestyle (what
a term for practically a computer
chess ! ) game Valori, New_Rybka
1. 1 32 - Heff alump, playchess.com
INT 20 06 , showed an easy equalis
er: 19 . . . cS 20 ttJe3 (20 d2 c8 ! 2 1
edl ltJg6f!) 2 0 . . .f4 2 1 ltJfS ( 2 1 ltJg4
fS 22 ttJh6+ xh6 2 3 xh6 ltJg6+)
2 1 . . . ltJxfS 22 xfS f6 ! = .
1 9 . . . g c s 20 c 4 bxc4 2 1 g h 3 h 6
2 2 lll xc4 lll xd 5 ! 2 3 \Wxf5 e 3 !
This break leads to mass elimi
nation and equality.
24 g xd5
The alternatives leave Black
more winning options:
a) 24 fxe3 ttJxe3 2S xcS ttJxdl
169
Part 8
2 6 ltJxd6 (26 ds elt) 26 . . . eS !
27 c80 (27 c6 dS ! 28 ltJe4 cS !
29 a4 cl-+) 27 . . . xc8 28 ltJxc8
e8+ 29 d3 ! ltJxb2 30 ltJd60 ltJxd3
(30 . . . d8 31 ds !f8 32 ltJe4=) 31
ltJxe8 !d4+ 32 hl ltJf2 + 33 gl
ltJe4+ 34 hl = ;
b ) 2 4 f3 e2 ! ( 24 . . . exf2 + 2 S
xf2) 2 S he2 xe2 26 xe2 ltJf4
27 e4 ltJxh3+ 28 gxh3 gs+ fol
lowed by . . . dS+ ;
c) 24 !d3 ltJf6 2S f4 exf2 + 2 6
xf2 ltJe4 ! ? ( 2 6 . . . c7oo) 27 he4
xe4 28 b3 (28 ltJxd6 !d4 ! - + ; 2 8
xcS e l+ 29 M2 xdl+ ) 2 8 . . . e7
29 hd3 e2 30 f3 gs 31 g3 ds 32
xdS xdS 33 xdS xa2 = .
24 . . . e 2 ! 2 5 xe2
2S xcS? dxcS ! (2S . . . el 26 dS
e6oo) 26 he2 xe2 27 bl (27 g4
el+ 28 g2 dl 29 d3 gl+ 30
h3 fl + 3 1 h4 !f6 ++) 27 . . . d4
28 ltJe3 d2 29 ltJfl el 30 d3
xd3 31 xd3 hb2+ .
2 5 .. . xe2 2 6 f1 Ye8 ! 27 g93
gxe3 28 li:) xe 3 'lb5+ 29 e1 =.
E. 1 4 c 3 .ig7 1 5 V9h5
lS el 0-0 16 hS e4 17 !fl
transposes to line 0 2 .
l S llJc2 i s a n introduction t o ano
ther plan, connected with a4. Black
can meet it with either . . . bxa4 or
. . . b6, for instance: lS . . . 0-0 16
a4 e4 17 !e2 bxa4 (or 17 ... b6 18
axbS axbS=) 18 xa4 b6 19 b4
cs 2 0 ltJe3 ab8 21 b3 xb4 2 2
cxb4 Sadvakasov-Khalifman, So
chi 200S, when 22 . . . d4 ! ? 23 dl
(23 ha6 f4) 23 . . . f6 (23 . . . xb2 24
170
xb2 hb2 2 S ha6 b8=) 24 h a6

f4 2S ltJc4 g6 would have offered
Black rich play.
1 5 ... e4 1 6 .ic2
16 !e2 transposes to line B .
1 6. . . 0-0 1 7 g ae1
In this line White is planning to
crush Black's centre with f3 or even
g4. Then the c2-bishop would be
come extremely awkward, hitting
h7. Therefore, 17 adl? ! is incon
sistent: 17 . . . c8 and then :
a) 18 f3? b4 19 llJbl bxc3 2 0 ltJxc3
b6 + 21 f2 xb 2 - + ;
b) 18 !bl cs 1 9 llJc2 ttJxdS 2 0 f4
( 20 xfS llJxc3 2 1 xcS ttJe2+ 2 2 hl
dxcS 23 xd8 xd8 - +) 20 . . . b6 2 1
hl a S 2 2 a 3 b 4 2 3 axb4 axb4 24
cxb4 ltJxb4 2S ltJxb4 xb4 26 xd6
bs 27 fdl xb2 28 h3 b3+
c) 18 !b3 f4 19 ltJc2 fS 20 ltJd4
hd4 2 1 xd4 (21 cxd4 ltJg6+)
2 1 . . . llJg6 22 el c7! 23 f3 e3 24
!c2 f6 2S d3 g7, Pucher-Lau
tier, Montpellier, 2 007. White is
in a difficult position without plan,
while Black is building up pressure
along the g-file .
'
1 7 ... Yc8
7 !gS a6 8 l!J a3 bS 9 !xf6 gxf6 10 l!JdS fS 11 !d3 !e6 12 0 -0
The queen defends the fS

pawn, thus enabling the e7-knight
to move. It is also eyeing the bish
op on c2 . White is unable to achieve
any advantage, so he keeps on try
ing new ideas:
El. 18 f3
E 2 . 18 g4
E3. 18 !bl
E4. 18 !b3
ES. 18 hl
E 1 . 1 8 f3
A risky continuation, which may
turn insufficient even for a draw.
1 8 . . . b4 1 9 cxb4
19 l!Jbl bxc3 20 l!Jxc3 hc3 2 1
bxc3 xc3 2 2 !bl l!Jg6 (23 fxe4
f4f!) 23 xfS exf3 24 xf3 xf3
2S :9:xf3 :9:ae8 was equal in Asrian
Wang Yue, Khanty Mansiysk 2 00S.
Perhaps Black could shape better
this idea by starting with 20 . . . l!Jg6
21 fxe4 and only then 21 . . .hc3 2 2
bxc3 xc3 23 :9:e2 f4 with dark
squared strategy.
1 9 . . . i xb2 20 fxe4 .ixa3 2 1
e 3 'Mfxc2 2 2 g3+ tiJ g 6 23 h 3
fd 80
24 'Mfxh 7 +
24 exfS l!J e S 2 S gS+ 8 2 6
:9:xh7 e8 ! - +
24 . . . t8 2 s h S !
O r 2 S exfS l!JeS 2 6 :9:xa3 e8+.
25 . . .f4 ! ?
This move was suggested by Ro
gozenko in CBM 104. Black is not
satisfied with a draw in the varia
tion 2S . . . l!JeS 26 E!:hxfS :9:a7 27 h8 +
e7 2 8 E!:xeS+ dxeS 2 9 xeS+ f8=
3 0 h 8 + e7 3 1 es+ f8 .
After the text, 2 6 ::fs llJ es
(26 . . . :9:a7 ! ? 27 xg6 !cl 28 hl
c4 Rogozenko) 27 :9:5xf4 :9:a7 28
h8 + e7 2 9 h4+ d7 30 h3 +
c7 31 xa3 b8 favours Black.
(Rogozenko)
E 2 . 1 8 g4
A very aggressive approach,
but Black easily achieves good play
thanks to his more active piece s:
1 8 . . . b4 1 9 cxb4 tiJxd5
19 . . . hb2 h a s been known to be
equal ever since the game Sax-Je
len, Medulin 1997: 20 gS + l!Jg6
21 gxfS ha3 22 fxg6 fxg6 23 he4
:9:f7 24 !d3 c3 2S :9:e3 xb4 26 :9:g3
!b2 27 :9:g4 lf2- l/2.
171
Part 8
Now Black is threatening with
18 . . . tt:JxdS.
1 9 f3 Wes+ 20 h 1 e 3
20 Wxf5
An attempt to improve on the
game Nunn-Reinderman, Leeu
warden 199S, which saw 20 gxfS
tl:Jf6 2 1 h3 dS 2 2 @hl h8 23 gl
g8 24 b3 b7 (24 . . . b8 ! ) 2S tl:Jc2
ad8 26 tt:Jd4 tt:Jd7 with good com
pensation.
20 ... \WxfS 2 1 g xf5 xb4 22
gxe4 g ab8 23 .i b 3
23 @hl? ! hb2 ! 2 4 gl+ @h8 2 S
f6 hf6+ would leave White won
dering why did he so generously
gave out his pawns. After the text
the position looks drawish.
23 . . . dS 24 ge2 gfe8 25 gfe 1
gxe2 26 gxe2 d 3 27 .ixd 5 f4
28 gd2 gxb2 29 gxb2 .ixb2 30
c4 x d 5 31 xb 2 @ g7=.
E3. 1 8 .ib1
White enables the manoeuvre
tt:J a3-c2-e3 which would enhance
the efficiency of pawn breaks like
f3 or g4. Black must hurry with his
counterplay before it became too
late. As the dS-pawn is still immune
in view of 18 . . tt:JxdS? 19 he4, he
chooses :
1 8 . . . ge8
172
No w 2 1 hfS? tl:JxfS 2 2 xfS e2

would be sad for White, so he might
want to blockade the passer with 2 1
e2 . However, Black has good play
then after either 2 1 . . . h6 ! ?oo 22 tl:Jc2
tt:JxdS 23 xfS es or 21. .. tt:JxdS 22
xfS h6. That's why we'll focus o n:
2 1 f 4 x d 5 22 Wxf5 f6 23
.id3
Black has a comfortable game.
He can choose between:
a) 23 . . . xfS ! ? 24 hfS b4 !oo 2S
cxb4 ab 8 ;
b) 23 . . . e2 2 4 xe2 xfS ! ? 2 S
xe 8+ xe8 26 MS e2 2 7 tl:J c 2
d2oo, (A. Sokolov) 28 gl dS ;
c) 23 . . . b4 ! ? 24 cxb4 xfS 2S
hfS ab8 26 f3 dS 27 tl:Jc2 (27
fxe3 xe3 2 8 xe3 xb4+) 27 . . . d4
28 tt:Jxd4 xb4 29 tl:Jc2 xb2=.
E4. 1 8 .ib3
White's idea is similar to the pre
vious line, but White protects the
dS-pawn. Its downside is that Black
can hinder 19 tl:Jc2 by:
1 8 . . . as
7 gs a6 8 '2J a3 bS 9 xf6 gxf6 10 'tJ dS fS 11 d3 e6 12 0-0

23 . . . f6 24 g4 d S
Here Nunn erred with 2 S a3? !
and would have been worse fallow
ing 2S . . . xgS ! 26 fxgS xd4+ 27
cxd4 '2Jc6+.
The fine point is that after

19 ttJxbS a4 (20 '2Jxd6? loses to
20 . . . d7) 20 dl cs 21 '2Jd4 (Or
2 1 e2 ttJxdS 22 gs h6 ! 23 g3 f4
24 h4 a3 ! 2S b4 c8 26 cl b8
27 '2Jxd6 e6 28 'tJbS xbS 29 xbS
'2Jxc3 30 xc3 xc3 31 xf4 xa 2=)
2 1 . .. xdS Black regains the pawn
with strong centralisation. Since
the game Nunn-McShane, Hastings
1997, White has not find improve
ments:
2 2 gS
The alternatives are in Black's
favour: 22 e3 fb8 ! 23 h3 (23
g3 xb2 24 gs es 2S f4 exf3 26
M3 c8+; 23 b3 f4 ! 24 xdS ttJxdS
2S xe4 '2Jxc3 26 f4 b4 !+) 23 . . .
h 6 ( 2 3 . . . xb2 ! ? 24 xh7+ cj{f8f!) 24
b3 f4 2 S xdS ttJxdS 26 'tJfS axb3 27
axb3 (27 xb3 '2Jxc3 2 8 '2Jxd6 b6+)
27 . . .xc3 28 c2 (28 '2Jxd6 e3f!; 2 8
'2Jxh6+ Ms 29 'tJfS eS+) 2 8 . . . e8
29 '2Jxd6 e3 ! 30 fxe3 e6+.
22 . . . es
22 . . . fe8 !? deserves attention:
23 f3 xa2 24 fxe4 xb2 2S 'tJxfS
ctJxfS 2 6 xfS a3 ! ? with initiative.
2 3 f4
Or 2 3 f3 f4 24 h4 '2Jg6 ! 2S h3
e3 26 c2 cS !+.
1 9 \Wg5 \Wb7 20 f3
Or 2 0 ttJxbS xbS 2 1 xe7 a4
22 c2 (22 dl xdS+; 22 c4 e8
23 xe8 fxe8 24 c2 hb2+)
22 . . . xb2+.
20 f 3 h6! 2 1 \Wg 3 !
a) 2 1 d2 a4 2 2 c2 b4 23 cxb4
xdS 24 xdS ttJxdS 2S fxe4 '2Jxb4
26 bl xb2 27 '2Jc4 c3=, M . Hoff
mann-Dub, Budapest 20 0 3 ;
b) 2 1 hS a4 2 2 c2 b4 23 '2Jc4
bxc3 24 fxe4 fxe4 2S xe4 (2S bxc3
fS 26 '2Je3 b2 27 h4 f7+) 2S . . .
cxb2 2 6 g4 xdS 2 7 xh6 '2Jg6+;
c) 21 f4 a4 22 c2 b4 23 cxb4
(23 '2Jc4 xdS 24 '2Je3 xa2 2S fxe4
'2Jg6 26 xd6 f4-+) 23 . . . xb4 24
fxe4 '2Jg6 ! 2S f3 (2S xfS xb2 26
eS ! xeS 27 'tJ b 1 ac8 ! 2 8 e4 ce8 !
29 cj{hl g7t) 2S . . .f4 26 e2 xb2
27 '2Jc4 xa2 28 '2Jxd6 a3oo .
21
. . .
a4 22 ic2 b4
Now Black has sufficient play.

173
Part s
ES. 1 8 h 1
The reason behind this move is,
besides prophylaxis, to prepare the
opening of the g-file with g4, while
keeping the third rank free for a
rook lift via e3.
1 8 . . . b8!
years ago . White won a great

number of games by pushing f3 or
g4 and destroying the helpless black
king. You can see a good example
of White's play in game 31 Kolev
Echavarria, Istanbul ol. 20 0 0 .
Then Leko introduced the strong
rook move and ever since White has
often been even struggling to main
tain the balance. It turned out that
White is unable to bring his knight
into play:
19 i.bl b4 20 cxb4 b4 21 f3
xb2 22 fxe4 c3 = ;
1 9 i.b3 aS 2 0 f3 b 4 (Rogozenko
suggests 20 . . . a4 !? 2 1 i.c2 b4 22 fxe4
bxa3 23 exfS l!Jg6 !) 21 cxb4 b4 2 2
fxe4 fxe4= .
Thus White began trying t o un
dermine the centre:
19 g4 b4 2 0 cxb4 l!JxdS 21 gxf5,
but the simple 2 1 . . . @h8 ! neutralises
allWhite's attacking chances:
22 i.b3 l!Jf6 23 gs ds 24 l!Jc2
bToo ;
22 he4 l!Jf6 23 f3 xb4= ;
2 2 gl hb2 23 h6 c3 ! 24
e4 f6 2S h3 g8t Topalov-Le
ko, Dortmund 20 0 2 .
A little more complicate is:
This line was very topical ten
1 9 f3 b4 20 fxe4
The inclusion of moves @hl b8 makes the capture 20 cxb4? bad
due to the long variation 20 . . . hb2
21 fxe4 ha3 22 e3 (22 exfS? xc2)
22 . . . xc2 23 g3+ (23 h3? fc8 24
xh7+ @f8) 23 ... l!Jg6 24 h3 fd8
2S xh7+ @f8 26 hS l!JeS 27 xfS
b7 28 xeS dxeS 2 9 h8 + We7 30
xeS+ @d7 3 1 xf7+ Wc8 32 e6 +
@b8 when White loses since he has
He only should keep the kingside

close, even at the cost of a few
pawns : 2 3 cxb4 xb4 (23 . . . l!JxdS
24 fxe4 f4oo) 24 fxe4 f4 2S gn4 l!Jg6
26 ffl hb 2= 27 l!Jbl ac8 28 e2
i.eS. Usually White prefers to bring
his knight into play:
23 c4 Wxd 5 24 fxe4
24 l!Jxd6 bxc3 ! favours Black: 2S
fxe4 (2S bxc3 fd8+ 26 dl cs+ 27
@ h l xc3) 2 S . . . d4+ 26 @ h l f4 27
xf4 l!Jg6 2 8 bxc3 cs+.
24 f xe4 Wxc4 25 exf5
This position arose in the game
Shirov- Kramnik, Wijk aan Zee
20 0 3 . Shirov suggests:
25 . . . xfS ! 26 xf5 ae8! (26 . . .
bxc3? 2 7 e4; 2 6 . . . xa2? ! 27 d3
fe8 28 efl e6 29 Sf2) 27 ef1
bxc3 28 bxc3 Wxc3 (28 . . . dS ! ? oo) 29
Wxc3 .ixc3 30 .ixa4 e5=.
174
7 gs a6 8 l2Ja3 bS 9 f6 gxf6 10 ltJdS fS 11 d3 e6 12 0-0

no check on b6 . (Rogozenko)
The game Shirov-Leko, Dort
mund 20 0 2 saw 20 ltJbl? ! bxc3 2 1
bxc3 hc3 2 2 l2Jxc3 ( 2 2 E: e 2 es
23 fxe4 f4oo) 22 . . . \Wxc3 23 fxe4 f4 ! ,
when even the best 2 4 bl (24 b3?
l2Jg6 2S E:cl \Wf6 26 \WfS \We7t turned
well for Black in the game) 24 . . . l2Jg6
2S eS ! is just about equal.
20 . . . bxa3 21 exf5 g 6 !
2 1 . . .f6 ! ? might b e good enough,

but it gives White much more chan
ces to put Black under pressure af
ter 2 2 E:f3 ! E:e8 (22 . . . axb2 23 E:g3 +
g7 24 E:xg7+ ! xg7 2S f6 + ! xf6
26 h4 + + -) 23 E:g3 + l2Jg6 24 E:fl
(24 E:gl E:b7 2S fxg6 fxg6) 24 . . . \Wc4
2S d3 ! ? (2S \Wdl ! ? is also rather
unclear) We could not find an ad
vantage for White here, but the var
iations are so long and complicat-
ed, that we cannot exclude possible

mistakes. 2 1 . . . l2Jg6 is more clear and
well tested.
22 f6
22 E:f3 E:xb2 23 e4 E:e8 24 fxg6
hxg6 2S h4 c4- + ; 22 fxg6 ? !
fxg6-+ o r 2 2 b3? ! xc3 2 3 fxg6 fxg6
24 \We2 eS-+ are clearly worse. 2 2
bxa3 ! ? \Wxc3 2 3 fxg6 hxg6 24 \Wdl
E:b2 is playable, with a possible
draw ahead.
22 . . J xb2
2 2 ... f6 is equal, 2 3 E:xf6 E:xb2
24 hg6 fxg6 2S E:xg6 + = , Musil
Leiner, Czechia, 2004.
2 3 fxg7
23 E:e3 ? ! looks appealing, but
23 . . . E:xc2 24 E:h3 xh3 2S xh3
E:xa2 ! 26 fxg7 E:b8+ is able to cool
down White's enthusiasm.
23 . . . xg7 24 .id3
Or 24 bl \Wxc3 2S \WfS (2S E:cl
eS-+) 2 S . . . E:xg2 !+.
24 ... '%Yxc3 2 5 g93 g 8!
If Black i s willing t o take risks, he
might prefer 2S . . . d2 26 gS ! f6 27
g3 with unclear position. After the
text, the source game Dominguez
Schandorff, Esbjerg 2 0 03 quickly
steered to a draw:
26 hg6 hxg6 27 E:xc3 gxhS 2 8
E:xa3 E:e 8 29 E:xa6 E:ee2 30 E:xd6
E:xg2 31 E:gl E:xgl+ 3 2 xgl E:xa2 = .
17S
Part 8
1 e4 cs 2 f3 c6 3 d4 cxd4
4 xd4 f6 S lll c 3 es 6 db S d6
7 ig S a6 8 lll a3 bS 9 1xf6 gxf6
1 0 lll d S fS 1 1 id3 i e6 1 2 0-0
COMPLETE GAMES
29 Kotron ias - S h irov

Calvia ol. 2 S . 1 O .2004
1 e4 cs 2 ll:) f3 ll:) c6 3 d4 cxd4 4
li:) xd4 li:) f6 s li:) c3 es 6 li:) db S d6 7
i gS a6 8 li:) a 3 bS 9 ixf6 gxf6 1 O
li:) d S fS 1 1 id3 ie6 1 2 0-0 ixdS
13 exdS li:) e 7 1 4 c4 e4 1 S ie 2 ig 7
1 6 'l;Yd2 b4 1 7 'l;Yxb4 gba 1 8 'l;Ya4+
@ta 1 9 g ab1 ix b2
of the bishop o n h5 by 23 . . . h5 ! ? 24
hl g7 25 Wffxa6 g as 2 6 Wff b5 .ia3 ,
obtaining a slightly better version of
the stem game .
2 1 . . .f4 22 'l;Yc2
Or 22 ll:Jf5 ll:Je7!
22 . . . 'l;Yf6 23 li:) d 1 f3
White is at a juncture. In all the
20 li:) c2
Or 2 0 Wffc2 .ie5 2 1 Wff d2 gg8 = .
2 0 . . . li:) g & 2 1 li:) e 3

In Leko-Kramnik, Linares 2003
White chose the more testing 21 ll:J b4
Wfff6 22 ll:Jc6 ge8 ! 23 f4 (23 Wffx a6?
ll:Jf4 24 .idl ll:Jxg2 ! ) 23 . . . .id4+ 24
hl g7 25 .ih5 .ic5f2 with un
clear position. We think that Black
should have a voided the appearance
176
lines he would be a pawn up in an

endgame, but the activity of Black's
rook should compensate it. Most
obvious is 24 bf3 exf3 25 g3 (or 25
gxb2 gxb2 26 Wffx b2 g7 27 '!9xf6 +
xf6 28 ll:Je 3 gb8oo) 25 . . . g7 2 6
gxb2 gxb2 2 7 Wffx b2 Wffx b2 2 8 ll:Jxb2
ll:Je5 29 gel gbs with sufficient com
pensation. In this variation the f3pawn is quite awkward, so Kotroni
as chose:
7 igS a6 8 ttJa3 bS 9 ixf6 gxf6 10 ttJdS f5 11 id3 ie6 12 0-0

24 gxb2 g xb 2 2S Yxb2 Yxb 2
26 xb 2 fxe 2 2 7 ge1 <j;>g7 2 8 gxe2
gb8
Again the control over the b-file
balances the game.
29 f3 ti)f4 30 gxe4 gxb2 31 gxf4
gxa2 32 gg4+
%-%
30 Efim e n ko - M o iseenko
Zl at ibo r 200 6
1 e 4 cs 2 ti)f3 ti) c 6 3 d 4 c x d 4 4
ti)xd4 ti)f6 s ti)c3 e s 6 ti) d bS d6 7
igS a6 8 ti) a 3 bS 9 ixf6 gxf6 1 O
ti) d S fS 1 1 d 3 ie6 1 2 0-0 xd S
1 3 exd S ti) e 7 14 c4 e 4 1 S e 2 g7
1 6 g b 1 bxc4 1 7 ti)xc4 0-0 1 8 f3
gb8 1 9 <j;> h 1 gbs
The principal aim of Black i n this

structure is to roll his central pawn
pair as further as possible. With
his 18th move White stopped f4, so
Black turned to the next step of his
to-do list. Now he wins the dS-pawn
while White captures e4. In result of
the clash in the centre most pieces
disappear f ram the board.
20 Yc2 ti) xd S 21 fxe4 ti) b4 22
Ya4 fxe4 2 3 ti)e3 ti)d S 24 ti)xdS
gxdS 2 S Yxe4 ges 26 Yf3 as 27
ic4 Ye7 28 b3 <j;>h8 29 g b d 1 ts
30 gd s t4 31 id3 ge3 32 Yhs es

33 c4 a4
Play is equal. Simplest for Black
is to seek exchanges or disturb
White's pieces.
34 gbs axb3 3S axb3 'ld7 36
Yh6 g7 37 Yh4 es 38 Yh6 g7
39 'Mfh4 es
%- %
3 1 Kolev - Echavarria
Ista n b u l ol. 2000
1 e4 cs 2 ti)f3 ti) c6 3 d4 cxd4 4
ti) xd4 ti)f6 S ti) c3 e6 6 ti) d b S d6 7
f4 es 8 igS a6 9 ti) a 3 bS 1 o xf6
g xf6 1 1 d S fS 1 2 d 3 e6 1 3 0-0
xd S 14 exdS ti) e 7 1 S c3 i g7 1 6
YhS e4 1 7 c2 Yc8 1 8 g ae 1 0-0
1 9 <j;> h 1 ge8
Black discovered the right move
19 . . . b8 ! three years later.
20 f3 b4 21 cxb4 xb2
22 fxe4 ! xa3 23 a4 ! (White's

attack is decisive) 23 . . .f 4 24 gxf4
b2 2s gh4 g7 26 Yxh7+ <j;>f8
27 gf1 ti) g 8 28 ixe8 Yxe8 29 Yg6
es 30 gh7 ga1 31 a4 gc1 3 2 g 3
Y e 7 33 bS a x b S 34 axbS f6 3 S
Yts 'Mi e s 36 'Mixes ixe S 3 7 b 6 g b 7
3 8 gfxf7+ gxf7 39 gxf7+ <j;>xf7 4 0
b 7 ti)f6 4 1 b81l ti) xe4 4 2 Yc8 1 -0
177
Part 9
1 e4 c5 2 lll f3 lll c6 3 d4 cxd4

4 lll x d4 lll f6 5 lll c 3 e s 6 lll db5
d6 7 lll d 5
QUICK R EPERTO I RE
This move looks shocking at first.

