A horri horribl blee blun blunde derr  now Black Black ns by force. Accordng to Nunn, the position after

Sp nish Fo Knigh ts with 4 . . . b 4

17 �3 is uncar but h ds nt gv any  ins. Ths is cvr as th psn s d nty vry mssy In a crrspndnc gam  ud prfr t pay Whit; n an vr th bard gam Bac 17.:g 18 h1 Jx3 (8.xh 19 h xd 0 xd Jb 1 g is a tuch bttr fr Whit) 19 3 'xh5 0 h 'h 1    xc ab8 3 b4 x3 4 xc7 xd3 5 ad  ith a mss

 .. .xc3

1 9. . .xe4 20 3 'g3+ 21 1 e 22 4 '3+ 23 g1 Je3 24 h4 'x4 2 J2 h3 0-1

9 bxc3

 bv a much bttr n s 8...c 9 c4 d5 (this ida sms s natra) 10 b3 (10 xd5 xd5 11 4 J7 1  is qua) 10.:d 11 h3 h1 xf'xf 13 h5 d4 (13..Jxc3 14 bxc3 'g5  15 'xg5 hxg5 1 f th quaty is as pssb) 14  d  and Bac s qua (anaysis by Krs) Bac ants t pay h7 and  g7 gr a my ..:g5. N frcng pay ds nt r 15 f4 xf4 1 c7 17 d4 1 7.e4! b18 5 g19 xgxg0 h 8 Wth ths Bac accrats hs attac 1 xf4 x5 and Bac is bttr t s intr 1  dxe4 e 1 9 d4 Or 19 d3 gxh 0 b4 b1 f sng that n CO th psin aftr 10 b3 and Bac ns s cnsdrd sghty bttr fr Wht

ame 3 AamsKornoi add 1996 1 e4 e 2 3 c6 3 c3 6 4 b b4  00 0-0 6 d3 d6 7 g e7

Th grat Pau Krs as cnncd that ths mv as th bst ay t quaity. 7 8 d5   xd5 9 xd5 7 10 c4 g11 �a4 s a tt bttr fr Wht

r  hav th psin frm th Pis buy Varian ..xc3 7 bxc3 d 8 g5 7 9 h4 9 ...g6

Or 9..cand n a) 10 a4 8 11 b3 th a sight dg fr Wht Nunn-Tatai ana 199 b) 10 c4 d5 11 Jb3 (11 xd5 d5 1 1 f 13 d h8 as v n ntr rin Via Bastr 001) 11.:d 1 f4 xf4 13 1 dx4 14 .xf4 c5 15 3 h5 1d4 th sm niav. 1 0 xg6

Ths s th natua mv. 10 .c4 f4 11 Or 8 �xf gxf 9 h4 g 10 xg   (11 xf4 xf4 1  Jg4 13 h3 Jh5 hxg11 f4 c 1 h g7 13 �c4 c14 14 b b15 d x 1 h5 gav Bac a th chancs n ansi-Lasr  xf4 15 xf4 d4 th quaity. Paris 1909) 11h 12 xf4 g4 13 3 xf4 14  g5 15 h3 1x17 f3 5 and hr Bac s bttr vn thugh th payrs agrd a a n arcy Psbury ndn 1899  h4

1 0.. xg6 1 1 c4+  1 2 4

With th t bshps and cntr vr th pan stuctr Whit has a sma dg 1 2. . .'e

1..h13 xfxf(13... gxf14 f5 8 15 'd gvs Wht sm chancs f stab  ishng an attac n th ingsd) 14 fx5 x5 15 'd and Whit ps a sight dg 11

The Fo Knights

due t his suei aw stuctue

g8 2   with a uclea si,   2 e6+ with a daw ) 17 gf6? 18 h6 �e7 19 f8 f8 20 ' ad Wie is ette 1  ix6 gx6 1 9 .7 e7 20 xg6 e6

20J g8 21   22  ves White seus wiig chaces. 21 b4

fte 21 a3?! g7 22 fS 23 efS g8 2 g3 hS yu wuld wde what Whites ieces thik they ae dng  the a le while Black is atackg his kig 2 1 . . .g7 22 .5

22hS? h3 23 g3 g ad Black ns 1 3 xe5

22 .. .g

fe 13 d2 S 1 3 a 15 eS eS 16 a a 17 a d7 18 d7 'd7 Black has gd cutelay.

22.fS? 23 efS g8 2 e! wuld ve Wite a clea edge. Nw it ecmes clea why he quee elged  the futh ak

1 3. . .dxe5 1 4 b1 !

White  was t kee a eye  the Ssquae ad get is quee i lay  the queeside. 1 ' d2 S 15 3 a wuld equaise 1 4. . .b6 1 5 b3 .d7 1 6 a4 h6 1 7 x6

Bad wud e 17e3? g 18d2 e7 f eamle 19 h3 'c5 20 h + 21 h2 h3 22 gh3? + 23  + ad Black ns

23 g3 c5

23 2 efS g 25  e3+ 26 f2 cS 27 c ad Black is a aw dw f  vey itte. 24 b1 .x5 25 ex5 h5! 26 1 g4 27 Wg2 g5 2 3 x3+ 29 Wx3 x5+ 30 Wg2

Wite might e sightly ette i tis ed ng u Kchi is quick  eutase the diffeece 30.. .c4 31 a5 e4! 32 axb6 axb6 33 dxe4 a5 34 b1 a2 3 5 xb6 xc2+ 36 Wh3 Wg7 37 c6 xc3 3 Wh4 c2 39 h3 c3 40 Wxh5  

Game 4 SpeelmanKarpov nae . 1 e4 e5 2 l3 lc6 3 lc3 l6 4 ib5 ib4 5 00 0-0 6 d3 d6 7 ig5 ixc3  bxc3 h6 1 7. . .x6

O a) 17gS 18  (18 f1 \e7 is als uclea) 18. 19  e6 20 f+ 'f8 21 e6 c5 22 h 1 \c3 23 h3+ 1

8..d7 9 a6 10a 8 113 h6 12h asses t 9.d7 elw 9 h4 e7

9...d7 10  a6 11 a 8 12 3 e7 13 e aS 1 d d8 15 h3 h7

Spnish Fo Knights with       

 gave an uncear pn n Shr-Speean, Lndn 1991 9g4!? 10 h3 Jx3 11 'xf3 g5 12 g3 h7 12.. d7 13 d4 f6 14 ' g4 h8 15 h4  gave he a cear edge n Capabanca aker, Peerburg 1914) 13 ab anw Tarach, Venna 1898) Nw 13 e7!?  pbab be n he game 13. c8 14 fd g7 15 d4 f6 wa bad fr Back) Flng 14 d4 g6 15 fd! 'f6! 16 '5 ' g7 we have an uncear pn.   pran ha aer 1 Jc4 b6 18 xe5 xe5 19 jxe5 xe5, 20 d7 fall hr  20f6 1 0 e 1 d

Or: a) 1b8!?  mar  the Breer Vaa n f he Ru pe 11 d4 c6 12 Jd3 e8 13 d2 13 h3 bd7 14 ' d2 f8 15 g4 g6 16 �g3 h7! 17 g2 'f6 18 e3 g5 19 xg5 hxg5 20 h f4 21  �e6 and Back  ghtl beer, traecu Tmhenk, Buchare 1993) 13.bd7 14 c4  15 f4 g6 16 e5 dxe5 17 �g3 wa a me n Vhn-Tkh, Vggrad 1994 b) nereg  10 g5!? 11 Jg3 11 xg5!?  uncear afer 11hxg5 12 Jxg5 d8!, bu n 12g7? 13 e3! fwed b g3, gvng \he a trng aack) 11..�g4 12 h3 �h5 13 d4 ad8 14 b b8 15 �d3 b6 6 h2 bd7 7 g4  g6 8 d2 fe8 and he pn  equal, AnkaGr, Val ev 2000

11 d4 e6?

Th na a ng ne f acc where he emere wh he advanage Beer wa 1.h7 12 'c1 �g4 13 'e3 and he  nl a tle beter 1 2 dxe

fer 12 g3 g5 3 xg5 hxg5 14 b a6 15 d3 c5 Back  equal 1 2. . .dxe 1 3 xe

k  3 g3 d8! 14 ' c1! h5! 14.. c5 15 2 a6 16 jf f4 1 c4 gve Back me prbem wth e5) 15 jxe5 15 xe5?? xg3 16 hxg3 c5!) 15 .. g5 16 �d4 xf+ 7 gx c5 18 e3 f6 and Back ha g aacng chance n he ngde 1 3.  c 1 4 x6 xb

Karpv ha awa been knwn fr h  grea accuac n defence Afer 14 .. gxf6! 15 g4 ' xb5 16 e5! fxe5 17 xe5 c4 18 xh6+ h7 19 h5
15.. f4 6 d2 'h5! 16.h5? 7 �xg7! Jxg4 18 xf8 xf8 19 h3 jc8 20 xh6 and he n) 17 xf4 xg4 18 'xg4 xg4 19 e3! gve he a lgh advanage.

1 6 e3?

Here he had he much mre accurae 16 e5!! hxg4 16 ... f4 17 h6+ gxh6 18 e4 a gve he a trng aack becaue afer 18.. g4 19 d2 he wn) 17 xg4 d8 13

The Fo Knights

18 3 f8 19 jxd8xd8 0 4 wth a  vr strong attak.

 39 d 4 40 d4 e6 41 d  

1 6.  gx6 1 7 d Jd?

Game 5 ulaevKolmov sa 1994

A msaluaon. Aftr 17..g7! 18 "f3 6 19 xf6 h7 20 f h6 Blak has ralis hans of holding. 1  x xd

18...g7? 19 " xdS 20 xdS f8 21 7 " 22 a1 givs it a nning attak. 1 9 exd g7 20 6  21 "x6

 aurat mov. Aftr 1 3? 2  a1 "x2 3 g3 jg6 24 h3 h7 2S g3 jg6 Whit ight not rak through vn though h is ttr. 2 .. .b6 22 "4

Whit is not gong to xhang quns  w h is attakng.

1 e4 e 2 3 c6 3 c3 6 4 b b4  0-0 00 6 d3 d6 7 g xc3  bxc3 "e7 9 e1

9 x6 x6 10 1 h6 11 d2 is qual 9 . .d

is is th orrt plan. 9..h6 10 h4  wold transpos to th prvious gam 1 0 d4 e6

1Q..g4 lads to a tpial posion whr Whit has a slight advantag aftr 11 h3 x 1 xf3 6 13xf6 gxf6 14 ad1.

22 . . .xc2 23e7 d6 24 c4 b

1 1 c1 2 cxb?

is is a wast of m. Whit is attaing and should hav onnud to do so aftr S a1 A possil in would  2S..4 6 "g g6 (6..aS? 27 + x8 28 x+ h7 9 4 g6 30 h8 and Whit wins) 7 h4 a 8 g4! and Bak annot suriv th prssur. 2 ...xd 26 a4

Not 26 x7?? S.

A frsh ida is 11 d2? 6 1..S 12 j4 is slightl ttr for Whit) 12 j 7 13 h4 xd4? (ttr is 13..8 14 fS g and Bak shoud  oka) 14 xd4 xd4 IS f4! 6 16 I g4 17 d3 " 18 3  19 xd4 xh4 0 x7 whh lad to a lar advantag for Whit n minmagin t Ptrsurg 1994. 11...d

Aftr 1..6 1f 7 13 h4 8 14 26.. .c6 27 bxc6 xc6 2 3 .g6 29 fS g6 IS g3 f4 Blak is n. Je c 30 ae1 a6 31 a e6 32 6 1..S 12 a4d8 13 4 h6 14 h3 8  33 xc6? xc6 34 d1 e6 3 4  wth th ida ..76 is unlar aording to �g7 36  7 37 d7 �6 3 e3 hirov. 14

Sp nish Fo Knigh ts with 4 .   i b 4

1 2 i1

1    e

r 12 a4 f 3 h4 c6 14 d3 (14  ?) 14..g4! an provement over 14.h6 15  e 16 f5 xf5 17 "xf5, when White was a ile beter  Adams Akopian Groningen 1993) 15 e2 e6 nd Black is ne 12 h4 d5 13 f5 "f is also okay for Black.

15g6 16 g5 xh4 17 xh4 e7 I f4 e 19 f5 gves hite a stong bass for an initiave on the kngside.

1 2c

12c6! with equlity s probably beter.

1 6 ig 6d7 1 7 'h b6

17h6 18 xh6 gxh6 19 g g6 20 xg6 g6 21 xg6 f 22 e3 <e7 23 e2 and Black  have to tend to an m meate and very fcu defence 1  a4

1 3 g3 'c7?!

e pophylacc move 13.f8 s beer as White s theatening d4-d5 and h4 with a strong poson. After he moves 14 h4 g4 15  d7 he poson is not com pletely cler. 1 4 d  1  4

1   id7

r .a5 19 b5 d7 20  xb5 21 f6! f6 21..xa4 22 h6 e6 23 xg7 and Whte wns) 22 h6 e6 23 axb5 nd We is pobably ust wnnng Whte does no acheve anytng from 15 g5? 8d7 16 b5  17 h4 h6 18 xd7 (I d2 c4 and Whte is forced to play xd7 or he  lose he bshop) I.hxg5 19 xc h4 20 h3.

1 9 a c 20  ix 21 ex e7 22 ib ec 23 'g4 xd 24 ad1 xc3 2 i6 g6

After 25g6 26 g5 xb5 27 h6 e6 2 e4! bd4 29 xh7+ <xh7 30 h4+ g 31 h Black s maed 26 xg6 gx6 27 gx7+ � 2 'e6 1 -0

15

he Fo Knights a

The pnish Fou Kngts wth b4 ves Wite soe cnces of ceng pessue  the n ine 6 xc6 does not pose lc n pobles so onl 6 d3 nd ten  g5 s coect bt then I believe tt e! s the splest ethod of solvng lcs pobles Howeve lc should lso be ne  the e wth xc3 8e nd 9d8 wen the ge cn be descbed s poson coplex 1 e4 e5 2 l3 lc6 3 lc3 l6 4 b5 b4 5 0-0 0-0 ) 6 d3 6 xc6  Game 1 6 . d6 7 ig5

 e2 ) Game 2  ixc3

..e Game 3  bxc3 e7  Game 5 8 6 Game 

5   00

16

7e

CHAR TWO Spansh Four Knghts wth 4d4 1 e4 e 2 3 c6 3 c3 6 4 b d4

Th  a lag dea b the great Akba Rube, the mot famou Poih plaer i che hor Actual he tarted wih 4..c5 5 xe5 d4, but later reed h to 4..d4 irectl. Toda   codered he m move afer 4 b5 ad clearl t  the mot damc road to equa. Iead of eekg h luck  a mmecal poo Black create acve couerpla, omeme at he cot of a paw.

6 .xb 7 xb d6  3 xe4+

Th  a ver old move, but alo a ver  good oe. New but ot o good  8..c6? 9 c3 xe4 10 0-0 xc3 11 c3 �d8 12 :e1.e73 "e2 g4 14 h3 x 15 x 0-0 wth equat horKhaifma Mocow 2001. 9 2?

9 e2 �xe 10 xe2  ecear but Black  l preferable.

am 6 Spelmann-Ruinsen BadBad 1925 1 e4 e 2 3 c6 3 c3 6 4 b d4  xe

Th  ot dagerou for Black, but mght be for Whe! Alo harmle  5 0-0 xb5 6 b5 c6 7 c3 d6 8 d4 c7 9 h3 e7 10 e3 0-0 wth equat  Kere  ...e7!

Clearl he oget move. 6 4

6 f3 xb5 7 xb5 xe4 8 �e2 xe 9 xe2 d5 10 :e f6 wth equa  wa alo poble The move c2-c4  al  wa be awered wih ..a7-a6 of coure

9  g4+!

The rect atack  er dagerou. Le trog  9.."c4? 10 e1 e6 11 a4 d7 12 d4 g4 13 e3 c6 14 b3 xf3 15 :xf3 d5 16 c3 �f5 17 h3 h5 18 d5 c5 19 d3 xd3 20 :xd3 h4 21 e2 :e8 22 c4 wh equat YeoHebde, Bh eage 2001. 17

he Fo nights

10 �g3

for the pawn deci

After 10
Worse is 10d8? 11 h3 h6 12 dfS 13 a3 e7 1 e c6 IS dS a6 16 c  wth an unclear game, Keitghaus-Meissner, Germany 2001. 1 1 e2+

The great Paul eres analysed he follow ing ines and, even today with the assistance of strong computer programs and SO ears of 1 6 b4 r 16
te is also worse after 12 h3 e3 13 h2 c2 1 b1 and now: a) Wining the echange is sucidal: 1d IS d b1 16 �c! c6 (or the entertang 16cS 17  e 18 c7+ e8 19 d3 e7 20 d6 d6 21 e1 e7 22 e7+e7 23 c6 f6 2 b8! and Black loses material) 17 7 cbS 18 e1d7 19 e c 7 20 a8 c1 21 e6 c6 22 f8 and th gh is over b) 1.c6! (the rght appoach) IS c3 e7 and Black has a pawn more and a better posion 1 2. . .d7 13 tbd4 te3+ 14 �2 txc2 1  txc2 xc2

White does not have rea compensaon 18

c3 Ja! 2 t3 gxh2 29 6+ �c 30 c3 c 31 d3 h 32 te

32...xg2

A power blow hat highghts the weak-

Spn ish Fo  nigh ts with 4   d4

ne ite i uffeng from on the ght quare. But not 32..dxe5 33 xe5 and after e+ it i te who n. 33 �xg2 dxeS 34 xeS g4+ 3S g3

Or 35 xh2 f4 36 g �xf6 and Black n 3S  xg3+ 36 �xg3 d6 37 �xh2 xa2 3 �g1 xd2 39 hS b6 40 eS cS+ 41 �f1 �b7 42 g3 bS+ 43 �e1 e2+  �d1 g2 4S nc3 e2+ 0-1

ame Nikolenko-Aleksanrov Smoes 2000 The Ruian che choo teache u that drawn poion do not ext, only equal poi ton do. Thi game i good illutraon of that pont of view. 1 e4 eS 2 f3 c6 3 c3 f6 4 bS d4  xd4

hipmanWeber New York 1985) 12 e5 f6 13 d3 g6 14 e2 and \hite i much better, Mev-Fudeder Amterdam 1956. Back never really got round to nihng h development and mot kely never   dxc3 .cS

The poion i ao a dead draw (orry, dead equal) after 8...e5+ 9 e2 (or 9 e2 c5 10 0-0 0-0 11 d3 d5 12  d6 13  g3 c6 14 d2 e6 15 5 �h3 16 xh7 h8 17 g6+ g8 18 h7 and the player agreed a draw n WittmanGreenfeld, The aloniki 1984) 9 ...xe+ 10 xe2 d5 11 f4 c6 12 c4 e6 13 cxd5 xd5 14 0-0 c5 15 fe1 00 16 d3 fe8 Alekine Capabanca, t eterburg 1914. After 8...c6 9 d3 e5+ 10 e3 d5 11 �f3 c5 12 0-0 xe3 13 fe1 0-0 14 xe3 d6 15 ae1 \hite i ighy better, ma naiev-Thipay, Frunze 1985. 9 00 00 10 .xd7

5 a4 i conidered in Game 912 whle 5 c4 i deat th in Game 8. S . .exd4 6 eS

Or 6d5xd5 7 exd5 �f6 8 00 je7 9 f4 (very riky 9 e2 00 10 c4 i equal) 9...00 10 3 c5 11 b3 d6 12 b2 jf5 13 ae1 d8 (thi move i worth remember  n 14 d3 a 15 e2 ae8 16 g3 (16  and hite hold, although Black re mn much better) 16..xd3 17 cxd3 (17 xe8 x 18 xf+ xf8 19 xf1 jxd2  give Black an ealy  g endgame) 17..e2 18 �xe2 f5 and Black i cloe to nnng, at leat in praccal play, Wolf Alekhne, Carbad 1923. 6  dxc3 7 exf6 xf6

Black  be punhed hard for h greed after 7..cxd 8 xd2 xf6 9 0-0 e7 10 c3  g5 11 e1 0-0 (it eem tempting to take a piece too but after 11...�xb5 12 g4 g8 13 xe7 ;xe7 14 e+ d8 15 h4 f6 16 xf6+e8 17 e17 [17..f8 18 xg7 xg7 19 e7 ;g6 20 xh7 18 e7 g6 19 e mate come o oon,

f \hite want a draw 10 xd7 i the mpet method wrote GM John Nunn about th oion . However, a we hall ee, there i l the little tak of obtaining thi daw. 1 0. . ..xd7?

10...d8 11  h5 xd7 1 2 xc5 c6 i dead equal once agin. 1 1 xd7 ae 1 2 e3

Or 12 jd2 e2 with an nave . 1 2.  xe3 1 3 fxe3 b6 1 4 ae 1 xe3

The poion i alo balanced after 19

he o nights

4 xb2 5 xc 5 x?! is risky be cause you have o calculae he pawn endng: 5..xc 6 xfxf  �e h  fxf+ 9 xf �f 20 xf6 gxf6 and Black looks beer here, bu probably Whie can keep he balance 5�xa2 6 �xb 1 xe3 xe3+ 16 �h1 e2 17 d3 e

f 36 c7 a 37 d4 xg2+ 3 �h1 a 1 + 01

ame  Ivanuk-Sviler ares 1999 The nnoaons o hs gme are based on Vassy vanchuks annoaions  h Ir mnt 5

1 e4 e 2 lf3 lc6 3 lc3 lf6 4 b d4  c4!?

As 5 a4 does no seem o prose Whe any advanage, his has also been ried

 hard o beliee ha an M wih an Elo rang of 2500 can lose a posion like his. Bu Black does have a hree agns o ma oriy on he kngside, and woud  any pawn endgame n a rook endgame, hough, he majoriy canno realy play a par. 1 �g1 h6 19 h3 e6 20 b3 e3+ 21  2?!

  c6

Back obviously has some oher possb iples was 2 xe xe 22 c4 e2 es: (22..f 2 d e 24 f2 is equa oo a 5c5 and now 2d xc2 24 dh 25 dxa2 26 a 6 0-0 d6!  xd4 xd4  d c6 9f xc and he game be drawn. b5 0 b a  a4 b4 2 d e6 wih 21 . .e7 22 a4 d 23 f3 c6 24 b4? equay accoring o Guman fer hs move he whie pawns on he a2 6 xe5 ih a rher branch: queenside are al weak. a2 6. e   d? d5  xd5 24 .. .d6 2 h xd5 9 xd5 �xe4+ 0 e d6  0-0 25 �f4 ih h2 o fllow looks much e6 2 e 0-0-0  c f5 gave Black he sronger Black s si beer, hough. upper hand in Nimzowisch-lekhne,  2 e6 Peersburg 94 d5  xd5 �xe Black is sarng an aack agans he whe ...xd5 9 xd5 c6 0 xd4 cxd5  b  weaknesses on he queensde and he open xe4 2 xc5 xc5  d4 exd 4 xd  whie king 00 was raher level n Janovsky-Mrsha, 26 1 e3 27 a? a6 2  g6! 29 Pars 905 9 e g4 0 e2 xe2  4 xc3 30 f2 h 31 d2 b 32 axb xe2 0-0-0 2 d e6  0-0 d5 and axb 33 �h2 c4 34 d4 xc2 3 e4 Black has quie srong compensaon  

Spnish Fo nights with    

(hen a human play uch a move we tak Kere. a22) 6d5? kachev paten) 7xd5 (7 about npiraon and tye, bu if a computer exd5 0-0 8 0-0 d6 9 e e8 10 d3 played it we would talk about lackng under g4 would ve Back an attack; 7 xd5 tanding or aume that it wa al calculated xd5 8 c3 i unclear accoring to Krank e bet moe i of coure he central puh and Domaov the main idea i to avoid 8 8..d5 9 exd5 xc3+ 10 bxc3 cxd5 11 b5+ xd5 g5 9 x e7 10 h5 xe5 d7 12 xd7 xd7 13 g5 "c6 14 0-0 and Black i a piece up) 7 ..xd5 8xd5 0-0 d7 th equaiy) 9 h3 (Not urpngl 9 c3 (9 00 e8 10 d3 d6 wh compen Leko decne thi intaon to a dance e aon  Kramk, Doatov) and now computer probably wa hopng for ome thing ke 9 g5 d5 10 exd5 g4 11 "g3 b6) 9.d5 10 exd5 xc3+ 11 bc3 cd5 12 b3  c7 13 0-0 00 he pawn on c3 ap pear to be poioned even for the computer: 3. xc3 4 a3 e6 5 g3 g8 6 fe1d7 17 ab1 f6 18 b2 "c6 19 d4 e4 20 g6+ and hite attack ha oy jut taed) 14 g3 n t poion it i cear hat the incluion of h7-h5 and h2-h3 ha been of benet to hite, eko-Deep unior (C) Dorund 2000. b) 5...xf3 i underemated e re ulng poion might be bong and oid, a221) e move 9...f5 and he folowng bu it alo appear to be equal wich hould analyi belong to Lev Guan In my opin appeal o ome people: 6 xf3 d6 7 d3 (7 on thi move i a miake but a very nter d5 c6 8 x+ xf6 9 xf6 gxf6 10 d3 eing mtake 10 cxd4 xd4 11 (11 0-0 g8 and Black i more than okay) 7.e7 8 e4 12 b6 xe5 13 xa8 h4 14 g3 0-0 00 9 e2 4 10  e6 11 b3 c6 12 "h3 15 d4 g4 and Black win my i f4 exf4 13xf4 d5 14 exd5 cxd5 15 d4 a6 16 provement woud be 11 c4 xf 12 h3 b5 17 a3 e8 and the player agreed a �xf2 e+ 13 �g 'xd5 14 e3 and the raw in J onkmanRomanhin Lvov 2001 compenaon for he pece i not enough) 6 xe 1xf 12 �x e4 13 f4 exf3 14  i he iht approach f Black ha b3+ h8 15 d4 g2 16 �xg2 g5 17 e6 freedom to o wha he ike, h  equaie xe6 18 'xe6 "xd4 19e "+ 20 h eaily After 6 d3 Black can chooe beteen the folong  ith pepetual check a222) 9...e8 10 cxd4 xd4 11 0-0 (11 a) 6.a5 7 d2 b4 8 b3 xb3 9  "xd5 12 d3 g4 13 e3 x 14 gx axb3 c7 10 0-0 e7 11 h3 0-0 12e2 d5 d6 15 xd4 xd4 16 e2 ad8 17 000 13g3 dxe4 14 e4e8 w equaity e6 18 e3 and Whie i a little beer ac b) 6..b5 7 b3 xb3 8 axb3 d6 9 0-0 coring o uan) 11..xe5 12 d3 c6 13 e7 10 h3 0-0 11 e2 a 12 g3 e6 13 f4 (hte hould be better here) 13...b6 14 h2 d5 and Back i cloer to beng beter c2 c5 15 e2 a6 16e3 "d6 17 ac1 than to beng wore xc1 18 xc1 e5 and a draw wa agreed, 6 ..d 7 exd d6  3 x3+ 9 x3 00 10 h3 b ShirovKramk Cazola 1998 a3 6 d3 c6 7 xd4 xd4 8 'f3 h5 Black ted 1Oe8 1 e2 cxd5 1 d4 

he Fo nghts

b6 13 00 �b 14 �b5 d n the game Podesnk. Pogar Lubana 1994 Whte shoud w after 15 �g5! d6 16 fe h2 17 f d6 1 e5 1 1 e2 b4 1 2 e4

Wte shoud not fa for 12 dxc6? bxc3 13 c7? (13 xc3 s better) 13..xc7 14 xa �a6 15 'xf xf 6 xa6 'e 17 e2 cxb2 1 xb2 'xb2 19 00 'xc2 and ns 1 2xe4 1 3 'xe4 cxd?!