White plugs himself the hole on dS,
which is the only major drawback
of the Sveshnikov in general . At the
same time, 7 ltJdS is the first choice
of ALL the engines? ! Where is the
trick? !
First of all, White gains space in
the centre. Second, he accomplish
es it with tempo, repelling the c6knight fram its good stand. Third,
he makes room for the bS-knight
on c3. In short, White gets a few
immediate benefits, which pleas
es computers . However, in the long
run, Black gets an easy and pleas
ant game, based on the full control
of the es-square . It allows him to
play all over the board, and every
too often to collect some queenside
white pawn.
7 . . . lll xd5 8 exd5 lll b 8 9 c4
Piece pressure on the queenside
is inefficient and might submit
White to a direct attack: 9 a4 ie7
10 ie2 0-0 11 0-0 tt:Jd7 12 ie3 a6 13
l2J a3 fS 14 f3 f4 15 if2 f6f!.
9 ... ie7
Now White must define his set
up. He chooses plans with f3 or f4 .
178
We'll examine them separately,

since the difference in the position
of White's f-pawn implies different
Black approaches .
A . f3-setup.
There are several key positions
we should remember:
1. 10 d3 0 - 0 11 0 - 0 a6 12
c3 f5 13 f3 d7
3 d4 cxd4 4 ttJxd4 ttJf6 S ttJc3 es 6 ttJ dbS d6 7 ttJdS

In this structure Black will be
happy to trade dark-squared bish
ops. That would secure the gS-h6
squares for his heavy pieces and
also would reduce the threat of cS .
White can avoid the exchange
by 14 ie3 igS lS if2 , but from f2
White's bishop only hampers the de
fence . The thematic break . . . e4 be
comes especially efficient: lS . . . Wf6
16 We2 e4 !
tive, but Black still has a chance to

balance the game :
18 ... a4! 19 li)xa4

Or 19 b4 ttJb3 2 0 bl c8 .
1 9 ... .ixa4 20 bxa4 b6 =.
This position arose in Kotroni
as-Eljanov, Warsaw 200S. White's
queenside pawn structure is static
and he is unable to make any prog
ress.
17 fxe4 f4 18 hl ttJeS 19 id4
We7. Black has full compensation.
Hence White usually chooses :
14 cbhl, (preparing a retreat
square for the bishop on gl) but
14 ... ig5 15 b4 a5 ! opens up the
a-file and gives Black counterplay
against the extended queenside
pawns. 16 a3 axb4 17 .ixg5 Vxg5
18 axb4 al 19 xal Wfe3 2 0
ie2 li)b8 ! = The knight is heading
for a6.
2. 10 .ie2 0 - 0 11 0 - 0 a6 12
li)c3 f5 13 f3 li)d7 14 cbhl (14 ie3
igS lS Wd2 he3 + 16 Wxe3 aS=)
14 ... a5 ! ? 15 .ie3 ig5 16 igl
li)c5 17 b3 id7 18 a3
This is a model setup for both
sides . White is looking forward to
b4 which would earn him the initia-
f4-setup.
1 0 ie2 0 - 0 11 0 - 0 a6 12 li)c3
f5 13 f4 if6 14 cbhl li)d7
a.
Black's plan is to take on f4 and

use the es-square as a strong out
post. (If White plays g3 and recap
tures on f4 by pawn, Black's knight
goes to cS and eventually to e4, even
at the cost of a pawn. That will un
derline the weaknesses in White's
179
Part 9
castling position.) The fine point
is when to trade the dark-squared
bishops. It should be done only if
White's queen is unable to occupy
some of the central dark squares,
especially d4. For instance, 15 c2
exf4 16 ixf4 .ieS ! shows an excellent
timing for that exchange because
the queen is on c2 . Play can go on
with 17 !!adl (17 g3 ltJf6 ! ) 17 . . . ixf4
18 !!xf4 ltJeS 19 b4 as 20 a3 axb4 2 1
axb4 .id7! 2 2 d2 b6= , see game
32 Spraggett-Yakovich, Santo
Antonio 200 1.
15 .ie 3 exf4 16 .hf4 e 5! 17
gel d7 18 b4 a5 19 a3 axb4 2 0
axb4 \Wb6 !
White has trouble defending his

pawns.
180
Important!
1. Black's primary aim is to com
plete development and try to ex
pand in the centre .
2 . Despite his pawn majority on
the kingside, Black rarely wins by
direct attack. You should aim first to
activate all your pieces. The previous
diagram shows one good setup.
3. Do not be too afraid of the
thrust c4-c5. It is dangerous only if
White dominates in the centre, as in
the fallowing example:
33 Yud a s in- K h a rlov
M oscow 1 99 1
Black i s deprived o f counterplay

and can only watch the opponent
preparing c5-c6. See the "Complete
Games" chapter.
Part 9
1 e4 c5 2 f3 c6 3 d4 cxd4
4 xd4 f6 5 c3 es 6 db5
d6 7 d5
STEP BY ST EP
7 . . . xd5 8 exd5 b8
8 . . . ltJce7 is a decent option, but it
is not in our repertoire.
9 c4
Unusual alternatives are:
a) 9 !e2 a6 10 ltJc3 !e7 11 0-0 fS
12 f4 !f6 13 !e3 0-0 14 ltJa4 exf4 ! 15
hf4 bS 16 ltJc3 ltJd7t Lilj a-Schan
dorff, Copenhagen 1996.
b) 9 f3 a6 10 a3 !e7
This original manoeuvre of

the White queen aims to hamper
Black's normal development by hit
ting d6. We'll soon see that it does
not achieve its goal. Besides, Black
can also respond with 10 . . . b6, taking
the sting of 11 !d2 due to 11. . . !b7.
ll !gS
11 f4? ! ignores development and
should backfire badly after 11. . . 0-0
12 fxeS axbS 13 xa8 dxeS ! with a

very strong initiative, for example:
14 d6 hd6 15 a7 b6 !? 16 a8 ltJa6
(16 . . . e4 !?);
1 1 !d2 i s similar t o the main
line : 11. . . 0-0 12 !as b6 13 !b4 d7 !
This standard motive neutralises
White's threats. Then 14 ltJc3 (14
hd6? axbS; 14 ltJxd6? aS) 14 . . . !b7
15 b3 c7 16 !e2 ltJd7 turned well
for Black in Gullaksen-Cherniaev,
Gibraltar 200 3 .
1 1 . . .f6 ! 12 !d2 0-0
13 !b4 (0 r 1 3 !a5 b6 14 !b4 d7

15 0-0-0 d8, Eisenbeiser-Nedev,
Mulhouse 2004.) 13 . . . d7! 14 c4
b6 15 ltJxd6 ( 15 ltJc3 f5 16 !e2 !b7
17 b3 c7 18 0-0 ltJd7 19 f4 exf4
2 0 M4 !gS is very comfortable for
Black, Solleveld-Alekseev, Santo
181
Part 9
Domingo 20 03) lS . . . aS 16 cS hd6
17 cxd6 ib7 18 ic3 d8 19 0-0-0
l!J a6+;
c) 9 a4
In contrast with 9.c4, here
White plans a piece attack on the
queenside . He wants to fix weak
nesses and gain outposts for his
pieces with the help of the a-pawn.
However this pawn has no impact
on the centre and Black gets a free
hand there and on the kingside:
9 . . . e7 10 ie2
10 ie3 l!Jd7 11 d2 (11 l!Jxa7??
xa7 12 ha7 as +) 11. . . a6 12 l!J a3
fS 13 f3 0-0 14 l!Jc4 b6 lS aS bS 16
l!Jb6 l!Jxb6 17 axb6
Thus White opens the a-file and
generates some threats. 17 hb6
e8 18 c4 (18 e2 g6 19 0-0 igS
2 0 b4 f4 2 1 c4 h6 2 2 h3 f6--+)
18 . . . d7 would be fine for Black: 19
ie2 (19 b3 ih4+ 20 f2 (20 g3 if6)
2 0 . . . e7 2 1 ixh4 xh4+ 2 2 f2
e7t) 19 . . . bxc4 20 hc4 bs+.
17 . . . b8 18 c4 (18 !d3 b6 !oo)
It looks like White has seized the

initiative, but it is Black's turn and
he hits first :
18 . . . h4 + ! ? (18 . . . bxc4 19 hc4
b6 was also good for Black in
Borisek-Wang Yue, Calvia ol 2004 .
182
Black can also shape the exchange

sacrifice as follows: 19 . . . h4 + ! ? 20
g3 f4 2 1 f2 gs 2 2 c2 xb6 ! ) 19
g3 f4 20 f2 gSt. White's setup is a
failure . Let's return now to the more
sound 10 ie2 :
1 0 . . . 0-0 1 1 0-0 l!Jd7
12 @hl
12 aS? ! a6 13 l!Jc3 fS 14 f3 gS lS
@hl ixcl 16 xcl l!Jf6 17 f4 e8 18
d2 id7 19 fxeS xeS 20 if3 c7
2 1 l!Je 2 ae8 22 b3 cS+ Grigorov
Vyzmana vin, Tbilisi 1986;
12 f4 a6 13 l!J a3 bS ! 14 @hl bxa4
1S l!Jc4 exf4 16 id2 Rowson-Adams,
London 1998 , when 16 . . . l!JcS seems
equal. Instead, the game went
16 . . . l!JeS 17 as e8 18 l!Jb6 f3 19
hf3 b8 with unclear position.
12 ie3 a6 13 l!Ja3 fS 14 f3 (14 f4
exf4 lS ixf4 l!JeS is good for Black.
Compare it to the 12 .@hl line) 14 .. .f4
lS f2 f6f! . Black has good chanc
es to build up a dangerous attack.
12 .. .fS 13 f4 a6 14 l!Ja3 exf4 ! lS
ixf4 l!JeS 16 l!J c4 l!Jxc4 17 hc4 if6
18 c3 gS ! 19 ie3 f4 2 0 id4 fS .
Black's attack is running faster, for
example: 2 1 hf6 ? ! xf6 22 b3
h6 ! ? 23 xb7 g4-+.
9 .ie7
...
3 d4 cxd4 4 l2Jxd4 ltJf6 S l2Jc3 es 6 ltJdbS d6 7 ltJdS

It is possible to start with 9 . . .
a 6 . Then 10 l2Jc3 e7 transposes
to the main line. Only 10 .a4 is of
independent significance, but in
our opinion it is not dangerous for
Black:
10 . . . l2Jd7 11 cs
Now Black can pick up the

gauntlet by sacrificing the exchange
with 11. . . dxcS ! ? or prefer the solid
11. . . b8 :
a ) 11. . .dxcS ! ? 12 d6 axbS 13 xa8
c4 In this highly unbalanced posi
tion Black has interesting active
play: 14 e3
14 d2? ! hd6 lS as xaS 16
has e7 lead to a better endgame
for Black in Pavlovski-Spasov, Sofia
1996 . Black has compensation also
after 14 a3 b6 lS e3 hd6 .
14 . . . hd6 lS 0-0-0 l2Jb8 16 a4
The fine point of Black's idea
is that he has good prospects even
without queens, e.g. 16 a7 l2Jc6
17 b6 xb6 18 hb6 eroo . The
other option, 16 a7, is bad due
to 16 . . . gS+ ! 17 e3 e7 ! ? 18 a7
(18 a7? 0-0 19 !xb8 gs+ 20 bl
hb8+) 18 . . . l2Jc6 19 b6 0-0 20
xbS fSt.
16 . . . bxa4 17 xa4+ l2Jc6 18 hc4
(18 xc4 0-0 19 d3 l2Jd4!oo) 18 . . . 0-0
with compensation d ue to the vul

nerable position of White's king.
b) n .. . gb8 12 l2Jxd6+ hd6 13
cxd6 0-0 14 d2 (14 .ie3 ? ! l2Jf6 lS
a7 a8 16 cS b6 17 a3 b7t, Wes
terinen-Kramnik, Gausdal 1992 ; or
14 d3 ltJf6 lS gs xd6 16 hf6
xf6 =) 14 . . . b6= .
We'll examine here two princi

pal plans:
A. 10 .id3 (intending f3) p. 184
B. 10 e2 (intending f4) p. 18S
a) 10 e3 0-0 11 d2 (thus White
aims to prevent the exchange of the
dark-squared bishops with e7gS) spends too much time and en
ables Black to launch a kingside of
fensive: 11. . . a6 12 l2Jc3 fS 13 f3 l2J d7
14 e2 f6 ! ? lS cl l2Jf8 16 b4 as 17
a3 axb4 18 axb4 g6 19 g3 (or 19
0-0 f4 2 0 .if2 h3 winning mate
rial) 19 .. .f4, Perez Candelario-Moi
seenko, Sanxenxo 20 04, with a nas
ty attack;
b) 10 cS is seldom seen because
Black is not oblige to take it. Af
ter 10 . . . 0-0 11 e2 l2Ja6 only White
might have problems in view of the
weakness of his dS-pawn.
183
Part 9
A. 1 0 .id 3 0-0 1 1 0-0 a6 1 2
c 3 f5
1 3 f3
13 f4 does not fit into White's set
up. When a black knight appears on
eS, the bishop will have to retreat,
as in the game Benhadi-Amin, Cai
ro 1999: 13 . . . ttJ d7 14 c2 g6 lS ie3
(Or lS bl if6 16 b4 c7 17 b3
bS ! t Vink-Harikrishna, Wijk aan
Zee 200 1 ; lS ctthl if6 16 a4 llJcS
17 i.e3 id7 18 as c8 19 b3 e8+,
Stanojoski-Nijboer, Plovdiv 2 00 3 .)
1S ... if6 16 adl e8 17 ctt h l exf4 1 8
M4 llJeS 19 ie2 id7 2 0 b3 c8 = .
1 3 d7 !
There is no reason to give White
extra options with 13 . . . igS 14 ixgS
xgS lS f4 ! ? exf4 16 e2 .
1 4 @ h1
White can preserve his dark
squared bishop, but after 14 ie3
i.gS lS if2 it only hampers the de
fence. In most cases its black coun
terpart turns to be more dangerous :
1S . . . f6 16 e2
Alternatively: 16 c2 llJcS 17 i.e2
as 18 a3 a4 19 ixcS (19 ael i.d7
20 ixcS dxcS 21 i.d3 h6 22 llJdl
ae8 23 e2 ie7 ! . Black relocates
184
the bishop to d6 , with a strong at

tack in Rahman-Spasov, Novi Sad
ol. 1990) 1 9 . . .dxcS 20 ttJxa4 id7 2 1
ttJc3 ie3 + 2 2 ctt h l a6t with a bish
op pair and bright prospects.
16 . . . e4 !
This thematic break is especial

ly efficient with the bishop on f2 ,
since it disrupts the coordination
of White's pieces . 17 fxe4 f4 1 8 ctt h l
ttJeS 19 id4 e7. Black has full com
pensation, for example: 20 dl f3
2 1 gxf3 ih3 2 2 ixeS xeS 2 3 gl
if4.
1 4 .igS 1 5 b4 a5 1 6 a3
axb4 1 7 xg5 Y;Yxg5 1 8 axb4
gxa1 1 9 Y;Yxa 1 Y;Ye3 20 .ie2
lll b8!
We can strike a balance here.
Black has almost forced draw with
2 0 . . . bS 21 cl (21 ttJxbS? xe2 22
ttJxd6 ttJf6+; 2 1 cxbS? ! ib7 22 dl
c8 23 d3 f2 24 gl h4oo)
2 1 . . . xcl 22 xcl bxc4 23 ttJbS i.b7
24 llJxd6 (24 ixc4 f6 2S ttJc7 ttJb6=)
24 . . . ixdS 2S ixc4 ixc4 26 c4 e4
27 fxe4 fxe4 28 ctt g l e3 29 e4 ttJf6
30 xe3 d8 31 e6 (31 llJe4?? dl +
3 2 M2 tlJg4+) 3 1 . . . b8 32 b S b6 3 3
g 3 h6=. White is unable to unpin his
d6-knight. However, the novelty of
3 d4 cxd4 4 l2Jxd4 l2Jf6 S l2Jc3 es 6 ttJdbS d6 7 ttJ ds

Valerij Filippov 20 . . . ttJbS keeps the
fight on:
21 Y;\fb2
White has also tried 21 W!cl b6
22 a3 .
22 l2Ja2 l2Ja6 23 c3 is passive,
23 . . . d7 24 g3 (24 gel gas 2S dl
d4 ! ) 24 . . . gas 2S g2 \Wd4 (it would
be interesting to try 2S . . . a7 ! ? 26
ltJcl bS 27 l2Jb3 E:cS 2 S ltJaSoo) 26
xd4 exd4 27 gd1 l2Jc7 2 S ltJcl ga4
29 bS gb4 30 f2 ltJaS 3 l gxd4 .ixbS
32 l2Ja2 gb2 =, Xu Yuhua-Stefanova,
Krasnoturinsk 200 S .
2 2 . . . l2Ja6 23 gbl \Wf2 2 4 b2
d7 2S fl d4. Black had good
counterplay in the source game Be
lozerov-Filippov, Tomsk 2004.
21 lll a & 22 gb1 id 7 2 3
lll d 1 Y;\fd4!? (23 . . . b6oo) 2 4 g3!
The endgame was difficult for
White in Svidler- Timofeev, Mos
cow 2004: 24 xd4? exd4 2S gl
gbs !+.
24 gba 25 Y;\fc3.
The game is balanced.
B. 1 0 .ie2 0-0 1 1 0-0 a6 1 2

lll c 3 f5
Bl. 13 a3 (preparing b4)

B 2 . 13 f3
B3. 13 f4
13 hl l2Jd7 is not of independ
ent significance, since after f3 or f4
play transposes to other lines.
13 b4 could be similar to line Bl,
but Black also has 13 . . . as which de
stroys the plan with c4-cS . Note that
13 . . . l2Jd7 is another good option: 14
b2 b6 (Sveshnikov recommends
14 . . . l2Jf6 ! ? lS \Wb3 hS 16 gadl f4 17
cS fSoo) lS \Wb3 hS 16 gadl f6
17 l2Ja4 (17 gfel e4) 17 . . . \Wc7f! with
sufficient counterplay;
13 a4? ! in conjunction with c4 is
slow: 13 . . . l2Jd7 14 as f6 lS d2 ? ! e4
16 l2Ja4 eS 17 f4 exf3 lS gxf3 gbS 19
e3 l2Jf6 20 h3 l2Je4---+ Kagan-E . Gel
ler, Skara 19SO .
8 1 . 1 3 a3 lll d 7 1 4 b4 if6 !
If White adopts here a waiting
strategy with lS hl e4 16 \Wc2,
Black builds on with 16 . . . eS . Then
White should anticipate the attack
with f4, when Black takes on f3 and
brings all his forces on the kingside:
17 f4 exf3 lS gxf3 l2Jf6 19 gs \Wes
lSS
Part 9
2 0 afl tlJg4 2 1 h3 Wg6t Adla
Kharlov, Maringa 1991 .
1 S e3 e4 with mutual chances .

Black should keep his dark-squared
bishop on. See the very instruc
tive game 32 Yudasin-Kharlov,
Moscow 1991 for detailed explana
tion of the ideas of both sides.
Black, but retreating to f2 would

not be any better due to the power
ful position of the g5-bishop: 15 !f2
Wf6 ! 16 ttJ a4 (Or 16 b4 Wh6 17 Wb3
!e3 18 ttJ a4 hf2 + 19 filf2 b6 2 0 dl
b8 21 Wc3 b7! ?. Black manoeu
vres the rook to c7, using that 2 2
c5? fails t o 2 2 . . . bxc5 23 bxc5 ttJxc5
24 ttJxc5 c7 !) 16 . . . Wh6 17 Wb3 f6
18 ttJb6 !f4 ! 19 !g3 (19 g3? Wh3 ! -+)
19 . . . !e3 + 2 0 !f2 !f4= .
1 4 . . . a S ! ? 1 S e3 g s 1 6 g 1
c s 1 7 b3 d7 1 8 a 3
8 2 . 1 3 f3 d 7
Black plans t o play . . . !g5, but he

wants to do it when White had al
ready moved his dark-squared bish
op. 13 . . . !g5 is less precise in view of
14 hg5 Wxg5 15 Wcl.
1 4 @h1
The point i s that 1 4 !e3 !g5 15
Wd2 he3 + 16 Wxe3 a5= is fine for
186
This i s a model setup for both

sides. White is looking forward to
b4 which would earn him the initi
ative, but Black still has a chance to
balance the game:
18 . . . a4 ! 1 9 xa4 (19 b4 tt:Jb3 20
bl c8) 1 9 . . .xa4 20 bxa4 b6=
Kotronias-Eljanov, Warsaw 200 5 .
White's queenside pawn structure
is static and he is unable to make
any progress. On the other hand, his
bishop pair should be able to par
ry an eventual Black attack on the
kingside .
8 3. 1 3 f4 f6 1 4 @ h 1
3 d4 cxd4 4 ttJxd4 ttJf6 S ttJc3 eS 6 ttJdbS d6 7 ttJ dS

White might prefer to control the
es-square by 14 g3 ttJd7 1S c2 exf4
16 gxf4, but then Black would have
active play after 16 . . . ttJcS 17 if3 bS !
18 b4 ttJe4t as in Erwich-Kulj asevic,
Stork 20 0 2 .
Another option i s 1 4 c2 ttJd7
when lS @hl transposes to 14 @hl.
Instead, lS xfS? ! would be ex
tremely risky: lS . . . exf4 16 e6 + E:f7
17 xd6 (17 hf4 ?? ttJeS 18 heS
dxeS- + ; 17 xf4?? ttJeS 18 f6
gxf6 -+) 17 . . . ieS 18 b4 (18 e6??
id4+ 19 @hl ttJeS-+) 18 . . . as 19 a4
6 1 ?c=:.a
oo .
14
. . .
!i) d 7
bishops, because White would get

full control of the centre: 16 . . . ieS?
17 hes ttJxeS 18 d4t.
17 E:cl d7
17. . . tlJg6 ! ? 18 ie3 ieS is a decent
alternative, e.g. 19 id4 if4 20 c2
h4 2 1 igl id7=.
1 8 b4 a5 19 a 3 axb4 2 0 axb4
b6 ! 2 1 b3 :gfc8 2 2 g3 a7+.
White has trouble defending his
pawns.
B3b. 15 c2 exf4 16 chf4

e5
B3a. lS ie3 ; B3b. lS c2
Rare moves:
lS g3 ttJcS 16 c2 a5 17 E:bl id7 18
b3 c7 19 a3 g6 is good for Black;
lS a4 exf4 16 !xf4 ttJeS 17 ie3
tlJg6 18cS? ! ieS+, Bologan-Hamdou
chi, 20 0 2 .
B3a. 1 5 e3 exf4 16 chf4

lll e 5!
Note that after the exchange ofthe
eS-pawn, Black should avoid trading
A good timing for this exchange !

Compare this position with the pre
vious diagram. Here White's queen
cannot occupy d4.
187
Part 9
Still, 16 . . . l!JeS is a decent alter
native. It gives good attacking pros
pects, e.g. 17 b4 l!Jg6 18 !g3 f4 19
f2 as.
17 g3
With this move White aims to
deprive Black's knight of the eS
square . It also restricts the light
squared bishop, although it is not so
bad on d7 either. Other moves show
White often losing the initiative:
17 adl hf4 18 f4 l!JeS is simi
lar to the main line, 19 b4 as 20 a3
axb4 21 axb4 !d7 ! = , see game 33
Spraggett-Yakovich, Santo An
tonio 20 0 1 ;
1 7 !d3 g6 18 ae1 M4 ( 1 8 . . . b6 ! ?
1 9 l!J e 2 Wfff6 2 0 Wffd 2 M4 2 1 Wffxf4
l!JeS 2 2 l!Jd4 d7=) 19 f4 l!JeS is
fine for Black as 2 0 cS? ! stumbles
into 2 0 . . . e8 +;
17 hes l!JxeS 18 b4 d7 19 c5
Wffh 4t Sarthou-Nataf, France 2003.
188
Perhaps the most challenging

answer is:
17 W1d2 M4 18 Wffxf4 Wfff6 19 acl
b6 20 b4. Apicella-Wagner, Clichy
20 07 saw further 20 . . . b8 21 l!J a4
!b7 2 2 cS bxcS 23 bxcS l!JxcS 24
l!JxcS dxcS 2S xcS, when 2S ... Wfe7
26 Wffc4 Wffd6 27 !f3oo would have
been roughly equal . We suppose
that Black should include . . . as at
some moment, (maybe 20 . . . aS) to
get rid of the weak a-pawn.
After the text we like:
17 f6 !
Black i s planning t o complete
development with . . . !d7 and . . . c8,
hoping to get to the weakened white
king in the future.
Instead, 17 . . . gS? ! is bad, since af
ter 18 !d2 Black faces development
problems. Conversely, 17 . . .Wfff6 is a
reasonable alternative and should
lead to a balanced game.
.
. . .
Part 9
1 e4 cS 2 f3 c6 3 d4 cxd4
4 xd4 lll f6 s lll c3 es 6 lll dbS
d6 7 d S
COMPLETE GAM ES
32 Yudas in - K harlov
M oscow 1 99 1
1 e4 c 5 2 tll f3 tll c6 3 d 4 cxd4 4
tll x d4 tll f6 5 tll c 3 e5 6 tll d b 5 d6 7
tll d 5 tll x d 5 8 exd 5 tll b 8 9 c4 ie7
10 e2 a6 1 1 tll c3 f5 1 2 0-0 0-0 13 a3
Usually White prefers to re
strict the mobility of Black's pawns
by f3 or f4. However, that weakens
the gl-a7 diagonal end especially
the e3-square. In this game Yuda
sin embraces the most straightfor
ward approach. He wants to push
c4-cS while staying passive on the
kingside . Since 13 b4 could be met
by 13 . . . aS, White starts with a3 . . .
1 3 . . . tll d 7 1 4 b4
1 4 . . . e4 ? !
This looks imprecise i n view of
lS f4 ! ? Then 1S . . . f6 16 gel es
(Blatny) 17 .beS ltJxeS 18 d4 wins
a clear temp o, compared to the stem
game.
This variation explains why it is
better to play first 14 . . . f6 ! .
We have also analysed 14 . . . ltJf6,
when lS e3 f4 16 d2 fS or lS f3
b6 + 16 hl d7 would be fine for
Black. However, lS f4 ! poses prob
lems . White retains an edge after
lS . . . aS 16 e3 exf4 17 E1xf4 ltJg4 18
.bg4 fxg4 19 xf8 + .
1 5 ie3 if6
lS . . .f4 is dubious, because 16
d2 e4 17 ltJxe4 ! ? .bal 18 xal ltJf6
19 ltJxf6+ xf6 2 0 xf6 xf6 2 1 cS
gives White an edge.
1 6 .id4
We have noted in the previ
ous chapters that White does not
mind exchanging bishops, provid
ed his queen could occupy d4. 16
cl ! ? would cost a piece after 16 . . .f4
17 .bf4 .bc3 18 .bd6 e8 19 xc3
f6 20 d2 xd6 21 cS eSoo, but
White would have been the active
side .
1 6 . . . ieS ? !
This i s a positional mistake.
189
Part 9
Black needs all his pieces in order
to retain more tension. We propose
16 . . . ltJeS ! 17 cS d7oo .
The exchange of the dark-squa
red bishops gives White a free hand
in the centre, but he immediately
stumbles into a tactical trap .
1 7 i.xe 5?!
Best was 17 cS ! with a n edge, e.g.
17 . . . f6 (or 17 . . . f6 18 c6 h6 19
g3) 18 xeS ltJxeS 19 Wff d2.
1 7 . . . li:) xeS 1 8 '%Yd4
1 8 . . . .id7 ? !
Kharlov misses the chance to
complicate things by 18 . . .f4 ! 19
ltJxe4 f3 20 gxf3 xf3 with perfect
compensation for the pawn.
1 9 cs
The opening is over and White
should be happy with his position.
Black is passive and lacks a clear
plan. He can only stay and wait, for
only one good piece is insufficient
to build an attack, especially when
the centre is so mobile.
1 9 . . .'%Yf6
It is difficult to resist such a
move . (seemingly winning a tempo
on the threat of 20 . . . ltJf3+) Follow
ing 19 . . . Wie7 2 0 fdl ac8 21 acl
(21 c6? bxc6 22 xa6 cS+) 2 1 . . . c7
190
22 a4 White can prepare c6 or fol

low up with as first.
20 gfd 1 gfc8 21 gac1 gc7 22
h3 '%Ye7
Probably Kharlov already real
ised that something with his setup
went wrong, and he begins to rede
ploy his pieces for defence . Alterna
tives do not change the character of
play: 22 . . . ac8 23 e3 f8 24 a4t;
22 ... f8 23 a4 e8 24 c6 bxc6 2S
dxc6 xc6 26 bS axbS 27 ltJxbSt.
23 @f1
This move is not a mistake, but it
looks artificial. White had more en
ergetic options, as 23 a4 b8 24 c6
bxc6 2S dxc6 xc6 26 bS. The bad
news for Black is that if White does
not like this variation, he can keep
on manoeuvring, seeking the best
timing for the breakthrough cS-c6.
23 .. . ge8 24 c6
Black was already threatening
to take on cS, so White must go for
ward.
24 . . . bxc6 25 dxc6 gxc6 26 li:) d S
'%Yf7 2 7 gxc6 li:) xc6 (or 2 7 . . . xc6 2 8
xa6)
28 '%Yb 6?
The critical moment ofthe game.
Yudasin apparently underestimated
3 d4 cxd4 4 l!Jxd4 l!Jf6 S l!Jc3 es 6 l!JdbS d6 7 l!JdS

Black's counterattack and his posi
tion quickly falls appart. He should
have prevented .. .f4 with 28 c3 !
(or 2 8 d2) keeping the edge, for
28 . . .f4?? would lose to 29 4.Jc7+ - .
2 8 ...f 4 2 9 .ixa6 YhS!
White's game is hopelessly com
promised . Ironically, the prophy
lactic move cj{fl only helps Black's
offensive.
30 gc1 f3 31 tllf4 fxg 2+ 32 @ g 1
Y g S 3 3 gxc6 Yxf4 3 4 g c 3 'ld2 ? !
( 2 4. . . eS ! was better, intending to
push . . . dS) 3S g g 3? (3S e3 !) 3 S . . .
e 3 ! 3 6 gxe3 Yc 1 + 37 @xg2 .ic6+
38 f3 gxe3
0-1
We have seen a very instructive
game ! In this example we have seen
most of the ideas of the 7 4.JdS varia
tion, together with typical positional
mistakes.
33 S p ragg ett - Yakov ich

Santo Antonio 2001
1 e4 cs 2 tll f 3 tll c 6 3 d4 cxd4 4
tll xd4 tll f6 s tll c3 e s 6 tll dbS d6 7
tll d S tll xdS 8 exdS tll b 8 9 c4 .ie7
1 O .ie2 0-0 1 1 0-0 a6 1 2 tll c 3 f S
1 3 @ h 1 tll d 7 1 4 f4 .if6 1 S Yc2 exf4
1 6 i.xf4 i.eS
Unlike the previous game,

here this move makes sense, since
White's queen is misplaced on c2 .
Keeping the dark-squared bishops
on leads to interesting and sharp
er play, for instance : 16 . . . 4.JeS 17 b4
l!Jg6 18 g3 f4 19 f2 as, followed
by . . . es, fs, l!Jh4.
1 7 g ad 1 ? !
With several common moves
White loses the initiative and his
queenside pawns become from a
major asset, a weakness. Spragget
should have directed all his efforts
to execute cS, even at the cost of a
pawn. In the game Apicella-Wagner,
Clichy 20 0 7 White's play was more
consistent: 17 d2 ! xf4 18 xf4
f6 19 E:acl b6 2 0 b4. Here, 2 0 . . . as
deserves attention, aiming to elimi
nate the weak a-pawn. Instead, the
game saw 2 0 . . . E:b8 21 l!J a4 b7 2 2
cS ! bxcS 23 bxcS tt.JxcS 24 tt.JxcS dxcS
2S E:xcS, when 2S . . . e7 2 6 c4 d6
27 f3oo would have been roughly
equal .
1 7 . . . .ixf4 1 8 gxf4 til e s 1 9 b4
aS!
A thematic move i n this pawn
structure. Black aims to activate his
rook, but he also exchanges his po
tentially (after cS-c6) weak pawn.
20 a3 axb4 21 axb4 i.d7 22
'ld2
Apicella-Saric, Cannes 2007 saw
22 cS ! ? dxcS 23 bxcS as 24 c6 bxc6
2S E:a4 cs 26 a2 8:xa4 27 tt.Jxa4=
with a level game. After the text
Black has a good game.
22 . . . Yb6 23 g b 1 gfe8 24 h 3
@h8
Black must be constantly on the
191
Part 9
watch for the pawn sacrifice c4-cS,
followed up by d6. Yakovich decides
to anticipate an eventual check from
dS . 24 .. . E:a3 was also possible, for
2S cS dxcS 26 ttJbS does not work
due to 2 6 . . . E:g3.
2 5 if1 g a3?
Now this is a tactical mistake.
2S . . . h6 was a much better option.
2 6 c5! dxc5 27 b 5
Suddenly the board turns to be
too small for the rook. 27 . . . 8:g3 al
ready loses to 2 8 Wfff2 , whereas
27 .. . E:aa8 28 d6t threatens with a
fork.
27 . . . ixb 5 28 bxc5 '%Yxc5 29
ixb 5 gfa 30 d 6
192
Black i s i n a very difficult situa

tion . His extra pawn does not help
at all, because the advanced d-pawn
effectively breaks the coordination
of his heavy pieces. Spraggett confi
dently leads the game to a deserved
victory until move 38, when his usu
al terrible time trouble causes a fa
tal mistake:
30 . . . gaa8 31 d7 g ad8 32 gd4
c 6 33 gd s '%Ya3 34 gf1 e 7 3 5
g95 h 6 36 '%Y d 4 c6 3 7 ixc6 bxc6
38 gexf5??
White blunders his d-pawn. Any
other was winning, e.g. 38 8:fel +
c S 39 WffdS a4 4 0 8:e7.
38 . . .gxfS 39 gxf5 '%Ye7 40 g h s
'%Y e 6 41 h 2 @ gs 4 2 g e s '%Yxd7 4 3
1Mfc4+ 1Mff7 4 4 g e 6 c s 45 h 4 gfa 4 6
g 3 h8 4 7 '%Ye4 1Mff2+ 4 8 h 3 '%Yd4
49 g4 c4 so gc 6 '%Yd 1 + 51 h 3
'%Yd7+ 5 2 @ g 2 c 3 5 3 '%Yc4 1Mff5 54
 h 2 1Mff2+ 55 h 3 1Mff3 56 h 2 g ea
57 gca gxc8 58 '%Yxc8+ h7 59 g4
1Mff2+
0-1
The moral ofthis game is that with
such a pawn structure Black should
be calculating the consequences of
the cS-break on every move!
Part 1 0
1 e4 cs 2 tt::lf 3 tt:Jc6 3 d4 cxd4 4

tt:Jxd4 tt::lf6 s tt:Jc3 es 6 tt::l dbS d6
Unusual seventh moves
QU IC K R EPERTO IRE
It has been long established that

rare moves do not work against
the Sveshnikov. White's knight's
long walk wastes too many tempi
and could be justified only by gain
ing stable control of dS with 7 gS .
Perhaps we should remember only
a few variations to get out of the
opening with a pleasant position:
manoeuvre of Black's bishop to the

queenside :
1 1 J.d3
11 c4 b4 1 2 tt:Jbl as 13 tt:Jd2 Wg6
14 h4 e7 1S g3 0-0 16 g2 dB !
1 1 J.e7 1 2 0-0 0-0 1 3 tt::l b1

J.d8!?=
C. 7 a4 a6 8 tt:J a 3 J.g4!
A. 7 tt:Ja3 J.e7 8 J.gS (8 tt:Jc4

tt:Jxe4!) 8 tt:Jxe4!?=.
B.
7 J.e3 a6 8 tt:Ja3 bS
This i s meant a s a surprise.

8 . . J;bs 9 tt:JdS tt:JxdS 10 exdS tt:Je7 is
good enough.
9 tt::l d S gbs 1 o tt:Jxt6+ Y*lxf6
An important fine point ! Now 9

e2 .ixe2 10 Wxe2 dS 11 gS tt:Jd4 12
Wd3 b4 ! is easy to play as Black,
so:
9 t3 J.e6
White is playing an obvious

ly worse version of the main line
Sveshnikov. Note the important
Black provoked a weakening of

the e3-square and his next task is
to push dS . In many lines he sac
rifices a pawn for the initiative, for
instance, 10 gS e7 11 tt:Jc4 dS 12
kxf6 dxc4 13 Wxd8 + z;xd8 14 .ixg7
193
Part 10
g8 15 h6 cSoo
These examples showthat White

cannot efficiently clamp on dS, so he
usually continues developing by:
1 0 i.c4 gca 1 1 o-o tll b4 1 2
d5 bxd5 1 3 exd 5 i.d7 1 4
c3 i.e7
Or 10 c4 c8 11 l!JdS xdS 12
xdS l!JxdS 13 xdS h4 + 14 g3
h3oo.
In this typical structure White's
pieces are awkwardly deployed.
The game Xie Jun-Kramnik, ra
pid Monte Carlo 1996 saw further
15 hl 0-0 16 e3 l!Jh5 17 d3 gs+.
White should better divert the black
queen fram the kingside by 15 b3
c7= .
194
Part 1 0
1 e4 cS 2 f3 c6 3 d4 cxd4 4
xd4 lllf6 s lll c 3 es 6 lll d bS d6
Unusual seven th moves
STEP BY STE P
A. 7 tlJa3
B. 7 ie3
C. 7 a4
page 195
page 195
page 196
A. 7 lll a3 i e7 8 igS
Otherwise White loses control of
d5 : 8 ttJc4 ttJxe4 ! 9 ttJxe4 d5 10 ttJc3
dxc4 11 xd8+ hd8 12 hc4 ttJb4+;
8 id3? ! d5.
8 xe4!?
14 ttJ e 3 ie7! 15 a 3 ig5 16 d3 ttJb3

17 cxb3 E:xd3 18 c2 d8+ B auer
Lautier, France 2 0 05 . Instead, 13
ttJd5 is more challenging.
9 lll x e4
Or 9 he7 ttJxc3 10 hd8 ttJxdl
11 E:xdl @xd8 12 ttJb5 @e7+.
9 ixgS 1 0 lll x d6+
10 ttJxg5 xg5 11 xd6 a6 ! ? 12
ttJc4 ie6 13 cs 0-0-0 14 ttJe3 @c7
15 id3 f5 ! t was good for Black in
Gusev-Timoshenko, Odessa 1975.
1 o ... @e7 1 1 lll ac4 ie6 !m
Black has good counterchances,
for example, 12 ttJxb7? ! xdl+ 13
E:xdl ttJd4 ! 14 ttJ a3 E:ab8t.
7 ie3 a6 8 lll a3 bS
All the books consider only
8 . . . E:b8 9 ttJd5 ttJxd5 10 exd5 ttJe7= .
It is a good line indeed, but most
likely White knows it better than
you. Let us surprise him !
9 lll d S gba 1 o lllx t6+
10 ie2 ie7 is innocuous ;
10 c4 b4 11 tlJc2? ! (it is better
to transpose to the main line by 11
ttJ xf6+) 11. .. tlJxe4 12 f3 tlJcS brings
about a position with dubious com
pensation for the pawn;
a.
Simple and good. It is possi

ble that 8 . . . 0-0 9 hf6 hf6 10
ttJc4 ie6 11 xd6 (11 ttJxd6 ?! ttJd4
12 ttJc4 E:c8+) 11 . . . c8 offers suffi
cient compensation, but there is not
enough practical evidence: 12 d2
(12 dl E:d8 13 cl ttJd4 14 ttJe3
ig5oo) 12 . . . E:d8 when the only game
in our database saw 13 cl? ! ttJd4
195
Part 10
10 g3 ttJxdS 11 exdS tlJe7 12 g2
tlJfS 13 d2 e7 14 0-0 hS ! 15 c4
h4 16 cl hxg3 17 hxg3 d7f! gives
Black active play.
1 o ... \Wxf6
White plays an obviously worse

version of the main line Sveshnikov.
1 1 id3
Alternatively:
11 tlJbl e7 12 tlJc3 g6 13 d2
0-0 14 0-0-0 e6f! ;
ll c4 b4 12 tlJbl a S 13 tlJd2 g6 14
h4 e7 15 g3 0-0 16 g2 d8 ! 17 0-0
b6+ Szalanczy-Dokhoian, Cattoli
ca 1993 .
1 1 ie7 1 2 0-0 0-0 1 3 b1
id8 !? 14 c3 e7 1 5 \Wd2 h 6
1 6 a 4 b 4 1 7 e2 a s 1 8 ic4
c 7
White is unable to prevent ... dS,
e.g. 19 adl b7 20 tlJg3 bd8 21 f4
dS= , Hort-Andersson, Amsterdam
1978 .
C. 7 a4
White restrains Black's queen
side expansion and aims to bind
the opponent with a defence of the
dS and b6-squares. This logical
plan of Schlechter requires precise
196
play. Black must oppose something

to White's unpleasant positional
threat of gS, c4 (or tlJc4-e3 first)
with a total control of dS . Black
should use the fact that the enemy
is lagging behind in development to
break this unfourable pattern.
7 . . . a6 8 a 3 ig4
An important fine point ! Now:

Cl. 9 e2
C2. 9 1Wd2
C3 . 9 f3
C 1 . 9 ie2 ixe2 1 0 \Wxe2 d 5 1 1
i g 5 (11 exdS ltJxdS 12 ltJxdS xdS=)
1 1 ... liJ d4 12 \W d 3
O r 1 2 !xf6? ltJ x e 2 13 !xd8 ltJxc3
14 c7 ltJxa4 15 exdS @d7 16 !xeS
e8 17 0-0 xeS 18 ltJc4 xdS 19
xa4 c5+.
1 2 . . .ib4 !
6 ltJdbS d6, rare seventh moves

Black wins the battle for the centre:
1 3 0-0
Or: 13 exdS Ek8 14 ixf6 (14 ttJ abl
xdS 15 0-0 c4t) 14 ... xf6 15
ltJ abl (15 0-0? ! xc3 16 bxc3 E!xc3 ! )
l S . . . gS 1 6 0-0 g6 ! 17 xg6 (17
E!el xd3 18 E!xeS+ d7 19 cxd3
l2Jb3 2 0 E!a2 tll cl 2 1 E!al l2Jxd3 2 2
e3 E!he8 ! + ) 1 7. . . hxg6 18 E!cl b6!oo.
White is tied u p and down.
13 xf6 xf6 14 exdS E!c8 - see
13 eds.
1 3 . . . .ixc3 14 .ixf6 'Mfxf6 1 5
bxc3 dxe4 1 6 'Mfxe4 tll c 6 1 7 g ab1
'Mf e7 1 8 tll c4 0-0=.
exdS (14 xdS? xdS 15 exdS c7+)

14 . . .fS with good play.
1 2 . .. 0-0
Black had a pleasant choice be
tween the text and 12 . . . l2Jxe4 13
xe6 (13 l2Jxe4 dS 14 b3 dxe4 15
xe6 xdl 16 E!xdl fxe6+) 13 . . . ltJxc3
14 hf7+ (14 hS g6 15 hf7+ xf7
16 f3 + f6 17 xc3 dSt) 14 . . . xf7
lS bxc3 dS+.
1 3 .ig5 tll xe4 1 4 tll xe4
14 xe7? l2Jxc3 15 xd8 ltJxdl+ ;
14 xe6? l2Jxc3.
1 4 . . . d5 1 5 .ixe7 tll xe7 1 6 ib3
dxe4 1 7 ixe6 fxe6 1 8 'Mf g4 'Mfb6i.
C 3. 9 f3 ie6
C2. 9 'Mf d2 .ie6 10 ic4
10 l2J c4? ! does not achieve the
aim to take control of dS in view of
10 . . . ttJ b4 11 tll e 3 (11 as E!c8 12 ltJ b6
l2Jxe4+) 11. . . dS+.
1 o . . . gc8 1 1 o-o
Black provoked a weakening of

the e3-square and h is next task is to
p ush dS. White has tried to contest
that plan by:
White's pieces are ridiculously

placed.
1 1 . . . ie7
11. . . l2Jd4 12 d3 c7 13 a2 was
equal in Fedorchuk-Kolev, 20 05.
1 2 'Mfd 1
12 E!el l2Jd4 13 ltJdS ltJxdS 14
C3a. 10 l2Jc4 ; C3b . 10 gS; C3c.