Back shoud pay th a hs peces: 13..e! 14  a6 aso possbe s 14.cxd5!? 15 00 e7 16 "e3 �xh3 7 gxh3 h4 1 3 xe2 19 xe2 xh3 20  c 21 'g3 22 f h 23 g! th a dra) 15 d3 cxdS 16 00 6 xdS?! c ooks too dangerous) 16.'e7 17 d e5 1 g3 �c? and Back has good com pensaon 1 4 '3 'c7

Btter as 4..e? 15 d4 �a6 16 e3 xe2 17 xe2 c 1 00 �b and Back s has some counterpay 1  00 d4?

s s smpy a tacca error I have no ced durng my ork th s book ha despte s great understandg of attack and ches n genera, ver makes some strange tacca sakes for s ee from me to e 15�b7 16 d3 d4 17 g4 ae gves compensaon for the pan %0 kos f t s enough? 1 6 'xa b7 1 7 'xa7  c



1  'xb7?

s s sucent bu hy not 1 a4 d3 19  g3 20 xb7 xe2 21 d4! xd4 22 �d2 exf' + 23 xf1 and a you have to do s count the pans.. 1  . 'xb7 1 9  d3 b 20 b3 4 2 1 b2  d6 2 2 g 3 'g 2 3 h2 h 2 4 4 'd 2 h4 Je 26 a4!

Whes pan s very prve Frst he cre ates some pased pans, then he pushes them to he eghh rank and promoes them o queens. 26bxa3 27 xa3 xa3 2 xa3 f 29 b4 7 30 b5 shos hat ths s a reasc pan 26 c7 27 a

Probaby t as beter to exchange rooks th 27 ae xe 2 xe1 g6 29 e4 hen Whe domnates. 27 .. .xa 2 xd4 xd4 29 xa c 30 e 

Whte shoud not fa for 30 xh5? xc2 3 �h7+ f 32 xc2 xd 33 h3 xc2 34 h e7 35 e1 d7 36 hS xb3 hen Back has been success n consg he matters. 30 'd6 31 e1 g6 32 

Cearcu s 321e2 g7 33 c4 c5 34 e7 c7 3Sxc7 xc7 36 c3 bxc3 37 xc3 32 . .. gx 33 x Jc 34 e+ g7 3 d3 d 36 e3 d4 37 h3 'd7+ 3 g2 d+ 39 2 c 40 1 e2 d 41 g2 d6 42 h2 c6 43 e4 'c  e?!

Sp nish Fo nigh ts with 4 . . d4

Hee White me a  wth 44 c4 f6 4 e d4 46 g2 a 47 g f 4 xh f1+ 49 h2  x 'xf 1 h3 xd2 2f xc2 3 x7+

dan 11 b3 e6 12 g3 h6 13 xf6 'xf6 4 'xf6 f6 and Black had equalied in NiboeHuzman Leeuwaden 199

   d 4 e3 d6 46 2 6 47 �g2 e 4 4 c6+ 49 �g1 

nteeng i 7 - d6  h3 e7 9 e2  1 c3 b 11 c2 c (11d 12 d3 e 13 g3 might be a lle bete fo White 12 d4 cxd4 13 cxd4 b7 14 dxe dxe 1 g3 c 16 b3 which wa played in taecu Bekovich el viv 1994 White eem o be a ittle bete placed but not much

Whte could  etan e maoiy of hi advantage with 49 e! "d6  g4 hxg4 1 xg4+  2 ff4 c 3 g3 b6 4 d4   gf4 when Black  unde tong peue 49  e3 0 dxe3 c3  �2 d2+ 2 .e2 xc2

Now White ha ome weaknee too and the poion i moe o le impoible to  3 xb4 + 4 4 c2  b4 �g6 6 b �g7 7 c4 +  J4 c2 9 g4 xg4 60 xg4+ �h6 61 4 �g6 62 h+ �g7 63 g4+ �7 64 c4 + 6 4 e 66 .d3+ �6 67 e2 �7 6 Jd4 + 69 J4  

ame 9 Kamsky-Aams Domund 1992

7 d3

7 ... d6  00

 b3 e7 9 e2   f4 g4 11 f2 e6 12 - exf4 13 xf4 d 14 ae1 dxe4 1 xe4 xb3 16 axb3 'd7 17 xf xf6 wa level in Anand-vanchuk Dot mund 1992  ...e7

fte b 9 b3 e7 1 h3 - 11 e2 d7 12 c3 c 13 c2 f 14 exf xf Black poion i pefect he playe ageed a daw in cKholmov adubice 21 bu nmay Black would play on White can impove on th of coue but an advantage pobably be had to nd 9 d d 7

1 e4 e 2 3 c6 3 c3 6 4 b d4  a4 x3+

 ne i of coue moe peacel han c and .c6 Black doe not wih to give up mateial imply take a defenive poiion and enoy the fact that Whte can no play d2-d4 6 x3 c6

ome black playe have alo expei mented with 6b4 fo example: 7 e2 afte 7  -  d3 e7 9 3 c6 1 b3 d6 11 e2 e6 12 g3 d7 it i doubtl if White ha any advantage at   7..  c3 a 9 d3 d 1 g (o 1 h3 dxe4 11 dxe4 e 12 g3 d6 13 c2 b6 14 a4 a 1 b3 c with equaity Niboe-Ciene aada Holand 199 1.c6 (1'd6 11 xf6 dxe4 12 dxe4 xf6 13 xf6 gxf6 14 g3 ve White an edge accoding to Yu

Black would loe afe 9xd? 1 exd b 11 dxc6! e4 (o 11bxa4 12 c7 12 xe4 d 13 c7 d7 14 d4 bxa4 1 xg7 f 16 f4 he lne 9..h6  xe7 xe7 11 c3 - 12 e d7 13 c2 fe 14 h3 ad 1 e3 c 16 d4 c6 17 dxc dxc 1 ad1 3

he Fo nights

lead to a daw in okopchk-Magomedov, Mocow 2002. Back cold alo  9..0-0 10 e7+ e7 11 g5 h6 12 d2 b5! 13 b3 a with a ne poion 12 f6 with a mall advantage i bette 1 0 xe7 xe7 1 1 c3 0-0 1 2 g3

Wite i a little bit bette hee Te two bio do cont for ometig

3 c3 0 4 c! 1 xd6 Jec7

51"h3 wold end wih a pepetal check after 52  ec7 53 d 7 54 f c1 55 + g 56 f 7 57 g 2 2 c2 3 b6 h3 4 c6 2xc6  dxc6 xc6 6 d2  - 

ame 10 MoylevGrisuk Moow 2001

1 2. . Ac 1 3 c2 e6

Black wold gt into roble after 13.f5? 14 g5 "c7 15 f4! fo eample: 15 e6 16 b3 h8 17 e5 5 I "g5 e5 19 ef5 and Wte ha won a pawn o 15.b6 16 f2 e4 7 e4 e6 I f5 g5 19 g5 wth an attack on the kingide 1 4 e3 c 1  ad1 b6 1 6 b3 b7 1 7 4 ex4 1  x4 tx4 1 9 x4 c

1 e4 e 2 t3 tc6 3 tc3 t6 4 b d4  .a4 c6

An inteeting pawn acice rt fond  te eary 1990 by Jeoen iket o fa the concion i tat Black obtai good con teplay for he pawn 6 xe

Th i te only way to ght for an advan tage l Whte hold emembe the phrae live by the wod e by te word. Alterna vely a) 6 0-0 'a5 7 e  a3 e7  b4 "c7 9 b2 0-0 10 d4 ed4 11 e2 e4 12 d4 d5 i eqal) 7.d6 and now a)  d4 ed4 9 d5 d7! 9.a4? 10 c7+ d 11 b3 a5 12 a b6 13 b2 b7 14 d4 a8 15 e5 e5 16 e5 wold gve White a very tong attack n the eavy piece ending the atack on for the piece) 10 b4 'd 1O.'a4 11c7+ he kng i not a impotant and Black a d 12 a8 b5 i a me) 11 f4 e5 moe chance of eaching a aw  with eqality 20 d b 21 d1 e6 22 h4 be a2)  h3 e7 9 a3 00 10 d3 o 0 b4 23 c4 xd 24 cxd
Spnish Fo nights with    

d8 19 d4 exd4 20 xd4 b 7 21 xf

'xf6 22 a3 and ite is better, ZAasi Rabiega, Bundeslga 2001  b) 6 a3? (this knd of move is usualy a waste of time and here is no excepon) 6 cS 7 0-0 0-0 8 d3 dS (as White does not want to take over he nave, Black feels himself compeed to do so) 9 h3 e 8 10 gS bS 11 b3 xb3 12 cxb3 d4 13 e2 f8 14 b4 a 1 bxaS xaS and Black was beer in RyzhovZubov, iev 2002 c 6 d3 bS 7 b3 xb3 8 axb3 d6 9 0-0 e7 10 e2 0-0 11 g3 and the posion is more or less level. ne wonders why hite chose to ve up the bishop  .. .d 7 f3

Completely hamless is 7 d3 bS 8 b3 a 9 a3 dS 10 eS e4 11 00 cS 12 xcS xcS 13 h1 00 (Adams-Kramn, ares 1999) After 14 d3 e8 1 f4 xb3 16 cxb3 d4 ts posion is eual accordng to Va ramni his is of course e, but to bishops are stronger than one, so I pre fer Black. 7 .. .ig4 a d3

Interesng is 8 h3 xf3 9 f3 g6 (Black plays on the dark suares the most impor tant ng is to contol eS and f4) 10 f1 g7 1 1 e2 xf3 1 2 d4 a 1 3 c3 gS 14 eS fe4 1 xgS xgS 16 g4 e6 with unclear play, Ponomariov-Gomez steban, Pamplona 1996. a    dS

In Acs-arakehajov, Athens 2001 Black had no success with 8.d7?, which does look a bit passive here foowed 9 e3 xf3 10 gx hS 1 d4 f6 12 g1 x? (better was 12 ..bS 13 gS x 14 x x 1 g3 hS 16 b3; Black has regained his pawn, but he is underdeveloped and ite has started an iniave) 13 d3 'h4? (ths gluttony wi be puished se verely 13 . h6 with .. e7 and . 00 was much better) 14 g3 xh2 1 d2 hS 16 dS S 1 7 ag1 (hite is aead nng ust try to nd a savng move for Black) 17 f6 18 xd7 �xd7 19  d8 20 'xb7 c8 21 xg7 e8 22 xa7 e7 23 b d7 24 bS 1 0. 9 00

r a) 9 h3 xf3 10  hS 1 1 exdS xdS 12 e4 e7 13 f1 0-0-0 14 c3 'c7 1 g2 e7 1 6 e1 gS and Black had a strong iniave  hoov-alaniu, t Peters burg 2001 b) 9 e3 xf3  gx hS 11 d4 e4 12 dxe4 xf3 13 xf3 'xd4 14 d1 e 1  0-0 d6 would be ne for Black 9  . dxe4 1 0 xe4

After 10 dxe4 xf 11 gx bS 12 b3 c 13 eS d7 Black has compensaon for the material according to zech G ave Blatny 1 0.  xe4

10...x? is one of my own ideas, un veried by practice to tis day. he plan is to give up the bishop  order to obtan some niatve It seems to give Black a good game, for example 1 1 gxf3 'd7 1 2 c3 e6 1 3 e1 (there  be a strong attack on he white kng, should his soldiers be so careless as to open he g-) 13..0-0-0 14 e3 dS and who knows  beeve in Black, but pracica games have yet to be seen 11 e1 ie7

After 11fS? 12 c3 xf3 13 gx hS 14 x4 xd1   xd1 Black has no real compensaon fo the pawn. However, after 5

The Fo nights

1.x3 12 gx 6 13 xe e7 14 e3 O-O-O Black does have real compesa tio or the paw But ot ith 4.x3 15 h g5 16 xg5 xg5 17 "e2 dS  S e1, ith a sog iave o te 1 2 xe4 xf3+ 1 3 gx3  1 4 e

1 9 �2

19 h4? with some aive ideas ke h5-h6 is bad: 19  d7 20 4 x4 21 x4  ad the ook is horribe o 4. 1 9   20 e6 xh2?

Ate 20d7 21 g5 aeS 22 e xe6 23 'xe xe6 24 xe6 c7 25 e3 Whte has a slghly bette edgame 21 e7 d6 22 Ja4

Bgg he worst placed piece ito ac tio Ater 22 xb7? eS Black has acve couterplay 22 . . 6

22 23 e6 c7 24 4 ad Whte wis 23 g g6 24 e+ �7 2 e7+ xe7 26 xe7+ �g

Ater 14  e1?! 0-0 the queso is how much bete is Black?

Wte also has tuly excelet wg chaces ater 26S 27 xh2 b5 2S a6 xg5 29 e5 eS 30 xe xeS 31 xc6 4 32 c7

1 4.  .e6

27 �x2 Jxg?

Black has two ateaves a) 14. d6? is met ith 15 "e2 "g+ 16 h 0-0-0 17 x7 h3 I S g5 (a mpor tat resource) S. �x5 19 g 6 20 b3 d7 21 e5! ad White has a clear edge due to hs bette pieces. b 14 d7! 1 5 "e2 S 16 d2 (16 4? 6 ad Whte loses the exchage) 16d6 17 e3 xh? (Whte has reer deveopmet so probably this is good) I S xh2 'd6+ 9 h1 �h6+ 20 g "g 21 h "h 22 g �gith a draw.

Better was 27 b5 2S a6 S 29 d7 xg5 30 xc6, though Whte has vey ser ous wig chaces. Now there is o sava o or Black

1  b3 b3 1 6 axb3 0-0 1 7 e2 d6 1  e4 7

Whte would be slghtly better ater  S!? 19 e6 "d7 20 g5 (20 d2 aeS 21 e1 xe6 22 'xe xe6 23 xe6 is also pleasat o Whte  a edgame ith a exra but weak paw is bette tha a level edgame) 20  eS 21 e xe6 22 xe+ 'xe6 23 xe6 c7 Ad ow 24 e3! with some chaces rather ha 24 e3  25 c5 xe6 26 xS, which would lead to a daw edgame. 

2 xb7 a 29 e7  30 f4 g4 31 f3 4+ 32 �g3 1 33 e a6 34 Jxf a6 3 Jx 12 36 axa 63+ 37 �g4 h4+ 3 � h6 39 �e4 1 0

ame 1 1 As-Kaliman Hoogeveen 2002 1 e4 e 2 3 6 3 3 6 4 b d4  a4 

Ths gambt has bee popular o almost a hudred yeas 6 xe

hte has o real ateave to accepg the paw as ater 6 d3 0-0 7 0-0 d6 S xd4 xd4 9 e3 xe3 10 e3 e6   b3 c6 12 d2 b5! (Le-schuk Chcago 2002) Black has equased completely e o Blacks ext moves  be .a7-a5.

Spnish Fo nights with     

6 .. .00

I order to get compeso fo the pw Blck hs to sh s deelopmet. After 6 .e7?! 7 d3 xe4  - xc3 9 dxc3 (9 bxc3? e6 1 xc5 xc5 11 ' c6 12 d '5 13 b3 d5 14 c lso ges  srog pressure 9 .e6 1 xc5 xc5 11 e3 Whites posio is more th sighly better 7 d3

bee pid to the following   f4 d5 9 d3 c6 1 h3 Me 11 - dxe4 12 dxe4 xe4 13 xe4 xe 1 c3 c7! (1 . e6 15 xd xd 16 c2 e wth ded euit .tosCmpor  gre 199 15 e3 b5 16 cxd4 bx d i my opiio Blck is sightly better but of course the positio is er close to eul. b  O-O? d5 9 xd5 xd5 1 exd5 xd5 11 f g5 12 d3 g 13 d5 "h5 14 e7 h 15 d2 c5 gies Blck  cler dtge ccordig to Keres  ..e 9 d

his hs bee estbshed oer e s the mi moe After 7 d3? d5!? (7.d6   g4 9 e3 c6 1 h3 x 11 gx d5 12 g5 h6 13 h4 'd6 1 g3 e6 ge Blck good counterply for the pw i the more recet Kmsky-okolo Brssels 1992  g5 c6 9 ' (9 h4 e 1 f4 '5 1 1 3 dxe 12 b xb4 1 3 xb4 xb 14 - xc3 15 dxe4 xe would lee Blck clerly better 9 e 1 f4 b5 1 1 b3 h6 12 h4 (12 xf6 gxf6 13  dxe 14 dxe4 b4 es Bck  srog ttck  e is er frle 12 .xe 13 xd xd2 1 xd2 xd 15 e2 xe2 16 xe2 e 17 f b7 1  c3 f6 19 g h5 Blck com pletely domnted the bord rrsch Rubistei  ebs 1 91 2 7 ....b6  e5

hs moe is er locl s Blck hs to udeelop s ight to the bck rk f course the epw work s  hook for destroyng the White cetre but l siler gs he  cloud ome tteio hs lso

his is lso cosidered he mi moe After 9 - d6 1 exd6 f6 White  come under hey ttc. hte ow hs to mi choices  11 d7 xd 12 xd7 xd7 13 e1 e 1 d3 g4 15 f3 x 16 'x x 17 d5 xd3+  xb6 d is  ittle bit better for Blck ccorg to Hb er b 1 1 dxc7 "d6! (other excellet moe from Keres 12 h! (12 Me xc7 13 e5 g4 d Wte is i troube 12 g 13 g3 (fter 13 f? xh2! hite is lost 1 xh2 h6+ 15 g3 f 16  h 17 g "h3+ 1 e4 g3+ 19 e5 xg 1 3. .. 'h6 14 h xc7 d Blck hs  lot of strog threts tht re ot esy to meet here re ides like .xf2 d ...b7-b5 d lso  tcky ittle ine like 1 5 g1 h2! 1 6 xh2 xh 17 g1 xg3 1  g3 'xg3+ 19 h h 2 g1 g4 21 e f3+ 22 x3 x 23 f4 h1 2  h4 25 f xf whe Blck is more or less win ing. 9 ...d6

A ew de s 9 c6? 1 e3 d5 11 c3 e6 12 c2 d 13 f5 g5 1  g6 15 g3 f6 16 e h d here in this ucler posio the pers greed  drw n Potkin Acs Prdubice 22. I he  feelng tht White might ed up being better i ts lie. 9 .d6 just seems more trl 1 0 e3 c? 

he Fo nights

Here Black has a wde range f alterna tves 10..dxe s cnsdered n the next game): a) 10  c6! 1 1 c3 f 1 0-0 c7 1 3 f4 xe3 14 dxe3 f 1 exd6 xd6 16 e e4 17 xdS axdS and Black has cm ensatin Jenni-Blaet nternet 00 b) 1O . g! and nw b) 11 exd6 xd6 1 0-0 4f 13 e1 xe3 14 e3 g4 I  h 6 �e c6 and the sn hlds chances fr bth laers accrng  Hbner. b) Nt s relable  11 f4! �g6 1 0-0 g4 th a strng attack 1  f6 13 exd6 xd6 14 f 4f 1  xe3 16 dxe3 f als ves Black cmensan fr the awn  Tmman) fr examle: 13 xg4 xc+ 14 c xc+  d4 xd4 6 e3 e4! 17 e 1 b 1S xb c; r 13 �e1 e and Black has a ver strng sin. b3) 1O . e7 11 exd6 xd6 1 0-0 dS 13 c3 e6 14 c 14 h !) 14 .c  e �g 16 g4 x4 17 x4 c4 1S de f6 19  c 0 e3 d3 1 xd3 cxd3  d g4 3 b3 c 4 c4 x  gxf3 f 6 b and Black wa a litle beter in Nnn-HitechC) age 199

c7 16 f4 f6 17 0-0 adS t s hard t shw hw Whte shld cmlete devel ment. 1 3 . tf6 1 4 00 ic7 1  g3

Black has mre than engh cmensa tin fr he awn. The ght fr the h1-aS diagnal  be a matter f lfe and death fr Whte  mst kel death   f4 xe3 16 dxe3 g4 17 e i f was mre nrma. 1 . . .b

A strng imrvement. Previs tred was 1..c6 16 xf xf 1 7 b4 �d7 1S xf xf 19 c adS 0 d4 feS 1 b4 cxb4  xb4 e4 3 d h and the black ave was ver dangers n Ydas Nelsen Kemerv 199 1h has been sested b Ydasn 1 6 f4

1 1 c3 f 1 2 exd6?

Ths wa te las wth re  he shld rbabl be cntent wh eqali. Bt if we remember that we acall have a jnr wrld chamn lang te here, aganst a frmer FIDE wrld chamin we can nderstand wh hite s adventrs Whte shld rbabl la smetng lke 1 b3 xe3 1h4) 13 dxe3 dxe 14 e4 �c7 1  c4 d6 1 6 d e6 17 00 with eqalt r 1 00 xe3 13 dxe3 c4 14 b4 dxe and the stn  eqal After 1 c xe3 1 3 dxe3 c4 14 f4 de 1  d f6 16 xb6 xb6 17 e4 hwever the stn shld be ne fr Black White has tw bishs bt the are nt imressve and Black has a tle sace. 1 2. . .xd6 1 3 ic2

After 13 e e6 14 x xf 1  e 8

t seems t be sme knd f accelerated maschism t enter sch a sin aganst a master f atack like Alexander Khalifman. 1 6.h4

Black s aing he canns directl at the whte kng akg mves lke this was called anchng b Ta wh cmared t  ice hcke where the ck cld be sht far int the nnts area wth n idea er than hng fr smething t haen 1 7 d4

Perhas beter was 17 h h3 17 .. b7! 1S f4! c6 19 xf6+ gxf6 0 xh7! hS 1 f   f c4 3 h3 f 4 g h6  h3 and the black attack s

Spnish Fo nights with    

hsory) 18  1 8 19 d4 (19 xf6 xf6! 20 xh7 h8 21 5 f3+ 22 h x 23 xh3 6 24 xh6 gxh6 25   ad8 and Black has a clar adanag) 19  cxd4 20 cxd4 4 21 f4 g5 and h aack gos on b h shas hs xra pawn  7b7   d ad  9 a4

Afr 19 2 f8 l h black pics play (b no 19 g5 20 h5! xd5 21 f4 x3 22 x3 g2 23 g5! and h ins) For hs h has only a pawn (and som problms dlopg h qnsd rook and bshop) y dgmn s ha s is no nogh. Play conns wh 20 d g5!. ow h saon is dffrn  Black s a ll rook br off han n h prios n bas i s now aking par n h acon 21 h3 (or 21 h5 xd5 22 xb5 f 23 h "h6! 24 xb7 xh5 and Black ns) 21g4 22 g5 h6 23 4 x4 24 xg4 g5 and h s rady for his own nral

 hs for fr ops sorry Black is f

cors mssin on of h il ons, b who s cong whn yor hag so mch n 21 3

r 21 f5 c7 22 xf4 �xf4 23 �d3 xd5 h an acclrang aack 21 . ..c7 22 d1 xh2+ 23 �1 a6!