10 c4
He can also ignore the enemy
plan, but then he faces problems: 10
e3 ltJ b4 11 l2Jc4 (11 gS c8) 11. . .
dS ! 1 2 b6 e7 ! , see game 3 4 De
Firmian-Vallejo, Selfoss 20 0 3 .
197
Part 10
C 3a. 10 lll c4 lLl b4 11 i.g5 (11
e3 dS!) 11 E:cS 12 chf6 %Yxf6 13
lll b6 E:c5 14 a5 (14 ttJbdS? hdS lS
exdS \Wg6) 14 d5! 15 lll bxd5 (lS
exdS ifs 16 d3 ttJ xd3 + 17 cxd3
xaS 18 xaS \Wxb6 19 a2 b4oo)
15 hd5 16 exd5 (16 ltJxdS ttJxdS
17 exdS \Wh4 + 18 g3 \Wb4+ 19 f2
\Wxb2+) 16 %Yg6oo.
11 d5 12 chf6
Black has the edge after 12 exdS
ttJxdS 13 he7 (or 13 ttJxdS hgS
14 l2Jdb6 b8) 13 . . . l2J cxe7 14 ttJxdS
ttJxdS in view of lS ttJxeS? ! \Wh4+ 16
g3 \Wb4+ .
12 dxc4
C3b. 1 0 i.g5 i.e7 11 lll c4

11 c4 could be countered with
11 . . . \Wb6 ! ? and the n:
a) 12 he6 fxe6 13 l2Jc4 \Wb4 14

\Wd3 dS ! ;
b) 1 2 hf6 hf6 13 he6 fxe6 14
l2J c4 \WcS ! lS \Wd3 (lS l2Jxd6+? e7
16 l2J xb7 ifMb4-+) 1S . . . l2Jb4 16 \We2
c8+;
c) 12 b3 dS ! ? (12 . . . l2Jg4 13 fxg4
hgS 14 he6 fxe6 lS l2Jc4 \Wb4 16
0-0 0-0-0 ! is a good alternative)
13 exdS 0-0-0 looks excellent for
Black;
d) 12 \Wd2 l2Jd4 13 e3 (13 b3
0-0 14 he6? ! fxe6 lS l2J c4 \Wes 16
aS dS ! 17 exdS exdS 18 ttJxeS ac8+)
13 . . . \Wxb2 14 f2 (14 0-0?? hc4
lS l2J xc4 \Wxc3 - +) 14 . . . ifMb6 lS l2Je2
\Wc6 16 l2J xd4 exd4 17 \Wxd4 0-0 and
Black is at least equal .
198
13 %Yxd8 +
In all cases Black gets full com
pensation for the g7-pawn, but with
queens White's defence would be
more difficult: 13 hg7 (13 he7
\Wxe7 14 ltJd S hdS lS exdS d8+)
g8 14 h6 ltJd4 lS e3 (lS \Wd2
h4 + ! 16 dl \Wf6t) 1S . . . \Wb6t.
13 E:xdS 14 hg7 (14 he7
xe7+) 14 E:gS 15 .ih6 i.c5
Black's pieces are very active,
so he has a lot of attractive ways to
develop his initiative, for instance,
1S . . . l2Jd4 ! ? 16 0-0-0 (16 cl fSoo)
16 . . . bSoo.
16 .id2
After 16 ltJdS hdS 17 exdS xdS
18 dl (18 hc4? d6 19 d2 xg2+)
18 . . . xdl + 19 xdl bS Black retains
the initiative with equal material.
16 .if2 + 17 @e2 i.h4oo.
C3c. 1 0 i.c4 gcs 11 0 - 0

An interesting position arises af-
6 tl:JdbS d6, rare seventh moves

ter 11 tl:JdS xdS 12 xdS tl:JxdS 13
xdS (13 exdS as+ 14 d2 tl:Jb4 lS
bl cs 16 c4 e7+)
Once again the weakness of

the kingside causes White trou
ble : 13 . . . h4 + 14 g3 h3 lS @f2
(lS d2? tl:Jd4) 1S . . . tl:Jb4 16 b3 dS !
17 exdS fS 18 g4 g6 . White's king
has not a safe haven.
11 gS leaves Black a strong in
itiative on the queenside, as in De
Jong-Kolev,
Hoogeveen
20 0S:
11. .. tl:Jb4 12 e2 e7 1 3 hf6 xf6
14 0-0 0-0 lS adl c7 16 b3 fd8

17 @hl cs+.
11 ltJb4 12 ltJd5 ltJbxd5 13
exd5 d7 14 c3 e7
In this typical structure White's

pieces are awkwardly deployed.
The game Xie Jun-Kramnik, ra
pid Monte Carlo 1996 saw further
1s @hl o-o 16 e3 tl:JhS 17 d3 gs+.
White should better divert the black
queen from the kingside by 15 b3
c7=.
199
P art 1 0
1 e4 c5 2 f3 c6 3 d 4 cxd4 4
xd4 f6 5 c3 e5 6 db5 d6
Unu sual seventh moves
COMPLETE GAM ES
34 D e F irm i a n - Va l l ejo Pons

Selfoss 2003
1 e4 cs 2 f3 c6 3 d4 c xd4 4
xd4 f6 S c 3 eS 6 d bS d6 7
a4 a6 8 a 3 ig4 9 f3 ie6 1 0 ie3
b4 1 1 c4
1 3 exdS bxdS 1 4 xdS xdS

1 s At2 gda 1 6 1Mfd2
In a bad position all moves are
bad. Mueller preferred 16 Wcl to
lose quickly after 16 . . . Wfc7 17 c3 i.cS
18 b4 hf2 + 19 ct;xf2 0-0 20 as e4
21 '2Je3 exf3 22 tlJxdS hdS 2 3 g3
i%fe8 24 i.d3 i.c4 0-1. White's prob
lem is that he is not only lagging
behind in development, but the b4
and e 3 squares are weak. The use
of these factors is the favourite mo
tif of Black's play in the 7 a4 line.
He must however be venturesome
since these advantages are dynam
ic and he could land in a dull posi
tion with a chronic hole on dS .
1 6 . . . b4 1 7 1Mfc3 YMgS
1 1 . . .d S !
I n this variation Black often sac
rifices a pawn on dS for an initia
tive. Here he achieves this break
through even for free. It is possible
because 1 2 ttJb6 stumbles into 12 . . .
d4 1 3 ttJxa8 Wxa8+.
1 2 ib6 1Mfe7 !
De Firmian i s not the first victim
of this surprising retrea, as Mueller
had already lost to Babula in 1998.
Now 13 tlJxeS Wfd6 14 i.d4 dxe4+
would be quite sad, so White chose:
200
17 . . . i%c8 ! ? 18 aS 4Jxc2 + 19 Wfxc2

Wb4+ was winning a pawn, but
Valejo was playing for the brlillian
cy prize that evening.
1 8 g d 1 x c2+!
Beautiful and best !
1 9 1Mfxc2 ib4+ 20 d2 ifS 21

1Mfc 1 gca 22 1Mf a1
Or 2 2 i.c4 i.d3 23 i.bS + ct;e7 24
Wxc8 i%xc8 25 hd3 i%cl- + .
2 2 . . . ic2 2 3 h 4 1Mff4 2 4 g3 1Mfxf3

2S gh2 1Mfe4+ 26 .ie2 ixd 1 27
0-1
1Mfxd 1 gda
PART 1 1
1 e4 c 5 2 lll f 3 tll c 6 3 d4 cxd4 4 tll x d4

tllf6 5 tllc 3 e5 unusu al lines
-
QUIC K R EPERTO I RE
In 1976, Bronshtein called into

doubt the move 6 ltJdbS and sug
gested that White might be overes
timating his position. His reasoning
was that the manoeuvre l2Jf3-d4-b5a3 costs four tempi and the knight is
still very bad on a3, requiring more
time investments. However, mo
dern theory has proved that the al
ternatives are quite innocuous and
might even lead White to trouble.
In this chapter we examine:
6 l2Jxc6, 6 ltJ b3, 6 ltJf3, 6 l2Jde2 , 6
ltJf5. Only the last of these options
poses some problems to Black and
requires some knowledge:
6 tllf 5 d5!
7 exd5
7 ltJxdS l2Jxe4 ! 8 ltJfe3 !e6 9 !e2

aS+ ! 10 @fl 0-0-0 earns the ini
tiative.
7 .ixf 5 8 dxc6 bxc6! 9 \Wf3
\Wd7 1 O .ig 5 e4
1 1 \We2
You might want to pay more at
tention to 11 dl since it was suc
cessfullyused in Volokitin-Kuzubov,
Moscow 2 007. We recommend an
improvement: 11. . . !d6 ! aiming at a
complex middlegame with a bishop
pair, harmonious development and
problematic dark squares in the
enemy camp : 12 !xf6 gxf6 13 d4
e7! with excellent prospects.
1 1 ... .ie7 1 2 gd1
Next, White will trade queens
201
Part 11
and the endgame should be rough
ly equal. Black must play energeti
cally and attack White's queenside
pawns. Possible continuations are
12 . . . b7 13 a6 b8 14 xb7
xb7 15 cl b4= ;
1 2 . . . e6 ! ? leads t o calmer, ba
lanced play after 13 c4 b8 , see
game 36 Akopian-Yakovich,
Rostov on Don, 1993.
6 llJ b 3 and 6 l!Jf3 do not create
threats, thus giving Black time to
develop his bishop to b4. (6 l!Jde2
b4 is also a fair option, but we re
commend 6 . . . cS as main line to
this knight retreat.) We'll consid
er two main plans of Black: He cap
tures the e4-pawn, hoping to ex
change pieces and get a better end
game, or bolster up his pawn cen
tre, trying to win the crippled White
pawns on the c-file. Let's see some
examples:
6 f3 b4 7 i.c4 0-0
O r 8 gS hc3 + 9 bxc3 d6= .

xc3 9 bxc3 xe4 1 0 i.a3
d6 1 1 ge1
11 Wel fS is in Black's favour.
1 1 gS 1 2 xg 5 9xg 5 1 3
i.xd6 g4 1 4 9c1 9xc 1 1 5
gaxc 1 gfe8=.
See game 35 Mukhin-Mina
sian, Leningrad 1990 .
6 b3 i.b4 7 c4 0-0 8 0-0

xc3 9 bxc3
The placement of White's knight

on b3 gives Black a good game with
simple development.
9 d6! 1 0 9d3 1 1 g d1 i.xc4
1 2 9xc4 gca 1 3 9d3 as.
It is fun to play this as Black.
a whole, Black has good and

easy play. We cannot think of some
particular traps which must be
avoided. It is important to lead out
the dark-squared bishop and com
plete development with . . . d6 and
. . . e6 .
As
8 0-0
20 2
Part 1 1
1 e4 c 5 2 lll f 3 lll c 6 3 d4 cxd4 4 lll x d4

lll f6 5 lll c 3 e5 un usual lines
-
ST EP BY STEP
A. 6 l2Jxc6
B. 6 l2Jb3
c. 6 l2Jf3
D. 6 l2Jde2
E. 6 ltJfS
page
page
page
page
page
203
20 S
20 8
20 9
210
A. 6 lll x c6
Such a n exchange is surprisingly
popular in the Sicilian lately, but in
the current situation it is rather in
consistent. White deprives himself
of his only trump - the command
over the dS-square.
6 bxc6
Capturing towards the centre is
a basic strategical rule. However,
we suppose that 6 . . . dxc6 is also suf
ficient for equality.
7 ic4
a) 7 gS? ! This aggressive move
weakens the dark squares and that
could be immediately exploited
with 7 . . . b8 ! 8 xf6 .
Or 8 bl aS ! 9 d3? xb2 ! 10
d2 xbl 11 xbl b4- + ; 8 c4
xb2 ! 9 b3 b4 10 d2 aS 11 f3
a6 ! 12 a3 0-0 13 a2 e7 with a
fine game, for example: 14 b3 (Or
14 tt:Jdl xa2 ! lS as aloe; 14 ltJdS
cs lS tt:Jxe7 Y9xe7 16 b3 b8 ! t)
14 . . . c7 1S. l2J a4 xb3 16 cxb3 dSoo.
8 . . . f6 9.c4 xb2 10 .b3 b4
ll.d2 dS- + , Mellgren-Alekhine,
Oerebro 193 S.
b) 7 f4 offers Black a choice :
7 . . . b4 8 d3 d6 9 0-0 0-0 10
fxeS l2Jg4 with typical Sicilian play,
or:
7 . . . c7 ! ? 8 f3 d6 ! 9 c4 (9 fS
b4 ! followed up by . . . d7-dS) 9 . . .
exf4 1 0 0-0 g S (in the King's gambit
style) 11 g3 g4 12 '!9f2 f3 and White's
compensation is dubious.
c) 7 d3 b4
Black has no difficulties after the
solid 7 . . . d6 8 0-0 e7 9 '!9e2 e6 10
c>hl 0-0 11 f4 exf4 12 hf4 l2Jg4 ! = ,
Kofidis-Andrianov, Athens 199 2 .
8 0-0 0-0
8 ... dS? ! is b ad d ue to 9 exdS
cxdS 10 el ! t (10 gs xc3 11 bS +
d7 12 xf6 gxf6 13 Y9xdS b8 ! 14
Y9xd7+ xd7 lS xd7+ c>xd7 16
bxc3 b2+)
9 c>hl
Or 9 f4 d6 10 fxeS dxeS
(10 . . . l2Jg4 ! ?) 11 gs (Black was in
tending . . . tt:Jd7-cS and f7-f6) 11 . . . e7
with a good game .
20 3
Part 11
9 . . . d6
9 . . . dS ! ? 10 exdS cxdS ll gS hc3
12 bxc3 h6 13 h4 Wffd6 is worth con
sidering, but we prefer the well test
ed pawn structures after 9 . . . d6.
10 f4 ltJg4 11 Wi el
Or 11 h3 exf4 12 f4 Wih4 13 Wfff3
ltJe S=
ll . .. exf4 12 hf4 ltJ eS 13 Wffg3 Wie7

14 ltJe 2 cS=
Black's position has a sound stra
tegical foundatio n. He controls the
centre and has no weaknesses on
the kingside.
7
...
.ib4
8 0-0
It was still early for 8 gS , for
White has no good answer to 8 . . .
h6: 9 xf6
Alternatives are :
20 4
9 d 2 d S (9 . . . hc3 10 hc3 ltJxe4

levels the game as 11 heS?? loos
es to 11. . . WiaS + , whereas 11 Wffg4 0-0
12 Wixe4 dS 13 Wie2 dxc4= reduces
the tension) 10 exdS cxdS 11 bS+
d7+;
9 h4? gS 10 g3 ltJxe4 11 hes
[11 Wfff3 hc3 + 12 bxc3 dS 13 hes
0-0 14 d3 (14 e2 e8 1S d4 cS 16
e3 ltJxc3 -+) 14 . . . e8 (14 . . . Wffe7 !?
l S d4 ltJg3+ 16 @d 2 ltJxhl 17 el
Wffd6 when Wh ite's initiative is in
sufficient because of the shaky po
sition of his king in the centre) lS
he4 xeS+, Cherkasov-Shariya
zdanov, Swidnica 1997] 11. . . Wffe7 12
Wid4 f6 13 Wixe4 WixeS+.
9 . . . Wffxf6 10 Wffd 3 0-0 11 0-0 d6 12
a3 cs 13 ltJ a4 d4 14 c3 b6= .
8 h6!?
Sveshnikov decorates this move
with an exclamation mark. He un
der lines its prophylactic function as
White is deprived of the gS pin.
Instead, 8 . . . dS 9 exdS hc3 10
bxc3 cxdS 11 bS + d7 12 ixd7+
Wixd7 13 a3 ! or 8 . . .hc3 9 bxc3
ltJxe4 10 a3 ! dS 11 d3oo favours
White.
More interesting is 8 . . . 0-0 9
gS !
The alternatives are worse:
9 @hl? ! c3 10 bxc3 dS+;
9 f4 Wffb 6 + (9 . . . Wffe7? ! 10 fxeS !
WixeS 11 Wff d3 ltJ g4 12 f4 Wies+ 13
@hl ltJf2 + 14 xf2 Wffxf2 lS floe) 10
@hl c3 11 bxc3 ltJxe4 12 Wfff3 dS 13
fxeS (13 d3 exf4 14 hf4 fS+) 13 .. .fS
14 d3 e6+.
9 . . . h6 10 h4 e7 Yakovich
claims that Black has no problems,
...
1 e4 cs 2 l!Jf3 l!Jc6 3 d4 cxd4 4 l!Jxd4 l!Jf6 S l!J c3 es

but we think that White's game af
ter 11 Wff d3, E:adl and solid central
play should be easier.
9 f4
Another possible setup is 9 Wff d3
0-0 10 E:dl d6 (10 . . . b7 11 a3 xc3
12 Wffxc3) 11 l!Ja4 (11 e3 l!Jg4 ; 11 a3
xc3 12 bxc3) 11. . . dS, when 12 exdS
is well countered by 12 . . . e4.
9 %Ye7 1 0 fxe5 V9x e5 1 1
h1
Or 11 b3? ! 0-0 1 2 @hl xc3 13
bxc3 l!Jxe4+.
1 1 ixc3 1 2 bxc3 0-0
6 tll b3
Tarrasch condemns this move,
calling the placement of the knight
lamentable. But, then, he claims
that a knight on b3 is nearly always
bad . . .
6 .ib4
a.
Bl. 7 d3
B 2 . 7 gS
B3. 7 c4
8 1 . 7 .id3 d5 8 exd 5
8 0-0? xc3 9 bxc3 dxe4 1 0 e2
Wffxdl 11 E:xdl 0-0+.
8 . xd5 9 .id2
White can sacrifice the exchange
with 9 0-0 l!Jxc3 10 bxc3 xc3 11
a3 , but his compensation is insuf
ficient, for Black can castle long. (He
h ad not a better choice though as
11 E:bl 0-0 12 Wfff3 b4 or 12 ... Wfff6 ! ?
favours Black) 11. . . xal (There is
no need to shy away fram the chal
lenge. In that case 11. . . b4 12 xb4
l!Jxb4 13 bS + l!Jc6 would be only
equal) 12 Wffxal Wffc7 13 f4 (13 E:bl b6
14 Wff c3 b7 lS fS g6 16 h3 fS ! + ;
1 3 l!JcS? 0-0 !) 13 . . . e6 ! (Enables
the queenside castling, making it
clear that the extra material should
prevail. This is not the only move
. .
1 3 if4
13 Wffd4 E:e8 14 xh6 (14 WffxeS
E:xeS lS xh6 E:xe4=) 14 . . . Wffxd4 lS
cxd4 E:xe4= leads to a roughly equal
ending, e .g: 16 d3 ! ? (16 c3 dS 17
d3 E:e8 (17 . . . gxh6 !? 18 E:xf6 E:e3
19 E:f3 E:xf3 20 gxf3 E:b8=) 18 f4
l!Je4=) 16 . . . E:xd4 17 E:ael d6 18 E:xf6
E:xd3 19 cxd3 gxf6 20 E:e8+ Wh7= .
1 3 V9xc3 1 4 es
This position arose in the game
Paiva-Van Riemsdijk, Sao Paulo zt
1972 , when 14 . . . l!JdS lS xdS cxdS
16 WffxdS (16 E:f3 Wffc6 17 E:g3 Wffe6 18
Wffd 2 Wh7=) 16 ... a6 17 E:fdl e2 =
would have been totally equal.
20 S
Part 11
though. 13 .. .f6 ! ? 14 fxeS l!JxeS lS
d4 l!J xd3 16 cxd3 f7 ! is not bad
either) 14 l!Jc5 (14 fS hb3 lS axb3
0-0 - 0+) 14 . . .exf4 lS V9xg7 0-0-0+.
9 . . . xc3 1 0 bxc3 .id6
1 1 0-0
White has also tried to deprive
Black of castling by 11 hS V!fc7 12
0-0 e6 13 gS ! ?
1 3 e3 0-0-0 i s good for Black
since 14 l!JcS?? loses to 14 . . .hcS lS
hcS dS- +
13 . . . h6 14 f4
14 adl is too slow: 14 . . . g6 lS
h4 e7 (lS .. .fS ! ? 16 f4 e4 17 f6
0-0 18 e2 f7 is also interesting)
16 he7 V!f xe7+.
14 . . . exf4 lS ael d7 16 fS
(or 16 xe6? ! fxe6 17 V!ff7+ c8 18
xe6 + V!fd7 19 V!f dS es-+)
This is the famous game Schlech206
ter-Lasker, World Ch. 1 910 . Now

the best move is 16 . . . V!fb6 + ! 17 hl
g6 18 he6 + (18 xe6 gxhS 19 e7+
d8 20 d7+ e 8 21 f6 es - +)
18 ... fxe6 19 V9xg6 hxgS 2 0 xe6+
c7 2 1 f7+ b8 with a big advan
tage. (Lasker)
1 1 . . . 0 -0 1 2 f4 f5 (12 . . . e8 ! ?) 1 3
fxe5 .ixe5 1 4 VNf3 .ie6 1 5 ae1
Perhaps White could maintain
the balance with lS l!JcS dS 16
l!Jxb7 V!fb6 + (16 . . . V!fd7 17 l!JcS d6
18 l!Jb7 hh2 + is unclear: 19 hl
es 20 V!fh3 g3 2 1 l!JcSoo) 17 e3
hf3 18 hb6 axb6 19 xf3 albeit
Black still has some threats.
1 5...g6
This i s better than l S . . . dS,
which was played in Br kic-Shariyaz
danov, Zadar 1999. After the text
Black's game is to be preferred, for
instance :
16 h3 d6+;
16 f4 dS ;
16 c4 V!fd6 (16 . . . V!fc7 ! ? 17 l!JcS??
d4+ 18 e3 l!JeS-+) 17 f4 (17
h3 ae8+) 17 ... hf4 18 xf4 xf4
19 xf4 ae8+;
16 xeS? l!JxeS 17 V!fe2 e8 - +.
82. 7 .i g 5 h 6 8 .ixf6
1 e4 cS 2 tLlf3 tLlc6 3 d4 cxd4 4 ttJxd4 tLlf6 S tlJc3 es