23f4 24 g4! wd g h som conrpay. 24 axb axb 2 g4 xg4 26 xg4 xd 27 xh7+ �xh7 2 xd c6 29 h+ �g 30 xd  1 + 3  �e2 xd 0-1

ame  Movsesian-lIIesas Coroa Elista Ompiad   e4 e 2 3 c6 3 b 6 4 c3 d4  a4 c 6 xe 0-0 7 d3 b6  e e 9 d d6 1 0 e3 dxe

1 9. . .d7 20 gxh4

20 h5 xd5 21 g4 (or 21 d3  22 h g5 23 g2 h3 24 f3 b4 25 f5 xh5 and ns) 21 �7 22 gxh4 xc3 23 bxc3 xd 24 xd xh4 and s. Afr 20 d3 xd5! 21 xc5 (21 h4 h3) Black has h bal 21..f4!! 22 gxf4 (or 22 xh7 xh7 23 xb7 h3 24 h x 25 xf2 xd1) 223 23 x7 xh7 24 c h8 25 f3 xf4 and h black aack is crashng hrogh. 20 .. .x4

hs is kown as h main n b I pr fr 1O c5! as n h prios gam.   xe g  2 c4  1 3 4

An rsing nw ida s 13 h4! 'f6 (13 .�g6 4 5 f6 5 f4 c6 16 xc6 bxc6 7 d4 d6 18 c4 4 19 b4 ga hi a clar dg  oylhiro os cow 2001)  f4 d6 15 c3 xc4 16 xc4 c6 1 d4 6 18 5 x5 19 fx5 "  wh an dg o h accordg o oyl. 1 3. x4 1 4 c3 h4 Te alrnas ar wrs

9

he Fo nights

a) 4e  4 h  g3 '8 7 g2! f  8 f4 f  9 f4 f 2 gf4 h 2 2 'f 22 e3 an Blac ha insufcient compensaon in Kobaija Kornee, t Petersburg 998 b 4..c!  4 e   h  g3 h3 7 b ab 8 b} h8 9  f4 2 g2 h 2 OO! gae Whte a clear ege in Alan-Varain, Ufa 999  b3 c 7 c4 h+ 8 3 e4 9  h8 2 e 'g4 2 4 g4 22 g2 an White has a posonal aantage, th the better pan strcture an the o bish ops, Campora-Hanoo, stanbul 2

White is rying to echange in orer not to be mate, bu no he is left th ery ea ght squares an a ar-square bishop  no a goo combinaon

1  g3 h3 1 6 cxd4 4 1 7 1 

2 . .xb3 26 axb3

s is sy Beter is 7 f4 h4 8 e2 g+ 9 g4 'g4 2 e 2 3 f 2 e2 [2 c3  22 4 & oul be a gft] 2g ras) 2h4 2 �e2 g+ 22 �e /2/2 l Kher-lsen, strp 2

1 7 .. .xe3 1 x+ �x 19 3+ 6 20 dxe3 xh2 21 2 h1 + 22 1 h2 23 2 h

23houl ra, bu Blac righty ecies o go for more 24 b3

Beter as 24 2, een though 24..e! 2 b ab 2 c2 a2 27 a2 a2 28 e c4 still poses problems for White 24 . ..e6 2 lxb6

2 a8  mate! 26  cxb6 27 g2 7 2 g4 h6 29 d2 �g 30 �e2

o protect the pan seems rong 3 a3 a 3 g hg 32 g e4 33 g2 f8 34 b4 e8! oul connue the attac for Blac 30 .. .e 31 g

3 a7 e 32 a3  as maybe a better chance, een though Blac remans much beter 31 ..hxg 32 h1 ?! g4 33 h2 xb3 34 'h+ �7 3 h4 b+ 36 �e1 xb2 37 1 b1+ 3 �e2 b+ 39 �e1 e6 40 h1 b1 + 41 �e2 b+ 42 �e1 e7 43 d d3 44 h+ g6 0-1

30

Sp nish Fo nigh ts with 4 . .  d4

ummary

4.. d4  reman te man lne for some me to come The mmedate captures ke 5 xe5 and 5 xd4 are haless or even dangerous for te. The man focus n s varaon s sl drected towads 5 ..a4. Here Black can  to equalse wth 5 ..x Though hs poson  be a ttle passve, s s a pefecy vable strategy and has been successl numerous es. oe standard s te gambt wt 5 ..c5, wch has gven Back good resuts over he years. The current status s at Khafmans choce Oc5! gves Black sucent compensa ton, whle oher moves n that poson could be sghy dubous The more mode dea wth 5 ..c6 aso seems to ve perfect cmpensaon for a pawn, so Whte mght choose to folow the footsteps of Ivanchuk and  5 c4 n e tre Black  probaby prove to be dong just as wel ere too, but tee  be chances for orgnal deas and openng suprses sometng at 5 a4 no longer seems to offer n te same way 1 e4 e5 2 3 c6 3 c3 6 4 b5 d4 5 a4 5 xe5 )  Gam 6 5 xd4  Gam ; 5 c4  Gam 8 5  .c5 5 .. x  Gam 9; 5 .c6 )  Gam 0 6 xe5 0-0 7 d3 b6 8 e5 e8 9 d 5 d6 1 0 e3 dxe5 )  Gam 

O . c5  Gam  

5 xe5

5. . . c6

10    dxe5

3

CHPTER THREE Spanish Four Knights: Sidelines 1 4 5 2 3 c6 3 c3 6 4  b5

In ths a haptr on th pansh Four ghts w w oo at unusua optons for Ba aftr 4 b5. n am 13 Ba m daty attas th bshop th 4a6 Ths s tmpng baus aftr 5 a4 th gam n trd a hass araon of th Ruy op Howr, aftr th strongr 5 �x6 Ba annot ah an qua poson, so ts df ut to rommnd Bas fourth mo In Gams 416 Ba pays 4..d6, an ntrsng da that aods th symmtry of 4 b4 and th dpy nstgatd man ns of 4  d4

ame  uasin-Saalik Kemevo  4 5 2 3 c6 3 c3 6 4  b5 a6

trnay a) 4. 5  d6 6 d4 xd4  xd4 d ads to th tntz Dfn to h Ruy Lop Ths usuay ass a 3 �b5 d6 and s outsd th sop of hs boo b) 4  5 appars o b qut natura, but wht has sra promssn ontnuatons b1) 5 x6 s an mprod rson of th assa Ruy opz xhang. Th man 3

rason s that ba aady has dopd th nght and bshop onto th nfror squars f6 and 5, . 5  dx6 6 d3  (Rs but f 6 g4  h3 h5 8 3 '  9 x5 'x5   '2 d 11 g4 g6 12  f8 13 h 4 h5  as  n Hug-Frnands, Thssaon 1 988  thn 14 d4! xd4 15 xd4  16 b3 xd  1 xd ' 18 '3
Spnsh Fo nghts Sdelnes

xf6 11 'd5 gives te a clear plus

1 0 e1 e6

6 txe5!

fter 1 .e7 1 1 g5! f6 12 e2 - (12 .. fxg5?? 1 d) 1 g 'x5 14 xe7 'a5 1 5 e Wit as an overem ing advantage

Te acid test. ite as no advantage af ter 6 - d6! 7 d4 (7 d is also armless) 7...d7 8 dxe5 xe5 9 xe5 (or 9 �f4 g4 1  �xe5 �xf 1 1 'x xe5 it equality) 9 xe5 1 xdxd8 1 1 d1e7 6 ...txe4 7 xe4 d4 8 00 xe5 9 d4 �f5

Or 9 ...d5 1 e e6 (1  e7? 1 1 �g5! e6  1 1 ..f6? 1 2 xf6!  12 �xe7 xe7 1 c5 gves ite a innng posi onal advantage) and no a) 11 g5 (is is not so clear) 11 12 xe6 e6 1 g4 xd4 (1. .. c5 14 xe6 xd4 15 e4 d7 16 xd7 x d7 17 e d8 18 xd xd8 19 d2 ves ite some cances  te endgame, even toug tey are sm) 14 xe6+ 'd7 (Black is even orse off after 14 .b8? 15 g5! �c5 [1 5  c8?? 16 xc xc8 17 e d7 18 d 16 xd8 xf 17  xd8 18 e2 '4 19 f1 d6 2 g d4 21 c 'b6 22 g2 c5 2  5 24 :xg7 a7 25  1- Teicmann-piemann, ipig 1914) 15 g5! �b4 16 �xd8 (16 c xe6 17 :xe6 d5 18 cxb4 xg5 19 e d8 ves Black enoug counterplay for equality) 16.xe 1 1 7 xd7 xd7 1 8 xe1 xd8 and Black sould dra b) 11 �g5! �d6 12 b �b4 (12-? 1 c4  14 xd6 cxd6 1 5 �e7 drops a pan and 12.f6?? 1  c4! f5 14 g4 'g6 14xg4 15 xg4 xg4 16 xf 15 xd+ cxd6 16 e+ d7 17 e2 ae8 18 e 1 drops everyting) 1 c4 (1  e �c! 14 xc 'xg5 1 5 e4 'g6 16 c5 -- is less clear) 1   and no b) 14 d5 cxd5 15 cxd5 �xe 16 dxe6 �xf+ 17 xf2 'x6 18 c1 c6 gives no advantage to ite b2) 14 e  (14f6? 15 �xf6! f6? 16 g) 1 5 a a5 1 6 d f6 1 7 4 ad8 18 b4 b6 19 c5 and ite as practicaly an exa pece

1 1 g5! d6

Tere are oer possibiies ere a) 11..'g6?? 12 d5! cxd5 1 'xd5 c6 (1 xd5 14 f) 14 d4! 6 (14..f5 15 ad1 and Black is aso lost) 15 ad ! xg5 16 'd xd7 17 f d8 18 xd7 c8 19 e8 mate. b) 11.b4!? 12 c d6 1 4 6 14 xd6+ cxd6 1 5 'b b5 16 a d7 1 7 d5! cxd5 18 g t a strong attack c) Te bes  must be 116! No te sould continue t 1 2 4! in order to play as in te main game. nstead 12 d?! d7  4 e 8 (1..b5 14 'd2 e8 15 b! c8 16 c4 '5 17 g d8 18 'c ves ite som pressure accordng o aul eres) 4 c4 c8! (14.�d6 15 Mac1 c8? [1 5 . g4 is bette] 16 d5! it an at tack  clecter) 15 g! (1 5 Mac1?! b4! 16 Me [16 e2?? 'f4!] 16 g4 17 �g f5 18 'b e7 19 Mce f4 2  f5 21 �xf4 'xf4 22 g f6 2 Mxe6 2 Me4? d6 24 c5 xb 25 xe Mxe8 26 Mxe d 27 cxd6 x8 28 axb cxd6 and Black as realisic ops of inning te endgame 2.. Mxe6 24 Mxe6 xd4 and Black as a good position) 55 16 4 leads to an un clear poson Tis is e only uscaon of 4  a6 I ave found so far 12 d2

Not 12 g4?! d5! (2 g6?  f4 f5  4 xd cxd6 15 d5! as a er  nosko BorovskyRubinstein, Ostend 197) 1 b b4 14 c4 'd7, i a most a very sigt advantage for ite 1 2. . .h6

12  -! is robably better Tis is a l non idea  Black sacrices a pan and in ret gets good draing cances 1 xd6 cxd6 14 �e7 Mfe8 15 �xd6 Mad8 16 �e5 (16 Me5 f6 17 'b4 g6 18 ae 'xc2 19 33

Fo nghts

d7 34 g6 e7 3 g7 xf6 36 g8' fN+ 37 f4 and ite ns b) 17...c! 18 dxc �xc (8. dxc 19 �b6 �c6 2 �a7 and Blacks positon looks ke a wiss ceese!) 19 'd2 d 2 b4 c4 21 �e3 d4 22 �e ves Wte good atack ing prospects

c7 d7 2 xb xd4 21 xa6 a4 is not better for Wite at a) 6.. c 17 c3 f6 18 f4 cxd4 19 cxd4 c8 2 g3 c2 21 4 'd and Back as cmpensaon, but must gt for te draw al te same. ven so it sould be possible to make it Te uncom fortable tng about Blacks poston is tat the game mgt last 1 moves befoe te daw is acieved  Wite  alwas ave an extra pawn. 12  d7 13 c xc 14 dxc �d  ad1 �xd2 16 xd c8 17 f 4 ves te good attacking cance  Ydasin.

After 18   19 c4 c 2 cxd (but not 2 dxc? �xc 21 xd6 �xb4 22 xb4 xd 23 xd1 d8 and Black as cances to draw) 2...cxb4 21 dxe6 fxe6 22 xe6 xg3 23 g3 Wite as good winnng cances

1 3 ih4 g5

1 9 a3 ;c7

After 13...�g4 14 xd cxd6 I g3 - 16 4 Wite as a trong attack

Te alternaves also lose: a) 19...b7 2 c4 bxc4 (2 . � 21 xd6) 21 bxc4 �xc4 22 xd6 and Blacks king is about to be mated. b) 19.4 2 c4 (2 �xa d7 looks tempg but it is not completel clear) 2...bxc4 (2.. � 21 xa) 21 bxc4 'xc4 22 xd6 �xa2 23 �c and Black is lost

1 4 txd6+ cxd6 1 5 g3 0-0-0 1 6 b4!

1 8 ad1  b5

20 c4! bxc4 21 bxc4 xc4 22 d5! ixd5 23 e7+ ;c8 24 ixd6 d7 25 b2 xe7 26 b8+ 1 0

ame  insnes-Heor reb  Now Wite is cea better. 1 6. . ."d5

1 e4 e5 2 t3 tc6 3 tc3 t6 4 ib5 id6!

A better tr was 16.. 17 xd6 �xb4 18 xb4 and Wite as te advantage  after 18 ..xd4?? 19 c3 \ites cosing in on te point. 1 7 b3 h5

Te folowin lines are qte entertainig a) 17 . f 18 c4 'xd4 19 ad 'c 2 �d2 wit te advantae for example: 2..f4 21 xf4 (21 xe6? 3 22 xg3 d ves Black counterpla) 21...f4 22 xe6  23 xd6 xd6 24 �xd6 'xd6 2 xd6 g8 26 g3 e8 27 4 e1+ 28 2 e2 29 g4 xf 3 g3 xa2 31 x6 (31 x onl gives a cear advantag) 31.. 32 f6 b2 33 g 3

5 d4

Spnsh Fo nghts Sdelnes

s is quie a local aack on he cene, bu no necessay he bes move. Whie has he folong aernaves a 5 0-0 is seen  Game 15 and 5 d3,  Game 16 b 5 xc6 dxc6 6 d4 exd4 6. .d  xe5 xe5 8 xe5 xe5  xd+ xd8 0 d2 e 11 0-0-0 d8 2 f4 gives Whe some advanage  he endgame  xd4 e 8 g5 c5  xf6 (9 d2! h6 10 xf6 xf6 1 1 00-0 0-0 is beer for Black  e o bshops domnae 9 ..xd4 0 xe xc3   bxc3 xe  2 0-0-0 ih equaiy c 5 h4! xe4! 6 f5 6 g4 xc3  dxc3 f8 8 'e4 d6 and Whie pobably does no have enough for he pan 6 ...f6  xe4  g4! xf2! 8 xg xg  xg f8 and Black s a pan up  ..xf5 8 xd6 cxd6 9 0-0 0-0 doesn offer Whie enough compensaon for he pan d 5 a3! h6 6 d3 b6  c4 b 8 h3 a5 8...c5 9 b4 d4 10 xd4 xd4 11 b2 gves Whie a sigh edge  a2 c5 0 e2! c!   c 1 1 g3 d5 ih a level game as beer   g3 f8 12 d2 d6 13 b4 c6 4 c3 d8 15 0-0 and Whie as a lile beer  Keinghaus-Pedersen, Germany 2002 e 5 g4!

Ths fanasc move is a compleey ne idea. o Black can pay e  5 00 6 g5 h5  d3 g6 8 e3 e8  d2 f8 10 0-0-0 a6 11 c4 b5 12 d5

fooed by d3-d4 ih advanage for Whie e2 5  h6 6 g d4  c4 and Whie has some advanage e3 5 d4 6 c4 and Black has achieved ile e4 5 xg4 6 g f6 6.h5  h3 xf2 8 x c5  d4 xd4 0 e3 give Whie he advanage  xg f8 8 g3 h5  g1 f6 (9 ..g8 0 xg xg8 1 xe5 ns a pan 1 0 d3 and Whie is be er e5 5 ...c5! is probably bes. No Whe has he fooing possibes e51 6 g5 g4  f1 h6   a6 8 xc6 dxc6  h3 x 10 x xf 11 xf2 xh3 12 g d 3 d3 0-0-0 h a grea mess as also possible 8 gxh6 h6  h3 f6 0 d3 d6 1 1 h looks abou level. e52 6 g ! d4  c4 es Black he chance o ake he niiave h  .d5! 8 xd4 exd4!  xd5 xe4 10 d3 d6   b3 0-0  Black is beer e53 6 h3 d4  c4 d6 8 d3 c6  g5 d 0 a3 ih an unclear game Bu no 10 xd4! exd4 11 e2 e5 2 f4 xc4 3 xc4 h6 4 d3 hxg5 5 g5 e and Back is beer 5    xd4 Also possible is 5  exd4 6 xd4 0-0 and no a  xc6 dxc6 8 d3 e8  0-0  g5 xe4!  g4 10 g3 0 h3 h4! ves Back a song aack 10  f6 1 1 f4 c 12 g2 6 h good play for Black b  g5 h6! or ..e5 8 xc6 xc6  xd8 xd8 0 d3 ih a level game 8 h4 e5  xc6! Black is beer afer  xc6! bxc6 10 d3 b8 1 c1 d5 and  f5! d6  ..dxc6 0 de2 h equaiy. c  e3 a6 8 e2 b4!  xc6 bxc6 0 d4! 10 e5 d5 11 d2 xc3 2 bxc3 a5 only ves Black a sigh edge 1 ..c5 1 d3 e8 1 2  d5! 13 0-0-0 c4 14 d2 b and Black is beer, arosGrabarcyk, Par dubice 2002 6 xd4 xd4 7 'xd4 '7   35

Fo nghts

Or 8 f 6 8�S 9 eS xeS 1 eS g 11 dS - 12 x b8 1 dS xeS 1 �f ves Whte some advantage) 9 �e2 �S 1 eS d6 11 xe xe 12 �d2 wth equaty. 8 .. .00

Ths s the most logal move but Bak has also ted 8  6 9 e2 - 1  d e 1 1 gS h6 1 2 h eS 1  d6 1 ad e6 15 g ad 16 e2 xg 1 xg dS  eS d 19 e g 2 e a6 21  th an even game eado Trvno Korneev alaga 1999 9 h3

9 �d? 6 1 a 1 e? �xh) 1 bS 1 1 b a also ves Blak the na ve but 9 dS xdS 1 exdS �eS 11 e e 12 �d d6 1 h wth equaty was beer but not 1 ? xh). 9 e5

Too mehanal Blak oud have taken over wth 9  6 1  e2 1 a �eS 1 1 e x 12 x xe and 1  d  1 1  bS 12 b d6 are both dsasters for Whte) 1..eS 11 e x 12 bx xe and Whte has some ompensaon for the pawn but hardy enough. 1 0 e3 c 1 1 d3 e8 1 2 e1 d5

Or 12�x 1 bx dS 1 exdS d  1 d2 xe1 16 xe xdS wth an unlear poson 1 3 exd5 cxd5

A taal oversght of huge dmensons 1 e2  15 f �d 16 f would be equal. 1 4. . .d4! 1 5 xd4 xe1  1  Wh2 h5 1 7 xd5 g3 1 8 Wg1 e1  1 9 .1 e 20 xb7 ne8 21 e4 g 22 d2 x4 01

ame 15 Gallaer-Cooper Bsh Lague 2001 1 e4 e5 2 3 c 3 c3 6 4 b5 d! 5 0-0 0-0  d3

6 d has not brought Wte a lot of su ess 6  exd  xd e and now a)  �x6 dx6 9  e 1 �e 1 gS? d 11  S 1 2 dS eS 1  xf6 gxf6 1 b xh would leave Blak wth the  pont) 1dS and Bak s smpy beter. b)  e1? xd 9 xd eS 1 d 6 11 a a 12 a3 bS 1 �b �a6 1  b wth a strong nave Barglowsk alanuk Bydgozz 21 ) 8 x6 dx6 9 �d g 1 g f6 11 f � 12 g2 h6 gves Blak good at takng hanes .h

6 ..b6  e2 e  g a6 9 x6 6 1 b d 11 �b2 S was agreed drawn n emmers-otwan Belum 1996 After 12 fS omes 12.f wth the dea ..e6-d 7 e2

Whte has aso ed  h? a ompletely unneessary move as �g snt a threat at he moment b6 8 e2 �b 9  e 1 g � 11 e a6 12 �a dS 1 e2 bS 1 2 d wth equaty equera-Hetor Bled 22. Ths poson oud just as easy have asen rom a Ruy Loez wth d2-d nstead of d2-d. 7  e8 8 g3 a6 9 xc dxc6

The poson s equal. 1 0 b3 e! 14 4

36

Ts move s very strange The bshop s

Spnsh Fo nghts Sdelnes

utterly useless o ths square, ad t eve blos the ghts usual route to d4. Better was 1Q...5 1 1 b2 b6 12 e2 d7! 1 3 d2 8 14 4 e6 wth a eve game

ame 16 Yemelin-Karlov Mosow 

1 1 ib2 d7 1 2 d4 exd4 1 3 xd4

ow hte has a sght pul. 1 3. . h4 1 4 4 g6

Playg ths s le poug petol o a e  t oos terrble ad t s terrble. 1 4...g4 15 �d3 5 16 h xd4 17 xd4 6 18 e5 h5 19 xh5 xh5 20 e6 gves Bla problems but mabe thee was aother way out o the mess. 1 S d2

Also teresg s the ouao 15 5!? gx5 16 dx5 x5 17 x5 e5 18 �h5 xh5 19 xh5 ad Blas gsde s very ragle.

 e4 eS 2 3 c6 3 bS 6 4 c3 id6  S d3 h6

Also e s 5 ...a6 6 a4 h6 7 h3 b5 8 b3 b7 9 a3 5 10 0-0 (or 10 e3 a 11 d2 00 12 xa7 xa7 13 0-0 a8 1 4 a4 b4 15 d5 b8 16 x+ x6 17 d5 d4 wth equalt Forto-Labolta, Bueos Aes 2002) 10 ...0-0 11 d5 d4 12 xd4 xd4 13 3 a7 wth a level poso Joma .oolov Leeuwarde 2002.

1 S  ad8 1 6 Wh 6

Better s 16..g4 17 3 6 18  x 19 x, but hte stl has a stable ave. 1 7 ae1 gS 1 8 eS

 order to aa o the gsde hte rashes through  the etre. 1 8 xeS 6 h3 O-O

1 9 dS

hte vades the ght squares ad has a wg atta. 1 9. . ..xS 20 xS hS 21 xgS hxgS 22 xd6 cxd6 23 S "h6 24 xgS+ Wh7 2S "d3+ Wh8 26 g3 cs 27 e2 1-0

Better was 6 ..a6 7 x6? (7 a4 b5 8 b3 a5 s equal) 7..dx6 wth a vey good verso o the pash Exhage or Bla: 8 e2 8 e3 �e 9 e2 e6 10 a3 b5 1 1 a4 0-0 12 00 d7 was eve  Kosels halhsh Dorud 2001) 8...5 9 b3 e6 (or 9 ... d7? wth the dea o ... b8-6d4) 10 b2 d7 wth equalty Podles Pavasov Doba 2002. 6.. b6? 7 x6 dx6 8 e2 �e7 9 a3 d7 10 d2 b5 11 g4 5 12 g3 g6 ave haes or boh sdes tma-Rabega, Graz 2002. 7 g4

hs lusrates he drawba o ombg ...h7h6 ad ... 0-0. he ow has a hoo to atta. 7

Fo nghts

7 .e8

...a6 8 xc6 dxc6 9 g! oo ow i 9 e2?!  10 g3 c 1 1 f e8 12  e3  f8 13 d2 "f6 14 4  1 g c 6 16 g3 g6 ad e poio wa ucear  e gae Podeik-Paaoic Ljubljaa 2002) 9xg 10 xg e 1 1 d2 e6 12 0-0-0 b4 13 dg ad \ie ad a rog aack i Podekebi Bled 2002 8 g5 hxg5 9 .xg5 d4

fer 9 ..b4 10 g1 d4 1 1 c4 b 1 2 b3 c6 13 a3 xb3 14 cxb3   1  d4 exd4 16 "xd4 Wie a a rog aack 1 0 d5 .e7 1 1 xe7+ xe7 1 2 xd4 exd4 1 3 g1 6 1 4 3!

Afer 16  a1 ? 1 xf6 g6 18 g4 we ae rapoed o e preiou oe 1 7 x6 h5+

Wie w afer 1 g6 18 g "b4 19 e Mxe 20 6 1 8 3 g6 1 9 g5 h7

Here Black could ae kep e gae alie w 19x3 20 xg6+ fxg6 21 xg6 f8 22 a3+ e 23 "f g8 24 xe f1 2 d2  26 c1 e1 2 b2 c3+ ad Black draw 20 xd4

Afer i ie a a ery rog aack o e dark quare

i i ber a 14 a4 d    ee oug i alo ake ee 1 4. . .b4+ 15 �d1 xb2

1  cxb?? 16 xf6 "f8 1  g3  18 "4 ad Black i aed

20 . . g7

20 d 21 b2 x3 22 d2 "6 23 ag1 dxe4 24 fxe4 i alo ery good for Wie 21 b2 xd4 22 ixd4 d6

1 6 1

ere We ie a reay rog cobi aio 16 xf6! xa1 (16gxf6 1  c1  18 xb2 cxb 19 xd4 e6 20 d2 b6 21 g4 followed b ag ad 4 wi ae) 1 c 1  18 g4 ad Black a o defece eg 18Me6 19 d& g 20 4 wi ae o coe. 1 6. . .xb5

38

fer 22  b6 23 4 c 24  b2 b 2  e6 26 d2 c4 2 ag cxd3 28 cxd3 c8 29 xg6 xg6 30 xg g6 3 xg Wie a ery reac wg cace 23 h4 �8 24 �d2 �e7 25 4 5 26 b2 �d7 27 5 gx5 28 h5! xe4 29 h6 �6 30 h7 e3+ 31 �e2 6 32 x6 .e6 33 h1 h8 34 xh8 xh8 35 �xe3 b5 36 a3 a5 37 h6 a4 38 xe6 xh7 39 gg6 1-0

Spnsh Fo nghts Sdelnes ua

Afr  a6?! I am vry sscos abo t Black oso Oly   wt 1   !? mg v Black a raso o co stdyg ts l bt I dobt t.  .. d6!?, owvr as so ar rov to b a vaao  ts ow rg o clar way to a advatag as b od or Wt ad lI ca o r s o rcommd  g wc wl a as srs yor oo t 1 4 5 2 f3 c6 3 c3 f6 4 b5 d6 )

 a6 ) - Gam 13

5 d3  Gam 16  d  Gam 14  00 )  Gam 15

   d6

   6

5 00

39

CHAPR FOUR Scotch Four Knights: The Main Line 1 4 5 2 f3 c6 3 c3 f6 4 d4 xd4 5 xd4 b4 6 xc6 bxc6 7 d3 d5 8 xd5 cxd5 9 0-0 0-0 10 g5 c6

Th diagram abv s th bas psn f th th ur nghts and has bn fr as lng as anyn an rmmbr. r a lng m is i was nsdrd uttrly harmss and drawish, but n th 990s auir am up ith 11 a4? and th i mmdatly bam fashnabl. Hwvr vr m t bam apparnt that t has n ra han f an advantag hr thr hs days this i s payd may by ths wh favur famiar trritry d nt ar abut pning advantag but hp that ppnnts hav nt truly gurd ut th s and uts f th psn

Blak mmdiatly prvks ht t x hang. 1 1 .8 1 .d6 and 1 1..7  b sussd n tr gams 1 2 xf6 1 h4? wud b a bad mistak. ftr 1..g 3 g3 g4 ht ny has n way t avd lsng a p 14 7 x  xd8 d8 16 gxf3. Hwvr Bak s has a ar dg. 1 2 :xf6 1 3 'xf6 gxf6

ame 17 Bykovsky-Howell Cappelle la nde 1 995 1 4 5 2 f3 f6 3 c3 c6 4 d4

xd4 5 xd4 b4 6 xc6 bxc6 7 d3 00 8 00 d5 9 xd5 cxd5 1 0 g5 c6 1 1 f3 This s th tradnal mv. 11  s nsdrd in Gams 3 and 11 a4, i Gams 48. 1 1    h6? 40

Th psn s mr r lss ual I  lav t b up t th radr t dd n whih sd h prfrs. hit has a bttr strutr and Blak has t bshps and t halfpn ls. 1 4 2 d6 ftr ths mv Wht an gan sm ad

Scotch Fo nghts The Mn Lne

 vantage. 14.b8! not commitg the bish ops too soon is better Then there are the followng possibes a) 1 c �d6 16 ab1 e8 7 d4 �d7 18 b4 hS! 19 fe1 h4 20 g h 2 f4 cS 22 xe xe8 2 bxcS c 24  b6 2 a6 �g4 26 e b8! with equaity. b) 1 g aS 16 ab e8 17 fS! �fS 18 fS e2 19 g4! b4 20 g2 f4 and Black was a ite better in Rubistein pieann Berin 1926. c) 1 d4 b6 16 c4 d8 17 b f8 18 cxdS cxdS 19 ac e6 20 fd a with counterpay Buxade Roca-kembris Vila de ges 1999 d) 1 a d6 with equal play  Keres 1 5 d4 Whte has also ted a)  c and now a) S...b8 16 ab e8 17 fd eS 18 d4 d7 19  e7 20 d4 eS 21 b4 hS 22 bS cbS (22...cS 2 c6! �xc6 24 bc6 c8 2 �bS and Blacks weakesses on c6 and dS give Whie good winig chances) 2 bS xbS 24 xbS xbS 2 xbS c and White is possibly sighy better a2) S...aS (iected against b2-b4) 16 g b8 17 ab  e8 18 fe  eS 9 e2 f8 20  This was played in Bykhovsky Khavsky Moscow 1999 Now Black sould not play 20...e2! as in the game but 20hS! 21 h4 g4 22  d7 and thanks to the weaknesses on the kngside Black has some initiave b)  g �eS 16 c b8 17 f4 c7 18 b e8 19 ae  xe1 20 e1 f8 21 hS �d8 22  a 2 h cS with good counter play veshkov-Yudasin UR 1986 1 5  5 S. d7 would alow White to play 16 fS dominang the Black bishop  the knight versus bshop endgame would be terrble After 16.e8 17 �h b8 18 b White is perhaps more than ust slghy bet ter. S..b8 6 xc6 b2 17 xa7 d7

18 a4! would not work eiher 1 6 f5? After this move chances of an advantage have basically gone. Better was 16 bS �eS (16...�b8 7 c4! a6 18 c dxc4 19 �xc4 eS 20 ac1 gves ite an edge) 17 f4 �xb2 (17 c4 18 fxeS cd 19 f6 xc2 20 c1 b8 21 a4 a6 22 d4 b2 2 c2 c2 24 xc2 e8 2 d6 gives \te good chances in the endgame) 18 ab1 a6 19 xb2 axbS 20 �xbS and the passed apawn gives White a slght advantage. 1 6  xf5 1 7 xf5 f8 17...ab8 8 b d4 19 fe fe8 20 xe e8 21 f was righty agreed drawn in olubovic-urkovic Kastav 2002.