8 . . ..ixc3+ !
The split queenside pawns are
the future targets in Black's war
plan.
9 bxc3 1Mfxf6 1 O .ic4
Alternatively:
10 e2 d6 11 0-0 0-0 12 d3 d8
13 fdl e6 14 f3 ac8+;
10 d3 tLle7 11 tLld2 0-0 12 c4 d6
13 tLlfl e6 14 tLle3 ac8+. Accord
ing to Sveshnikov, Black will dou
ble his rooks on the c-file, gaining
an edge.
1 0 . . . 0-0 1 1 0-0
White's pressure along the d-file
is inefficient, due to the weakness of
the g2-pawn: 11 d2 d6 (11. . . g6 ! ?
12 f3 tLl e 7 1 3 0-0-0? ! bS ! 14 hbS
b8 15 c4 b6t with powerful ini
tiative in Andrade-Vallejo , Menorca
1996) 12 dl g6 13 f3 (13 0-0 xe4
14 dS a4+) 13 . . . e6 . Black is al
ready better: 14 d3 ? ! gS ! ? lS 0-0
(lS xgS hxgS+) lS ... xd2 16 xd2
ac8 17 e2 tLle7+, Butt-Ochoa, The
ssaloniki 1984; 14 dS hdS lS exdS
tLle7+; 14 he6 fxe6 lS 0-0 ad8 16
c4 b6+.
1 1 . . . d 6 1 2 1Mf d3 .ie6 1 3 gad 1
gfd 8 1 4 .id5 gac8 1 5 gd2
1
Or 15 bs c7 16 d3 hds 17
xdS tLle7 18 d2 b6 and White's
game is difficult.
1 5 . . . 1Mfe7 1 6 gfd 1 .ixd5 1 7
1Mfxd 5 1Mf c7i , Andreas-Tsermiadia
nos, Singapore 1990 . White has too
many weaknesses to defend.
83. 7 .ic4
White is ready to part with a
pawn for initiative, but Black should
not oblige him by grabbing it . He
can follow a clear strategical line,

based on a better pawn structure.
7 . . . 0-0
8 0-0
White can also protect the e4pa wn by:
8 gs h6 9 h4 d6 ! (Suggested
by Sveshnikov. 9 . . . hc3 + 10 bxc3
d6 is also appealing, e.g. 11 f3
e6 12 dS c8+) 10 f3 (10 xf6
hc3 + 11 bxc3 xf6 or 10 a3 hc3 +
11 bxc3 e6 12 dS hdS 13 exdS
tLlb8 14 tLld2 tLlbd7 15 tLle4 e7 16
bl b6 17 b3 fc8 give Black a clear
positional edge .) 10 . . . e6+;
8 d3 dS ! ? (8 . . . d6 9 gS h6 10
h4 e6 11 0-0 hc4 12 xc4 hc3 !
13 xf6 xf6 14 xc3=) 9 exdS (9
xdS tLlxdS 10 exdS tLle7 11 0-0 fS
12 g3 f6+) 9 . . . e4 10 g3 tLle7 ll 0-0
tLlfSoo. (Rogozenko)
8 . . . .ixc3 9 bxc3 d 6 ! 1 O 1Mf d3
White's compensation for the
central pawn is not so clear after 10
a3 tLlxe4 11 d3 (or 11 el tLlf6 12
dl tLl e8 13 f4 b6 + 14 h l exf4 1S
xf4 tLle S) ll ... tLlf6 12 adl (12 xd6
regains the pawn, but loses the in
itiative: 12 . . . xd6 13 hd6 d8 14
adl fS+, or 12 hd6 fS 13 xf5
xd6 14 adl c7+) 12 . . . tLle8 13
20 7
Part 11
dS (13 xd6 l!Jxd6 14 xd6 fS+)
13 . . . g4 .
1 o . . . i.e6
1 1 gd1
After 11 xe6 fxe6 1 2 a3 E:f7 ! 13
c4 (13 xd6? ! xd6 14 xd6 l!Jxe4+;
13 hd6?? E:d7 14 E:adl l!Je 8 - +)
13 ... E:d7 14 E:adl b6 ! lS E:d 2 e8 16
E:fdl (16 hd6?? E:ad8-+) 16 . . . E:ad8
17 e2 e7 18 f4 l!Je 8 ! 19 hS f7 ! +
White's pawn weaknesses become
tangible, Forgach-Luther, Zwesten
1999;
Or 11 a3 xc4 12 xc4 c7 13
E:fdl E:fd8 14 E:d2 bS ! lS e2 (lS xbS
l!Jxe4 16 E:d3 b6 17 xb6 axb6 18
cl (18 f3 E:xa3 19 fxe4 @f8+) 18 . . .
dS+) l S . . . aS ! 1 6 E:adl b 4 1 7 b2 (17
cxb4 axb4 18 cl l!J e7+) 17 . . . bxc3 18
xc3 l!Jb4+ and White was worse in
Crepan-Dobrov, Garica 2004.
1 1 . . .i.xc4 1 2 '%Y xc4 gca 1 3 Wf d3
as 14 g b 1
White could fight for the draw
with 14 l!JxaS xaS lS c4 a6 16 gS
xc4 17 xf6 xd3 18 E:xd3 gxf6+;
or 14 xd6 xd6 lS E:xd6 l!Jxe4+.
1 4 . . . c4 1 5 .ig5 h6 1 6 .ih4
16 hf6 xf6 17 d 2 b6 ! 18 l!Jfl
a4 19 c4 l!JcS 20 e3 (20 xd6
xd6 21 E:xd6 xe4 22 E:d3 (22 E:d7
20 8
l!JcS) 2 2 . . . b6+) 2 0 . . . b6 2 1 g3 l!Je6 !

with an edge, Crepan-Grosar, Celje
20 0 3 .
1 6 . . . b 6 1 7 d 2 xd 2 1 8 gxd2
Wfe 7i.
Black is a little better thanks to
his superior pawn structure.
C. 6 f3 i.b4 7 .Ac4
Alternatively:
7 gS h6 8 xf6 xc3 + ! 9 bxc3
xf6 10 c4 0-0 11 0-0 d6 This po
sition is dangerous for White who
has no compensation for his split
queenside.
7 d3 dS 8 exdS Here Black can
choose between 8 . . . xdS ! ? 9 0-0 (9
d2? xc3 10 xc3 e4 11 hf6 exd3
12 xg7 e4+ 13 fl E:g8 14 c3 (14
cxd3 g6- +) 14 . . .dxc2 lS el fS+)
9 . . .xc3 10 bxc3 0-0 1 1 l!J gS E:d8 12
e2 e6 ! = and the natural 8 . . . xdS
9 d2 l!Jxc3 10 bxc3 e7 11 0-0 0-0
12 E:el f6 with a good game.
7 . . . 0-0
With White's knight on f3, it
would be risky to capture on e4:
7 . . . l!Jxe4 8 dS l!Jd6 9 b3 as 10
0-0 0-0 11 d3 ! a6 (ll . . . xc3 12
l!JgS ! e4 13 xc3 xc3 14 bxc3 h6 1S
E:dlt) 12 xa6 ! bxa6 13 l!JdS as 14
E:dlt with clear compensation.
1 e4 cS 2 ltJf3 ltJc6 3 d4 cxd4 4 ttJxd4 ltJf6 S ltJc3 eS

8 0-0
We are already familiar with the
pawn structure after 8 gS xc3+
9 bxc3 d6= . Black follows up with
c8-e6, enjoying a good game.
8 Wffd 3 ? ! allows the central break
8 . . . dS ! After the farced sequence 9
xdS ttJxdS 10 WffxdS WffxdS 11 exdS
ltJd4 12 ttJxd4 exd4 13 a3 aS ! 14 b4
e8 + lS ltJe2 (lS @dl dxc3 16 bxaS
g4 + 17 f3 ad8 18 M4 xdS+ 19
@cl fS+ Gligoric) 1S . . .b6 16 @d2
(16 b2 fS 17 @d2 xc2 18 hel
b3 ! 19 ttJxd4 xdS+) 16 . . . d8+
Black is in command.
8 .ixc3 9 bxc3 ttJxe4
Black might try to followthe same
plan as in the 6. ltJb3 variation, but
the knight is actually much more ac
tive on f3 so the effect is rather dif
ferent: 9 . . . d6 10 Wffd3 e6 11 xe6
fxe6 12 a3 ltJe8 (12 . . . WffaS 13 xd6 !
ad8 14 c4 ! ttJxe4 lS Wffxe4 xd6 16
cS ! t) 13 fdl f7 (13 . . . WffaS 14 c4t) 14
Wffc4 Wie7 lS abl d8 16 d3oo .
1 0 i.a3
10 el gives another pawn
without clear compensation after
10 . . . ttJxc3 11 Wffd6 (Or 11 Wffd3 dS 12
b3 d4 13 ttJxeS ttJxeS 14 xeS Wfff6 1S
f4 g4 ! 16 h3 e6+) ll . . . Wfff6 12 a3
e8 13 e3 bS 14 b3 aS ! lS xc3
(lS ttJxeS ttJxeS 16 Wffxf6 gxf6 17 xc3
b7! 18 g3 + ltJg6 19 c3 e2 ! t) lS . . .
b 4 1 6 Wffxf6 gxf6 1 7 xc6 (17 ttJxeS
fxeS 18 g3+ @f8 19 b2 a4 20 dS
as 21 c4 ltJe7! 22 xeS ttJxdS 23
g7 + @e7 24 cxdS d6+ leaves White
a pawn down) 17 . . . dxc6 18 cl a4
19 c4 e6 . Commonly, a rook and
2 pawns are quite stronger than a
knight and bishop in an endgame.
1 0 d6 1 1 ge1
11 e l fS is in Black's favour.
1 1 lll g S 1 2 lll x g5 %Yxg 5 1 3
xd6 .ig4 1 4 %Yc1 %Yxc1 1 5
gaxc1 gfe8=
See game 35 Mukhin-Mina
sian, Leningrad 1990 .
D. 6 lll d e2 .ic5
The e2-knight takes the sting of
6 . . .b4, but nevertheless that is a
good and popular alternative.
7 ti)g 3
7 e3? ! covers the d4-square at
a high price . Black obtains a good
game, playing in the spirit of the Si
cilian: 7 . . xe3 8 fxe3 0-0 (8 . . . dS ! ?
9 exdS ltJb 4 when 1 0 e4? fails to
10 . . . ltJg4 ! 1 1 Wff d2 ltJe3 12 cl ltJc4 13
V!ff dl ttJxb2 14 Wffd 2 ltJc4 lS V9dl g4+)
9 ltJg3 d6 10 V9d2 Wff b6 11 0-0-0 e6 ,
Rodriguez-Spasov, Tunj a 1989.
7 d6
7 . . . Wffb 6 is too hasty in view of
8 Wffd 2 ! ltJ g4 9 Wff gS ! xf2 + 10 @dl
ltJe3 + 11 xe3 V9xe3 12 Wffxe3 xe3
13 ltJdS b6 14 ltJfSoo.
8 i.e2
8 gS is a consistent attempt to
occupy dS: 8 . . . h6 (8 . . . Wffb6 9 Wff d2
20 9
Part 11
l2Jg4 is very sharp, but White is po
sitionally better after 10 0-0-0 d4
11 l2J a4 ! 13 f3 ! l2Je3 14 xe3 xe 3+
lS bl) 9 xf6 Wffxf6 10 Wffd 2 b4 !
(10 . . . 0-0 11 ltJdS Wffd8 12 c4t) 11 a3
xc3 12 Wffxc3 0-0 13 c4 (13 dl
l2Jd4 14 Wffd 3 d8) 13 . . .e6 14 dl
l2Jd4 lS xe6 fxe6 16 Wffd2 ad8= .
8 c4? ! can b e attacked by
8 . . . l2Jg4 ! 9 fl (9 0-0? Wffh4 10 h3
l2Jxf2 11 hf?+ d 8- + ; 9 e3? !
l2Jxe3 10 fxe3 0-0+) 9 . . . 0 - 0 10 h 3 (10
l2J a4 Wff aS+ 11 c3 bS 12 ltJxcS bxc4 13
l2J a4 d7! t) 10 . . . e6 !
the bishop pair and good develop

ment, he should be confident about
the future :
1 0 i.b5+
10 xf6 Wffxf6 11 bS + 8 12 0-0
l2Jf4 is similar to the main line .
1 0 ... @fS 1 1 xf6 xf6 1 2
0-0 lllf4 1 3 lll d 5 lll x d5 1 4 xd5
g6 1 5 lll e 2 @g7 1 6 @ h 1 .ie6
Black seizes the initiative as 17
Wffxb7? is bad due to 17 . . . a6 ! 18 d3
(18 xa6 hb8 19 Wffc7 xb 2 - + ; 18
a4?? a7- +) 18 . . . hc8 + .
1 7 d2 ghd8 1 8 lllc 3 .id4!t
Nikac-Komarov, Niksic 2000.
E . 6 lllf 5 d5!
11 ltJd S (1 1 xe6? xf2 + ! 12 xf2

l2Jxf2 13 xf2 Wffb 6 + 14 e2 fxe6 lS
Wffel dSt White's king is vurnerable
in the centre, e .g. 16 exdS exdS 17
ltJxdS? WffbS +- +) 11. . . l2Jf6 12 c3 (12
gS? walks into an amazing queen
sac: 12 . . . ltJxdS ! ! 13 xd8 b4+ 14
e2 l2Jd4 + lS Wff xd40 ltJf4+ 16 3
exd4 17 b3 l2Jg6 18 gs h6+) 12 . . .
h 6 . Black has fair chances i n this
complex position, for example, 13
b4 b6 14 a4 aS lS bS xdS 16 xdS
(16 exdS? ! l2Je7+) 16 . . . l2Je7 17 xb7
b8 18 dS ltJexdS 19 exdS Wffc7oo
with nice compensation.
8 ... lll d 4 9 .ig 5 lll e 6
With this manoeuvre Black un
pins the f6-knight. He loses the
right to castle indeed, but having
210
El. 7 li:}xdS
E2 . 7 exdS
E 1 . 7 xd 5 x e4 !
The best decision. Play is rather
dull after 7 . . . ltJxdS 8 WffxdS
Or 8 exdS? ! xfS 9 dxc6 Wffxdl+
(9 ... bxc6 10 Wfff3 Wffd7 11 c4 d6 12
0-0 e4 13 el 0-0+, Sveshnikov) 10
xdl 0-0-0 + ll d 2 bxc6 12 c4 (12
a6 + b8 13 e l f6 14 e2 cs+)
12 . . .f6t
8 . . . Wff xdS 9 exdS hfS 10 dxc6
1 e4 cs 2 ll.Jf3 ll.Jc6 3 d4 cxd4 4 ll.Jxd4 ll.Jf6 S ll.Jc3 es

bxc6 11 i.d3 i.xd3 12 cxd3 0-0-0 13
c;t>e2 i.cS 14 i.e3 i.b6 1S gacl c;t>b7 1 6
gc4 gd6 = .
8 fe3 ie6 9 .ie2 Wa5+ !

Other good options are 9 . . . i.cS 10
0-0 0-0+ or 9 ... ll.Jd4 ! ? 10 c4 W! a S + 11
i.d2 ll.Jxd 2 12 \Wxd2 W!xd2 + 13 c;t>xd2
gcs 14 gacl i.d6 lS b4 Mitzka-Per
ov, Budejovice 1996, when 1S . . . b6
16 ghdl 0-0 17 c;t>el gfdB+ would
have underlined Black's advantage
in the endgame.
1 0 f1
Or 10 i.d2 ll.Jxd2 11 \Wxd2 W!xd2 +
12 c;t>xd2 0-0-0+.
1 0 .. 0-0-0 1 1 c4 .ic5+
.
E2. 7 exd5 .ixf5 8 dxc6 bxc6!
9 i.d3 is also imprecise as it gives

Black a tempo and spatial advan
tage: 9 . . . e4 10 i.e2 (10 i.c4 i.d6 11
h3 0-0 12 i.e3 Wle7 13 Wf e2 i.eS 14
0-0 gfe8 lS gf dl h6 16 i.b3 gac8
17 ll.Ja4 i.b8 ! with attack in Galego
Antunes, Portugal 1 993) 10 . . .i.d6
11 igS h6 12 i.h4 0-0 (12 . . . i.f4 ! ? ,
Sveshnikov; 1 2 . . . Wfe7 ! ? intending
to cramp the opponent even more
with . . . gds) 13 Wfcl gS 14 i.g3 ll.Jds
lS i.xd6 W!xd6 16 \Wd2 Lutikov
Sveshnikov, 1976. Here 16 . . . \WeS+
reatains a slight edge .
Hence White's best is:
9 Wf3 Wd7 1 0 ig5

10 ia6 scatters White's pieces
all over the board and in the game
Garbarino-Cifuentes, Casilda 1984
Black used it with 10 . . . i.e7 11 0-0 e4
1 2 \Wg3 0-0 13 i.gS (13 i.h6?? ll.JhS
+) 13 . . .i.d6 14 Wlh4 ll.Jg4 lS gfdl (lS
h3 ll.JeS 16 gfdl \Wc7) 1S . . . W!c7+ 16
ll.Jxe4? ! i.xh 2 + 17 c;t>hl when 17 . . .i.eS
18 c3 gae8t would have been clearly
better for him.
1 0 . . . e4
Black leaves the choice to the op

ponent. Now the endgame would be
level:
9 \Wxd8+ gxd8 10 i.e3 when the
most aggressive is 10 . . . i.b4! 11 ixa7
(11 a3 i.xc3+ 12 bxc3 gd7+) 11. .. gd7
12 i.e3 ll.JdS 13 i.d2 ll.Jxc3 14 hc3
(14 bxc3 i.cS lS i.e3 i.xe3 16 fxe3
hc2+) 14 . . . i.xc3+ lS bxc3 i.xc2
16 a4 c;t>d8 ! + . Black's pieces activi
ty should prevail over White's extra
pawn.
1 1 We2
a) 11 Wlg3 i.d6 12 \Wh4 i.eS 13 ic4
0-0 14 0-0 h6+;
211
Part 11
b) ll e3 ? ! ltJg4 12 d2 (12 g3 ? !
cS 1 3 ltJ d l h 6 14 f4 g S l S es o-o
16 e2 ltJxeS 17 xeS d4+, Serp
er-Kasparov, Internet blitz, 1998)
12 . . .xd2 + 13 hd2 cs 14 ltJdl
ttJxf2 lS ltJxf2 e3 16 he3 he 3 17
d 3 e6 18 @e2 d4 19 c3 b6 20
hdl 0-0-0 21 b3 Teske- Krasenkow,
Bundesliga 2 00 3 , when 2 1 . . . he8
22 @fl fS+ highlights the power of
the bishop pair.
c) 11 dl is a rare move which re
centlywas successfullyused byVolo
kitin at the Aeroflot Open. Follow
ing the logical:
11 . .. d6 ! Black aims at a complex
middlegame with a bishop pair, har
monious development and prob
lematic dark squares in the enemy
camp. (Instead Volokitin-Kuzubov,
Moscow 2007 saw 11 . . . c7? ! 12
xf6 gxf6 13 d4 eS? ! 14 a4 c8
lS a6 c7 16 dl d6 17 ltJdS+.)
12 xf6
Or l2 d4 e7 13 dl eS 14 a4
0-0 and Black is 0 K, for instance, lS
xc6? ! b4 ! 16 bs xbS 17 hbS
ab8t or l3 c4 0-0 14 xc6? ! ac8
1S a4 b4+)
12 . . . gxf6 13 d4 e7! with excel
lent prospects:
14 0-0-0 es lS a4 0-0t;
14 a4 0-0 lS g4 (lS xc6? ac8
16 a4 b4+) 1S . . . g6 16 h4 h6 17
0-0-0 ab8t;
14 bS c8 (14 . . . 0-0 ! ? deserves
attention: lS hc6 es 16 e3 ac8
17 dS fd8oo) lS a4 0-0 with a
preferable game.
1 1 . . ..i e7 1 2 d 1
I t i s risky t o give Black a strong
212
initiative for merely one pawn: 12

xf6 ? ! hf6 13 ttJxe4 0-0 14 ttJxf6+
gxf6 . White does not manage to cas
tle : lS d2 fe8 + 16 e2
16 dl is hardly any better, as
16 . . .c7 17 d3 ad8 18 @cl (18 b4
cs 19 bxcS xcS 20 bl ds 2 1 b3
d4 22 c3 b8+ Mellado-Cam
pos Moreno, Hostafrancs 2002 also
gave Black a deadly attack.) 18 . . . cS
19 b3 (19 el xel+ 20 xel hd3
21 cxd3 c4 ! 22 d4 xh2+ Faisst
Hohm, corr. 1993) 19 . . . eS 20 bl
c4 ! 21 bxc4 b8+. Black's attack is
irresistible, Rogers-Volzhin, Aosta
2002.
16 . . . e7 17 @fl ad8 (17 . . . eS !?)
18 d3 es 19 bl bs !
1 2 . . . '%Yb7
This suggestion of Sveshnikov
looks the most consistent continu
ation. The alternative 12 . . . e6 ! ?
leads t o balanced play after 1 3 c4
b8 , see game 36 Akopian-Yak
ovich, Rostov on Don, 1993.
1 3 '%Ya6
The greedy 13 xf6 xf6 14 ttJxe4
0-0 ! (14 . . . b4 + lS ltJd2+ e6 16
e4 ! ) lS ttJxf6+ gxf6 is similar to
the game Mellado-Campos Moreno,
but the difference favours Black.
1 3 . . . b8 1 4 '%Yx b7 xb7 1 5 c1
=
1 e4 cs 2 lt:Jf3 lt:Jc6 3 d4 cxd4 4 tt:Jxd4 lt:Jf6 S lt:Jc3 es

1 5 .ib4
The diagram position is satisfac
tory for Black. His piece activity and
space advantage amply compensate
the split queenside pawn structure .
Instead of the text, lS . . . 0-0 also re
tains a slight initiative in more or
less balanced endgame: 16 a6 (16
c4 e3 ! 17 d3 exf2 + 18 xf2 cs+
19 fl Dusper-Thurmer, Harkany
2001, 19 . . . .bd3 + 20 xd3 e8 and
White should be careful as 21 lt:J a4
would walk into 2 1 . . . be7! 22 d2
. . .
lt:J e4+) 16 . . . b6 17 c4= .

1 6 .id2 @e7 1 7 .ic4
After 17 a6 d7 Black is more
active although the draw is the most
probable result. Play could continue
with 18 tt:Jxe4 .bd2 + 19 lt:Jxd2 hd8
20 d3 .bd3 21 cxd3 xd3 22 e2
lt:Jg4 23 h el f8 24 lt:Je4 xdl 2 S
xdl e 8 26 f3 fS 2 7 fxg4 xe4+ 2 8
f3 xg4+ with a material advan
tage in a drawish ending.
1 7 J d8
The game is level.
..
213
PART 1 1
1 e4 c5 2 li)f3 li)c6 3 d4 cxd4 4 li)xd4

li)f6 5 li)c3 e5 u nu sual l i n e s
-
COMPLETE GAM ES
3S M u kh i n - M in a s ian
Len in g rad 1 990
1 e4 cs 2 ti:)f3 ti:) c 6 3 d4 cxd4 4
ti:) xd4 es s ti:) f3 ti:) f6 6 ti:) c 3 ib4 7
ic4 0-0 8 0-0 ixc3 9 bxc3
9 . . ti) xe4
In line B3 we recommend in
nearly the same position, except for
the knight being on b3 , to play 9 . . .
d 6 , threatening with . . . t2Jxe4. Here
Black levels the game by capturing
the pawn immediately.
1 O ia3 d6 1 1 e1 ti:) g S 1 2 ti:) x g S
YMxg S 1 3 ixd 6 ig4 14 YMc 1 YMxc1
1S axc1 fe8 16 id S
White has the bishop pair so he
should be trying to open more oper
ating space for it. 16 f4 exf4 17 hf4
ie6 18 ib3 would have preserved
.
2 14
some tension in a balanced posi

tion.
1 6 . . . ie6 1 7 ixc6 bxc6 1 8 a3
18 E'!:xeS xa2 19 E'!:cS aS 20 8:xc6
a4 2 1 ia3 8:ec8 = is a dead draw, but
in the game White is gradually be
coming worse. Tournament prac
tice has seen many examples when
the opposite coloured bishops do
not guarantee a draw. In combina
tion with rooks, one extra pawn or
even just a passer, commonly brings
a full point.
1 8 . . . f6 1 9 f3 ed8+ 20 ed 1
ic4 21 f2 f7 22 icS a6 23 b 1
 e 6 2 4 e1 dS 2 S xdS xd S 2 6
ie3 i b S 2 7 b 4 g S 2 8 c4+ .ixc4
29 b7 h 8 30 ia7 hS 31 b8 h7
32 d8+ e6 33 i c S i d S 34 c4
ixc4 3S d 6+ f S 36 xc6 ibS+
Black is already much better, but
37 8:d6 would have offered chances
for salvation. Instead White loses in
a couple of moves.
37 b6 c7 38 ib4 c1 + 39
f2 c2+ 40 g 1 g4 41 d6 gxf3
42 gxf3 ic6 43 d3 h4 44 ie 1 h 3
4S .i d 2 .i b S
0-1
White resigned because h e has
no useful moves, e.g. 46 8:d6 8:a2 .
1 e4 cs 2 ll:Jf3 ll:Jc6 3 d4 cxd4 4 ll:Jxd4 ll:Jf6 S ll:Jc3 es

36 Ako p i an - Yakovich
Rostov on Don, 1 993
1 e4 cs 2 f3 c6 3 d4 cxd4
4 xd4 f6 s c3 e s 6 ts dS 7
exdS ixfS 8 dxc6 bxc6 9 \Wf3 \Wd7
1 0 igS e4 1 1 Ye2 ie7 1 2 g d 1
\We6 ! ?
I n 1993 this move was a novel
ty. We chose as a main line 12 . . . b7
13 a6 b8 , but the text is a good
alternative. White will trade queens
anyway, the question is what square
Black should prefer. Yakovich has
an original idea in mind . . .
1 3 \Wc4 g b a 1 4 Yxe6
1 4 . . . fxe6 !
I n such positions one should
not think about pawn weaknesses.
By capturing with pawn, Black en
ables the move . . . ll:JdS which will
underline how vulnerable White's
queenside is. There are two oth
er points in favour of 14 . . . fxe6: the
light-squared bishop remains on
active position, thus rendering lS

ll:J a4? ! dubious due to 1S . . . e3 ! ; the
e6-pawn will be a good shield of the
king in the centre . 14 . . . .ixe6 lS ll:J a4
would have been roughly equal.
1 S b3
The computer contemplates a
dumb defence with lS icl ll:JdS 16
ll:Je2, which is, of course, the last
thing a human would choose. Then
16 . . . e3 ! would be as good as ever.
1 S. . . d S !
1 S . . . ib4 16 id2 d8 17 ll:Jbl defends everything.
1 6 ixe7 @xe7 1 7 a4 e3 1 8
c4
18 id3 exf2 + 19 @xf2 E:hf8
would allow Black to activate his
second rook.
1 8 . . . exf2+ 1 9 @xf2 b4 20
ie2
White has finally completed de
velopment and the imminent elimi
nation of the queenside pawns pre
determine the draw.
20 . . . xa2 21 ga1 b4
Yakovich could have posed more
practical problems with 21 . . . hf8,
but 22 if3 (22 ll:JcS @d6+) 22 . . . ll:Jb4
23 ll:JcS a6 24 E:hdl ! eS 2S El:aS=
would avoid any danger.
22 c s a6 23 ghd 1 ! ghd8 24
gxd8 gxd8 2S g a4 ! =
Draw, in view of 2 S . . . b8 2 6 ll:Jxa6
ll:Jxa6 27 xa6 xb3 28 xc6=.
21S
Part 1 2
1 e4 cs 2 lll f 3 lll c 6
3 c3
3 lll c3 lll c 6 4 es
Rare Li nes
QUIC K REP E RTO I RE
The repertoire with 2 . . . l!Jc6 is per

fect against anti-Sicilians. You do
not need to learn anything specific,
as, for instance, is the case with 2 . . .
e6. Perhaps the only variation of in
dependent significance is:
3 lll c 3 lll f 6 4 es lll g 4 s e2
d6 6 exd6 e6
Black ensures normal develop

ment of his army. This allows him
to fight for the initiative while White
tries to disentangle his pieces. Now
7 g3 seems a realistic approach, al
though Black would be fine after
7 . . . xd6 8 i.g2 0-0 9 0-0 es 10 h3
l!Jf6 11 d3 h6. Should White attempt
to snatch a pawn with 7 \Wc4 eS ! 8 h3
l!Jh6 9 \WxcS? ! , he will have a diffi
cult time neutralizing Black's initia
tive following 9 . . . xd6 10 \Wc4 0-0 .
2 16
Against 3 c3, we recommend a

very well tested system:
3 ... lll t6 4 es lll dS
Now 5 d 4 cxd4 6 cxd4 d6 7 i.c4

dxeS 8 dxeS l!Jb6 9 \Wxd8+ l!Jxd8
10 i.bS+ i.d7 11 l!Jc3 e6 is level, so
White usually prefers:
S .ic4 lll b 6 6 .ib3 dS 7 exd6
xd6
White is at a juncture here . He
1 e4 cS 2 tlJf3 tlJ c6 rare lines: 3 c3 ; 3 tlJc3

can sacrifice a pawn, trying to ex
ploit his lead in development, or
castle.
Sometimes he plays 8 tlJ a3 as
well, but the fact that White has not
castled yet makes possible 8 . . .e6 ,
when 9 0-0 hb3 1 0 axb3 d3 !
gives Black an easy game.
a) 8 d4 cxd4 9 0 - 0 e6 1 0
a3 dxc3 1 1 V:Ve2 hb3 1 2 b5
\Wb8 13 axb3 g6 and Black was
slightly better in the game Matsuu
ra-Leitao, Santos 2006.
b) 8 0 - 0 i.e6 9 he6 11Mxe6
1 0 d4 cxd4 11 xd4 xd4 12

11Mxd4 E:d8 13 11Mh4 11Me2
The latest occurrence of this var
iation was in Sveshnikov-Kobalya,
Dagomys, 0 2 . 04 . 200 8 , where Black
preferred 13 . . . g6 14 !e3 g7 15 tlJ a3
0-0 16 ::fel and here simplest would
have been 16 . . . f6 = . The text is ex
tensively tested and practice has
proved that Black is not worse at
all. We should not be afraid of tak
ing the two minor pieces after:
14 d2 8:xd2 15 ttJxd2 xd2 16
::fdl ifMh6 ! 17 g3 ifMc6 and gradual
ly Black completes development af
ter 18 b4 e6 19 8:d.3 f6 2 0 8:adl tlJdS
21 b8 + f7, Ossa-Marin, Sebas
tian, 2006. Another variation on
this theme is:
14 e3 ifMxb2 15 tlJd 2 8:xd2 16 8:abl
xc3 17 hd2 ifMxd2 18 ::fdl ifMaS 19
ifMe4 f6 ! 20 xb7 f7, Lintchevski
Kuzubov, Kirishi 2004.
The onus is on White in this
line.
2 17
Part 1 2
1 e4 cS 2 f3 lll c 6
3 c3
3 lll c 3 lll c 6 4 es
Rare Lines
STEP BY STEP
A. 3 c3 page 2 18
B. 3 l2Jc3 l2Jf6 4 eS page 2 2 2
A . 3 c3
Black has not committed him
self with any pawn moves like 2 . . .
e 6 o r 2 . . .d 6 , s o h e keeps all options
open. Whatever you play against 2
c3 should be also good on move 3 .
We'll restrict t o some brief recom
mendations about one of the main
lines against 2 c3 . It has been very
well tested and White seems unable
to demonstrate new ideas here :
3 lllt6 4 es lll d S
A1 . 5 ic4 b6 6 .ib3 d 5 7 exd6

7 d4 cxd4 8 cxd4 is pointless as
Black's bishop can land on fS or g4 .
7 . Yxd 6
. .
Ala. 8 d4
Alb. 8 l2J a3
Ale. 8 0-0
Ala. 8 d4 cxd4 9 0 - 0
9 ltJ a3 e6 1 0 ltJ bS V!id7 transpos
es to Alb ; 9 cxd4 e6 10 l2Jc3 hb3
11 xb3 e6=.
9 .te6 10 lLla3 dxc3 11 e2
After 11 ltJbS V!ixdl 12 xdl c8
13 he6 fxe6 Black's extra pawn is
doubled, but it controls the impor
tant dS-square: 14 l2Jxc3 (14 bxc3
l2Jc4 lS bl es 16 el b6 17 e4 l2Jd6+
Morozevich-Topalov, rapid Monte
Al. S c4
A2 . S d4
218
1 e4 cS 2 l!Jf3 l!J c6 rare lines: 3 c3; 3 l!Jc3

Carlo 20 02) 14 . . . g6 15 el l!JdS+.
ll hb3 12 tllb 5 11Mb8 13
axb3
White plays a risky gambit. In
practice Black obtains good results
with normal development:
13 g6 14 ie3 tll c 8 15 ga4
15 bxc3 g7 16 l!Jbd4 0-0 17 l!Jxc6
bxc6 18 c2 as 19 d4 l!Jd6 20 hg7
xg7 21 a4 cS+, Manca-Kotronias,
Cork 2 0 05.
15 ... g7 16 bxc3
Or 16 f4 eS 17 l!JxeS l!JxeS 18
hes hes 19 f4 0-0 20 fxeS cxb2
21 e4 a6 22 l!Jd4 l!Je7+.
16 0 - 0 17 gh4 tlld 6 18 if4
e5 19 tllxe5
We have been following the game
Matsuura-Leitao, Santos 2006 , when
19 . . . l!JxeS 20 heS e8 21 l!Jxd6
\Wxd6 22 hd6 xe2+ would have
been in Black's favour.
. . .
..
White has tried here: 16 e3 c8

17 l!JbS cS= , 16 l!JbS \Wc6 17 ie3
cs 18 hcS \WxcS= and 16 l!JfS Wic7
17 l!J xe7+ \Wxe7 18 e3 l!JdS=, Han
sen-Timofeev, Skanderborg 200 5 .
9 hb3 1 0 axb3
Black should meet 10 \Wxb3 with
10 . . . \Wd3 ! 11 el e6 12 e3 \Wd7 13
l!Jc4 l!Jxc4 14 xc4 !ie7= .
1 0 . . . %Yd3 !
..
..
Alb. 8 tll a3 ie6

8 . . . a6 is the other popular op
tion. It also brings Black good re
sults.
9 0-0
9 d4 cxd4 1 0 l!JbS d7 ll l!Jbxd4
hb3 12 xb3 l!Jxd4 13 l!Jxd4 e6 14
0-0 ie7 1S dl 0-0 leads to an equal
position:
11 E:el
In Sveshnikov-Gallagher, Calvia
2004 White failed to remove the
blockade on the d3-square after 11
c2 d8 12 xd3 xd3 13 l!Jc4 l!Jxc4
14 bxc4 e6 15 el e7 16 l!JeS l!JxeS
17 xeS a6 = .
11 e6 1 2 E:e3 \Wd7 13 tll c4
tllxc4 14 bxc4 E:d8
Black deployed his pieces on
good squares and does not have any
problems.
..
Ale. 8 0 - 0 ie6
8 . . . c4 ! ? 9 ic2 g6 10 b3 ig7 11 l!J a3
cxb3 12 axb3 0-0 13 d4 ig4 leads to
a very solid position for Black, for
instance, 14 h3 hf3 15 xf3 es 16
llJ bS (16 l!Jc4 l!Jxc4 17 bxc4 exd4
18 ia3 \Wc7 19 hf8 xf8 with ex219
Part 12
cellent compensation) 16 ... dS 17
xdS (17 e2 exd4 18 llJc7 d7 19
ttJxa8 xa8oo) 17 . . . llJxdS 18 !e4 a6 19
hdS axbS= .
9 he6
Or 9 a4 hb3 10 xb3 e6 (10 . . .
c 4 1 1 bs ds 1 2 llJ a3 0-0-0oo) 11
llJ a3 (1 1 aS? ! dS 12 bS c4 ! 13 llJd4
llJd7+) 11 . . . !e7 12 llJbS b8 = .
9 Wfxe6
1 0 d4
10 a4 d7 11 aS llJdS 12 d4 cxd4
13 ttJxd4 e6 14 a6 b6 leads to an un
explored position, which looks ac
ceptable for Black, for example:
lS c4 ttJdb4 16 llJxc6 xc6 17 llJc3
e7f: .
1 0 cxd4 11 lllx d4
Or 11 cxd4 d7 12 llJc3 = .
1 1 lllxd4 1 2 Wfxd4
12 cxd4 d7 13 ttJc3 e6 14 g4
ttJdS ! was excellent for Black in
Ofek-Gruenfeld, Ramat Aviv 1998.
12 gds 13 Wfh4 Wfe2
The latest occurrence ofthis vari
ation was in Sveshnikov-Kobalya,
Dagomys, 0 2 . 04 . 20 0 8 , where Black
preferred 13 . . . g6 14 e3 !g7 lS llJ a3
0-0 16 fel and here simplest would
have been 16 . . . f6 = . The text is ex
tensively tested and practice has
proved that Black i s not worse at

all:
a) 14 d2 xd2 lS llJxd2 xd2 16
fdl h6 17 g3 c6 18 b4 e6 19
d3 f6 2 0 adl tlJ dS 2 1 b8 + @f7 2 2
a 3 (22 xa7 e7- +) 2 2 . . . g 8 (22 . . .
bS- +) 2 3 xa7 e7 24 d4 c8 2 S
el fS- + Ossa-Marin, Sebastian,
2006 ;
b) 14 e3 xb2 l S tlJ d 2 xd2 16
abl xc3 17 hd2 xd2 18 fdl
as 19 e4 f6 ! 20 xb7 @f7 2 1 el
gS 2 2 e2 hS 23 bel h4 24 h3 (24
xe7+ he7 2S xe7+ @g6 26 e4+
@h6 - +) 24 . . . tlJdS 2S b3 @g7 26
e4 llJf4 27 d l fs 28 f3 cs+ 29
e3 xe 3+ 30 xe3 e6-+ Lintchevs
ki-Kuzubov, Kirishi 2 004.
A2. 5 d4 cxd4 6 cxd4

6 c4 llJb6 7 b3 d6 8 exd6 xd6
9 0-0 e6 transposes to line Al.
6. . .d 6
220
7 c4
7 exd6 xd6 8 llJc3 fS= is trivi
al .
7 llJc3 is an old move which leads
to an equal endgame. In principle,
when playing against an eS-pawn,
Black benefits f ram exchanges, es-
1 e4 cS 2 l!Jf3 l!Jc6 rare lines: 3 c3 ; 3 l!Jc3

pecially those of knights. Throwing
queens in the deal is even better.
7 . . . dxeS 8 dxeS l!Jxc3 9 xd 8+
l!Jxd8 10 bxc3 d7
Black has many good plans here.

We'll note a fresh idea, co 6 nected
with an attack on the eS-pawn. It
involves g6, g7 to provoke f4, and
then .. .f6 .
ll e3
Or 11 l!Jd4 c8 12 e3 g6 13 cl
g7 14 f4 f8 ! ? lS e2 f6 16 exf6
xf6 17 0-0 h6+ IKARUS-ZAPPA,
20 06 .
11. . . g6 ! ?
11. . . e6 i s solid and good : 1 2 l!Jd4
c8 13 @d2 cs 14 d3 l!Jc6= Svesh
nikov-Korneev, Sochi 2006 .
12 l!Jd4
Alternatives: 12 0-0-0 g7 13
c4 c8 14 dS xc3+ lS @b2 c8
16 ha7 l!Jc6 17 b6 g4 18 hc6 +
bxc6t Afek-Gruenfeld, Tel Aviv
199 2 ; 12 h4 h6 13 e2 g7 14 0-0
l!Jc6 lS f4 0-0 16 abl b6 17 bs
ac8 18 fel l!Jb8 19 hd7 l!Jxd7
20 bdl l!JcS+ Pomar-Polugaevsky,
Palma de Mallorca 1972 .
12 . . . c8 13 @d2 g7 14 f4 f8 ! ?
planning f7-f6.
7 .ic4 dxe5
Players who like the French De

fence, might prefer 7 . . . l!Jb6 8 b3
dS, e.g. 9 l!Jh4 e6 10 g3 d7 11 l!Jc3
l!J c4 12 0-0 bS 13 a4 b4 14 l!Je2 e7
lS l!Jg2 l!J4aS 16 c2 b3 17 hb3
l!Jxb3 18 xb3 0-0 with excellent
play, Rausis-Chomet, France 200 0 .
O r 8 bS dS ! ? ( 8 . . . dxeS i s well
known from thousands of games)
a) 9 h3 fS 10 l!Jc3 c8 11 0-0 e6
12 gs e7 13 he7 xe7 14 l!Jd2? !
0 - 0 l S l!Jb3 l!Jc4t Bryant-Nakamu
ra, Las Vegas 2007;
b) 9 l!Jc3 g4 10 h3 xf3 11 xf3
e6 12 0-0 c8 13 dl (13 e3 a6 14
e2 h4 lS fdl e7=) 13 . . . a6 14
fl e7 lS g4 g6 16 h6 f8oo ;
c) 9 l!Jh4? ! d7 (setting the trap
. .. l!JeS) 10 e2 e6 11 g3 e7 12 l!Jg2
f6 13 f4 c8 (hindering White's cas
tling, in view of . . . l!Jxd 4) 14 l!Jc3
0-0 lS e3 @h8 16 exf6 xf6 17 b3
l!Jxd4 ! 18 hd4 xc3+ Cherniaev
Zhigalko, Moscow 200 8 ;
d ) 9 0-0 g4 10 e3 e 6 1 1 l!Jbd2
e7 12 Wfe2 c8 13 fcl 0-0 14 a3
d7 lS h3 fS 16 l!Jb3 l!Jc4 ! 17 hc4
dxc4 18 xc4 b6oo, Tiviakov-Topa
lov, Wijk aan Zee 200 6 ;
8 dxe5
8 xdS? ! WfxdS 9 l!J c3 d6 10
221
Part 12
d S ttJd4 11 ttJxd4 exd4 12 xd4 es
13 d3 d7 14 0-0 fS has long been
found to be good for Black.
8 dxe5
8. . . b6
8 . . . ttJdb4 9 b3 e6 can be used to
complicate things.
9 \Wxd8+ xd8 1 0 .ib 5+ .id7 1 1
lll c 3 e6
This endgame is rather drawish,
but, amazingly, White often mana
ges to lose it.
a) 12 xd7+ ttJxd7 13 tlJbS c8
(13 . . . b4+ 14 e2 e7 lS a3 cs
16 b4 a6 17 bxcS axbS 18 e3 ttJc6
19 hbl b4 20 axb4 xal 21 xal
ttJxb4 22 bl ttJa6 23 xb7 c8 24
a7 ttJ axcS 2S tlJgS h6 26 ttJxf7 xf7
27 xcS xcS 28 xd 7 + g6 29 f4
cMS with a draw in the rook end
game, Pavasovic-Sveshnikov,Ljublja
na 1996) 14 tlJxa7 c2oo;
b) 12 0-0 a6 13 xd7+ (13 e3
xbS 14 tlJxbS tlJdS lS tlJbd4 h6=)
13 . . . ttJxd7 14 f4 ttJc6 lS fdl ttJb6
16 acl b4 17 ttJe4 tlJdS=.
Black ensures normal develop

ment for his pieces . In this pawn
structure he might gradually gain
some spatial advantage thanks to
his control of d4 .
7 c4
7 ttJe4 xd6 8 ttJxd6+ xd6 9 g3
es 10 h3 ttJf6 11 g2 0-0 12 d3 fs
13 0-0 fe8= ;
7 d4? cxd4 8 tlJbS xd6 9 ttJfxd4
ttJxd4 10 ttJxd4 b4+ 11 d2 xd4
12 c3 (12 bS + d7 13 xb4 xf2 +
14 dl ttJe 3 + lS xe3 xe3 16 xb7
0-0 - +) 12 . . . xf2 + 13 xf2 ttJxf2 14
xf2 cs++;
7 g 3 xd6 8 g2 0-0 9 0-0 es 10
h3 tlJf6 11 d 3 h6= .
7 e5! 8 h 3
8 ttJe4 xd6 9 ttJxd6 + xd6 1 0
tlJgS 0 - 0 11 ttJe4 g6t 1 2 ttJxcS? ttJd4
13 d3 fS+.
8 lll h6 9 d3
Or 9 xcS? ! xd6 10 c4 0- 0t;
9 ttJe4 xd6 10 ds (10 ttJxd6 + ? !
xd6 1 1 tlJgS 0 - 0 12 ttJe4 g6t)
10 . . . e7 11 xd 8 + xd8 12 bs
ttJb4+.
9 lll f 5 1 0 i.g 5 xd6 1 1
0-0-0 i.e6 1 2 a4 f6 1 3 lll e4
c7t Vachier Lagrave-Spasov, Tu
rin 200 6 .
3 lll c 3 lllt6 4 es lll g 4 5

e2 d6 6 exd6 e6
a.
222
Part 1 3
1 e4 c 5 2 f3 c6 3 d4 cxd4
4 xd4 f6 5 c3 es 6 db5 d6
7 .ig 5 a6 8 a3 b5 9 .ixf6 gxf6
1 0 d5 .ig7
The Novosibirsk Variation
QU IC K R EPERTO I RE
This variation offers a different ap

proach to the Sveshnikov. Instead
of immediately attacking the cen
tre, Black wants to kill first White's
knight on d5 with CiJc7-e7, and only
then to think about activating the
poor g7-bishop . White cannot ef
fectively prevent it, since 11 CiJe3
CiJe7 12 d3 f5 ! 13 exf5 d5 or 11 h5
CiJe7 12 CiJ e3 f5 13 exf5 e4 14 0-0-0
0-0 opens play favourably to Black's
bishop pair. Other attempts like 11
g4 CiJe7 12 CiJe3 do not deserve atten
tion either, because Black will com
plete development, e.g. 12 . . . b7 13
g2 0-0 14 0-0 cj{h8 15 c3 g8 16
d 3 CiJg6 when the whole White
kingside will be weak.
In practice White often tries to
refute Black's construction by the
sharp c4, but then Black's bish
op pair comes to life and his piec
es are like unwinding spring. The
most unpleasant approach is when
White calmly develops, reinfarcing
the key point e4 by f3 , and prepar
ing to produce a passed pawn on the
queenside by c4 at an opportunity.
Note that in the latter case,
it is very difficult for Black to
obtain counterplay.
That is why we do not recom

mend the Novosibirsk variation as
a main line.
1 1 .id3 e7 1 2 x e7
V!fxe7
A. 1 3 c4 (or 13 0-0 0-0 14 c4)