1 8 d 1 d8 This rook looks more natraly placed on b8. 18...d4 19 f 1 eS is leve 19 f4 7 20 g3 5 21 f3 b8? 21...c4 22 a b6 2  f8 was s ne even ough the posion no longer can be called completely drawn. 22 3 1 +?? This is a crime the ending without the a pawn i ust lost. ecessary was 22.b6 2 dS e2 24 h 1 be8 2 f c4 26 f hS wth counterplay 23 x1 x1 24 f1 d2 25 2 1 26 x7 8+ 27 �f3 xb2 28 4 4 29 5 d4 30 Jb7 3+ 31 g4 1 32 6 3 33 7 g7 34 .6 d3 35 xf7+ 10 1

The Fo nghts

ame 18 Kounz-Van en Doel Dei;au  1 4 5 2 f3 c6 3 c3 f6 4 d4 xd4 5 xd4 b4 6 xc6 bxc6  d3 d5 8 xd5 cxd5 9 00 00 1 0 .g5 c6 1 1 3 8 1 2 h3 %ite has n eally pmisig way hee: a)   xf6 �xf6 3 �xf6 gxf6 is eqa b)  is  fe e6 3 e h6 4 d d6  ae b8 6 b3 d  h3 alic ichta Thessalki 88 1 2..b8 Ateately a) h6? 3 xf6 3 f4 d4 4 e4 d  h f 6 3 g6  e f6 8 c4 e6  ad  c s als ghy leel)  3 xf6 4 xf6 xf6   e b8 6 ad d6  b3 a 8 3 b4 with eqal ty in GegieYsp naes  b  d6! 3 fe e6 4 a4 b8  b3 h6 6 f4 c  ad is Gtmans analysis He beliees that White is bette hee bt  beliee that afte xf4 8 xf4 a! the psin is eqa 1 3 b 1 d6 1 4 2  4 a4 h6  xf6 xf6 6 xf6 gxf6 with eqality 1 4. . .h6 1 5 f4?! Bette was   xf6 1 5. . .c5

w Back is bette placed f the mid dlegame 1 6 b3 5 ls inteestng was 6 d4!?  xd6 �xd6 8 fe b  �g3 e fllwed by be8 when Black is a little bette 1 bd1 .b 18 .xd6 xd6 19 g3 5 Black aids exchanng qeens i de t eta his attacking chances xg3?!  xg3 is ny eqal  g nt eally a sft spt anyme 20 c3 b8 Back is still in the blding p phase t is t sn t n t fcig meases: afte h  h4 d4  b �c6 3 f3 g White has 4 �e4 �xe4  e4 g3 6 fe xe4  xe4 xb 8 d3 b8  e wi cmpensan f the pawn 21 d1 d4 22 b5 c6 23  c Afte 3  e6? 4 xe xe  xe xe 6 b a6  d6 xd3 8 cxd3 e  c Whte btains cntechances 24 b5 b8 25 5 x5 26 3

26 .. .d? Hee Black cld hae hed n t the in tiate with 6h! leaing White in t ble His pssibe answes ae: a  �f? f4 8 h h  e4 c8! 3 a4 xh3 3 xh3 xh+ 3 g e and Black wis b)  h4 g 8 a3 f4  g4 e 3 c4 e6 and Back wins wth g Thee is 4

Scotch Fo nghts: The Mn ne

no w to pevent his, eg 31 h d3 32 cd3 e2 33 I f 34 g g5. c) 27 'h2 g5! 28 3 dS!. White wold ie to hve is knight on c4, t Blcks knight is f speio to ites ishop so fte 29 c4 c4! 30 c4 d6 Blck hs  cle dvntge 27 1 d 28 4 Wite cnnot pl 28 c4?! fte which the d4pwn hs ncesed ts vle  2-3 ponts! Afte 28 . c6 29 d f6 30 d2 h5! Blck is dong ve wel 28.. g6 29 t3! f o do not hve  pln, t is often  good de to impove o wost piece 29.. 1+ 30 x1 t 31 4 b4!? 32 'xb4 White hs o ltenves ) 32 c? e4 33 e4 c4!? 34 c4 c4 35 c4 c4 36 ' g7 37 h f6! th  pomisng endgme fo Blck. ) 32  e1 323 33 c3 '3 34 5 'c1 35 d g7 36 'd5 is pol somewht ette fo Blck) 33 e e4 34 e4 �f8 35 �d2 e7 36 c3 e6 37 5 c6 38 �d3 e5 with n even endgme. Bt not 38.. e5?! 39 �c4! �e4 39.c3 h  sight white dvntge is ete) 40 c5 d3 4 d6 f4 42 g3 �f3 43 c4 d8  �d4 e2 45 2 2 46 �c4 d' 47 d d 48 4 nd it s Blck who hve to ght to w 32 . cxb4

33 tb?? Now Wite loses! White onl hd one choice, to pl 33 d5!! 3 34 4! 4 35  c6 36 6 7 37 � �f8 38 �e2 �e7 39 �d3 d6 40  c5 4 f4 wth chnces fo oh sdes. Bt not 33  ?? d3 34 cd3 3 35 d4 f5 36 2 4 37 d5+  nd Blck wis. 33 . .d3 34 cxd3 xb3 3 d4 35 f4 d7 36 d6 4 37 7 c2 38 5 3 39 c4 c5 40 2 e4 41 de4 e4 ves n esil winnig endgme. Blcks kig  id the kngside while Whites knght s occpied doing nothing on he eensde. 3.. .x4 36 td6 tc6 37 d t 38 tb7 f 39 b1 b3 40 d6 4 41 tc �f7 42 f4 tc4 43 d7 �7  d3 tb6 4 b d1 46 td3 b3 47 t b2 48 txg6+ �d8 49 t txd7 01

ame 19 NunnSulskis Mosow Oiad 1994 1 4  2 tf3 tc6 3 tc3 tf6 4 d4 xd4  txd4 b4 6 txc6 bxc6 7 d3 d 8 xd cxd 9 00 00 1 0 g c6 1 1 f3 d6!? 1 2 f1  ) 12 f6 f6 13 f6 f6 4 d 4 e2?! c5 15 c3 8 16 3 e8 looks even ette fo Blck) 14.. 8! 15 3 e 6 e2 e8 with eli  Gn. ) 12 h3 8! 3 4 13  h6 14 f4 e8 1 5 fd e6 is hmless) 1 3 e6 4 d h6 15 f6 'f6 6 'f6 f6 17 3 fd8 wih elit Fiedich-Devshi Aco 2002. c) 2 4 e8! 2..8!?  3 c4 e6 1 4 c1 h 6  5 f6 'f6 16 'f6 f6 1 7 3 ws geed dwn in tovsk-Gel olon 1995) 13 h3 5 14 f6 '4 15 3 f4  6 f4 f4 1 7 d4 d6  8 fe e1 9 e  e6 wih eit in Adms Pet Dond 1992 3

The Fo nghts

1 2. b8 12.h6 13 xf6 xf6 14 xf6 gxf6 1 e2 b8  6 ab  e8 i a little better fr te accrding t Gutman. 13 4 h6 14 xf6 xf6 1 5 xf6 gxf6 1 6 Jd1 g4 1 7 f3 6 1 8 g3 Thi mve create weaknee n the kingide fr n rean Simler wa 18 c3 h 9  h4 20 de2 c 21 b3 d 22 b2 with euaity 1 8.  g7 1  b3 h5

Black nw ha a target n the kingide, and White  have t be accurate in i defenive ay t kee the balance. 20 Wg2  21 c4 4 22 J2 dxc4 23 xc4 xc4 24 bxc xg3 25 xg3 Jfd8 26 %xd8 After thi Black rk enetrate. The  ght track wuld have been 26 b  f 2 b3 f6 28 c a 29 xb8 xb8 30 b2! c3 31 c4 b 32 d6 xc 33 g4 e 34 c2 d 3 xf d 36 xd2 xd2 3  f4 38 e2 e 39 d3 with a draw. Of cure t i nt frced, but it de hw a igly lkely way t euality 26 .. .xd8 27 Wf2 2 b2! f 28 2 f6 29  f4 30 f4 f 3 c d 32 a4 d2 might be better, but Black l tand very well 27  d1 28 b2 c5 2 Wg2 c1 30 g4 Or 30 d3 xc4 31 xc xc and the extra awn mght nt guarantee a , but 

t a heavy brck n the wall. 30 .. .d4 31 Jd2 c5 Securing the bih n d4 and thereby igighng the weaknee f c4 and the knight n b2 32 Wg3 Wg6 Better wa 32.f! 33 f f6 34 g4 (34 f4 e1 3 d e+ 36 g4 g1 3 h4 xf and Black wi  Sulk) 34... a 3 f4 g1 36  xf and Black ha made imrvement. 33 2 5 34 Wg2

34 ..xb2 Thi i a wrng decn  after 34...d4 3 g3 f Black l ha a lt f reure. hing t enter the rk ending hw a ittle lack f techniue The ld Ruian ma ter alway thught that yu huld imrve yur in lwly 35 xb2 xc4 36 Jb5 f5 37 gxf5 Wxf5 38 Wg3 W5 3 J5 f5 40 x7 c1 1 7 Wd4 42 Wf4 f1 43 4 c4  5 J1 45 d7 Wc3 46 Wxf5 x5 7 W6 Wb4 48 f4 c3 4 Jd1 6 50 W5 J2 51 1 f2 52 f5 c2 53 Jc1 2 54 Wd6 Wc3 55 f6 Jf2 56 W7 Wd2 % - %

ame 20 BerSokolov Malm 2001 1 4 5 2 f3 c6 3 d4 xd4 4 xd4

Scotch Fo nghts The Mn Lne

f6 5 c3 b4 6 xc6 bxc6 7 d3 00 8 00 d5 9 xd5 cxd5 1 0 g5 c6 1 1 f3 7 Black does not wsh to accept doued pawns 1 2 f1 For 12 h3 see Game 21 Better, however, s 12 Mae1 Then 12.eS 13 d! S 1 h3 e6 15 3 d 16 e3, 12..e6 13 e2 c5 1  g3 h6 15 f g 16 e3 MeS 1 , and 12S 13 d e6 1 h3 eS 1 5 xe �xe 16 Me2 d6 1 Mfe1 al seem to e eer for \hite. 12h6? is now a very rsky move; in my opinon it is actally a mstake 13 xh6! (13 f d6 1 h3 MS 15 a M 16 xd6 �xd6 1 3 s a little eter for \hie  Gutman; I am not completely sure aout his evaluaon  here are no ovious plans) 13..h6 1  'e3

and now: a) 1 d6? 15 xh6 S 16 Me3  (16 .. g 1 1 x2 1S Me5 xe5 19 Mxe5 e 20 xe f5 21 e6 and \hite ins) 1 �g5 hS IS h6+
'd2 MS 2 g3 1-0 Ardeeanu-Vajda, as 1999. c) 1e6? 15 xh6 d6 16 'g5+ hS 1 f!, MeS 1 S M3 g 19 h5+ g 20 Mg < 21 xg&
The Fo nghts

- Guman.

1 3 xh6? Maybe not te best opon. After 3 �d2 �d6 14 h3 b8 1 5 b3 c7  6 f4 b4 7 xc7 xc7 he paers agreed a draw in VydesaverLev, Israe 992. Probaby best is 3 f4! with a sma pus  Guman 1 3  gxh6 1 4 3 d6?? This is bad defence ternavey a) 4.e8  5 xh6 d6 ( 5 ..�f8? 6 xe8 and the knight and queen are over oaded)  6 e3 g4 and nw a) 17 g3 eS 8 h3 c5 9 xg xg4 20 g4 g7 2 xc6 xf Ths is te key defensive move afer wch i is Back who deivers te perpea Wiout s move \hte woud have exceent com pensaon for he exchange and probaby a arge advanage 22 xf2 d 3 g3 e5 24 h3 h 25 g3 e5 26 £ d4+ 27 f f4+ a2)  7 g f8  8 h3 d4  9 h6 g8 20 g3 dxc3 21 xg4+ and Back is maed. b) Correct is 4 .. d4! None of he winning atempts works for exampe 1 5 xh6 d6  6 g h8 and now b) White can head for a perpeua wit 1 7 xe7 xe7  8 e4 g8  9 h g7 20 f6 xf6 2 g5. b2) But he cannot o over the op wth  7 e5? h7. Nw Back is cose to wnnng afer 8 xe7 xe7 9 xe7 dxc3 20 bxc3 e6 even though Junior  connues to see 6

somehing good  Whte's posion What exacy a is is hard to e. 1 5 xh6 b8 Or  5..e8? 6 xe xe8 7 xf6. 1 6 g5+? Wite can pay  6 e3 drecy of course But he very young wedish man wsey pays a few checks rst  order o ocae e cor rec wnnng manoeuvre. 1 6  h8 1  h4+ �g 1 8 g5 �h8 1 9 h6+ �g8 20 3 g4 Or 20 ...xb2 2  and i is over 21 g 5+? Wie has no rush The me might pass but when te posion does no change there is nong to worry abou. 2 1   �h8 22 h6+ �g8 23  1  Cassica aacng chess Wte introduces another piece ino the atack and Back can st not bring any pieces into te defence

23  h5 23...xb2 24 e xe 2 xe e4 26 xe4 and Back mus part wth is queen and s bishop wih check in order to avoid being maed nstany. 24 g5+ �h8 25 h3 1 0

ame 21 Pavasovi-Beliavsky Pooroz 1999 1 4 5 2 t3 tc6 3 d4 xd4 4 txd4 t6 5 tc3 b4 6 txc6 bxc6  d3 d5

Scotch Fo nghts The Mn Lne

8 xd cxd 9 00 00  0 g c6 1 1 f3 7 1 2 h3!? Aordng to Gutman ths move gves te an advantag 1 2. . .h6 Or a) 2. e6  ad h6  4 f4 d6  5 a4 b8 6 b was sgh beter for Whte n anto Roman-Boudre Frane  99 . b) 2..b8  b h6 4 f4 d6 5 e2  5 fe b4  6 xd6 xd6  7 a4 f4 8 g h5 s gven b Guman; here 9 e gves Whte a sght edge) 5.e8  6 g 7 6 .. xf4  7 xf4 and Wte s a ltle beter) 7 xh6 .e5  8 ae h6 9 xf6 and Whte won urllo-Ugalde ngo 999. 1 3 f4 h7!? Bla s tryng to manoeuvre the nght to e6 where t  be deall plaed proeng 5 and the ng. t snt for nothng that n Beavss reent boo about non he and s o-author halsn suest ou mprove our worst-plaed pee when no other dea exsts. Also possble was   .e8 4 fe 6 5 b a6 6 a4 a5 6.7 7 xe7 xe7  8 5)  7 xa6 xa6  8 e2 wth a sght edge for Wte.

1 4 f1 g 1  g3 6 1 6 . Or 6 xh6 h4 7 g4 x  8 x f6 9 g xh6 and Bla s no worse 1 6. . .J8 1 7 d 1 .d7

Bla would be worse after 7h4 8 g4 g5 9 xg5 xg5 2 b4 1 8 f f8 1 9 4 g 20 d3 Better was the appaen llogal ex hange 2 .xd7 xd7 2 4 ad8 22 xd5 xd5 2  7 24 e2 and Wte has a sght pul 20 . . 21 b3 d6! 22 h4 22  woud be met wth 227 22.. xg 2 xa5 d6 24 4 xe  25 xe e8 s equa) 2 h4 5 and Bla has the nave. 22 .  Bla ould aso have ted 22 . f6 2 f4 e5 24 g5 hxg5 25 xg5 wth a mess  Probabl he felt nmdated b Whtes queen and the possble atang hanes on e ngsde for example 25 ...e4 26 xe4 and Bas queen s learl unproteted. 23 x c7 24 hxg x 2 h4 f4 Interesng was 25 d8 26 f4 .d6 27 e 5 28  6 29 e2 and the pos on s unlear. 26 gxh6 xh6

27 1?! mpler was 27 5 e8 28 xd7 xd7 wth omplet equaty 27 ...d6 28 7 f4 29 c?! Wte should be more ave After 29 4 e6  xd6 xd6  xd5 xd5 2  he eeps the balane. 29 . ..x7 30 x7 ..g4! 31 4? 

The Fo nghts

Not the coect epone to the theat of d6. Bette wa 31 e d2 32 b e8 33 b4 f8, when Blac ha ome chance to outplay hie in the technical phae. 31    32  d 33 b7 .c8 34 b2 xb7 35 xb7 b8 0

ame 22 LuzYusupov Munih 1992 he annotaon fo t ame ae baed on Atu uupov note n hess Inan 4

1 4 5 2 f3 f 3 c3 c 4 d4 xd4 5 xd4 b4  xc bxc 7 d3 d5 8 xd5 cxd5 9 00 00 1 0 g5 c 1 1 2 hi move mae moe ene than 11 f3 becaue Whe e no eal advantae fom inc doubled pawn on the f-le Now the nht i hean fo d4 and, in ome lne, 3f

11...h Alo poible ae a) 11e8 eem vey natual, and in pacce i a vey elable poibiity. 12 d4 (12 c4?! b8 13 cd cd 14 d4 b6 1 c2 4! ave Blac a bete ame in 1996) AavinBezodov, Kanoda 12..'d6 13 c3 c and now Whie ha to plauible opon a ) 14h4 e4 1  h6 16 e4 h4 8

17 a4 (17 f3? f! ave Blac a clea ad vantae in eo-Winan, Nettetal 1992 the point i of coue 18 ef? �d4+ 19 cd4 d4+ 20 h e3 and Blac win)  7de4 18 c6 d4+ 19 cd4 f 20 f ed3 21 c3 ad8 22 d3 d 23 d4 d4 24 £ ed8 h a daw in Nzotchaach, ambu 1910. a2) 14 a4 4!? (14..d7 1 b3 e4 16 c  17 d4 e6 18 e6 e6 19 ad ve Whte a ttle omethin 1  ' c7 16 h3 e (16£ 17   18 f2  19 f1 'b2 wth an unclea ame i alo poible) 17 e e 8 c6 h3 19 fe1 e1+ 20 e1 e1+ 21 h2 c8 22 d �e6 ith a anc me b) Blac ha ao ted 11...d6 12 d4 and now b) Afte 12c 13 b e7 14 c4! Whie i a tle bete becaue 14..e6 1 c2 dc4 16 c4 c4 17 c4 ant White a vey ton poion in which he can play fo a in withou any  b2) 12c7!? i moe iy and ambi ou, but alo uite poible, 13 e1 (13 h4!? i ineen emembe that 13 ..h 14 h2 4+ doe not wo due to 1 4!) 13. ..d6 14 3 4 1 d2 ae8 16 c3 e4 17 e4 e4 18 �f4 d7 19 �c7 c7 20 c6 e (Blac obviouly ha compenaon fo the pawn, but i it enouh?) 21 c3 f 22  h3 23 e2 h8 24 ae e7 2 d 'f6 26 d7 h with a hn ame, feoolov, Potooz 1993 b3) 12...d7! 13 f3 �e 14 c3 'b6 1 fe 4 16 'e3 fe8 and Blac i no  woe 1 2 h4 .d 1 3 d4 c5 1 4 f5 xf5 14..e 1 c3 f 16 f 'd6 17 �3 �3 18 h3 fe8 eache a poion imla o the ame and one pica fo the 11 e2 ne Pobably it i moe o le eual, but I denitely pefe the blac ide. 1 5 xf5 b8

Scotch Fo nghts The Mn ne

1 .e was also eogh for eqaity bt st Black asks a mpotat qeso 1 6 b3? This is a serios weakess of the qee side  the kight  d a vey valable sqare o c3. Black also has some iave after 1 6 b 1 b4 1  g3 g3 1 8 g3 b6 9 b3 d4 so 16 f6 f6 7 d fd8 18 f3 e 1 9 ad1 ith approi mate eqaity was pobaby better  6  5 17 b1 "d 6  g3 xg3 19 hxg3  The weakess of c3 secres Black a clear edge. 20 1 x1 + 21 "x ] 22 "d2 " 5

The qee eeds to take cotrol of the dark sqares 23 4 "c 3 Blacks qee ca ict more damage so he avoids echages 24 n g6 25 d3 �g7 26 3 5 27 g4 c4 2 bxc4 g5 29 c7 fter 29 g3 c4 30 f d 31 d1 f4 there is o defece agaist the vasio ith  e2. 29  "d4+ 30 �h2 A vey ce g ie is 30 h1 h 31 gh h 32 g3 e! (ctg off the qee) 33 g (33 �g2 g3! 34 �g3 f4 3 �h3 g4 ad Wite loses his qee) 33 .. g3 34 g3 e 3 �h2 or 3 g2 g 36 �h2  3 �h3

h 38 g4 d) 3 ...f 3 6 h3 h 37 g4 d 38 e4 h4 39 �g f6 mate 30  h5 31 gxh5 "h4+ Bt ot 31..h? 32 g3 ad o direct  is apparet 32 �g txh5 33 "b6 After 33 cd d 34 �h2 h8 Black has a ig attack. 33 tg3 34 cxd5 34 b1 dc4 3  f4 aso leaves Black i perfect cotroL 34 tx  Nice ad cea f 34.e2? Wte  be able to give may checks after 3 . 35 x1 "b4 Yspov is kow for is great eperse i the edgame. The qee has doe her destrcve deeds ad ow she is ready to be echaged. 36 xb4 xb4 37 b5 1 + 3 �2 d1 39 d3 �6 40 �2 g1 41 �2 c  42 d6 White caot save the edgame 42  �6 43 c4+ �xd6  x7 xc2+ 45 �g3 �5 46 b3 d2 47 g �d4 4 b3 �3 49 c4 d4 50 b5 �d2 51 �h3 � 52 �g3 d5 53 6 d2 54 c4 c2 55 b3 2! 56 �h2 d2 57 �g1 �2 5 c4+ �3 59 b3 �4 60 �h2 d6 01

ame 23 Crisansen-Gelan Munih 199 1 4 5 2 t3 t6 3 tc3 tc6 4 d4 xd4 5 txd4 b4 6 txc6 bxc6 7 d3 d5  xd5 cxd5 9 00 00 10 g5 c6  1 t2 g4 Ths is ormaly cosidered the mai lie bt as we have see Black has may good opos after 1 1 e2  2 c3 d6 ess accrate is 12 ... e7 13 a4 d7 14 d4 e4 (14...c8 1 Jfe1 d8 16  9

The Fo nghts

h6 17 f4 wit a sight edge  Gutma)  5 �xe7 xe7 ad ow \hte ca keep some pressure wth 16 b3!. 1  2 Alteratvey a) 13 h3 �xe2 14 xe2 e 15  ad ow 15..b is the easest way to equality, eve tough Black was also e after 5h6 1 6 �e3 d7  7 �c2 e5 1   g, Ato shin-holmov, UR 957 b) 13 a4!? �xe2 14 �xe2 h6 (4 e 1 5 �d3 b 16 ae 1 e5 17 xe5 xe5   e c7 was equal in Bykovsky-MMler, Groninge 991) 15 �h4 g5! (Black kes over te iniave, obtaining the typical paw fomao ad the favourable kght agast bishop) 6 �g3 �xg3 7 hxg3 b6  �d3  19 ae1 xe1 20 xe e 2 xe xe was a ittle better for Black, who ma aged to wi i Kotsur-Barkhage, Dusburg 992 1   h6 1 4 h4

1 4  d7?! This alows White to obtai a favourable exchage of bishops As we have see aga ad agan, the d3-bhop is te problem chld of the family. The rght approach was 14.�xe2! 15 xe2 (15 x e2 b  6 ae g5!  7 g3 e   d2 xg3 9 hxg3 d6 wt the threat of simply exchagig rooks is upleasat for \hte) 15 ...g5 16 �g3 �xg3 1 7 xg3 d6 ad Black is e 1 5 h7+ 50

The kig ca ofte d itsef i a worse posio o h tan o g. We remember how much such a check ca matter fom Kotovs classc Thnk e a Grandmaste  5  �h8 1 6 f5 xf5?! Better was 16..c 17 d4 g4 (17e4!? I �xe4 xe4 19 xe4 c5 wth compesao has bee suested by Gut ma  t s truy easier to sacrice other peo ples pieces  but \hte is probably better here)   h3 e5 1 9 f4 c4 2 ae c5 21 �xd7 xd7 22 f5 ae 23 b3 b6 24 xd6 xd6 25 'f5 c7 26 .£ wth a sght edge for \hte  7 xf5 8 After 17e  d4! \hite is doig wel. The folowig tactcs are importat  e4 9 f3 d2 2 fe1 s 21 xc6 xh 22 h f4 23 3 c7 24 b4 ad \hite wns  8 x8 x8 1 9 xf6 gxf6 20 td4 Now t is hite who has a good kght against a less effecve bshop. 20  f8 2 f 5! 22 �f 5 This march of the a-paw brigs Black lit tle reef Black had the chace to go ito a rook edgame wt 22 .. c5! 23 f4 xe1 24 xe �xd4 25 cxd4 b 26 e2 b4 27 d2 h7 Ths should be possible to draw because of the acve ook 2 g 4? Black plays wth the pressio tat he has all the e i the world. Necessar was 23 .. c5 24 ! (24 f4 xe1+ 25 xe xd4 26 cxd4 b 27 b3 b4 would act ally jus .a7-a5 ad Black should daw easy) 24 . e6 25 h4 Xce 26 f5, whe \hite has some advatage, but ot much 24 f4 x1 + 25 x1 The edgame clearly favous \hite Now he brings te kig to the queesde to protect his oly weakness. Black should probabl have ied to do something acve qucly but instead he remais passive ad loses without a ght 25  h5 26 tf5 f8 27 2 �h7 28

Soth Fo nghts The Mn ne

 1 g6 29 th4+ h7 30 d 1 Jb 31 c2 c5 32 d3  33 c2 d 34 b4 b3 35 b3 c5 36 Jd2 c4+ 37 bc4 dc4+ 3 2 J+ 39 3 c5 40 t5 1 41 d5 1 + 42 2 2+ 43  1 b6  b5 7 45 b7 2 46 7+ g6?! 47 td6 h4 Or 7..f5 8 xf5 xh2 9 g f6 50 'xc and Whit wins 4 5+ h5 49 Jg7! hg3 50 hg3 b 51 t! 5 52 Jg6 c3+ 53 1 1 + 54 g2 1 0 'g7 mat cannot b dayd much ongr.

ame 24 Lauier-Sokolov Coze 199 I hav rfrrd to Jo aurs annota ons in he Iant 54 in ordr to anays ts gam. 1 4 5 2 t3 tc6 3 tc3 t6 4 d4 d4 5 td4 b4 6 tc6 bc6 7 d3 d5  d5 cd5 9 00 00 10 g 5 c6 1 1 t4!? Ths mov and gam wr th rasons for th nw attnion pad to h cotch our nghts vn yars ago. Th main da is to pay for th c5-squar and mayb gan a vry sght pu n my opion this shoud not  gv Back caus for conc but s ts s t most promising pan hr

1 1      7?!