1 3 fS 1 4 0-0 0-0 1 5 V!fh5
The other major option is 15
f3 and then Radjabov's 15 . . . d5 ! ?
works fine:
16 cxd5 fxe4 17 he4 b8
223
Part 13
Black has a good game : 18 fdl
fS 19 d6 Wfff6 20 c6 ie6 .
1 S WhS gb8 !
Intending to take on e4 a nd push
. . .fS.
1 6 exfS e4 1 7 gae1 i.b7
game 37 Bologan-Nedev, Panormo 20 0 1 .
Currently we are unable to show

a clear-cut plan for equality here.
1 4 Wb7
14 . . . Wff g S 15 0-0 ! ib7 is yet an
other critical line .
1 S f3 !
15 V9f3 0-0 16 ttJe3 f4 17 tlJdS
ie6 18 g4 b4 ! is a typical break
trough, which ensures just enough
counterplay.
1 S f4 1 6 lll b4
Move order is not critical. White
can play first 16 0-0.
16 0-0 1 7 0-0 ie6
This position is very sharp and

both sides have plenty of possibil
ities, but Black's play is easier. The
key position arises after:
1 8 f3 dS! 1 9 fxe4 dxc4!
White's attack looks ominous,

but he has nothing decisive.
20 t6 ixt6 21 es Wes+ 22
@ h1 cxd3 23 gxf6 Wb4! 24
gef1 i.e4
We prefer Black's game, although
the position remains complex.
1 3 c3! ts 1 4 lll c 2!
14 0-0 0-0 15 tlJc2 b8 ! , prevent
ing a4, is satisfactory for Black. See
a.
224
1 8 @ h 1 !?
White prepares to play on the
queenside where he will make a
passed pawn.
18 f2 is also logical , because the
d6-pawn seems an accessible tar
get, but Black manages to activate
7 igS a6 8 l2J a3 bS 9 ixf6 gxf6 10 ltJdS ig7

the dark-squared bishop and obtain
counterplay:
18 . . . if6 ! 19 d2 id8 20 ic2
ib6 + 21 hl icS 22 ltJdS hdS ! 23
xdS b4 24 cxb4 xb4 25 bl aS
26 a3 (or 26 g3 h8 ! ? 27 gxf4 g8
28 d2 xd2 29 xd2 exf4 30 ib3
g7) 2 6 . . . b6 = .
1 8 ... as 1 9 d5 ts 2 0 c4
The most purposeful move. It
ensures a pawn majority 2 : 1 on the
queenside since 20 . . . b4 is strategi
cally bad due to 21 exfS.
20 ... bxc4 21 i.xc4 fxe4 22
fxe4 h8
This type of positions is crucial
for the assessment of the Novosi
birsk variation.
Although it seems that Black

does not experience difficulties, his
game is not very pleasant. In fact, he
can hardly hope to win at all . White
will soon make a passed b-pawn and
he will try to convert it by combin
ing play against the more exposed
Black king.
225
Part 1 3
1 e4 cS 2 f3 c6 3 d4 cxd4
4 'Llxd4 'Llf6 s c3 es 6 db S d6
7 igS a6 8 'Lla3 bS 9 ixf6 gxf6
1 0 dS ig7
The Novos ibirsk Va riation
STEP BY STEP
With this move order, Black aims to

sidestep some sharp variations, for
instance, the piece sacrifice on bS,
which is possible after 10 .. .fS. One
of the authors, Nedev, often plays
it, so we shall examine it in de
tail as a backup line. However, our
main repertoire is more active, and
it is currently in a better theoreti
cal shape . There is one specific line
in the Novosibirsk variation (B2b) ,
where Black's game is rather boring
and not suitable for playing for win.
The situation might change though.
If Kasparov chose it twice not so
long ago, Black certainly hides some
trumps up his sleeve.
1 1 i.d 3
a) 11 c3 fS transposes to other
parts of the book, for example, 12
exfS is covered in Part S while 12
d 3 e6 is considered in Part 7.
b) 11 l2Je3 l2Je7 12 d3 fS ! 13 exfS
dS 14 g4 g8 lS h4 d6 16 00-0 h6 ! t, Fernandez Aguado,EM. Kuijf, Sitges 199 2 ;
c) 11 hS l2J e 7 12 l2J e 3 fS 1 3 exfS
e4
13 . . . dS? ! brought Black a victory
in Papaioannou-Spasov, Korinthos
226
2001, 14 gS 0-0 1S f6 ! would have

been rather unclear: 1S . . . l2Jg6 16
llJxdS xdS (16 . . . h6 17 l2Je7+ h7
18 h4 ! +) 17 fxg7 xg7 18 d3 f6 19
g3 e4 20 e2 e6 21 0-0 d4oo.
14 0-0-0 0-0 (14 . . . aS ! ?)
Black's bishop pair and mobile

central pawns should prevaili n the
ensuing sharp clash .
lS g4
lS f3? ! b6 16 gs dS ( or 17
llJxdS? llJxdS 18 xdS h 8 ! Klinger
Vaisser, Szirak 1985) 17 f6 xf6 18
xf6 xf6 19 llJxdS llJxdS 20 xdS
e6 with the bishop pair advantage
and better development.
lS . . . dS 16 g2 (16 c3 b4 ! 17 cxb4
c7+ 18 bl es 19 d2 b8 ! )
16 . . . c7 ! 1 7 xdS !
Alternatives are : 17 llJxdS? tlJxdS
7 !gS a6 8 l2J a3 bS 9 xf6 gxf6 10 ltJdS !g7

18 xdS !b7 ! 19 d2 es 20 c3 e3 !
2 1 fxe3 xg2 2 2 xg2 xe3+ 2 3 d2
ad8 24 hdl xd2 2S xd2 d8 26
ltJbl h6- + ; 17 f6? ! xf6 18 ltJxdS
ltJxdS 19 xdS xg4 20 dgl ad8
21 xe4 d4 22 e3 b4 23 e4 hS
24 h3 eS ! 2S hxg4 xe4 26 gxhS+
h7- + .
1 7. . . ltJxdS 1 8 ltJxdS es 1 9 f6
xf6 20 xeS (20 xe4? fails to
20 . . . xb 2 + 21 d2 d8 22 e2
!g7! +) 20 . . .xeS 21 l2Je7+ g7 2 2
xe4 b8 23 l2Jxc8 bx98+, Emeli
anov-Tolstov, corr 200 2 ;
d) 11 g4 l2Je7!
Trading off White's most power
ful piece is certainly a good reaction
to White's flank strategy. 11. . . hS is
also worth considering: 12 gxhS fS
13 gl (13 !d3 l2Je7 14 l2Jxe7 xe7
lS c3 !b7 16 f3 gS+, Johnsen
Royset, Tromsoe 1999) 13 . . . !h6 !oo.
12 !e2
12 l2Jxe7 xe7 13 c3 hS ! ? or
13 . . .!b7 14 g2 !h6 lS l2Jc2 fS ! 16
l2Jb4 ! fxg4 17 ltJdS xdS 18 xdS
c8f! are both pleasant for Black;
12 l2Je3 !b7 (12 . . . hSf!) 13 !g2
0-0 14 0-0 h8 lS c3 g8 16 d3
l2Jg6 17 ltJdS d7 with an edge .
12 . . . ltJxdS 13 xdS !e6
14 c6 +
Or 14 d2 dS+; 14 b7 aS + ! ?
l S c3 c8+.
14 . . . e7 lS b7+ (lS 0-0-0
b8 ! - + ) 1S ... d7 16 xd7+ hd7
17 c4 h S ! 18 cxbS
White cannot maintain the grip
on the centre: 18 f3 hxg4 l 9 fxg4 h4
20 h3 !c6 21 !f3 fS 22 gxfS he4+;
18 gxhS c6 19 gl !h6 20 cxbS
axbS 2 1 ltJxbS hb8 2 2 l2Jc7 a4+.
18 . . . hxg4 19 bxa6
This position has been reached in
Sulskis-Nedev, Gothenburg 2 0 0S.
Black has the initiative a nd the best
way to develop it would have been
19 . . .fS ! 20 exfS (20 !d3 fxe4 2 1
he4 d S 22 hdS xa6 2 3 l2Jc4 e4+)
20 . . . e4 2 1 l2Jc2 (21 0-0-0 xfS+; 21
l2Jc4 xa6 2 2 xg4 d S 2 3 l2Je 3 !c6+)
2 1 . . .xb2 22 bl !c3 + 23 fl .ixf5
24 l2Je 3 !e6 2S b7+ f6+.
e) 11 c4 fS 12 cxbS
12 !d3 l2Je7 13 l2Jxe7 (13 cxbS
ltJxdS 14 exdS e4 lS e2 as+ 16
Ml 0-0 17 !c4 axbS 18 ltJxbS d7
19 l2Jxd6 b4 ! 2 0 ltJbS fc8-+)
13 . . . V9xe7 transposes to 11 d3 .
12 . . . l2Jd4
13 !d3
Alternatives are:
13 bxa6 0-0 14 !c4 fxe4 lS 0-0
2 27
Part 13
ha6+; 13 exfS b7 ! ? (13 . . .hfS 14
c4 0-0 lS bxa6 gS ! ? 16 0-0 h3 17
ttJe3 dS 18 hdS xa6 19 c>hl d7)
14 ttJe3 (14 c4 axbS lS ttJxbS hdS
16 hdS ttJxbS 17 ha8 xa8 18 0-0
ttJd4f!) 14 ... axbS lS hbS+ c>e7 16
d 3 e4 17 c4 as + 18 d2 dS + ; 13
b6? 0-0 14 d3 b7 lS ttJ c4 (lS ttJc7?
fxe4+) lS . . . b8 ! 16 ttJce3 (16 0-0?
fxe4 17 xe4 fS-+) 16 ... fxe4 17 he4
fS 18 d3 f4 ! 19 hS h6 2 0 g6 f7!
21 e4 fxe3 22 fxe3 ttJc2 + - + .
13 . . . 0-0
13 ... e6 was played in the famous game Anand-Lautier, 1997
and it is supposed to be good for
Black. We do not share this opinion,
since after: 14 0-0 0-0 lS ttJc2 ttJxc2
16 hc2 fxe4 17 bxa6 xa6 White is
somewhat better with both 18 a4 ! ?
fS 1 9 b 4 o r 1 8 he4 fS 1 9 c2 ! c>h8
20 b3 e4 21 d2 a7! 22 acl es
23 g3 g7 24 f4 ! ? exf3 2S xf3 h4
26 ttJf4 hb3 27 axb3;t.
14 ltJc2 (14 d2 fxe4 1S he4 e6 !
16 ttJc2 h4 gives Black a strong in
itiative, for example: 17 ttJc3 dS 18
hdS ad8 19 ttJe3 axbS) 14 . . . fxe4 lS
xe4 b8 16 0-0
16 b6 h4 17 f3 e4 !oo is excel
lent for Black; 16 ttJxd4 exd4 17 0-0
xbS 18 b4 (18 cl b7 ! 19 f3 e 8
20 a4 cS=) 18 ... b7 19 b3 is balanced after 19 . . . e8 ! ? or 19 . . . gS.
16 . . .xbS 17 b3 b7 18 ttJc3 dS ! ?
This exchange sacrifice i s more
enterprising than 18 . . .he4 19 ttJxe4
ttJxc2 20 xc2 dS 21 adl d7= .
19 ttJxbS dxe4 2 0 ltJc3 gSoo.
Black has an obvious compensa
tion. He could also try 20 . . . c7 with
tempo .
228
11
..
e 7 1 2 xe7 VHxe7
Now White chooses plans with

13 c4 or 13 c3 . Only the latter pos
es problems to Black.
A. 1 3 c4
White often starts with 13 0-0
0-0 a nd then 14 c4. Only 13 0-0
0-0 14 f3 is of independent signifi
cance: 14 .. .fS lS exfS dS 16 xdS
b7 17 b3 e4 18 fe l ! ? (18 e2
gs 19 adl e3 2 0 f3 es 21 ttJc4
f4 22 d4 ad8 = Anand-Kramnik,
1998) 18 . . . ad8 Black has very ac
tive pieces and he quickly regains
one of the pawns, retaining the in
itiative, e.g. 19 fl (19 c4? b4; 19
adl f6 ! 20 fl xb2 21 c4 b4 22
ltJbl aS+) 19 . . . h4 ( 1 9 . . . dS ! ?) 20
adl d4 2 l d 2 fe8 ! 2 2 g3 f6 23
de2 xfS 24 g2 e3 ! .
1 3 fS 1 4 0-0
Black has a fine game after 14
cxbS dS lS 0-0 fxe4 16 c2 e6 17
bxa6 0-0 18 ttJbS xa6, Jobava-Ya
kovich, Moscow 2 007.
14 hS fails to 14 . . . dS ! lS cxdS
fxe4+ due to the check on b4.
Or 14 exfS e4 lS 0-0 exd3 16 el
es 17 f4 MS 18 fxeS 0-0oo.
...
7 gS a6 8 l!J a3 bS 9 !xf6 gxf6 10 l!JdS g7

14
...
0-0
Al. 15 hS
A2 . 15 f3
15 cxbS is thematically met by
15 . . . dS ! .
15 e2 is also bad, because it
cannot stop .. .fS: b7 16 f3 fxe4 17
fxe4 f5 18 exf5 e4 19 cxbS axbS 20
!xbS dS+.
A1 . 1 5 \Wh5 g b8 !
Black intends to take o n e4 and
push .. .f5 so White has not a big
choice :
1 6 exf5 e4 1 7 g a e1 ib7
White has tried here nearly all

legal moves, but Black always re
tains good chances .
1 8 f3
Alternative s:
a) 18 l!Jc2 WeS ! + ;
b) 18 b 3 bxc4 19 bxc4
Or 19 l!Jxc4 dS 2 0 l!Je3 (20
e 3 f6 ; 20 l!J aS a8) 2 0 . . . exd3
(20 . . . c3 ! ? 2 1 l!Jg4 f6) 2 1 l!Jg4 Wd6
2 2 f6 .ixf6 23 h6 e7 24 xe7
xh6 25 l!Jxh6+ g7 26 l!Jf5+ @f6
27 g4 c8+
19 . . . h8 ! with strong counter
play, for example:
20 l!Jbl eS ! 21 f6 xf6 22 !xe4
!xe4 23 xe4 b2 24 e2 fb8 25
l!Jd 2 (25 xb2 xb2+) 25 ... g6 ! + ;
2 0 e3 eS ! 2 1 h3 f6 2 2 e2
c6 23 l!Jc2 b2+;
2 0 l!Jc2 g8 21 f3 dS ! 2 2 fxe4
dxc4, when 23 e3? loses to 23 . . .
cxd3 24 h3 h6 ! 25 xh6 g 2+
26 hl g7- + ;
c ) 1 8 cxbS dS ! 1 9 bxa6 c6 2 0
bl ( 2 0 e3 xb2 2 1 h3 h 6 2 2 bl
a8 23 g3 Wf6+) 20 . . . xb2 21 e3
f6 22 l!Jc2 d4+, T. Horvath-Nedev,
Fuegen 2006 ;
d ) 18 !xe4 !xe4 1 9 g4 ( 1 9 f3??
W a7+ 20 hl d3-+) 19 ... fe8 20
f6 xf6 21 xe4 Wxb2+;
e) 18 e3 bxc4 19 h3 (19 xc4
WeS ! 2 0 h3 h6 2 1 g4 c8 ! 2 2 hS
xb2 23 b3 e3 ! 24 fxe3 xe3+ 25
hl f2 !+) 19 . . . h6 20 !xc4 gS ! 2 1
WxgS hxgS. White has to defend a
grim endgame: 22 dl (22 g3 dS
23 xgS f6+) 22 . . . dS 23 !xdS !xdS
24 xdS xb2+;
f) 18 Wg4 (18 ... fe8 ! ? is also a
good choice) 18 . . . h8 ! ? 19 !xe4
(19 cxbS dS) 19 . . . fe8 2 0 e3 (20 f6
i.xf6 2 1 d3 g8 !) 2 0 . . . !xe4 21 fel
f6 2 2 xe4 xe4 23 Wxe4 xb2+.
229
Part 13
1 8 . . . dS 1 9 fxe4 dxc4 !
In Gouliev-Nedev, Illkirch 2004

was 19 . . . cS+ ! ? 20 hl dxc4, when
the best 2 l eS ! cxd3 2 2 f6 xf6 would
have transposed to the main line.
20 f6
Or 20 bl? d4 + 21 hl f6+.
20 . . . ixf6 21 es Wes+ 22 @ h 1
cxd3 2 3 xf6 Wb4 ! 2 4 ef1 ie4
We prefer Black's game, al
though the position remains very
complex. For instance :
2S l!Jbl (2S 6f4? ! g6 ! ; 2S e6
xb2 26 exf7+ h8t) 2S . . . xb2
26 gs + ( 26 6f2? g6 ! 27 gs
d4 28 l!Jd 2 be8 29 l!Jf3 b6 30
h4 f6 3 1 exf6 xf6 - +) 26 ... g6 27
h4! e2 ! 28 6fS hS 2 9 xf7 (29
l!Jd2 b6 !) 29 ... xf7 3 0 xf7 g4 31
xg4 hxg4+.
A2. 1 S Wf3 d S ! ?
This plan has been developed
by Radj abov. It consolidates the
queenside and shifts the focus onto
the centre where Black has an initi
ative. The bishop pair compensates
the sacrificed pawn.
1 6 cxdS fxe4 1 7 .ixe4 b8
230
18 fd 1
White assigns the queen's rook
to the c-file. Alternatives are :
a) 18 adl b6 19 d3 ! (19 e3
d6+; 19 fS f6 2 0 d6 xd6 2 1 xd6
xd6 22 l!Jc2 xfS 23 xfS g6 =)
19 . . . d7 20 hh7+ (20 g3? ! fS 2 1
g2 h 6 2 2 f4 exf4 23 gxf4 hb2 24
l!Jc2 g7 2S l!J b4 d6+ Leko-Rad
j abov, Linares 2 0 04) 2 0 . . . h8 21
d 6 ( 2 1 e3 h6 2 2 c2 d6oo, Smir
nov-Radjabov, FIDE-Web, Tripo
li 20 04) 21 . . . d8 22 e4 ( 2 2 e3
xd6 23 xd6 xd6 24 bl fS 2S
l!Jc2 e4 26 b4 es 27 g3 f6 2 8
e2 g'Too Lutz-Radjabov, Plovdiv
2003) 22 . . . h6 ! 23 fel (23 f4? !
b7!+ ) 23 . . .fS 24 xeS+ ( 2 4 b4
aS ! 2S xaS xd6 is risky for White)
24 . . . xh7 2S f6 (2S e7+ g8 ! +)
2S . . . b70 26 l!Jc2 e8 ! 27 l!Jd4 e4 !
2 8 l!Jf3 g7+;
b) 18 acl fS ! 19 xc8 bxc8
20 Ms gs 21 e6 + h8 22 dl.
Black has the initiative after either
22 . . . cs, Ganguly-Venkatesh, Rajendran 2004, or 22 . . . c7, Reinaldo
Castineira-Yakovich, Paris 2 0 0S;
c) 18 fS f6 19 hc8 xf3 20
gxf3 fxc8 2 1 acl e4 22 xc8 +
xc8 23 fxe4 h b2 2 4 l!J b l and now
7 gs a6 8 tt:J a3 bS 9 xf6 gxf6 10 tt:Jds g7

24 . . . E:c2 ! 2S E:dl Af gives Black
enough activity for the pawn : 26
g2 f6 27 E:d2 E:c4 28 f3 b4.
1 8 . . . fS 1 9 d 6 Wff6 20 c6 e6
21 d5
Or 21 E:d2 e4 22 hS E:fd8 23
E:adl E:b6 24 ds E:bxd6+.
21 .. Jbd8 22 Wfb3
The endgame after 22 he6 +
xe6 23 dS (23 b3 xb3 24 axb3
E:f6+) 23 . . . E:f6 24 d7 xdS 2S E:xdS
E:c6 26 E:adl ( 26 f4?? b4 ! ) 26 . . . f7
is equal: 27 E:d6 (27 tt:Jbl E:c7 2 8
tt:Jc3 e7=) 27 . . . E:xd6 2 8 E:xd6 f8
29 E:xa6 E:xd7= .
22 . . . f7 ! 23 li:) c2
Otherwise White would be
worse:
23 E:acl E:xd6 24 hf7 + E:xf7 2S
E:c8 + f8 26 E:xd6 xd6+;
23 b4 e4 24 E:abl h8 ! (threate
ning . . . xd6) 2S b7 (2S b3 hb3
26 axb3 E:f7 27 tt:Jc2 E:fd7 28 tt:Je3 f4
29 tt:Jfl g6) 2 S . . .ha2 26 E:al g8
27 ha6 xb2+;
2 3 ... e4 ! 24 g ab1
After 24 tt:Je3 f4 2S hf7+ (2S
tt:Jg4 g6) 2S . . . E:xf7 26 tt:Jds xb2
27 xb2 (27 E:acl xb3 28 axb3
E:fd7 29 E:c6 es 30 E:xa6 f7 !t)
27 ... hb2 28 E:abl a3 and Black is
at least equal. Or 24 a4 xb2=.
2 4 ... xd6 2 5 xf7+ gxf7 26 li:) e 3
f4 27 li:) g 4 gxd 1 + 2 8 gxd 1 Wfe 7.
Black has sufficient counter
chances .
B. 1 3 c3! f5
Bl. 14 0-0
B 2. 14 tt:Jc2 !
8 1 . 1 4 0-0 0-0 1 5 li:) c2 g b 8 !
A useful move, which prevents

a4 and threatens to take on e4, fol
lowed up with .. .fS .
1 6 exf5
White cannot hold e4 because
of 16 E:el fxe4 ! 17 he4 fS 18 dS+
h8 19 tt:Jb4 Wfc7, when 2 0 a4 bxa4 !
2 1 xa4 cs 22 a2 (22 E:edl? ! e4)
22 . . . e4 ensures counterplay, where
as 2 0 b3 as 21 tt:Jds b7 22 a3
e6 23 tt:Je3 hb3 24 xb3 f4 2 S
tt:JdS E:bc8 26 E:adl E:cS was level
in Anand-Radj abov, rpaid, M ainz
2006 .
1 6 . . . e 4 1 7 ge1 xf5 1 8 li:) b4
Practice has also seen:
a) 18 tt:Je3 g6 19 tt:Jds Wies 2 0
c2 ( 2 0 fl? ! i s a positional mis
take, since White needs this bishop
on b3 in order to protect dS. 20 . . .fS
2 1 f4 e6+ and Black is fine: 2 2 tt:Je3
f7 23 g3 h8 24 h3 g8 2S hS
g6+) 20 c2 as 21 f4 e6 22 a3 ( 2 2
tt:Je3 i s well countered with 2 2 . . . b4 !
and White i s worse despite that he
can win a piece with 23 g4 bxc3 24
b4 E:xb4 2s fS es 26 ds E:d4 27
W!xeS heS+) 2 2 a 3 h8 2 3 b3 Wfc8
24 d2 cs + 2S hl E:fe8 26 E:adl
hS 27 E:cl fS+ ;
23 1
Part 13
b) 18 ttJd4 hd4 19 cxd4 dS 20
d2 b6 ! 2 1 acl g6 2 2 fl h4
23 e3 gS ! 24 el ! (Or 24 g3 hS
2S cs e6 , Short-Illescas Cordo
ba, M adrid 1997 and Black had a
dangerous attack; 24 cs h3 2S
c3 hg2 2 6 g3 f3 !oo) 24 ... f6 ! ?
( 24 . . .g4 2 S g3) 2S b4 g4 26
g3 hS 27 h3 e6 2 8 d6 xg3 29
xg3 + @h7 30 e3 c 8 3 1 xc8
hc8 = ;
c) 1 8 fl as 1 9 a3 g6 2 0 ds
fc8 21 adl es 22 d2 cSf!.
1 8 . . . aS! 1 9 d5 es 20 ic2
ig 6
Black has a bishop pair and
good centre . See for further detail
game 37 Bologan-Nedev, Panor
mo 2 0 01.
82. 1 4 c2 !
Currently we are unable to show

a clear-cut plan for equality here.
We shall focus on the problem po
sitions, and you could be checking
for improvements for Black.
1 4 . . b7
14 ... 0-0 lS hS and 14 ... b7 lS
exfS are not appealing, so the only
.
232
alternative is:
14 ... gs, when lS 0-0 ! is t he crit
ical line. The other options are un
der control:
a) lS exfS xg2 16 fl h6 ! 17
a4 g8 18 axbS axbS 19 hbS+ d7
2 0 xa 8 + xa8oo for example : 2 1
hd7+ (21 xd6 e4+ 2 2 e2 xc2
23 b8 + @e7 24 xeS+=) 2 1 . . . @xd7
22 d3 a4 23 ttJb4 e4 24 dl
xdl + 2S @xdl gS = ;
b ) l S f3 f4 1 6 h 3 0 - 0 1 7 0-0
b7! 18 a4 (18 ttJ b4 aS) 18 .. .fSt;
c) lS ttJ e3 is a continuation that
has not been tested in practice. lS . . .
f4 16 h 4 g6 1 7 h S gS 1 8 tlJdS
xg2 19 fl leads to a complicat
ed position, but the following end
game seems equal: 19 . . . g4 20 f3
e6 2 1 e2 (21 ttJc7+ @e7 2 2 l!Jxa8
xb2 23 cl xcl+ 24 xcl xa8oo)
2 1 . . . xe2 + 22 @xe2 hdS 23 exdS
b8 =
d ) l S e2 dS ! 16 exfS (16 exdS e4
17 l!Jb4 0-0 18 c2 as 19 l!Jc6 b4f!)
16 . . . e4 ! 17 f3 (17 l!Jb4 0-0 18 c2
b7 19 0-0 ad8 20 f3 d4) 17 . . . 0-0
18 fxe4 hfS 19 e3 (19 0-0 dxe4
20 e3 xe3 + 2 1 l!Jxe3 g6 22 c2
b4+) 19 .. h4 + 20 g3 h3 21 fl
g4 2 2 g2 he4 2 3 he4 dxe4 24
0-0 g6 = .
Now let u s return t o l S 0 - 0 b7
'
7 gS a6 8 ttJ a3 bS 9 xf6 gxf6 10 ttJdS .ig7