This mov was atr rpacd by bttr op ons hat promis quai (s Gams 2528) Othr movs that hav bn d in cud a) 11.8? 12 c (Whit wnt astray aftr 12 c?! d6 1 h c5 1 c2 b7 15  7 16 h h6 17 ad ad8 18 f g5! 19 f2 g7 and Back was bttr i Rubvsy Zor, Disburg 1992) 12..h6 1 h b8 1 c1 6 (1d6! 15 b ads to Gam 28) 15 cxd5 xd5 16 c! 16 b d6 17 g b 18 a b8 19 d xg 20 xg a 21 f h5 22 x& x8 2 d2 was vn i Mosamsky, Maia 1992) 16 ..6 17 xd5 cxd5 18 a d6 19 b and Whit is a itt bit bttr  utman. b) 11.6!? 12 c 7 1  wou ranspos to th gam but wthout th un pasant 12 c. 1 2 1 Latr it bcam popuar to si th mo mnt stan wih 12 c! 6 1 c1 �a5 1 d2 �c7 15 cxd5 xd5 16 c ad8 17 xd5 xd5 18 'c and Wht was a tt bttr ZifroiOischuk Budapst 199 1 26 1 3 c3 c7 1 4 3  Aso n is 1c5 for xamp 15 f5 a8!? (15.c6 16 b a8 17 c xf5 18 xf5 xc 19 ac h6 20  cxb 21 axb  woud gv Wt good, acv pay) 16 x6?! 16 b is bttr) 16..6 17 x6? ' and Back wis. 1 5 3? 15 xf6 xf6 16 'c5 d6 17 'x6  with quaiy and 15 b 'd7 16 x7 x7 17  a 18 'c 'xc5 19 xc5 a7 20 �d6 c7 th th sam assssmnt wr both btr 1 5  g4?! 15g 16 � xg 17 hxg s qua but srong woud hav bn 15a5 16 d 16 xf6 xf6 7 c2 d is unpas ant) 16..c5 17 c2 ad8 and Back s btr  autir 1 6 4 16 x6?? oss to 16�xh 51

The Fo nghts

1 6. . d7 17 2 c5 18 b3 f6 19 h3 d6! he simplest and safes Afer 19 .. d4!? 20 cxd4 d5 21 'e3 c6 22 e5 ad8 Black has good play, but 20 c4! ves Whte the advanage. he ie 9...c4 20 c2 f5 21 xf5 'xf5 22 e5 also looks bete for White

20 1 c7?! After 20.xf4 21 xf4 ac8 the posi ton is completely even 21 g5 d7 22 h5 f8 23 d2 Or a) 23 b2?! ab8 24 c2 d7 25 d3 woud only bring he kgh rouble afer 25..xe2 26 xe2 c4! Back has a queenside attack b Best was the stereotpica 23 c4! d4 24 e4 ab8 25 b2 h6 26  d7 27  when Whte s a itte bt better. 23 . d7 After 23.c6?! 24 c4! a6 25 cxd5 �xd3 26 dxe6 xe6 27 e3 e5 28 g4 e4 29 b2 d5 30 f3 Mae8 31 c4 Back wi have to ght ong and hard i o de to reach a draw Whte s also doing wel after 23..ac8 24 b2 foowed by c3c4. 24 f4 t was possibly bete to bing the knght no he game agai with 24 b2!? 24 .c4 25 c2 f6 26 f5 b8? Overlooking a combiation but so does Whte 26..g6 was both necessary and 

enough for equat after 27 xe6 xe6 28 h4 £!

27 h1 ? Bad calculaion. Also wrong would be 27 xe6 xe6 28 f5  29 xh7 cxb3 30 axb3 xf4 31 �h f8 32 xe6 xe6 33 xe6 'xe6 34 xf4 xb3 35 h  36 �a8 �e4 37 xa7+ g8 38 d6 b1 39 h2 £1! wth a ely draw as a result Best was 2 xe6! Mxe6 (27..xe6? 28 xh7 f8 29 g6 28 xh7! xh7 29 xe6 �xe6 30 xe6 d8 31 bxc4 f8 32 Me2 dxc4 33 b2!  Whe wi wi the c pawn eventally and have a very pomising endgame. 27 .. .xf5? 27 .. ! 28 xe6 xe6 29 xd5 xe2 30 x+ x 31 xe2 cxb3 32 axb3 xb3 woud suddeny give Back chances in he endgame. 28 x8 b7? One senses me rouble around hee Af ter 28.g6 29 �xd5  30 xf xf8 31 d4 d8 32 e3 cxb3 33 axb3 xb3 34 c5 the endgame wi end n a draw. 29 f3 g6 30 8 cxb3 31 xb3 c6 32 f5 f7 33 f4 Cleaer was 33 e3! a6 34 �e2 b5 35 a d7 36 �g4 h5 37 �d1! b8 38 c4! �d6 39 .g1 xa7 40 cxb5 xg1 41 xg1 axb5 42 b2 'c 43 h1 ' 4 �e2 �xf5 45 'xb5 and White has a coseto winnig poson.

Soth Fo nghts The Mn Lne

33 .. .d7 3 xd6 x8 35 .xf8? ng th lat chanc t iz  inia v. Aftr 3S g3 c8 36 cS 8 37 6 x6 38 6 x6 39 7 8 0 xf6 7 1  Whit a a lang a vantag an g chanc f crang a mat ing attac. 35 .  :xf8 36 b 7 37 g3 x 1 + 38 x1 c8 39 7 d 0 cxd c1 + 1 �h2 f 2 �h1 f1  3 �h2 ��

7 with a rng iiav, Rublvy Gavrilv, t Ptrburg 1993 1 c3

Game 25 Arkipov-Zaisev Moow 1992 Th anntan  ti gam ar ba n gr aitv anntan in th b eets m ssa.

1  5 2 t3 tc6 3 tc3 tf6  d xd 5 txd b 6 txc6 bxc6 7 d3 d5 8 xd5 cxd5 9 00 00 10 g5 c6 1 1 t h6 Ti  th main cnnuan Blac  nt pay paivly but thratn amngt thr tng t play ..g7-gS an crat cun trplay 1 2 h d6! Thr  n quaity fr Blac aftr 12 6  3 c3 7 14  8  S c2 7 (r IS.. b8 16 4 a  7 3 wi an atac 16 x7 x7 1 7 3 f6 1 8 b4. 1 3  1 13 c3 with th ia f c2 an 3 i  lw  Blac ha n prblm a : 3.8 14 b S I S   a 1 6 a3 axb  7 axb g 18 c2 6 19 g3 xg3 20 xg3 7 21 �2 :x+ 22 x gS 23 c2 g7 2 �2 "f6 2S  1 S an Blac ha g canc f crang an aac but nta ffr a raw n annin-Timman Elta 998 1 3. . .b8 uch bttr an 1 3. ..cS? 14 b3 (ti tp Blac pawn 1 ..6 I S  b8 16  b 17 g3 xfS 18 �xfS g6 19 f3 g7 20 a xg3 2 xg3 aS 22

1 g!? Anthr gra ia cpright by t fa mu gr Ztv (mainly nwn fr i wr wit Anay Karpv n i anna n h writ that h fun thi mv vr th bar 1 5 c2! Th mt bviu quar but thr mv rquir ur atnn t a IS 2 'c7 1 6 g3 xg3 1 7 xg3 �a 1 8 c2 cS 9 ab 1  (an imprvmn vr 9 b3? 7 20 b2 4 21 c4 �xc3 22 xc3 xc3 an Blac i bttr 19 f8 20 x& x8 21 b4 cxb4 22 cxb 'a6 an Bac ha ufcin cuntlay b I S 'xg4 xg4 (but n I S.xh 16 f xg4 17 x8 8 8 g3  19 2 f6 20 g2 3 2  an Wit wn 6 x8 8 (th pin i qual 7 h3 S 18 c2 8 19 b3 7 20 x& x8 21 f g6 22  x1+ 23 x1 g7 2 3 f6 _ N.aurtn-Nrn Dnmar 1 996 c IS xf6 x 6 x8 xa4 17 7 x7 18 x7 xb2 9 xa7 c2 an Blac i btr t ha mr wan an Blac r ar air t acvat 1 5.g5 1 6 g3 th5 1 7 xd6 Or 17 fS xfS 18 xfS xg3 19 hxg3 8 an than  th night n a4 3

The Fo nights

lak is sghtly better 1 . . :xd6 1  3?! n a later game ite fun 8 b4 f4 9 5 be8 20  8 2  f6 22  7 23 g3 e6 24 e6 e6 25 4 g7 26 e3 f5 27 ae e3 28 :e3  29 a4 "6 30 e8 an White ha sme nia ie sratesuatse Rmana 992 1  1 9 d2 fter 9 ae e3 20 e3 f4 2  4 22 4  23 f5 4 24 3 5 25 3 e8 lak has sufient un terlay 1 9. . /f 20  x3 21 fx3? s me is ery unnatural  nw ite has sme serius weanesses n the kngsie fr itle reasn. Naural an g was 2 "e3 4 (2"6 22 e6 e6 23 5 e2 24 e2 e 25  3 26 xe6 is kay t) 22 e4 22 4? b4 23 5 8 24 b3 a with a srng iniae) 22  f5 23  h8 24 4 4 25 4 8 26 a3 4 27 b3 2 28 a2 an Wte hls hs wn 21 . ..6 22 2 f5 23 f1 fter 23 3  24  e5 laks aantage shul nt be uneresate  Wte has n unterlay. 25 h3 h3 26 h3 g3 27 h xh3 28 g g4! wth ieas f ..g4-g3 an eseiay  g5 ges lak a wnning attak 23 .. .f? n my inin hs is t early mler was 23 ..."e5 24 b2 f4 25 ef4 f4 26 h e8 27 3 'e2 28 'e2 e2 29 f4 2 an lak  mst ikely win the engame 2  f3 Nw the sin eseres a lt f sru ny 25 d1? s lses by fre. es lk at altea ies a) 25 g3? h3 26 f2 'e5 27 e5 x5 28 b3 g7 an lak is ing high b) 25 g?  26 f3  27 ex5 

f4 28 b3 5 ith grae anger fr tes king ) 25 e5 is he best hane. 25 ..f4 te is better after 25...5 26 g f3 27  f3 28 e3 while 25 .5 26 "5 5 27 h3  28 f5 f5 29 5 5 30 b5 f6 3  6 is rawn 26 xf3 h3 27 g2 f5 28 e4 f 29 h 5 30 5 h3 an lak has an aantage but nt a eisie ne. 25   fxg2 26 xg2 Or 26 f  27 g2 f4 28 '2 e4 an Wtes king is  big truble. 26 .. .xd1 2 xd1 tf 2 'f3 g6 29 1 g 30 g3 th3+ 31
32 .. .x+!! 01

ame 6 Malakoviner Balatonbere 1995 n hs game we sha ne again see lak in with nigh ersus ght-square bish, althugh when mare t Game 22 it is less lear that lak was better a the way thrugh 1  5 2 f3 6 3 3 f6  d xd 5 xd  6 tx6 x6  d3 d5  xd5 xd5 9 00 00 1 0 g5 6 1 1  h6 1 2 h d6 1 3 1  3 4 is nsiere in the net game. 1 3. . 1  3 5!?

Scotch Fo nghts The Mn Lne

nother good move 1 5 2

2 9. . .d 4 3 0 de1 8 31 f4 gxf4 3 2 gxf4 g4 33 f1 6 34 d 1 'g6 35 h3

With the hrat of f6.   e6 6 c2 g  7 g g  8 hg 'c7 i s eqal  Gman

 �g6 fg6 6  c7 7 g4 is not so cear as after  h 8 g g4 Blac has good coterplay

1 5. . .d7 1 6 b1 g5 1 7 g3 .xg3

35 . .. d3 36 e3?

 7..'c7  8 4 a4  9 a4 fd8 20 a a as eqal in Georev-Noc Brssels 992

Proaly e trole as playing a part here 6  g 7 e wth rher comp caons as etter no Blac taes over the iniave

18 hxg3 xa4 19 xa4 d6 20 d3 �g7 21 e2 b7 22 be1 fb8 23 2

Otherise Whte never get the ishop home to c2.

36 d2 37 f3 d3! 38 e2 d6 39 d1 g3 40 x5?

40 d d 4 d e6 42 d2 f4 sil offered some chances of srival he 40 d 6d 4 cd g4 42 h d 4 h f4 leaves none.

23 .. .�f8?!

Blac cold also play for a da ith 2.2 24 e7 ith the threat of �g) 24  2   26 'g f8 27 'h  and Whte has nothing more than perpetal Bt Blac chooses to ght for an advantage. Proaly White is etter after his t n pracce this is somemes part of a g strategy Rsy t f yo need a w, yo need a w Hoever 2..c4 as proay etter in order to  the -pan or mae sre the c pan e ea. 24 b3 b6 25 f5

Planning e6 in some positions 25 .. d7 26 'd3 8b7 27 e5 d6?

27..c7 old e etter 28 4 7 29 d1 ?

29 cd d 0 'c gives White some advantage

40  .xf3

he siple ay to w 41 xf3

r 4 c g7 42  �e 4 �f e 4 f2 g4! 41 e1+ 42 �h2 d3 43 8+ �g7 01

fter 4 d g4+ 4 h f every thig is lost

ame 7 Goluev-Malaniuk Alushta 1994 1 e4 e5 2 f3 6 3 3 f6 4 d4 exd4 5 xd4 b4 6 x6 bx6 7 d3 d5 8 exd5 xd5 9 0-0 0-0 1 0 g5 6 1 1 

The Fo nghts

a h6 1 2 ih e8 1 3 c

his mov bcam th ashion latr r jcg c3 as bing too slow. n my opinion Wht should not obtain any advantag atr this but sll som playrs hav good practi cal rsults with his ln

Wht rsignd in dlr-alanuk t P trsbug 1993 Bttr than th txt is 16 bS 6 1 7 x6 x6 1 8 xdS xdS 19 xdS x2 20 a1 S 21 d3 with complt qualiy. 1 6.  .ib7 1 7 d2 b6

aking conrol o l th dark squas  7 d6 1 8 g3 xg3 19 hxg3 d4 20 bS 'dS 21 3 x1+ 22 x1 'xa2 23 7 would giv Wht xcllnt play or th pawn. 1 8 ig3 xg3 1 9 hxg3 e

19d4 20 S x1+ 21 x d8 would giv Black a plasant poson too 20 ixe dxe 21 ad1 ad8 22 c1 xd 1 23 xd 1

1 3. . .d6

13  b8 14 c1 d6 tansposs to Gam 28 tr 13  6 hit can obtain th ad vantag with 14 cS 7 15 c1 aS 16 c2 (Gutman) 16..ad8 17 2 gS 18 g3 4 19 x4 dx4 20 d6 6 21 �c2 1  cxdS

hit also has no advantag atr 14 c1 4 15 c2 �d6 (1 S  g4 16  6 should also b okay) 16 x6 x6 17 cxdS cxdS 1 8 3 xh 19 xh2 h4+ 20 g1 xa4 21 xdS 6 22 '4  a daw is vry likly 1 . . .cxdS

 cannot s why ht should  bttr hr. 1 S c3

Or a) 15 c1 4 16 c6 (16 cS d 6 17 g3 xg3 1 8 hxg3 d7 19 4 ad8 20 a3 S wth quaity) 16gS 17 g3 d7 18 a6 and th posion  s balancd b) 1 5 bS S 1 6 c3 also looks n or Black atr 16b8 1 S. . .eS 1 6 e1 ?!

16 xdS 'xdS 17 x6 b7 and 6

Whit has not bn vry succssl in th opning and now has to dnd but whth Black is br is hard to say.  th potntial whit waknsss ar on dark squars so proably it s not so much o a problm. 23 . . aS 2 c3 b 25 d2 e6 26 a3 b6 27 b a6 28 d8+ �h7 29 d7 xa3 30 xb7 xc3 31 xe % - %

ame  Peersen-Krusiov Bled Ompiad  hs gam is includd to show how Whit can us th cotch Four Knights as a wapon. Whit rcivs no advantag hr ut his prsistnc lads to winnng a pawn and  gam 1 e e5 2 f3 c6 3 c3 f6  d exd S xd b 6 xc6 bxc6 7 d3 d5 8 exdS 0-0 9 00 cxdS 1 0 igS c6 1 1 a h6 1 2 ih e8 1 3 c .d6 1  c 1 b8 1 S b3

15 a3 4 16 c2 d6 17 g3 g4 gav Back a slightly bttr gam in ouhk Norr Hsnki 1993 1 S. . .ieS

Or a) 1S6 16 cxdS cxdS 17 c b 4 18 x6 x6 19 x6 6 with quality 

Scoch Fo nghs The Mn Lne

rua-leksadrov, Batm 2002 b 15f4? 16 Jc2 (16 g3? g4 17 �c2 xg3 18 hxg3 d4 19 �d2 d6 20 ce1 c5 21 �a5 ec8 22 b2 d7 s sligh beter for Black Rublevsk Obukhov urga 1994 16..d6 17 c5 d8 18 e2 wih a eve game te would be better if hs kight were o d4 but by he tie it arrives there, Black  have estabished couerpla.

26 nd1 g5 27 'e3 f6?!

Blacks best draig chace is 27xe3 28 Jxe3 c2 29 c3 Jc8 30 d5 f8 31 a4 Jb2, wth some couterpla. 28 5 'b6 29 nd5 g6 30 a3 na8 31 b4 a7 32 '3 ae7 33 ne3 f6 34 'd4 f7 35 d6 '7 36 a4 f5 37 a5

16 g3 'd6

so eual is 16..xg3 17 g3? d6 18 cxd5 cxd5 19 c5 1 7 ne1 g4 1 8 '2 xg3 1 9 hxg3 bd8

fter 19 .. d4? 20 c5 �d5 21 c4 �xc4 22 xc4 Jbd8 23 b2 ite s a lite bit beter but Back coud also have plaed 19 ...b4 20 �c3 �xc3 21 xc3 d4 22 a4 d7 th compete equalty Whte has but up a ig advatage ad, despite a few stmbles, gets the l poit i the ed 37 .. .h5 38 3 b8 39 b6 'a8 40 a6 h4 41 e3 ne5 42 nd6 'a7 43 d8?

hy ot 43 gxh4 here?

20 3 e4 21 'd4 f5 22 xd5 xd5 23 xe4 dxe4?

fter ths Whte has a sght pull Better was 23  e4 24 Jxe4 (or 24 xa7 d4 25 �a5 xe1 26 xe1 c2 wth good com pesaio 24. dxe4 25 �xa7 d2 26 Jf (26 �c5?? loses to 26..c8 26e6 with compesao for he paw 24 'xa7 d2 25 5 8?

Ths shows that Black is drifg After 25 g6 26 a3 (26 �c2?? e3 ad s 26 c3 d3 27 Jed 1 b2 with pla 26 .Jd3 27 b5 f6 Black has pa for he paw

43  .'7  d7 'b6 45 f4? 5e7 46 nxe7 xe7 47 gxh4 '7 48 ng3 h5 49 'd5+ f7 50 'd4?! h5 51 'd2?! e8 52 n3?! 'e7 53 'e3? 'xh4 54 b5 nd8 55 1 'f6 56 b1 'b6 57 �f2 f7 58 xe4 'a5 59 '3 a4 60 d 2 'xf4+ 61 f3 ne8 62 e1 a8 63 d4 b8 64 5 a7 65 1 e6 66 e5 b6+ 67 e3 'xe3+ 68 �xe3 d7 69 b1 xb5 70 xb5 xa6 71 xf5 ]a2 72 g3 �h7 73 �f4 �h6 74 �g4 a4+ 75 f4 a6 76 e4 g6+ 77 �h4 f6 78 g5 f 79 e6+ g6 80 h 3 �g7 81 f4 �f7 82 na6 nf3 83 �g4 b3 84 xg6 nb5 85 g5 b1 86 f5+ �g7 87 �g5 nb6 88 g4 a6 89 h5+ �h7 90 f7+ �h8 91 nd7 a4 92 ne7 b4 93 f4 �g8 94 e6 e4 95 �f5 a4 96 g5 a5+ 97 �f6 a8 98 d7 e8 99 d8 �h7 100 g6+ 1-0 7

Th e Fo ngh s ar

te as o cace of geg a advaage solely fom is ow sog play i te coc Fou igs    is ot dageous fo Black, uless e plays   e to avod e uina o of is paw sucue we e sould be at leas mildly cauous ad peaps cao cou o complete euality   e2 s adly ayng o be afad of ad Blac sould ake it easy ad develop Lutz-Yusupov sows a ealty oad o equiy fo Blac ad maybe eve a advaage fite s o cael eoug Fiy   a4!? as bee tested fo moe ta e yeas ad te cocluso as bee te same amos all te tme afte  6 Blac as o poblems euaisig e sould be a ie cael as see  te last game of tis cape bu all i al is poso is e 1 e4 e5 2 3 c6 3 c3 6 4 d4 exd4 5 xd4 b4 6 xc6 bxc6 7 d3 d5 8 exd5 cxd5 9 0-0 00 1 0 g5 c6 1 1 a4

  e2 6  Game 22 g4  Game 23   f3    d6  Game 19   e8  Game 1   6  Game 17 e7 2 fe  Game 20 2 3  Game 21 11...h6

e ()  Game 24 1 2 h4 e8 ()

2 d6 3 e  b8 4 c3 4c5  Game 26 4g4  Game 2 1 3 c4 d6 1 4 c1 ()  Game 28

4 cd5  Game 2

1 1 .  e 7

8

1   Je8

1 c 

CHAR VE Scotch ou Knghts: Sdelnes afte 7 d5   

1 e4 e5 2 f3 c6 3 c3 f6 4 d4 exd4 5 xd4 ib4 6 xc6 bxc6 7 id3 d5

In ts cpt w  nvstgt som mpotnt sdns t  ...dS n  cotc Fou Kngts s ncluds 1 .. nd 1...J (Gms 29-32, wc  wot wl ltnvs to 1..c. El on t cn dvg wt 1 bS (Gm 33, wl t 8 xdS Blck cn tow n n ntsng cck wt 8 ..?

ame 9 Spanener-Yemelin Bralava 1993 1 e4 e5 2 c3 f6 3 f3 c6 4 d4 exd4 5 xd4 ib4 6 xc6 bxc6 7 id3 d5 8 exd5 0-0 9 0-0 cxd5 10  g5

s s o cous on o  stndd pos tons n t wol o t ou Kngts. H 1..c nvsgt n t pvous cpt s t most common mov 1 0  ie6

A snsbl-lookng tv to 1..c. Blck dvlops  pc nd lnds suppot to t dS-pwn. 1 1 e2

s s not t most dngous mov. 11 ' s trtd n t nxt gm nd 1 1 bS

n Gm 31. 11h6?

t s ntsng tt ssy myslov trd 1 1..g4? gnst Axnd t Hstngs 1954/55 nd t 12 c3  13 c2  14 3 cS  cd n qul poston. 11...  s lso bn plyd, o xm pl  12 4 g4 13  .d 14 x (o 4 'd2 c wt ql py 14. . .x 15 xdS Jxb2 s qul. b 12 d4 Jd (ccodng to Gutmn ts s  mstk  gvs 12...'d 13  8 4 '3 cS 15 x  1 3 wn Wt s n dg 13  (13 x? xf6 14 's g6 15 'xdS .a4 1 6 d8 xd8 gvs Blck  good gm 13 ...b8 nd t s d to s wy Blck sod b wos Anot opon s 11..d? 12 d4  13 J cS 14 x  15 '2 S 1 Jg c4 (1 ..b8? 1 c4 gv t t dvn tg n ov-Nunn Wjk n Z 1991; p ps Wt s vn clly btt 1 b3 c 1 8 1 c3? wt qul cncs 1 2 ih4 ig4

Blck s to quy good ltnvs  12 ..J 13 d4 Jd 4 S xS 15 Jx S b8 1 b1 b4 1 Jx x  qulty ws Oco- l Vll Bclon 9

he Fo Knghts

1990 b) 12 .. d6 13 d4 d7 14 5 5 15 5 Mb8 16 b3 5 17 b1 'd6 18 g3 g3 19 hg3 RossttoBolbochan, ar dl Plata 956 h posion is mor or lss lvl, bt  wold prr to play Black hr.  a sr that a plar with prct tchniq, sch as arpov or Andrsson, wold mak 7/10 rom hs posion. R mmbr that atr th possibl chang o th rooks, Black has qn and kight aganst qn and bishop in th ndgam, which is nomay an advantag as thy compmnt ach othr bttr

Ncssar was 2 5 b5 cb5 26 Mb5, whn Wht has actv contrpla  h nvr gts his rooks playng, h wl los

1 3 c3

25 ...d6 26 b5?

Atr 13 h3 2 14 2 g5 (14  7 15 3 Mb8 16 b3 c6 7 c4 wold v Wht a slght ps) 15 .g3 4 Black is q  16 5 8 dos not achv any thing.

hs is bad now Black taks ovr Ncs sary was 26 b1  with th ida o b4-b5 Yo shold always conss yor ss and admt yor mstaks

1 3. . .id6 14 a4 ixe2 1 5 ixe2 b8 1 6 ab

28 'd2 '6 29 a g5 istrats th prs sr monng p on Whit Evn so, that was prrab to th tt

Atr 16 b3 g5 17 g3 4 18 'a7 c3 19 3 b6 Whts qn is ot o play, h knight s strong on c3 and Black has a strong passd pawn n th middl O cors Whit also has a good bishop on  and a passd apawn, bt sll Back s slghy bttr

bc3 Mb 20 b '2 wold hav vn Back a good ndgam. A in a Whit has bn drng and ptng too mch ath in hs two bishops and now h is wors 1 8 ig3 e6 1 9 i3

As  wrot n th not to Backs 12th mov, 19 d6 'd6 is qal bt not a draw. 1 9. . .ixg3 20 hxg3 e8 21 d4 b6 22 d1 c6 23 b4 b7 24 a4 be7 25 bc1?!

26. c5 27 4 e5 28 c4?!

28 .. .d4 29 a5 g5! 30 d2 e4 31 d3?!

ady, ncssary was 31 4 M4 32 M 5 whn Black has a clar dg Now th nal atack sts in. 31  5! 32 g4 xg4 33 ixg4

1 6. . .e8 1 7 d 1

1 7  .e5!?

17...4 wld aso hav bn a vry srong mov bcas 18 'd5 c3 19 60

33.. .x2! 34 x2 e 3 35 b1 d3! 36 b2 4+ 37 g1 e2! 38 ixe2 xe2 39 xe2 dxe2 40 e 1 g4 41 b 1 d4+ 42 h2 2 01

cotch Fo Knghts delnes fte 7 .  . d5

ame 30 Aekine-asker Moow 1914

ateria. 1 2. . .h6

1 e4 e5 2 3 c6 3 c3 6 4 d4 exd4 5 xd4 b4 6 xc6 bxc6 7 d3 d5 8 exd5 cxd5 9 00 00 1 0 g5 e6 1 1 f3? e7

Th i the naural ove, and lealy the bet. Alo poible are: a)  ..jg4  2 jxf6 and Whie n. b) 1..e? 12 jb5! f or 12..jd7? 13 jxf6 gxf6 14 jxd7 xd7 15 xd5) 13 j6 with advantage for White. ) 11..d6?! 12 b5! e7 13 d4 and White ha iproved h poon Move uh a xe6 and f5 look proiig d) 1...6!? 12  xf6 xf6 13 xf6 gxf6 wih equality. e) 1...b intenng 12 b5 je7! with unlear play ut not 12..5?! 13 a3 a5 14 b4 and White i better Guan). 1 2 fe1

Alternatively: a) 12 e2 h6 13  xf6 xf6 14 3 b i even. b) 12 h3 h6! 13  f4 6, th the dea of ..d6, alo guarantee equat. Even 13...5 14 fd1 d4 15 e4 d5 16 jd2 b6 i ne but not 16...b4?! 17 jxb4 xb4 1 g3 when the wo bhop are woth litte, but he pawn and the kng are ightly weak ened. ) 12 ae b 13 d and now 1) 13...6 14 h3 e?!. Guan beleve that thi i an exelent ove gvig equaity, but a ove ke ...e i oete a ign that he piee are le than ideally plaed. Therefore I prefer Whte after 15 h5 g6 16 xe7 xe7 17 6. 2) 13...e 14 b3 h6! and now 15 f4 �d6 i equ, while 15 h4? g5 16 g3 �g4!  worth reebering After 17 e3 �b4 1 xa7 jxe1 19 x7 a 2  xd xa7 21  d7 22 g4 b4 Blak wi

1 3 xh6!?

Thi lead to a draw by fore. The po tion would ao be equal after 13 jf4 6 but not 13 ...d6?! 14 b5 xf4 15 xf4 6 16 7  17 xe6 fxe6 1 g3 and Whte i better  Guan) 14 d d6. 1 3. . .gxh6 1 4 xe6 fxe6 1 5 Vg3+ �h8

ot 15..?? 6 g6 ate. 1 6 g6 Ve8   

ame 31 MaakovYemein Eatenbu 1996 1 e4 e5 2 f3 c6 3 c3 6 4 d4 exd4 5 xd4 b4 6 xc6 bxc6 7 d3 d5 8 exd5 00 9 00 cxd5 10 g5 e6 1 1 b5!

Prae ha hown that hi ove i the tronget. Thi i logial beaue allowing b5 i the only lear drawbak of 1...e6 over 1..6, o if the ove hould n any way be inferior, thi would be the reaon. 11..c5

Blak ould  to iprove on h poon here.  ee no reaon why Blak hould be any wore at a aer 11...e7!? and now a) 12 e2 5 12...e?! 13 f4! 5 14 4 would gran hte a all plu) 13 4 d4 14 e5 i ghtl better for White aoring to Gutan, but it i hard to detet why after 61

The Fo Knghts

14...g4 IS 'e4 g6 16 e7 'e7 when Black has an ecellent posion b) 12 d4 is the loical move Now Black has the foowng choices b) 2. 'd7 13 :e1 ! ab8 14 e2 b6 IS bS! c6 16 a4 e4? (16 .. :e8! 17 e6 �d6 18 ' e6 19 �b3 :f8 ves Black conterplay) 17 c6! and White is wnnng, Hamdoc-Pian, Castelar 1996 b2) 12 d7 13 :e cS 14 fS fS 1S �fS :b8 16 b3 h6 17 f4 d6 18 'd2 �f4 19 'f4 was roghly eqal n Denny Henao, t Marn 1992. b3) 12 d6 13 e6 13 c1 cS 14 fS fS I S  e4 16 e3 'eS is ne; so is 1 3 fS fS 4  b8 I S b1 h6 1 6 e3 e S and 13 h4 cS 14 g3 'b6 1S e6 fe6 16 b3 d6) 13 e6 1 4 h4 'b4 IS g3 d6 16 e 1 'e1 1 7 e 1 g3 18 hg3  wit an eqal ening, Mratov-hekhov Moscow 1975 1 2 a3 a5 1 3 b4!

This is the key idea 1 3. . .cxb4

Wte was a itle better after 13. b6 14 c3! h6 IS h4! c8 1 6 a4 a 1 7 baS aS 18 b  VGeorgiev-olds, Mangala 1992. 14 e1

There s only eqaiy after te materais c 14 ab4 �b4 1S a7 a7 1 6 a7 h6 17 h4 (17 c6 b6 18 �f6 c6 19 'f3 d6 is siiar) 17...'b6 18 f6 'a7 19 '! and now 6

a) 19 gf6? 2 g h8 21 f4 a8! (the qeen needs to be protected  21..:b8? 22 h6 g8 23 h7+ h8 24 f5 g8 2S h7+ f8 26 e6 and Wite is cose to wnnin 22 b4 and Wte is beter b) 19 d6! 2 hS d7 21 d4 g4 22 h4 fS wh a lkely aw 1 4.  .b8

Or a) 14  ba3?! 1S f6 f6 1 6 e3 �b6 17 'h6 (17 d4 �d4 18 'd4 ves White a sma pls accoring to akovich) 1 7...fS 18 a3 f6 19 e1 d7 2 c3! and Wte has a pls  a posion where Black mst defend and can therefore nd tle pleasre in his etra pawn. b) 14...b3 IS f6 gf6 16 'e3 bc2 17 6 fS 18 d4 c3 19 fS ves Wte a strong atack c) 14..h6!? makes a lot of sense, and mght actaly be qte good, even thogh Grischk later abandoned it Ths may have been to do with avoiding opponents preparaion rater tan a lack of condence  the move tself 1S ab4 b6 1 6 h4 and now c1) 16...a6 17 c3 a 1 8 bS gS 1 9 �g3 ab4 (19 . a4?! ops a pawn o 2 a3 'd7 21 'a1) 2 'b4 and Wite has a signicant advantage, HtyaMateny Hngary 1998 c2) 16 ..:e8 17 d2 a6 18 c3 gS (18..aS!?) 19 g3 e4 2 e4 de4 21 e4 d2 22 d2 ec8 with sfcient compensaon for the pawn  Malahov Grischk Moscow 996 Pobably White shold now play 23 c4 c4 24 c4 c4 even togh Black is better  the endgame. 1 5 x6 gx6 1 6 axb4 b6 1 7 d2 �g7?

Later Black improved wh 17e S!? 18 6 fS! with he dea of g7 This pro tects e king mch more than playin ..g7 Frolov-katchkov Rssa 1998 ended in a raw after 19 d6 �c7 2 ae1 'd6 21 g5h8 18 c3

fter ts Wte is better

o th Fo Kng h ts delne s  fte 7   . d

ame 32 Kuera-Oral rerov 1995 1 e4 eS 2 3 c6 3 d4 exd4 4 xd4 6 S c3 b4 6 xc6 bxc6 7 d3 00  00 dS 9 exd5 cxdS 10 gS

1 ..:eS 9 h1

ih h ida of advancing on h kng id wh f4fS 1 9. . e 20 ae1 gS 21 4 hS 22 e2

h no  fS! d7 3 d x 4 x c7 (4c S bS c   h4 7 g3 h3  f h 9 bxc xd 0 xdS and h i co o winnin S  h4  g3 gS 7 xdS xd  c4 d 9 xd  30 x x 3 �f4 hr?  ha ra winning chanc in th ndgam 22 . . h6 23 d4 d7 24 bS

4 f3, imp o kp h prur on woud hav bn br. 24  xe1 2S xe xd4 26 xd7 eS 27 g3 d 2 xdS?!

Pang ik a compur, bu hr i no win Th imp  c d4 9 d3 c7 30 cxd4 b 3 dS woud hav kp ong-rm winning chanc in an ndgam ha i vry uncomforab for Back 2  xc3 29 e7 g6 30 S h5 31 g4 h6 32 gS?

h i pang wh  bu urpriing thr i no win afr 3 x7 h 33  x 34 x g7 3S + h  h gam mu nd in a draw 32 . .. xgS 33 x7+ h 34 c g7 3S 6 x6 36 c7  37 e6 .g7 3 dS g6 39 x+ x 40 g2 g7 41 xg7+ xg7 42 bS   

 0 . .e7?

Or 0...xc3?  bxc3 h  h4 ( 3? g4 3 c  4 f3 S S g3 xd3  cxd3 fS wa v in pi 95) mann-khn, Badn-Badn  d 3  g4 4 " f S  ( S ab 4  x4 x4  7 x4 d4 i qua) S..d   hS ( aS!?) 7 x x   '3 and th poion i vy co o quait, vn hough h i ha h gh advanag of h o bihop. 1  x6

Thi i th drawback of  O 7 1 1 . . .x6 12 xdS

L ffciv i  �hS g  3 xdS b 4 ab (4 �d xd S d   b cS 7 b3 c4   4 b4  9 f3 wa pad n gnanko-Likov, mrovo 99 afr 9..d Back ha vr good compn aon for th pawn) 4 xc3! S �xd xd  bxc  wh a draw n ndra Borg Cuba 99 Afr 7 c4 cS h xra pawn i no oo rvan vn hough  woud invga a mov ik  a4!? wh th ida of 9 bS ab ..b4 poi 63

The Fo Knghts

m ia 1 2. . .xb2 1 3 b1 b8

13 ...5? 14 �h5 f5 15 4 h8 16 f 6 1 b3 woul gi Whit a ry trong iniati 1 4 4

Gutman gi Wit an aantag aftr 14 3 but thi an b bat 14...b (4...6 1 5 b4 x3  6 xh+ xh 1 � g8 1 8 x3 wa Gutman ua tion but n hr it i not o obiou that Whit i btt aftr 1 8..5)

an now a) 5 3 x3 (1 5 ..a3 6 b3 6 1 xh+ xh 18 + g8 9 xb gi Whit an xtra pawn) 16 xh+ xh 1 � f5 1 8 �x3 f4 1 9 g4  0 5 g5 gi an intrng poion with han for both i b)  5 x  .xg 16 xh+   xg 'x (..g5 18 h  �f6 19 �h5 g8 0 5 6 1 f  Wht a lar plu) 8  f5 19 xb �f4 an hr Blak i lo to winning Aftr probably th only mo 0 g om 0..f6 with a trong attak. An if 0 xb8? thn 0  g4+ 1 h1 f3  g  3 1 6 4 f1  1+ 5  6 an Blak win o probably 14 3 i not aantagou for Wt n though it ony bom wor aftr bing too puhy 1 4.  .6 1 5 3

6

Aftr 15 3 h6 16 4  1 �f3  18 f5 f8 th poition i ompltly qua 1 5. . .6?

Blak for a poion whr h ha a pawn l but xnt rawing han t work out w in pra but wa hary nary. Aftr 15. f5? 16  b4 1 f  (1  5 6 18 xb xb 19 �xb x3 i n) 1.a3 18 b3 xb3 19 axb3 .6 Blak ha an xlnt gam

16 2 .5 17 x5 x3 18 xh7 �h8 1 9 x3 x3 20 e4 5 21 d5

Whit of our ha th han n though ty ar not noou 21 .. .�g8 22 1 e8 23 �1 d4 24 h4 �8 25 .b7 e6 26 a4 be8 27 2 b8 28 a5 a6 29 1 g6 30 d5 xb1 31 xb1 6 32 3 3 33 b6 �g7 34 b7 .xa5 35 xa6 xa6 36 xa6 d8 37 g3 g5 38 h5 6 39 �e2 e5 40 g4 6 41 �d3 �h6 42 �e4 d4 43 �5 3 %%

ame 33 nyiukKiselev Zabani Weien 1994 e4 e5 2 3 6 3 3 6 4 d4 exd4 5 xd4 b4 6 x6 bx6 7 .d3 d5 8 exd5 xd5 9 00

Or 9 � (thi o not look o natural) 9 . (for 9 ...  Gam 34-35)

cotch Fo Knghts delnes fte 7 .   d

 well a 10 g5 hie ha alo ed a) 0 e2?! e 11 c3 d6 12 4 c5 3 xd6 xd6 wih the idea o b ad Black i a litle better b) 10 d2?! e 11 a3 g4 2  d6 1 3 4 c6 14 xd6 x d6 1 5 e1 e6!? ad Black i o wore GabizaGarcia Novi ad 1 990  0    g4!

ad ow a) 10 g5 0-0 11 000 e6!? (y ew idea 1 1 ..c6 1 2 he d6 1 3 h3 h6 14 h4 b 15 3 b4 16 x6 x6 1 x6 gx6 wa equal  Zakic-Gligorc Ni 199) 12 he h6 13 h4 b 14 5 b6 15 a4 b4! (1 5c6?! would be awkward ad he would be beer aer 16 d3!) 16 x6 (16 xe6 xh4 1 h3 xa4 1 xe xa2 give Black he edge) 16x6 1 xe6 e6! 1 xe
The o accurae 1 . c5?! 1 1 g5 e6 would apoe o Gae 31 whle 10  c6?! 11 c3 c5 12 d4 g4 13 �a4 d 14 g5 would ve he a ligh edge. 1 1 f3 he houl o play 11 e2? xe2 12 xe2 e 13 e3 c6 14 d4 c 15 5 e4 whe Black i beer due o hi geaer obi 11...d7

 good aeave ha ga equali i 1 1 ..c5? 12 h 1 d 13 c3 h6 (or 13  c6 14 d4 d6 15 g5 e wih couerplay  Gua) 14 e1 (14 b4 b6 15 4 h5! i equal  uza 1 6 e5 e! 1  4 6 would ve Black the advaage) 14c6 15 d4 e 16 h4 (16 'g3 h5 1 2 '6 i good or Black e -bihop i ll o playi 16  (16..d6! wa roger) 1 d2? (1 b3! with equal wa the oly ov) 1 c5! I  5 c4 19 c2 e2 20 ad x5 21 x5 b 22  c5 ad Back wa cleary beer i ahe uza ijk aa ee 1992

9    00 0 tb5

 cao believe ha hi ove  good he kgh doe o ee o be he piece o i eed o iprovee 6

he Fo Knghts

1 2 c3

White w in trouble fter 12 g5 b8 13 c3 c6 14 h e8 15 �c �e7 16 h4 c5 1 7  e1 e3 18 xf6 f6 19 �d f 2 f4 d4 21 2 4 22 f c5, when the frilty of lck pwn trucre i irrelevnt while hi to bihop nd tck ig propect re hghly reevnt. Rozentli Klovn Gronngen 1992 1 2  e7 1 3 d4

te would lo hve no dvntge fter 13 f4 xb5 14 xb5 c6 1 4 d6 6 xd6 ( 6 g5 h6 1 7 h4 b8 i  lo even 16xd6 1 3e8 1 4 e1 cS 1 S fS f8 1 6 xe8 xe8 1 7 e3

ite could hve tried 17 c2 but fter 17...c7 18 'd3 g6 19 h6 (19 e3 'h4 would give ck ome ttckg chnce) 19...xh6 2 xh6 f5 21 'e2 xc2 22 xc2 'f6 lck keep the blnce

c7 with trong compenon for the pwn. 24 . ..e8 2S xe6 xe6 26 cxd4 c4

26.cxd4 27 x5 would leve White ig nicntly beter. 27 dS?

he deciive mitke White hould hve pyed 27 xe6 xe6 28 xc4 (28 bxc4 b4 29 �c2 e1 3 f1 xf 31 �xf2 �d6 32 h3 d8 33 d5 c8 gve lck he dvntge due to the blockde 28 . f5 29 d3 h6 3 'c2 �d7 31 c4 h7 32 d3 �c7 33 4 nd ite h no obviou weknee, o it i hrd to beieve tht he hould be much wore if wore t  27 .. ..b4 28 dxe6 xe6 29 b2 xe1 30 xc4 d 1 31 f1 xf2+ 32 xf2 c6! 33 f4 c2 34 e3 c1 3S f2 h4 36 fS gS 37 h3 b1 38 f6 c1 39 a3 xb3 40 d2 b6+ 41 �h2 c7+ 42 �g1 cS+ 01

ame 34 Sirov-Karpov Mosow 199

1 7. . .hS!?

7 ...b8 i lo equ. 1 8 d2 g6 1 9 g3

19 h6 look intereting but nlyi how tht it i  mitke 9 ..7 2 b4 c7 21 e f6 22 f4 d6 23 xd6 'xd6 24 g5 e6. he kight i not rely doig nytg good for White t h6 but it remin tck there. 1 9. . .g7 20 e1 .e6 21 e2 d7 22 b3

he nnottion to hi gme re bed on he thoe by exei hirov in hes nan 56.

1 e4 eS 2 f3 c6 3 c3 f6 4 d4 exd4 S xd4 b4 6 xc6 bxc6 7 d3 dS 8 exdS e7!?

22 f4 weken the contro over the centre nd i reted by 22d4 23 f2 x2 24 b4 b3 25 bxc5 xc5 when Blck i better 22. . .a 23 f4

Now lck h n intereing pwn cri ce 23 .. .d4!

23...4 24 b4 cxb4 25 cxb4 d4 26  would grnt White  ight edge 24 f2

ntereng lo w 24 e4 d8 25 f2 d6 26 xe6 xe6 27 cxd4 cxd4 28 x5 d3 29 b6 (29 xd3 c 29d2 3 d1 66

One of Ef ogoljubow ide rt plyed n hi gme gint piemnn t the

cotch Fo Knights idelines fte 7    d

hoiday resort of Pistyan  1922 t is a rather rare line but on the other hand it can be used as a surpse.  trough the 1990s his was former World Champion atoly Karpovs choice. 9 e2

he obvious move. Some uprepared players have been known to play werd alter naves: a) 9 e2?! (some commentators has marked this move th ?, sometng it does not desere) 9 . xd5 (he obvious move  now Whte cannot p wth g5 9..cxd5 10 00 00 is equa but praccal examples are needed) 10  d2 and now a) 1 ..a6 seems to be quite good for Back. 11 xd5 cxd5! (much beter than 11... d2?! 12 d2 cxd5 13 0-0 wit good pros pects for White  Tartakower) 12 c3 00 1 3 e3 (13 cxb4 fe8 14 e3 xb 15 f xe2+ 16 xe2 d4 17 d2 xb2 18 d c5 gives Black a cear edge) 13 .xe2 14 'xe2 fe8 15 cxb4 d4 th equiy in Po gona-Dmovska, Oropesa del ar 2001  a2) 1..xc3! 1 1 bxc3 a6 12 f1 xe 13 'xe2 xe+ 14
and hte was clearly better Veimrovc Lukc, Vjacka Banja  962) 1 1 bxc3 00 12  e8 1 3 h4 e4 14  xe7 d+ 15 e2 x 16 x xe7 17 he f8 wth an even ending in Odeeveksanov Voskre sensk 1993. 9    cxd5

lso ntersng is 9..xe? 10 xe2 and now a) Sufcient is 10..xd5!? 11 xd5 (11 e4 0-0 12 a3 e7 13 c4 b6 14  f4 f5! with euai  Guman. robably it is in his e that Wite should look for an advantage, if there is any to be found 15 b3 folowed by a4 seems to be the best  11..cxd5 12 b5+ d7 13 xd7 xd7 14 d c6! (an mprovement over 14 .. he8+ 15 e3 c6 16 c4 xc4 17 d4 wth a sight advan tage for White  Oll-Beliavsky aila 1992) 1 5 e3 c5! 16 c4 dxc4 1 7 ac1 x e3 18 e3 ad8 19 xc b7 and the endgame should be a draw. b) 10 ..cxd5 wth a rther branch: b1) 1 1 d2 ansposes to the game. b2) 1 1 f4 c6 12 a4 0-0 (or 12 ..e6!? 13 hd 0-0 14 a6 g4! 15 f3  f5 and Black is okay) 13 hd e8+ 14  g4 15  e6 16 a6 h5! t an unclear posi tion for example 17 c7 e7! and Black is dong e. b3) 11 b5 d8 12 d c6 13 d4 and now: b31) 13..c7 14 c4 d6 15 h3 d7 16
Or 10 b5+ d7 1 1 xd7 xd7 12 xe7 xe7 13 d2 b6! (13..f6 is equal, but the knght has more acve oppor 67

The Fo Knghts

nes on e eensde) 14  d8 15 Me f8 16 5 xd 17 Mxd c6 1 8 d4 c5 19 e?! (19  keeping e al ance was stonger) 19.e8  dd Mad8 1  xe 1  xe1 g5 and Black was a tle etter  VGrevcSman Cappele la Grande 1 995

clear poson If 18xd3 te openng of te cle wold expose te cpawn

10  xe2+ 1 1 �xe2

..d7 1 g3 wold e an provement for te as Black s now ed to te defence of c5

11 xe xd+ 1 xd 8 13 3 c5 14 e1 d7 s ea Te ng  e ver we placed on c6.

1 7 ad1 e5 1 8 f1

18 �c5? s also possile. 1 8. . .f5 1 9 d2 h5

Black as eased 20 a3 7

21 b4?

1 g3 f6  3 was stonger t after .g5 Black looks oka. 21 4 22 x4 dx4 23 g3 a5 24 ib6

Sov gives 4 6 c3 5 Mde 6 6 �d4 (6 c4 ax4 7 ax4 �e6 8 �d4 M8! 9 e3 x4 3 �xc3 M5 s etter for Black as te e6isop s clearl speor to te e3gt) 6 Mxe 7 Mxe �e6 8 c8 ax4 9 ax4 7 3 �xc3 g5 31 e7 �d6 3 �f6 �x4 (3 g4?) 33 �xg5 g7 wt nclear pla 24 .. .xe1 25 �xe1 3 11...6

Oer moves look less natra. a) 11.? 1 5 e 13 d1 xd 14 xd 8 15 xc7 e7 16 5 and \te as a clear advantage ) 1 1 .d7?! 1 a3 e7 (1 �a5? 13 xd5 wold drop a pawn) 13 5 and te s somewat ette 1 2 a4 d6 1 3 e3 00 1 4 f3 d7 1 5 �2 e8

15 ... c5 16 c3 c4 17 c 7 18 4 wod gve \te a sgt edge. 1 6 he1

Or 16 Md1 6! 17 x6 ax6 18 �x6 M8 19 d4 c5 and Sirov eeves tat ts poson s sg eter for Black. He proal intended someng ke  �c3 d4 1 �d MX  a4 and now  .. �e6 3 a c4 4 �c1 8 5 �e4 f5 6 �c6 Mec8 7 a4 a8 sees \ite clear strng. 1 6. . .g6

16 e5 17 �c5! �c7 18 4 gives an u68

5 ax4 6 �xc7 xa4 7 ax4 x4 8 �e5 wold e eal. 26 e2 xb6 27 xb6 a6 28 bxa5?!

8 c4 ax4 9 ax4 Ma 3 e3 �e6 31 d1 �c4 3 e4 �a6 33 e8+ g7 34 Ma8 a1 gves eait  Sov Now Black s etter 28 .. .xa5 29 4 a4 30 d6 ie6

33 31 e4 (31 !? Mxa3 3 g4 xg4 33 g3 f5 34 4! gves \ite conter pla for te pawns  Karpov) 31..xa3 3 e3 Ma 33 e 1 34 xc3 x+ 35 f �d7 36 f1 4 37 g4 3 38 g1 g 39 f (Srov) wod proal end in a draw 31 e3 xa3 32 e4 5 33 �e2 g7 34 �d3 g5 35 �d4 g4 36 fxg4 hxg4 37 x3 ix2 38 e5

te s ale to draw ts endgame e case of s conterpla against Blacks weak pawns 38 . .ig6 39 5 a6 40 �e5 b6 41

co tch Fo Knghts delnes fte 7    d

�4 b4+ 42 �e3 b2 43 e2 b3+  c3 b6 45 �4 b2 46 �xg4 xh2 47 xc6 2 48 �h4 i5 49 g4 ie6 50 �g3 d2 51 e4 d3+ 52 �4 id5 53 c1 �g6 54 �e5 % - %

ame 35 Yanemirov-Aleksanrov 5t eterbu 1996 1 e4 e5 2 3 c6 3 d4 exd4 4 xd4 6 5 c3 ib4 6 xc6 bxc6 7 id3 d5 8 exd5 e7+ 9 e2 xd5 10 xe7

16 e ves White a substanal advantage accorg to Karpov while 1c3 13 baS dS 1 00 b8 1 + e6 16 f gave White a clear edge in Czerniak Fenogio Buenos Aires 199 13 e f6 1 b e8 1 00 and Whte is a cerinly better b 11c3  bc3 e6 (1c3! 13 b d  g7 g8 1 e is better for Whte as is 1e8 13 OO! 13 00 (13 a a 1 a d7 1 00 f 16 fe1 d3 17 cd3 he8 was rightly agreed drawn in Afekvkov Wijk aan Zee 000 13f6 1 c b6 1 e c 16 e  17 c6 ad8 18 e3 dS! 19 b c 0 c+ c 1 a7 dS and Black held in LauerKarpov Be 199 White cannot keep the pawn on a3 no mater how uch he would ove to 1..xc3

White had a clear plus after 11c3! 1 bc3 e6 13 c b6 1 b+ d7 1 000 c8 16 c d 17 a6 b8 18 aS c8 19 c e6 0 a6 c8 1 c e6  he1 in Yandovleksandrov Kstovo 199 1 0. . .�xe7

1 2 a3 ia5

lost everybody plays ths but why Why not 1Oe7! instead (theres sl pleny of room in chess for noveles The hreat of b and he seopen be makes this posion hard o assess 11 d 00 1 000 b8 13 d cdS 1 f b7 1 he e6 was equal in omRoanov Kstovo 199. 13 d sees to be innocu ous but how shoud Whte prove 13 e f 1  e6 woud be unclear (unforu natey 1b does not work

his sees to be he only ove although previous pubications have claimed ffer ently a Accoring to Gutan Black has a good game after 1.e8 13 ab f but the to bishops and stronger pawn strctre make this posion clearly better for White 1 f1 e 1 f d6 16  shows how b 1d6 13 c3 e8 1 000  1 he d7 16 h7 g6 17 h and hite has an obvious advantage Frolovhibaev amara 00

1  id2!?

Anoher ove order that leads to a wel known posion This move order seems to be he ore accurate than the noral 11 a3 for eaple: a 11.aS 1 b (ths move was found by M Moshe Czerniak 1b6 (1b 13 ab c3 1 b d 1 g 7 g8

1 3 b4

13 c3 c3 1 bc3 was equal in Tepitzchonau piemannRubintein 19 1 3. . .ixb4?!

ery risky. 13.e8! was probaby he best 1 ba (1 00 b6 1 fe1+ e6 16 69

h e Fo Kngh ts

xc3 f8 and the posion is equal  Gut man) 14f8+ 15 f e4 6 e (16 f4 c5 with equalty is an important dif ference from the notes on move welve;  6 b4 c5 1 7 e d6 is ne too) 1 6g8 17  c5 and Black is no worse 13  b6 1 4 xc3 f6 15 - e6 16 fe  also looks e The game Arkhipov Kholmov Azov 1993 concluded in a draw after 1 7 e2 g4 1 8 c4 g6 19 e 7 f5 2  h6 21 d g5 22 c3 hf8 23 b3 ae8 24 ae xe7 25 xe7 d8 26 f1 d 27 e d6 28 h3 c5 29 b5 c4 3 xc4 xc2 31 c1 a4 32 b4 d4 33 e2 g6 34 c5

cxd6 24 d6 e7 25 c5 woud put Black under pressure) 23 xc6 xd 24 xd c8 25 c5 f6 26 ¥c3 e7 27 e4 would ve Whte a clear advantage. b) So would 22 xd7 xd7 23 xc6 ad8 24 xd7 xd7 25 f when the end game is very unpleasant for Black

14 axb4 d5 1 5 e4!