Our analysis suggests that White
is able to retain control over the criti
cal square e4. Thus his chances re
main preferable, as he has a clear
plan of attacking the queenside by
a4.
Perhaps most challenging is 16
f3 ! ? , but 16 f3 0-0 17 exfS dS 18
f4 ! also looks unpleasant: 18 . . . exf4
19 f3 eS 2 0 a4 (20 ael fe8 2 1
ttJd4 f6=) 2 0 . . . fe8 2 1 axbS axbS
22 hbS xal 23 xal b8 24 tlJd4
hd4+ 25 cxd4 xfS ; On the con
trary, 16 e2 0-0 (or even 16 . . .fxe4
17 he4 d5 18 f4 f6 ! ) 17 f3 d5 looks
fine for Black;
After 16 f3 Black tested 16 .. .f4
17 a4 0-0 18 axbS axbS, when Acs
recommends 19 xa8 xa8 20 al
fS 2 1 xa8 ! xa8 2 2 exfS d8 23
ttJb4. The problem with this po
sition is that Black plays for only
two results since White is in total
command over the light squares.
We tried to improve his play with
19 . . . xa 8, when 20 ttJa3 fS ensures
Black an initiative: 21 ttJxbS dS 2 2
ttJd6 fxe4 2 3 he4 dxe4 2 4 dl f8
25 ttJxb7 f3 . However, White has a
better option: 2 0 ttJb4 f5 2 1 e2 dS
22 c2 and our b7-bishop is hang
ing in some variations. As a whole,
Black's pawn cluster in the centre
looks awesome, but the exposed
diagonal a2-g8 tips the balance in
White's favour.
After 15 . . .b7 the main line
branches to :
B2a. 15 f3
B2b. 15 f3 !
White can also defend the e4pa wn by 15 e2 fxe4 16 xe4 dS 17

b4 f8 1 8 h4 e4 19 e2 g7+, or
take on fS: 15 exf5 xg2 16 fl b7 17
ttJe3 c6 18 g4 f6 19 0-0-0 hS!+,
but in these cases Black is on top.
B2a. 15 'ti'f3 0 - 0 16 ltJe3

The mundane 16 0-0 would face
16 . . . dS ! 17 exfS e4 18 g3 hf5 ! ? 19
e2 b4 ! 20 ttJxb4 as 2 1 tlJc2 xb2 2 2
ttJd4 g6 ft .
16 ... f4 17 d5 e6 18 g4 b4!
This thrust ensures just enough

counterplay.
19 c4
Or 19 @fl ! ? bxc3 20 bxc3 ab8
21 @g2 d7 22 h3 b2 23 hbl tb8
233
Part 13
(23 . . . E1d2 ! ? 24 dl ! b2=) 24 b2
xb2 2S bl xbl 26 .bbl d8 27
d3 hS 2 8 ha6 (28 gxhS gs + 29
cM"1 h4 30 g2 gs+) 28 ... gs
29 e2 hxg4 30 hxg4 hdS 31 exdS
fS 32 h3 e4 33 f3 fxg4 34 xg4
xg4+ 3S fxg4 hc3 !+.
19 ... hd5 2 0 exd5
2 0 cxdS f6 ! occurred ion Dol
matov-Topalov, Groningen 1993
21 gS d8 22 fl ! , when 22 . . . c8 ! ?
( 2 2 . . . h8) 23 .th 3 c2 2 4 fS b6
2S hS f6 is double-edged.
2 0 . . . e4 ! 2 1 he4 hb2 22 :Sbl
ie5.
Black is going to produce a passed
pawn on the queenside while he will
defend the kingside along the sev
enth rank after a possible . . .f6 .
B2b. 1 5 f3 ! f4
This is a typical position which
requires a good understanding of
the main plans and manoeuvres.
Black has two active plans:

1. He launches a pawn storm on
the queenside with . . . . as, . . . . b4, and,
hopefully, . . . b3.
2. He activates his dark-squared
bishop. That could be achieved by
234
manoeuvring it to b6 via f6-d8, or

with the help of the pawn sacrifice
. . . . eS-e4 after an exchange on dS .
Unfortunately, with precise pro
phylactic play, White is able to re
strict both of them :
16 b4
16 0-0 should later transpose to
the main line .
16 e2 0-0 17 g4 unnecessari
ly weakens the kingside and loses
a tempo without achieving obvious
benefits :
17 . . . e6 18 ltJb4 fc8 ! The a8rook is needed to support the a-pawn,
whilst the f8-square might prove
useful for the g7-bishop. Now, 19 0-0
aS 20 ltJdS loses a pawn to 20 ... hdS
2 1 exdS xdS 2 2 e4 (22 e4 xe4
23 he4 ab8+) 22 . . . cS++, where
as long castling is even worse: 19
0-0-0 aS 20 ltJdS a4 ! 21 bl (21 a3
hdS 22 exdS b4 ! ) 2 1 . . . a3 22 b3
hdS 23 exdS f%xc3 24 e4 c7+.
It remains to examine :
19 gS aS 20 ltJd S b4 2 1 c4
2 1 . . . b3 ! ? Fixing a target on b2.

2 2 a 3 xdS 23 exdS e4 ! This sacri
fice is a must in such positions ! 24
fxe4 (Or 24 he4 a6 2S cl b6
26 fl d4 27 c3 b6 28 cl
d4=) 24 . . . e7 2S 0-0-0 (2S h4 f3 !)
7 igS a6 8 l!J a3 bS 9 hf6 gxf6 10 l!JdS :Ag7

2S .. .xgS 26 hgl es 27 g4 (27
g2 h8 28 dgl g8=) 27 ... h8 28
bl fS 29 exfS xe2 30 he2 eS= .
16 ... 0 - 0
Black can also start relocat
ing the bishop with 16 . . . !f6 17 0-0
ie6 18 !c2 0-0 19 ib3 (19 xd6?
a7+ !+) 19 . . . as 2 0 l!JdS d8 .
17 0 - 0
17 ic2 is rather pointless, since
after 17 . . . aS 18 l!JdS ie6 White's
bishop is at least not better than it
had been on d3 :
19 0-0 fS 2 0 W'd2 hdS 21 xdS+
W'xdS 22 exdS b4 ! f! 23 cxb4 axb4= .
There i s more reason i n 1 9 g4,
hoping to organise an attack with
out castling. We answer 19 . . . fc8 !
(threatening . . . b4) 20 e2 (or 20
d3 b 4 2 1 c4 hdS 2 2 exdS e4 ! t)
2 0 . . . b4 2 1 c4 hdS 22 exdS a4 ! 23
0-0 (23 e4? ! b3 ! t) 23 ... b6+ 24
hl e3f!.
17 ... i.e6
the dark-squared bishop and obtain

counterplay:
18 . . . if6 ! 19 d2 d8 20 ic2
b6 + 21 hl icS 22 l!JdS hdS ! 23
xdS b4 24 cxb4 xb4 2S bl aS
26 a3 (or 26 g3 h8 ! ? 27 gxf4 g8
28 W'd2 W'xd2 29 xd2 exf4 30 !b3
g7) 26 . . . W'b6= .
1 8 . . .a 5 19 d5 f5
Apparently 19 . . . fc8 is not
enough for complete equality. 2 0
d2 (after 2 0 b3 ab8 White can
not organize a passed pawn: 21 a4
b4 22 c4 a7f! ; 2 0 f2 ! ? however
deserves attentio n. It defends b2
a nd prepares a4) 2 0 . . . ab8 21 a3 ! ?
hdS ( 2 1 . . cs 2 2 b4) 2 2 exdS cs
23 b4:t and the position is similar to
the main line.
2 0 c4
The most purposeful move. It
ensures a pawn majority 2 : 1 on the
queenside since 20 . . . b4 is strategi
cally bad after 21 exfS .
2 0 W'b3 allows 2 0 . . . a4 21 W'b4
(21 W'xbS xbS 22 hbS hdS 23
exdS e4 24 ic6 ab8 2 S abl a3 =)
2 1 . .. hdS 2 2 exdS xdS 23 hbS
W'cS 24 W'xcS dxcS=
2 0 .. . bxc4 21 hc4 fxe4 22
fxe4 cbh8
18 cbhl! ?
White prepares to play o n the
queenside where he will make a
passed pawn.
18 f2 is also logical, because the
d6-pawn seems an accessible tar
get, but Black manages to activate
23S
Part 13
This type of positions is crucial
for the assessment of the Novosi
birsk variation. Although it seems
that Black does not experience dif
ficulties, his game is not very pleas
ant. In fact, he can hardly hope to
win at all. White will soon make a
passed b-pawn and he will try to
convert it by combining play against
the more exposed Black king. Play
maight continue:
23 b3 f7 24 \We2 \Wa7 ! ? 25 acl
White should not exchange his
light-squared bishop . After 25 l2Jc3
hc4 26 bxc4 ( 26 xc4 ac8 27 d3
d4f) 26 . . . e3 27 c2 f3 f Black is
236
back into play again.

25 ... ac8 26 c2 cs 27 fcl
fc8 28 a3;t.
White is ready to push b4, with

some pull .
Part 1 3
1 e4 cS 2 f3 c6 3 d4 cxd4
4 xd4 f6 s c3 es 6 db S d6
7 .igS a 6 8 a3 bS 9 xf6 gxf6
1 0 d S g 7
T h e Novosibirsk Variation
COMPLETE GAMES
37 Bologan - N ed ev
E U - C u p P anorm o 28.0 9.2001
Comments by Nedev
1 e4 cs 2 f3 c6 3 d4 cxd4
4 xd4 f6 s c3 es 6 d bS d6
7 g s a6 8 a3 bS 9 xf6 gxf6
1 0 d S g7 1 1 d3 e 7 1 2 xe7
V!Jxe7 1 3 0-0 0-0 1 4 c3 f S 1 S c2
gb8 1 6 exfS e4 1 7 ge1 xfS 1 8
b4 a s 1 9 d S V!JeS 20 c2 g6
21 f4
No one has tried to maintain
the tension in the centre with 22 a3
cj{h8 23 d2 fc8 24 cj{hl .
21 d2 , apart from 2 1 . . .fS, al
lows 2 1 . . . b4 ! ? 22 adl bxc3 23 bxc3
b2 24 cl bSoo.
21 . . . V!Je6 !
The queen must defend the d6-
pawn. In case of 2 1 . . . e8, 22 ttJe3 !

hits immediately this sensitive
point.
22 V!Jd2
22 a4 ! ? would have made a
passed pawn since 2 2 . . . bxa4? ! 2 3
xa4 f5 ( 2 3 . . . xb2 24 xe4) 24 b3 !
favours White. So Black should go
on with 22 . . . b4 ! 23 cxb4 axb4 24
bl (24 g4 fS 25 b3 cj{h8 26 gS
xb2 27 a2 g7 28 ttJxb4 c8+)
24 .. .f 5 (24 . . .fc8 25 b3 d7 2 6
g4 ! ) with a tangled position: 25 b3
f7 26 e3 d7oo .
22 . . . @ h 8 23 g a d 1
I n Anand-Radj abov, Mainz
2006 was 23 a3 fS 24 b3 c8+
and Black was already better. The
queen finds a perfect place on cS.
23 a4 fc8 ! ? would also leave Black
with very active major pieces.
23 . . . fS 24 V!Je2 V!Jf7 2S gd2
Bologan put all h i s pieces in the
centre, but they come under the
sway of Black's bishop s. Now 2 5
h 3 would have parried the threat of
. . .hS , but at the cost of exposing
the king : 25 . . . b4 26 cxb4 axb4 27
b3 hS 28 g 4 fxg4 29 hxg4 g6
30 :d2 ! (30 ttJe3 f4) 30 . . . be 8t
237
Part 13
25 . .. :gfc8
In this structure, it is vital to ob
tain counterplay on the queenside
before White redeployed his pieces
to attack d6. The only way to display
activity is 25 . . . b4 ! ? and I could have
pushed it right away. Stayed White's
king on hl, Black's task would have
been much more complicated, but
in the current situation Black's in
itiative develops smoothly, for ex
ample:
26 \Wa6 bxc3 27 bxc3 b2 ! f
2 6 !b3 ! ? a4 ! (thanks t o the check
from a7 !) 27 !c4 bxc3 2 8 bxc3 W'a7+
29 @hl \Wes+;
2 6 cxb4 axb4 27 !b3 W'a7+ 2 8
@hl !f7cc .
Anyway, White is unable to pre
vent it:
26 a3 b4 ! 27 axb4 axb4 28
tll xb4 d5 !
Black's pieces are too discoor
dinated for the otherwise thematic
exchange sacrifice : 28 . . . xb4? 29
cxb4 dS 30 \Wa6 ! + .
29 tll xd5
The position of White's king on
gl rescues Black in many variations.
For instance, 29 !b3 W'a7+ 30 hl
d4 31 !e6 dxc3 32 bxc3 fuc3+.
29 .. Jxb2
23 8
Finally Black's bishop p a ir is

unleashed a nd the enemy's extra
pawn does not play significant role .
Even without queens White's de
fence would be difficult: 30 !bl ! ?
hS ! (30 . . . W'a7+ 3 1 @hl xd2 3 2
\Wxd2 \WcSf) 31 W'f2 (31 \Wa6 xd2
32 W'xc8 + W'f8 33 W'xf8 + xf8+)
31. . . xd2 3 2 W'xd2 \Wd7 !+. Bologan
decides to keep them on:
30 1Mf a 6 \Wf8 !
My idea is to kill the dS-knight
with . . . !g6-f7xd5 and conduct the
attack with opposite coloured bish
ops. It is unclear how White could
save the game. For instance, 31 W'a7
!f7 32 cl xdS 33 xdS fuc3 34
W'aS !f6+. Bologan tries to simpli
fy with :
31 tll e7? !
Now the obvious 3 1 . . . xc3 wins
easily, because of the cross-pin of
the c2-bishop : 32 W'xg6 hxg6 33
llJxg6+ @g8 34 l!Jxf8 xf8 35 cl
!cS+ 36 @fl !e3 - + . Instead I made
a terrible move, which also effec
tively finished the game, but with
another result . . .
3 1 . . J; d a ? 3 2 tll x g 6+ hxg6 3 3
g xd8 \Wxd8 3 4 \Wx g 6 xc2 3 5 \Wxf5
ixc 3. Draw, due to 36 . . . xel 37
W'xc2 \Wd4 + .
Index of va ri ations
1 e4 c5 2 ll:)f"3 tll c 6
Part 1 The Rossolimo Variation 3 bS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
4 d3
16
4 hc6 dxc6 5 d3 g4 6 h3 hS ! . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
7 g4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
7 ttJbd2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
7 ttJc3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
7 {4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4 eS ttJdS S ttJc3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
5 0-0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2
4 'We2 g6 S e S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 3
5 0-0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 4
4 ttJc3 g6 S hc6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 5
5 h 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
5 e S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Part 2 The Positional Variation 7 gS a6 8 ttJa3 bS 9 ttJ dS e7 . . . . . . . . . . . 46
10 ttJxe7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 7
10 hf6 hf6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1
1 1 ttJbl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 2
11 c4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 3
Part 3 7 gS a6 8 ttJa3 bS 9 ttJdS e7 10 hf6 xf6 11 c 3 0-0 . . . . . . . . . . . . . . . 6 8
12 ttJc2 gS 13 a4 bxa4 14 xa4 aS 1 5 bS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1
1 5 c4 b8 1 6 a2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
16 b3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Part 4 12 ttJc2 gS 13 a4 bxa4 14 xa4 aS 15 c4 b8 16 b3 @h8 . . . . . . . . . 96
17 ttJce3 g6 18 h4 hh4 19 g3 gs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
20 a2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
20 {4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
2 0 'W'e2 ! ? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 0 1
18 0-0 fS 1 9 -ex.fS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 0 2
1 9 V!ffd3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 0 3
18 'We2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 0 5
Part 5 Alternatives to the Main Line after 9 hf6 gxf6 1 0 ttJdS fS . . . . . . . 1 1 7
1 1 'Wd3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 9
1 1 ttJxbS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
ll h bS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
ll g 3
122
11 -ex.fS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
11 c 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Part 6 11 exfS MS 12 c3 g7 13 l2Jc2 0-0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131

14 l2Jce3 e6 15 d3 fS 16 a4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
16 'WhS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
16 c2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 135
16 0-0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
Part 7 The M ain Line 9 hf6 gxf6 10 ltJdS fS 11 d3 e6 . . . . . . . . . . . . . . . . . . . . . 144
12 c3 g7 13 ltJxbS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
13 hS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
12 'WhS g8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
13 c3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
13 f4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
13 g3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
Part 8 12 0-0 hdS 13 exdS l2Je7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
14 ltJxbS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 3
14 'WhS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
14 c4
166
14 el
167
14 . . .g7 lS bl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
15 c3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
14 c3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
14 . . . g7 1S 'WhS e4 16 c2 0-0 17 ael 'Wc8 18 f3 . . . . . . . . . . . . . . . . . . . . . . . . . 171
18 g4 . . . . . . . . . . . . . . . . . . . . . . . . 171
18 bl . . . . . . . . . . . . . . . . . . . . . 172
18 b3 . . . . . . . . . . . . . . . . . . . . 172
18 hl . . . . . . . . . . . . . . . . . . . . 174
Part 9 6 ltJdbS d6 7 ltJdS ltJxdS 8 exdS ltJb8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
9 c4 e7 10 d3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
10 e2 0-0 11 0-0 a6 12 l2Jc3 f5 13 a 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
13 f3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
13 {4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
Part 10 Unusual Seventh Moves 6 ltJdbS d6 7 l2J a3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
7 e3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
7 a4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
Part 11 Unusual Sixth Moves 6 l2Jxc6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
6 l2Jb3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 2
6 l2Jf3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 8
6 l2Jde2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 9
6 ltJfS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
Part 12 Rare Lines 3 c 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
3 l2Jc3 ltJf6 4 e s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222
Part 1 3 The Novosibirsk Variation 9 ixf6 gxf6 10 ltJ dS g7 . . . . . . . . . . . . . . . . 226
11 d3 l2Je7 12 l2Jxe7 'Wxe7 13 c4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
1 3 c3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
240
ISBN 978-9-5487826-1
9 7 8 9 5 4 8 7 8 2 6 6 1

Chess

Enviado por

Dados do documento

Título original

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Chess

Enviado por

Direitos autorais:

Formatos disponíveis

The Easiest Sicilian

A Black Repertoire with 1 e4 cs 2 tlJf3 tlJc6

Current Theory and Practice Series

Translation and editing by Semko Semkov

Foreword by Atanas Kolev

The Rossolimo Variation 3 bS

The Positional Variation 3 d4 cxd4 4 lt:Jxd4 lt:Jf6 S lt:Jc3 eS

12 lt:Jc 2 gs 13 a4 bxa4 14 ruca4 as lS !c 4 !gb8 16 b3 c>h8 17 lt:Jce3 92

Alternatives to the Main Line after 9 xf6 gxf6 10 lt:JdS fS

9 hf6 gxf6 10 lt:Jd S fS 11 exfS MS

The Main Line 9 xf6 gxf6 10 lt:JdS fS 11 d3 e6

Part 10 Unusual Seventh Moves 6 lt:JdbS d6

Part 11 Unusual Sixth Moves

Part 12 Rare Lines 3 c3 ; 3 lt:Jc3 lt:Jf6 4 eS

Part 13 The Novosibirsk Variation 9 hf6 gxf6 10 lt:Jd S g7

About the Title

11 c3 i.gS 12 tlJc2 0 - 0 13 a4 bxa4 14

The book was already finished, but

!!xa4 as lS i.c4 !!b8 16 b3 <jf h8

We are already in the middle

would be playing for only two re

not a Sveshnikov fan. Then comes

1 e4 c5 2 lll f 3 lll c 6 3 ibS

The most frequent move you are go

restrict White's choice, because he

We propose to unbalance abrupt

1. Black wants to play l!Jd7, f6,

2 . Should White attempt to pre

The idea is to attack the enemy

Superficially, White's _pawn cen

9 .. .f6 10 0-0 d6 11 el 0-0 -0 12

9 gS lLJd7 10 dxc4 ihS ! 11 d3

3. If White advances h is g-pawn,

Black solves the opening prob

Now we are going to exam

11 4Je4 4Jd4 12 4Jxd4 W!xd4 13 4Jg3

A typical position. Black should

Here, instead of 17 . . . hS, Black

I n a n earlier stage of the open

Current practice i s favourable

g4, the move order of the latter vari

b4 ! 14 e3 xb2 with an attack.)

Black is well coordinated and

White is unable to keep control

While preparing to castle, White

Now, the bishop is pinned to cl

1 e4 c5 2 llif3 lll c6 3 .ibS

itiative b y opening the centre with

This modest move poses some

B. 4 xc6 dxc6 5 d3 .ig4!?

A key point in our repertoire.

This looks like a n overoptimis

8 2. 7 ll) bd 2 tl)d7 8 tl) c4

12 a3 hS 13 gS h4+!) 10 . . . V!JaS 11 V!id2

Killing two birds with one shot.

We see here a typical method of

terattack all over the board , trying

hf3 20 gxf3 b4oo) 18 l2Jxa4 V9xa4

Now I had lS .. . a6 ! ? (control

Pushing e4-eS is commonly good

C 2: 5 0-0 c7 6 .ixc6 dxc6 7

7 d 3 i s seldom seen, probably

shop ! ) 9 l2Jc3 l2Je6 10 .ie3 l2Jd4. Now