15 c4? b6 16 f4 d8 17 xc7 xd3 8 xb6 e6 would allow Black to get counteplay

Better was 29 h4!? h6 3  g6 3 c5 b2 32  b  33 d3 b2 34 g3 b3 35 e4 f5 36 c2 b2 37 ¥d3 and Wite sl has a itle pressue.

1 5. . e8 1 6 0-0 �f8 1 7 f3 e6 1 8 c4

29.. .b2 30 c5 g6 31 g4

White has an advantage somewhere be een slght and large  a medium advantage If 2..e7 2 d4 f5 22 xa7 xa7 23 xa7 g5 Black keeps good drawng chances

31 g3 h5 32 g2 b  33 h2 b2 34 xc7 xb4 35 xc6 xc6 36 xc6 h4 37 gxh4 xf4 38 g3 c4 would be easily drawn for Black And after 31 h2? f2 Black would have good counterplay

21 d1  ed8

31 . .. h 32 g5 h5 33 h4

1 8. . le7 19 fc1 lf5? 20 c3 d7?

22  .e8 23 xd8 xd8 24 xa7 ld4 25 xd4

After 25 g4? f5 26 xd4 g4 27 c5+  28 hxg4 d7 the endgame is most probably drawn. 25 . .xd4 26 e2 d2 27 f1 b2

Black has now found some counterplay 28 b7 b1 29 f4?

Maybe stronger was 33 g2 h4! for ex ample 34  d7 35 xc7 b 36 f2 b+ 37 e3 b3 38 e2 b 39 d3 b3 4 c2 x 41 xd7 xf4 42 c3 f+ 43 d3 Whites posion looks good but Black can keep his counterplay alive wth 43.g3 After 4 c4 xg5 45 d6 he poson is not clear at a 33 .. d2 34 g2 d4 35 f5 gxf5 36 xc7 xb4 37 .xc6 xc6 38 xc6 xh4

After 21..e7 22 xd7 xd7 23 xc6 ad8 24 xd7 xd7 25 1 Wte has seri ous winig chances 22 h3?

A sloppy mistake Two alternatives were better: a) 22 e5!? e6 (22.d6 23 xd+ 70

Black is better The daw occurs ony be cause the rook endgame with f- and h-pawns are theorecaly drawn 39 U6 g4+ 40 �h2 xg5 41 c6 f4 42 c7 c5 43 f5 xc7  xh5 c2+ 45 �g1 �g7 46 f5 c4 47 �f2 �g6 48 a5 c3 49 b5 f6 50 a5 e3 51 b5 e5 52 b8 �f5 53 a8 b5 54 �f3 b3+ 55 �f2 f3 56 �g3 e3 Y -Y

cotch Fo Knigh ts idelines fte 7   . d

mmay

One agan t makes no real sense for Whe o avod he man lnes but Blak has an nterest ng alternave n  .. xe7 whh sl onans many unred deas Later Blak should be arel beause as opposed o he noa  0...6 neher 10 ....e7 nor 10 .. e6 an ly gar antee equaty.  e4 e5 2 3 c6 3 c3 6 4 d4 exd4 5 xd4 b4 6 xc6 bxc6 7 d3 d5 8 exd5 cxd5

 . .e7 ) 9 e2 9...xd  Game 34 9 .. .xd  Game 35 9 0-0 0-0  0 g5

10 b   Game 33 1 0 . .  e6

1O . e7  Game 32   b5   Game 31 1 1 '  Game 30 1 1 e2  Game 29

8.   �e 7 +

1 0 

1 1 

71

CHAPR SIX Scotch Four Knights: Sdelines after 5 xd4 1 e4 e5 2 f3 c6 3 c3 f6 4 d4 exd4 5 xd4

hs posion is one with which many e nners are amiar And conrary to most symmetrical systems that egnnes opt o this ie is also played at the highest level. n this chapter we shal invesgate unusual ideas or oth sides his includes (ater 5 . 4 6 c6 c6 7 d3) he interesng 7 .d6? (Games 3637), wch can e used as a geat suprse weapon t is ittle known looks passive ut is oth good and acve ts no a very popular move, ut quie a good ater nave to 7 . d5 which is in my opiion a it overplayed and dul  it is very hard to nd new ideas. Aer 7 . d6 you  have a more dynamc game in uinhiited terrtory. Also making ts chapter is Hedens 5 .. c5? wch is no ly eliale and  nally 5 ..e4?, whch is somewhat duious.

ame 36 Goena-Beliavsky Reio Emilia 199/96 1 e4 e5 2 f3 c6 3 c3 f6 4 d4 exd4 5 xd4 b4 6 xc6 bxc6 7 d3 d6? 8 00

he es move,  g5 is seen in he net game 7

8   g4!

A good aggessive move a) Ater ..c3? 9 c3 Black cannot equalse a) 9  d7 is too slow 1  e3 -  4 c5  2 e5 7  3 h5 g6 4 3 5? (4.e7 5 ed6 cd6 6 ae wih a wte iiave was preerale) 5 e6 6 6 5 gave Whte a clear advantage Baldin-vanov, Chicago 1997 a2) 9  h6 1 4   1 3 e 12 Me ld7 13 e5 lxe5 14 xe5 xe5 15 �xc6 is

somewhat eter or White ) nteresng is  . �e7? 9 e2 (9 g5 h6 1 h4 g5   g3 c3  eter is   ..g4 as n he man game  2 c3 g4 3 �d2 h5 4  e5 15 4 h4 6 2 4

cotch Fo Knghts delnes fte  xd

17 x4 c5 18 ab Jd7 was unclear n otleokrak Bransk 1995) 9 ... Jc5 1 c3 (1 g3? g4 s dangerous: .4 and 54 are comn 1 " 1  d4 Jd7 12 3 6 13 J4 and Wte appears to ae a sgt space adantage.

c) 1 3 6 (1.e5 tansposes to note c2 to Wtes 9t moe) 11 'd3 a wit good counterpla or Black; e cances are approxmatel equal. 1 0 a4!

9 e2

e best e alternaes are: a) 9 J4 6  d2  11 ae1 6 12 g3 'd4 13 a3 Jxc3 14 xc3 xc3 15 bxc3 e5 16 Jxe5 dxe5 wit equat S.Lalcaric, Hasngs 1994/95. b) 9   1 g 3 e5 11 e2 c5 12 J6 g6 13 .g5 6 4 Je3 xe3 15 xe3 e8 was unclear n Nepena Leconte aric galo 1994. c) 9 3 e5 and now c1)   4? xd3 (a bsop s a bsop) 1 1 cxd3 6 1 2 e5 (12 'e d4 1 3 'e3 Jc5 s a lttle better or Black) 12"dxe5 13 e5 'g6 14 e2  15 4 5 16 d4 e4 17 d3 5 8 xe4 xe4 19 .e3 ae8 and Black wo plans .76, s better Grabcs kalcsn Noa Gorica 2. c2) 1 �e2 4 (1" 1 1 a4 a5 12 b3 w a smal adantage or Wte  Gut man)   4 (1 1 .e3? xc3 12 bxc3 xe4 13 b 1  and te dd no ae ull com pensaton or te pawn Ruceaarc at 1995) 11.g6 (11Jx3? 12 e  and Whte wins a pece) 12 'e .c5 13  'e7 (13"xe s equal) 14 'g3 d4 wt an unclear poson 9 .. .e5?

9"5?  d4 b8 11 'xg7 would just drop a pawn but 94? needs to be nesgated: a) 1 xg4 xg4 11 xg4 xg4 s equal or mabe een sgtly better or Black. He as to bsops and can ulse te e and b es. b) Interesng s 1 J4? Jxc3 1 Jg3 e7 12 bxc3 6 wen 13 ? 5 es Black sucent counterplay. Ater al Wtes sructre on te queensde s sigtl awed.

Aer tis Wte  eentualy gan te wo bsops, and ten te 24 pus  secue m a sgt adantage 1 0. . .00 1 1 a3 .a5 1 2 b4 b6 1 3 c4

13 xb6 axb6 14 4 g6 15 Jb2 looks good too, but Wte s plang a srateg o restant were Black ust wonder wen or  s bsop  be excanged 1 3. . .e7 1 4 b2 b8 1 5 d2 c5

15"e8? 16 c3 wins e excange ater 165 17 c5 dxc5 18 bxc5 Jxc5 19 xc5 xb2 as ater 16.c5 7 xb6 cxb6 18 4 te kngt s lost 16 bxc5 dxc5 17 c3 a5 18 d5 d6 1 9 c2

Wte onl as a sgt edge ater 19 xe5 �xd2 2 Jxd6 cxd6 21 ad1 b2 22 b1 . 1 9.  c 6 20 ad1 f6 21 f4!

Black s up aganst t and does not nd te ree e could ae ater 21 b6? 'c7 22 xc8 c8 23 4 g6 24 g4 wen Wte s better, but not b muc 21 . . .cxd5

Not 21..g6?? 22 x6. 22 fxe5 b6

22"e5? 23 x x8 24 xd5 'b6 25 xe5 'b1 26 d xc2 27 xc2 would lead to a wnnng endgame or Wte 73

The Fo Knghts

dcomes nex

35 x f5 c8 36 f1 ! c 37 d1 ?

23 b1 d

Black is given more and more chances Whie ins aer 37 Mb7 Ma4 3 Md h6 39 Mdh7 40 Mdd7

23 . dxe4 24 ex6 gx6 25 x6 �d6 26 Mxb xb 27 c1 ns he exchange and leaves Black in rouble 2 xd c7 25 xb8 cxd!

Trying o complicae maers. 2S ..�xb 26 xcS is hopeless. 26 b3 d3

26 ..xeS 27 MbS 'c7 2 h gves Whie a clear advanage 27 c5+ h8 28 xd3 xe5 29 b2?

 as beer o play acivel Aer 29 MS d4+ 29e7 30 MdS c7 31 a e6 32 Mx �x 33 i xdS 34 exdS ould  or Whe 30 h c7 3 a �a+ 32 b d4 33 dS �2 34 d Back is under aack and don on maera.

37 .. h6! 38 b7 a6 39 a7 a

o i is oo lae  Whie can no longer in he a-pan. 0 d3 b2 1 g3 xa3

o he posion is close o eual 2 dd7 f5 3 d5 h7  xf5 g6 5 d5 f6 6 a8 Ja1 + 7 g2 a2+ 8 h3 a3 9 h8 g5 50 d6+ f6 51 d5 g5 52 d6+ if6 53 a8 h5 5 dxa6 xa6 55 xa6 f7 56 a7+ g8 57 a5 g6

29 ...xc5+ 30 h1 d6

Ths posion s a heorecal da  i is a orress here Whie canno ener  Black has o do is preven he he kng rom ge ng o V. 31 b3?

Whe could s n aer 31 c4 c3 32 b eS 33 a xh2 34 xa7 e 35 �dS b6 36 MV MXV 37 x �d 3 e2 olloed by Md 1, and Back i have o ener he hopeless endgame here he a pan decide he game 31 .. .ic7 32 e5 xe5 33 xe5 xe5 3 .f5 xf5?

Aer 34  a6 35 d g6 36 e4 MV 37 d g7 3 Mb c7 39 e b6 40 3 c7 41 be h6 42 e7 c2 Whe re ans pracical chances o ning bu he posion is probably an objece dra 7

58 a7 f8 59 g2 g8 60 f3 f8 6 1 e g8 62 d5 c3 63 e6 ib2 6 b7 ic3 65 h3 a1 66 b5 g7 67 g5 ic3 68 h e1 69 e7 f2 70 d6 f7 71 d5 e3 72 e5   

ame 37 BezmanVaravin Pe 1997 1 e e5 2 f3 c6 3 c3 f6  d exd 5 xd . 6 xc6 bxc6 7 d3 d6 8 ig5 h6 9 h e6!

er 9 .. 0-0 10 00 Me   4 b7 2

Scotch Fo Knights Sidelines fte  xd4

e2! �e7 1 3 g3 e6 14 5 e5  5 xf6 xf6 6 3 White has stong aaking hanes  Guan. 10 00?

is uns into a sto of anient popo ons Bette is 10 f4 Now thee is a ot of logi n Bak onnung aessively with 10g5!  1 g3 (1  g5 g4 12 ' g8 13 h3 xg5 4 g3 e5 gves Blak a good game)  1 f4 (Blak must at; afte 1.8?! 12 0-0  3 h g4 4 f5! e3 15 f3 xf1 16 xf1 8 7 e5 e stands muh ette eause Blaks ooks an hadly e aled usel hee) 12 xf4 g8

and now: a) 13 O-O?! g4? (3 .  1 4 h g4 ight move ode) 14 h ? (4 a4! a5 1 5 3 e5 16 4  17 x6 ax6 18 a4 with unlea play) 145 15 e2 h4 6 g3 e7 (16h3!? and 16.h5!? ae equy viale opons) 1 d e5?! (17h5! wth a mess is poay ette) 8 xe5 g4 ( 8 ...dxe5 19 4 is ette fo Wite) 19 xd6 xd6 20 g2 and tee is not enough ompensation fo the pawn ) 13 d2 h5!? 14 e3 (14 O-O-O? x3!  5 x3 xf4 1 6 xf4 g5 s lealy ette fo Blak) 14 15  wi an unlea poson But note that 15..g3?? s not an opon eause of 6 OOO! and White WlS 

1 0. . .g5

With is move and ndeed Blaks whole plan, Vaavn s lealy nspied y a ass fom moe than 50 yeas ago Hee it s:  e4 e5 2  6 3 4 5 4 0-0 d6 5 d3 f6 6 g5?! (6 3 6 3) 6h6! 7 h4? (7 e3 7 xf6 xf6 8 3 is level) 7 ..g5! 8 g3

8..h5!! 9 h4 (9 xg5 h4! 10 x7 g3   xd8 g4!! 12 d2 d4 1 3 h3! lak wns afte oth 13 3 f3! 14 gxf3 x3 and 13  x2! 13. .e 14 xe2 xe2  5 e6 6 16 3 x 17 xf d7! 8 g3 ag8 19 x7 xg3 20 e2 g6 21 d5 [2 e6 g4! and Blak wns] 21xd5 22 xd5 f) 9g4 10 3 d7 11 d4 exd4 12 e5 dxe5 13 xe5 xe5 4 xe5 f5!  5 xg4 xg4  6 d3 d5 17 4 OO-O!  8 4 6  9 x5 xh4 20  8 21 g4 e8 (21 .e4! 22 xe4 %h 23  xe4 was quke) 22 e  e 23 �xe3 e3 24 g3 h  25 g2 8h 26  xf1 27 xf1 : 28 xe3 x 29 a4 d7 30 d3 xg4 31 3 e3 32 a2 x 1 33 d 6 34 e2 1 35 d2  36 xe3 xe 37 xe2  38 e3 x5 39 d3 f4 01 Duoistent, Lon don 1862 1 1 g3 h5! 1 2 f4

 a) 12 e5? d7! 13 f4 gxf4 14 xf4  15 h1 dxe5 and Blak is lose to winning ) 12  h4 3  g4 4 d4 g3 15 h3 5!? 16 e2 xd4 17 xd4 8 18 75

The Fo Knghts

xe6 e6 19 e5 d5 2 g6 d7 a unclear, but probably good for Blac in Frolov-Varavin, Eaterinburg 1997. aybe 16 xc5 xc5 a forced, even though Blac can connue ith  c5c4 c) 12 h3 and no c1) 12...d7 13 d2 g8 14 a3 c5 15 a4 b6 16 xb6 axb6 17 'c3 with ome advantage for White. c2) 12...h4 13 h2 g4 14 e5 xe5 (14xc3? 15 bxc3 gxh3 16 exf6 hxg2 17 e1 probably doe not offer enough com penaon for the piece) 15 xe5 d6 16 e1 ith a very unclear ituaon. c3) 12 ...d7 13 a4 f6 14 a3 a5 15 b4 b6 16 xb6 axb6 17 a4 h4 18 h2 g4 19 xg4 h3 and Blac ha a very trong at tac White ould lie to play 2 g3 but then the bihop i imply forgotten and buried, and after a move ie 2.. c5 Blac i much better.

Wte oud have been beer off trying 18 x xg 19 xg2 xg2 2 xg2 x!? (2x 21 f1 e7 22 g d7 23 xh4 xc3 24 bxc3 xa2 i le clear  White ha ome counterplay th 25 g3 and he h-pan, even though Blac chance ae peferabe) 21 a3 (21 g5? xc3 22 bxc3 b8 23 xh4 b2 24 g3 xa2 25 xa2 xa2 26 a6 c4 27 xc4 xc4 28 g5 a and Blac win; 21 b5 cxb5 22 xb4 e7 i jut very pleaant for Blac) 21... c5 22  d7  give Wte ome problem to olve, but at eat hey ould not be ovepoeing. 1 8.g4!

1 2. . .h4 1 3 .e1

o poible a 13 f2? xc3 (the right moment 13 gxf4 1 4 d4 c5 1 5 xf6 xf6 16 d5 xd5 17 exd5 xb2 18 xf4 c3 19 e i not o clear) 14 bxc3 gxf4 15  (15 d4 c5 i le attacve no) 15...g4 16 d4 g8 17 xf4 (1 7 h3 e3 18 xe3 e3 19 xe3 g3 i dangerou for Wte) 17.. c5 1 8 b5  19  b8 2 a4 h3 21 g3 g5 and Blac i cealy better 1 3gx4 1 4 Jx4 d7! 1 5 h1

15 d2 e5 (15 ..g5? 16 xh4 cot a pan for no reaon) 16 a3 c 17 h1 h3 1 8 g3 g8 ould give Blac good attacing chance, but White alo ha eaonable idea and the game i ll open 15 d5?? c doe not or 1 5.  g5 1 6 d2

16  b8? 17 a3 c5 18 d2 g7 give Blac the tin threat of the imple xb2 and the poional ..h4-h3. 1 6. . .g8 1 7 e2

17  h3 18 g3 e5 i ao uncomfort abe. 1 7. . .e5 1 8 a1 76

1 9 43??

A fatal blunder. 19 f2 c5 2 f6 g7 21 xh4 e2 (21 ...  22 g3 xg3 23 hxg3 g4 24 gx xf6 i hadly better for Bac a all) 22 g3 xf 23 xf1 h8 24 f4 g4 give Blac a clear edge, but l there i a ght left No it i lover 1 9. . :g7! 20 e3 h3! 2 1 g3 x3+ 22 x3 x3 23 x3 xc3 24 bxc3 xc3 01

ame 38 Bonarevsky-Bronsein Moow 1945 1 e4 e5 2 3 c6 3 c3 6 4 d4 exd4 5 xd4 b4 6 xc6 bxc6

cotch Fo Knghts delnes fte  xd

e4 but 9 O-O woud give White a good game. Therefore  ... cS with a sight edge should be preferred. 8   . d5

Here the standard move is 7 d3. his is certainly the most logic and natra move but even so other moves have found thei way nto the mnds of chess players from time to time. 7 d4

r a) 7 eS e7  e2 dS 9 d2 c3 1 bc3 c is simpy good for lack b) 7 d2 0-0  d3 e! ..dS 9 00 is equal) 9 0-0 c3 1 c3 e4 11 hS 11 d4 dS 12 e fS looks good fo lack too) 1...dS 12 ae g6! double edged) 13 e2 fS and White has to prove compensaon that is obviously there c) 7 gS h6  d2  h4 gS 9 g3 e4 1 d4 c3 11 bc3 OO gves lack the advantage) ..O-O 9 d3 e is an improved version for lack on the 7 d2 lne ven above. d) 7 d3! O-O! the strongest 7 ..e7  f3 dS 9 gS de4 1 f6 ed3 1 1 e7 e7 12 Jd3 is oy equal) and now d)  JgS c3 9 bc3 e 1  dS leaves White horribly paced  he centre. d2)  Jd2 c3 9 c3 (9 bc3 e4) 9 . e4 1 e4 e 11 JeS dS 12 e3 f6 and White is  touble d3)   dS 9 e2 a and ack clearly has the ntatve. 7 . . :e7 8 3

 d3 is known to be bad because of  ..dS 9 gS cS 10 b f 1 1 d3

here are some relevant aternaves a)  ...cS 9 d3 b 1 a4 Jb4 1..0-0 is very sightl better for White after 1 1 cS c 12 e3 b4 13 c3 c3 14 bc3 a S d4) 11 c3 Jd6 12 Jd2 12 Je3! 0-0 3 00-0 e 14 c2 dS! gave lack a strong attack  on darevsky-myslov Moscow 1944) 12 . 0-0 1 3 0-0-0 with good attacking chances for White  Gutman. b)  ..cS! 9 f2 (9 e3 00 1 Jc4 e 1 1 00 c6 12 �gS c3 13 bc3 dS 14 dS cdS S bS d7 16 Jd7 d7 and ack is a tle better  Gutman) 90-0 1 Jd2 1 Jd3 dS 11 0-0 c4 12 Je2 e is very pleas ant for lack) O ...dS 11 00-0 transposes to the game continuation. 9 d2?

Here White had a chance to escape the horrible posion  which he soon ends up. Afte 9 JgS lack has the foong choices a) 9 .. de4 1 O-O-O and lack is suddeny far behind in deveopment and also strctur aly damaged. b) 9 .cS 1 b f 1O .. Jd7 1 1 d7 d7 1 2 e+ e6 13 Jf6 gf6 4 dS dS S edS 0-0-0 16 0-00 is close to nnng for White) 1 1 d3 d4 1 1 ...de4 12 e4 c 12 ..e4! 13 e4 e4 14 c6 c3 S Ja e4 16 c3 c3 17 bc3 gS  b 1 wth advantage for White was the in ven by alvoi] 13 bc3 Jb7 14 00 e4 S g3 and hite has obvious comensation for the materia) 12 0-00 c3 12..h6 13 dS e6 14 f6 gf6 S Jc4! �d6 16 b4 cb4 17 �d4 and Whie ins) 13 bc3 h6 14 Jf6 �f6  S c4 g6! 16 cd4 g7 17 eS e7 1 a3 wth a disnct advantage for White c) 9 ..00 0 0-00 cS 11 f6 f6 12 d3  12 a4 e3 1 3 b d 4 14 d3 14 e2 cS S c Je6 16 Jc4 b 17 d3 77

The Fo Knghts

b6 and Blac can organse a strong attac] 1dxc3 1  xe3 cxb s very unclear) 1.e6 13 f (13 exd?! cxd 1 xd?? xd 1 xd fd8 would be rather aw ward to say the least) 13.d 1 e fd8 with consion in which lnes lie 1 f xa! 16 'a6 d3 can be very hard to evalu ate White is probably n trouble; his ingsde is not payng and his own ing is under at tac. 9 ...00  0 OOO?!

Whte should st play 10 g  better late then never ts unclear after 1..dxe 11 xf6 gxf6 1 000 xc3 13 'xc3 e3 1 d3. 1 0. . .5   Vf2 d4

c3 and it is very unliely that Whte  survve the attac Probably  chances of suival lie in the endgame after  a!? 18 ...a5?

Easier was 18...a 1 c e6 and Whte s in a deep crisis . 1 9 d3 a4?

is was obviously Bronsteins plan but it woud have been better to change diecon and play 1 ...c 0 f1 c3 1 bxc3 dxc3  f3 "xe 3 'xc3 f when White s completey outpayed. 20 e5 ie6 21 he1 a8

1..c   d 3 c3 xe  xe dxc3  xc3 d 6 'xc b6 would s have gven Bac a very strong attac. 22 b4??

A terrible blunder whch Blac does not exploit 22 .. .4?

What does White do after ...axb3 here? 3 e bxc  'xa8 'xa8  xa8 cxd1' 6 xd xa8 is just one of many wins This phenomenon is caled mutual blindness; probably a very beaul woman had just entered the playing ha! 23 e4 d5 24 f3? 3

Blac has clearly taen ovr the iniave.

Bac now has a completely g at tac 25 b1 V4 26 x3 dx3 27 xd5

1 2 b 1 ?!

is s rather passve. 1 d xd 13 exd xd 1 xd d6 1 c b7 was aso rather good for Blac, though. 1 2. . ..xd2+ 13 xd2 d7! 14 f4 b8 1 5 Vf3 b6  6 Va3

Bnng the ueen to the defence of the ueensde. After 16 d3 e6! the move c-c  come wth great power One possible line s 17 f c 18 e6 e6 1 "g3 cxd3 0 xd3 a 1 b3 c with great attacng prospects. 1 6 . :f6  7 3 V6 1 8 h3

18 g g 1 de 1 a 0 c (0 b3? c! 1 xd "b6  c3 xb and Blac  crash through) 0...'b 1 b3 78

27.. .xd5?

A terrible game wth many mistaes  maybe me trouble was playng a part here.

oth Fo Knghts delnes fte  txd

27.xa2 28 xc3 xdS smply leaves Blac a roo up

Jd6 3 d4 d7 s bete fo Blac, bu no much) 9xc3 0 bxc3 Jxc3 and there 28 xd5 'xb4 29 xa8 d8 s no real compensation for he paw n. Afe Bac plays a good move agan, and he Voava, Rshon e Zon 992 b) 7 Jd2 0-0 8 d3 eS 9 0-0 dS 0 poson s sl nning 30 d5 c6 31 e6 fxe6? exdS (0 xdS xdS  exdS xd3 2 cxd3 Mssng he specacular n 3xdS 32 Jxd2 3 'x2 xdS and Blac s much xdS cxdS 33 e7 s pobably wha Bac was beter because of the weaness of d3, bshop afrad of, bu afer 33.b 34 d b+  versus ngh and bee coodnaon) 35 e2  36 d e8 37 e3 f6 he Oxd3  cxd3 h6? (bad w oud be  be compleey nnng, as moves e  .xdS? 2 xdS xd2 3 xc7) 2 e2-d3  n no .. and e8 f4? (2  xc3 3 xc3 xdS th heefoe Bac can smpy pove hs equaty was beter) 2xdS 3 xdS stuaon, fo example 38 a3  39 fS g6 40 xdS 4 xc7 fS 5 a3 Je7 6e f6 g4 gx 4 gxfS xe7 42 xe7+ xe7 43 7 d4 fe8 8 e8 xe8 9 c  JgS 20 e2 d6 4 d3 eS 45 xc3 xfS and c3 g4 and Blac had he nave n Lac-Hebden, London 988 Blac wns 32 'e4 'a3+? 33 d1 xd5+ 34 e2 d6 35 f3 f7 36 xa4 d2 37 e4 g6 38 e3 h2 39 d3 'c5 40 d7+ f8 41 'd3??

A horble msae Afe 4 xh7 dS 42 xdS cxdS 43 a4 xc2 4 c7 e8 45 a d8 46 c d7 he endgame s easly dawn

6 ... bxc6 7 d3

 s oo opmisc o  o punsh Bac wh 7 eS? because afer 7..e7 8 e2 dS 9 e4 (9 d2? b8 0 O-O-O? Ja3 woud be a serious dsaser, whle Blac s better afe 9 xdS? cxdS 0 Jf4 b8) 9.Jb4+ 0 c3 aS Blac s a ittle beer accodng o Gutman.

41 . . d2 42 d8+ g7 43 d7+ f6  g4 'f2+ 01

ame 39 RozenalisAams ge 1997 1 e4 f6

hese two payes are used o geng the coch va he Alehne Defence; s also happened n RozenalsAdams, Elsa 998 2 c3 e5 3 f3 c6 4 d4 exd4 5 xd4 c5?

GM Ma Hebdens favoure 6 xc6

Ths s he naual move, bu no he bes. 6 e3 s consdered n he nex game. Whe has also tied 6 b3?, bu ths loos oo passve 6Jb4 and now a) 7 d3 dS 8 exdS xdS 9 O-O? (9 Jd2 xc3 0 bxc3 e7  'e2 xe2 2 xe2

7 .. .d6

r a) Afer 7.dS Whe should consder 8 exdS, bu he can also choose 8 00, which can tanspose no he man lnes f he bshop eus o e7 For example, 80-0 9 JgS Je7 0 exdS cxdS eaches Game 32 b) 70-0 s a ticy move order Blac 9

The Fo Knghts

waits before decdng on d6 or .dS. 8 0-0 e8 9 JgS h6 10 Jh4 d6 would take us to note a to tes 8th move 8 a4!?

is does not turn out to be successl  his game, but should not be dismssed ther opons are a) 8 JgS h6 9 Jh4 0-0 (9. e7? 10 e2 b8 11 a4 Jd4 12 c3 [12 f4? gS 13 gS g4 14 000 hxgS was beer for Black  the game AkerstKaciani, Baden-Baden 1 993 12 . JeS and Black wl foow up with gS) 10 0-0 e8 1 1 h1 gS 12 Jg3 g4 13  hS 14 e1 d7 1 ' b8 16 ab1 Jd4 17 f4 Jg7 18 b4 gxf4 with an unclear ame, PotknLasn, Moscow 1999. b) 8 00 is probably the best  8 d7 9 e2 0-0 10 a4 (here ts dea works bet ter) 10 .Jb6 1 1 b3 h4 12 xb6 axb6 1 3 f3 Ja6 14 c4 (to keep the advantage of te wo bishops) 14. cS 1 b2 with a very sight advatage for ite, Mes-Hebden, London 1994

ae1, e3, fe1 and after that e4eS f Black queen remains on d8) or f4-f. Im connced that e6 s the only sensbe move for Black  ths poson.

16 a3!

ite frees te rook from a 1 6 .e8?!

After is exchange Blacks pawns are no longer weak

I do not ke this move; the alteaves ook better 16..c6 17 f e 1 8 ae1 h7 (but not 18f8? 1 9 xf6 gxf6 20 f4 g7 21 e3 wth a strong attack) 19 e3 e7 (19 .. d7? 20 xf6 f6 21 h3 g8 22 f4 is very dangerous) 20 g3 e8 21 xe xeS and Black has good compensaon for the exchange 16 e7? 17 f e 18 g3 e8 would also secure Black compensaon but probaby ite just connues with 19 ae1 in order to delay he acceptance of he sacrice.

1 2. . .axb6 13 4!

1 7 f5 xe4

afe play does not work fter 13  b7 14 c4 d 1 c2 d6 16 fd1 dxe4 17 xe4 c 18  g 19 d3 c Black is at least eual Hs bshop on b7 s the best pece on the board and .f6-h4-f4 attacks the weakest spot  front of tes king. Best now is probably 20 d2  order to reach an endame, but Back is e after 20.h

Because of te naccuacy of hs preious move, Back s now forced to capture on e4 Back would e to play 17e?, but 18 Jxf6 xf6 19 f4 s very dangerous.

8 . .b6 9 g5

te stands sightly better after the stronger 9 O-O?, tenng xb6 and b2-b3. 9 . .00 1 0 00 h6 1 1 ih4 e8

Black searches for counterplay aainst the e4-pawn. 1 2 xb6

1 3.  .b7 14 4 5 1 5  2 e6

s move, preparing a posional ex change sacrice, is necessary because ite s threateing the foowing plan a2-a3, 80

1 8 ixe4

r 18 Jxf6? g4 19 ae1 c6 20  gxf6 21 fe2 d (21h7? 22 Je4 and te is obously better) and now a) 22 e7? d6 23 e2 (23 h3 g 24 cxd Jxd 2 7e2 d8 26 e4 Jxe4 27 xe4 d4 28 xd4 cxd4 29 g4 hS 30 e4 g4 31 xg4 c foowed by .  and .. g8 would ve Black a cear advantage)

cotch Fo Knghts delnes fte  xd

23  h! 24 cxd xd 2 �e4 d 26 h �xe4 27 xe4 xe4 28 xe4 xe4 29 xe4 d8 would gve Black realisc g chances. b 22 cxd xd 23 �e4 "d 24 h �xe4 2 xe4 xe4 26 xe4 "d7 and  sure ha a plaer e Adas woud w such an endgae Perhaps it could be rawn  correspondence hess but over he board the ask of the defender is oo great. 1 8. . .xe4 1 9 .g3?

Black of course has copensaon, but this ove does no ake sense Bes was 19 ae! e 20 �g3 "d 2 � d2 22 b f6 (22..x? 23 x d4 24 f6! is unheath for Black) 23 c1 xc1 24 xc1 g4 2 fe e. This endgae is where he nal udgeent on 16..e8 has to be ade  beieve Whie has soe chances to w it, while Black has ver few acve ideas

c h5 27 .h4 d2 28 c3 xc4 29 f6 e5 30 e1 b7 31 fxg7 xg7 32 g3+ g6 33 d8 xa4 34 xc7 d4 35 b3 d5 36 b2 e6 37 c2 f4 38 f d3 39 .d8 e4 40 e7 b4 41 c3 c4 42 xd6 V -V

ame 40 Reea Bin SaarHeen Dhaa 1995  e4 e5 2 f3 c6 3 d4 exd4 4 xd4 f6 5 c3 c5 6 ..e3

The ost accurate reaction against  ..c? 6 ..b6

1 9. . .b5 20 b3 bxc4

Black slowl proves hs posion 20  xg3? 21 hxg3 e3 22 h2 �e4 23 b2 bxc4 24 bxc4 f6 2 ae "d3 26 f4 would give White a clear edge 21 bxc4

7 d2?

21  :a4

21..xg3?! 22 hxg3 e 23 f2 xa3 24 xa3 xa3 2 f6! is dangerous for Black 22 xa4 xa4

n his enng he white pawns on a3 c4 and  are ver weak A draw i a far result. 23 fb1 a7 24 e1 h7 25 a4 ..a6 26

A upse  hite wants to buld an at tack! The on queson is what about the possibiit of g4 here Whites othe optons nclude a 7 g3 ust e rather haess. 700 8 g2 d6 9 h3 e8 10 00 �d7 1 1 e h6 was unclear n LaneHebden, London 1994, a gae Black later went on to w b 7 e2  a not afrad of this ove  where is its sng?) 70-0 8 d2 (or 8 0-0 d6 9 xc6 bxc6 10 �g e7 1  h e 12 f4 d4 13 �d3 g4 wih counterpla  Dzndzichashvi 8 d!? (after ths classic counterattack in the centre Black should equalse 8.. .e8 9 xc6 bxc6 10 g h6 11 h4 g 12 g3 xe4 13 xe4 xe4 14 81

he Fo Knghts

0-00 gives White good atacking chances  Guan) 9 exd5 xd5 9 ...xd 10 xd4 xd5? 1 1 xg7 xc3 1 2 �xc3 e8 13 h6 �d 1 �g g 15 0-0-0 �xg3 16 g3 e6 gives Whte a cler exra pwn) 10 xc6 xc6 11 0-0-0 11 g5?) 1.a5? 2 d e6 wth a doule-edged sitation. c) 7 xc6 xc6 and now: c1) 8 x6? just cnnot e good. 8 ...x6 9 d3 0-0 1 0 0-0 d6 1 1 e1 g 12 h3 e5 13 4 xd3 1 xd3  gve Black an excellent game in Kozkov-Adams Frnce 1997 c2) 8 e5? ths is an interesng possily) 8 ..xe3 9 e3 afer 9 exf6? the wo ish ops and he open le is too much 9 h6 10 �e f8 1 1 fxg7+ xg7  2 d 8 13 0-0-0 d5 and Blck has a very stong at tack, Munns-Boras, Zagan  995) 9  d5 0 xd5 cxd5 11 d4 11 �xd5 h 12 d 8 must e good or Blck)  ...0-0 1 2 0-0-0 wit unclear play.

eachng for couneplay sitaon



a dicult

 4 d2

akig the pawn was also interesng 1 xd6 �a5 15 g5 not 15 a3? d8 16 �f xd1+ 17 xd a6 and Black has compensa ton) 15..d8 16 gxf6 ths idea mght have evaded White) 16..xd6 17 xg7 8 18 xd6 x2 19 xh7 g8 20 g7+ 8 20 h8? 21 x2 x2 22 d5 wns or Wte) 21 g e6 Blck needs o e relly carel 2 ...e6 22 h g8 23 d2 and Wte wins) 22  g8 23 d1 xf6 2 e5 x 25 a4 c7 26 x2 xe5 27 1 a6 28 g2 J 29 4 xa 30 xa 5 31 c3 x2 32 7 nd White hs good chances o wn ths endgme. 1 4"a5  5 a3 a6

Unfortnately Black hs o lose  tempo w ths move. 1 6 5!

7 .. .0-0

7.g ooks like the most ovious move. Now 8 f5 is te correct reply 8 g5? f6 9 e3 xe3 10 e3 e5 is ovously ad) Play can connue 8...xe3 9 xe3 00 10 0-0-0 d6 1 g3 and, wit te idea ofg2 f2f, he1 and , Whites posion can e charactesed as sold nd rich in possiiles 8 0-0-0 e8?!

his is ad Necessary was 8  g4 9  and now 9 ...xe3 would retu to the pre ous noe, while afer 9  d5? 10 xd5 x5 1 1 ex5 xe3 12 e3 Blck hs no compenson. 9 f3!

As you  see ths is a vey dngerous t tcking plan. White plays n  similar ashion to the Yugoslv Atck against he icilan Dragon. In his game he GM achieves  drw only ater exreme dfcultes. 9 . d6 1 0 4 xd4 1 1 xd4 e6 1 2 1!

A song, atackng move 1 2  xd4 1 3 "xd4 c5!? 8

Wes tck rves s. 1 6 . d7

1 6.. h5 17 f g6 1 8 � 5 19  give Whe  song atack too. he nght  soon rive on d5.  7 f4

Connng the attack. 1 �xd6? d8 18 �g3  5 would llow undesered counerply. 1 7. . .b5

1 7 6 18 5 d7 1 9 �xd6 c6 20 g6 s n example o how ar We s wth s ttack.

Scotch Fo Knghts Sdelnes fte  xd

1 8 5 c4 1 9 g6!

arel.

Blak ingide i being broken up while Blak ha not been able to reate any teat of hi own. 1 9  xg6 20 xc4+ bxc4 21 fxg6 h6 22 xd6 e5 23 d5!

bjetively th hould be the end but omeme it i hard to win even the mot ompletely wnning poion. Epealy if the opponent i a rong andmater, who ha deided not to go down thout a ght. 23   d8 24 'xc5?! 'h4 25 dd1?! ac8 26 'a7?! 'xh2 27 d5 c3 28 bxc3?

Probably White wa drifng into e trouble. He hould have played afer wih 8 x3 f (8 . 4 9  h8 30 h and it i a over) 9 b f6 30 dS when White i  g. 28 . .'h4 29 d1 �h8

But not 9 ... xe4?? 30 

6 xe4

6 x6?! x3  xd8 xd 8 x x! (8..xf? 9 xh8 xh 1 f4 d6 1 1 d3 i unomfotable for Bak who ha poblem on he ingide) 9 4 f6 1 xd 6 wih equaity  Euwe. 6 ..'e7 7 3!

Thi i the tandard way for Whte to ght for an advantage. lternavely a)  �e? xe4 8 bS b4 9 3 aS  0-0 0-0 doe not ve hte enough om penaon for the pawn, even though the poiion i not ntirely lear. b)  d3 xd4 8 OO e6! give Blak a lear advantag aoring to Guan but after ometing ike 9 e  White hould have ome ompenaon. ne e goe 9.. d8 1  e  1 gS and Blak ha problem wih niing hi development ) Intereting i  d3!? xd4! (Blak od quy end up in trouble with  ..dS?! The poiion i not lear Whte ng  8 x6 bx6 9  fS?! 9.b8!?] 1 3! onue to ive dangerouly to he end. d 11 gS e6 1 OOO e4 3 e4 dxe4 14 4 � S h4 and White ha a ame   wnnng attak) 8 xd4 fS 9 d3 e4 1 KoalijaSevosianov xe4 b+ 1 1 xb4 xb+ 1 3  and Blak i only vry ightly wre. Moow  30 'd4 c6 % -%

1 e4 c6 2 d4 e5 3 3 exd4 4 xd4 f6 5 c3 xe4?!

Thi idea look ike a paer mitake but it i not o tupid a it look White hold be

7.  d5 8 b5 d7 9 xc6 bxc6 1 0 0-0 dxe4 1 1 fxe4

Gutman beieve that 1 1 e  !? i bet. Pratie how u that it i not o lear. However hite i better and doe have very 83

he Fo Knghts

gd attacng chancs. Blac has th flw ing ptins a 11..f6?! 1 4 7 13 f4 but nt 1 3  O-O! 14 7?! d and Blac is n 13..00 14 .5 g6 15 c7  1 6 h 1 ac 1 7 .g3 c5    .f5 19 f4 c6 0 :5 g6 1 d5 f6  f6 wth a cla advantag f Whit Fandz Rm-ands ant Antn 00. b 11?! 1 4 4 wth a th banch b 1 3 c3?! c5 14  5 15 g3 7 16 4 O-O-O! 1 6.. c6?? 17 g3! 'f6  'h5 was nnng f Whit n zindzichashvvl Nw Y 1994 but that ds nt say a lt abut th vaia n 17  �c6  g4 d7 19 f4 �g4 0 5 hg and Blac is nly sightly ws h b 13 '! g6 4 4 4 15  d 16 g5 c 17 f6 g   a  wth a cla plus c 11...0-00! 1 :4 f6 13 1! 13 '?! c5 14 b3 .c6! 15 a6 .b7 16 'f6 gf6 vs Bac lt f acv cunt pay vn thugh his stctu is awd 13. �c5 14 'a6 b 15 3 h!  15...b6?! 16 a4 wth an attac  Guan 16 c3 4 17 4 b6? 17g6! was much btt  a4 c5 19 a! wth an atack Timmman-Mkhalvski Din 1996 1 1 . .g6 1 2 3 ig7 1 3 e3!

Accuat play 13 c3?! 00 14 f4?! 14 3! is s pssibl 14 . c5 15 b3 c6 1 6 g3 s gvn as slghty bt f Wht n CO by Ngias but aft 16 c4! 17 d fd   c4 c5 19 3 4 th pw fu bishp n 4 givs Blac th dg

g4 1 'd3! 6  h3 'g5 3 b4 d8 4 d4 and Whit has a cla advantag.

1 7 id4 ixd4+ 1 8 xd4

Wht has a cla advantag Bac has a wcd pawn stct and n acv pay t cmpnsat f it. 1 8. . .a5 1 9 a4! b4 20 d 1 ab8 21 c3?!

M pcis was 1 c5! 7  b4 :b4 3 c3 and Wht is vitually wning 2 1 . . .e7 22 xb4?!

Th wng chang. Pbably th bst mv was  h3! t cat an ai-hl f th ing Lns such as  a5 c5! 3 d5 :4 4 b3 and  5 5 3 b4 c3 4 bc3 ab4 5 cb4 b4 6 a a4 7 :a a1 8 a c5 9 c4 f8 30 b3 7 31 c5 d6 3 d3 thugh favu abl f Wht a nt cmpltly cla. 22 . . xb4?

Bac shuld hav td  'b4! with th da f 3 d4 d 4 d4 f6! and th bishp cms int play via . l hit shuld p a cla dg th 3 5! 23 tc5 6 24 b3 g5?! 25 h3 h5

1 5 c5 e6 1 6 ad 1 ie8

Havng nthing t ls in this psin Bac stms fwad l a b wth his gad dwn hpng t gt a lucy punch n bf th bl gs As s ftn ths nly spds up hs wn dfat

Blac is als in tubl af 16  b 17 c3 a ths th bishp  b tappd and lst 1 8 :f a4 1 9 b ab3 0 ab3

26 e1 g6 27 e5 xe5 28 xe5 d8 29 te6 6 30 txg5 4 31 t3 e4 32 e8+ g7 33 g8+ 7 34 8+ 1 -0

1 3. . .00 1 4 tb3! b8

 14..b 1 5  c5 5 16 f a1  16 f8 17 ad 1 with an chang f nthn 1 7  h8 1 8 d7 and wins

8

cotch Fo Knghts delnes fte  xd4

ummary

Whte has seeigly o other soud opos etwee 4 d4 ad is teth ove 7 d4?! was learly a great part of Bodarevskys dowfall ad siiar oves will eet silar fates or Blak there are soe alteraves 5 xe4 ad 5 c5 do ot lead to eualty ad shoud e igored y Bla However 7   d leads to a gae wth haes for oth sdes ad plety of roo for origaty  e4 e5 2 3 c6 3 c3 6 4 d4 exd4 5 xd4 b4

5xe4  Gam 4 1 5 5 )  x  Gam ;  e3   Gam 40 6 xc6 bxc6 7 d3 7 'd4  Gam  7    d6 8 00  Gam 

8 g5   Gam 

  . c

6 e3

 g



CHAPTER SEVEN 4 d4 b4

1 e4 eS 2 3 c6 3 c3 6 4 d4 b4

n e poson afer 1 e4 e5 2 f 6   f6 4 d4 e naa and probaby aso e bes move s 4ed4 movng o e o For Kngs or e Begrade amb (see Caper 8) However ere are ose wo e o ompiae maers w 4b4! w s probaby no good enog for eay b w an ead o more neresng psons n is aper we sa ook a ese ines As ey are very aa in eir orenaon no posona nrodon  reay make any snse so nsead we jmp srag no e games

ame 42 Sano Roman-Heor Manila Ompiad 199 1 e4 eS 2 3 c6 3 c3 6 4 d4 b4 S dS

Afer s move Bak as a good game Te man reason s a e poson s os g and Bak as s bad bsop osde s pawn formaon we We as is nsde Of orse  s aways ard o say wen a bsop s ry good and wen  s ry bad b generay kgs ae beer sed o osed posons even og e 86

bsops s ave par of eir sreng S ..e7 6 xeS

Ts move ogeer w 7 b (ame 4) s very rsky Bak deveops ky and ges a srong aak Oer possbes nde a) 6 d2 s raer peae 6 d6 7 e2 0-0 8 a a5 9 b4 b6 10 a4 d7! (10d7 11 b6 ab6 2 g5 g6  g 6 14 f6 f6 was ea in pangenberg ea Benos Aes  99)  1 b6 ab6 12 g (12  w eve anes was beer) 12 g4 1 g5 ed5 1 4 ed5 e4 1 5 f6 e 16 4 e 17  gf6 18 "g g4 w a ear advanage for Bak  pangenbergRbe an M 199 b) 6 d s raer passve and ess s very m e ar of aggresson 6  g6 7 0-0  ( bd yo farewe bad bsop) 8 b d6 (s poson s  my opon sgy beer for Bak) 9 :b 00 10 e s move ooks wrong Wa s e rook spposed o do ere? Te ony sensbe pan was e 1 foowed by g2g and g2 o ge some onro of e mporan sares on e kingsde) 10b6 1 1 E 6 (No so m o pay 7 b more o preven g5e6 aer wi e bsop wod move more freey on e board g now Bak as

 d b posoned hs pawns to ght against the darksquared bshop but there s no reason to take lbertes with the ght squared bshop) 12 g3 h7 13 d2 S 14 exS xS 1 5 e4 6 16 g2 d7 17  xe4 18 xe4 g4 19 d3 6 20 e3 a and Black has a clear advantage, Wee-Yande mrov Russa 2001 c) 6 d3? 00 and now c1) 7 d2 ntends the very sky plan o castlg queensde. t seems hat Black has a more dynamic poson and thereby can cre ate attacking chances much easer. 7.d6 8 0-0-0 g4 9 e2 S 10 exS xS 11 h3 6 12 b (ths move also seems a bt out o touch h reaty; takng on b7 s not realy a viable ambon) 12. xc3 1 3 xc3 e4 14 e (now he queen is just a hook that Black explots) 14 .. c6! 1 5 dxc6 bxc6 16 b dS 17 c4 b8 18 xd5 h8 19 c4 cxdS 20 xdS b6 and Blacks attack clearly compensates or the sacriced pawn. annayer, Germany 1 988. c2) 7 gS (hs s saer) 7 ..g6 8 g3! (t s mportant to keep control o he 4square) 8 ..c6 9 d6!? (An nteresng pawn sacrice. Wheher Whte new the pawn was doomed here, or only decded to sacrce t later s not relevant. ore relevant s that Black has problems wth completing his development He could have avoded ts th 8d6 o course.) 9 ..h6 10 .x6 x6 11 0-0-0 e8 12 .h3 (12 g2 e6 would be a avourable verson or Black) 12xd6 13 xd6 x3 14 h 8 1 5 S g6 1 6 h3 b6 17 d2 hS. Ths poson s very teresing Unor unately, the to strong Russan grandmas ters elt i a peacel mood that day and agreed a draw alakhov-Turov, Russa Ch 2001) Ater 1 8 g2 e6 19 d3 a 20 e3 .a6 21 h4 the award poson o Blacks queen s good compensaion or the pawn. One ne s 21.x? 22 . and Whte s better. 6 d6

Worse s 60-0? 7 d4 xc 8 bxc3

and now a) Adventurous s 8..e8? 9 gS d6 (9 £? 10 exS d6 1 1 4 dxeS 1 2 eS xdS 13 0-00 xd4 14 cxd4 e4 15 h4 c3 16 d3 xa 17 b2 b4 18  ves Whte an oveheng advanage n the endgame wh the to bshops and the song cenal control) 10 x6 cS 11 dxc6 gx6 12 cxb7! (12 3 xc6 13 e3 S 14 0-0-0 and according to Guan Blacks ing s more ragle than hites) 12..xb7 13 g4 S 14 x6 e4 15 .e2 aS 16 0-0 xe2 17 + h8 18 h6 and Whte has a clear advantage ater 18...dS 19 xS eS 20 h6 b) 8 .. d6 9 c4 cS (9 e8 s a possble al ternave here) 10 d3 g6 1 1  bS 12 e3 a6 13 g3 e8 14 g2 and accordg to Lev Gutman Black does not have enough com pensaon or the pawn. This s probably rght. 7 f3

Weaker s 7 d3? xc3+ 8 bxc3 xe4 9  £, when Black s beter so possble s 9 ..xc3 10 4 a4 1 1 d3 c, whch s also better or Black.  he needs to do s to avod 11..0-0? 12 e4 g6 13 xa4 (wch actually happened i peann Bogolubow tockholm 1919). 7    x4 8 d4

8 xc3

The alternave 8 xc3 looks stronger. Ater ths move \hte needs to be very care87

The Fo Knghts

fu for eample 9 bc3 (Whte can also  9 �b4 ed5 10 d4 'e7 11 e3 e3 12 e3 e4 13 g7 'f6 wch leads to equaty and now a Ecng s 9 ..c5!? 10 g7 g8 11 'h7  12 'h4 (12 b c6! 13 dc6 h7 14 cb7  1 5 h6 g7 16 g7 g7 17 ba8 a8 does not gve Whte any reason to connue 12...d5 13 �d+ d8 and Black has a lot of play for the pawn probaby more than nough to jus the sacrce. b 9 ...a5 10 d3 (dubous s 10 g7?! g8 1 1 d4 d5 12 d2 [12 d5 c3+ 3 d a 4 c4 e6 15 e e does not work out] 12 ... e7 1 3 e2 c6 14 f1 b6 and Black s clearly beter due to hs coordnaon and ead n development 10 ...0-0 1 1 0-0 f5 (1 1 ..f5 1 2 c4 f6 13 d2 d7 14 a4 c6 leads to a preferable poson for Black 12 g5? Qosng valuable tme; 12 c4 b6 13 f4 s more or less equal 12 .. .f6 13 d2 'd7 14 ae g6 1 5 f5 f5 16 'e4 'e4 17 e4 fe8 18 fe 1 e4 1 9 e4   Black had a much better endgame  acek-Trnoc Prage 1946. 9 bxc3 6 10 c4

fter 10 g5 ed5 11 0-0-0 b6 tere s no real compensaton for the pawn. Also possble s 1 1 ...e6!? 12 c4 c5 1 3 f6 cd4 14 d8  d8 15 d5 d5 16 d4 00 17 f3 a2 wth a very good end game for Black, Reefscaer-Neunhoffer Germany 1990. S the compicaons seem unnecessary. 1 0.  .0-0 1 1 .2

Or a 1 1 g5?! c5! (1 1 ...f5 12 b2 e 1 3 e2 h6 14 f6 'f6 15 'f6 gf6 16 d2 s a ttle better for Whte accordng to Gutman 1 2 d1 (12 f4 g6 13 �d2 e+ 14 e2 h6 suts onl Black whe 12 dc6 c6 13 b2 e+ 14 e2 'a+ 15 d2 'h5 s etremely dangerous for Wte 12..e4 13 d3 g5 14 g5 g6 15 88

h5 'e7 16 f1 h6 and Black has a clear advantage. b Also risky s 1 1 d3 f5 12 �f4 e+ 1 3 d1 and now: b 1 3...h5 14 'g5 'g5 15 g5 h6 16 d2 wth equaty Caro-Schechter V enna 1898. b2 More aessve s 13..g4!? and: b21 14 f5? s somethng for fans of Mkal Tal. After 14 ...  15 d2 h1?! 16 h7 h7 (or 16..h8? 17 b2 and Black s soon mated 17 g5+ g6 1 8 ' h6 19 d3 e7 20 e4 h7 21 g h6 22 e+ ends  a draw Unfornately there s 15.. (as wth many of Tals spectacular stunts ths also has ts dawbacks 16 f5 g6 17 f4 e4 18 h6 f6 wth a strong compensaon for the pece  Gutman b22 14 f1 f6 15 b1 wth unclear play. 1 1   S

1 1 ... e8!? looks ke a local move too. The dea of .. .ed5 s obvous . 1 2 d1 4 1 3 .b2 cs 1 4 b1 h4

ere Black could have found a great square for hs queen wth 14..e8? 15 0-0 a4 16 a3 e8 after whch the poson s probably equa, but t s more pleasant to play Black. 1 S xh4 xh4 1 6 00 fS 1 7 d2 8 1 8 b1 h6 1 9 .d3

Wth ths move Whte more or less keeps the balance. 1 9 hS 20 3

Bad would be 20 f5? �f5 after whch Black would play ... and echange the rooks. The queen and knght versus queen and bshop scenaro would hen together wth Whtes wrecked strcte gve Black a better endgame. 20 x3 21 x3 i4 22 d4 g6 23 f3

More accurate was 23 c5! dc5 24 f4 e8 25 e4 e4 26 'd3 wh equay. 23 .. .hS 24 f4

  i

4 g3 g6 5 xc5 dxc5 6 xg6 g6 wold b bttr for Black 24.. .ig6 25 h4?!

hr wa no raon to crat a wakn lk th. 5 c3 xd3 6 cxd3 f6  b3 wa qal, thogh not drawn 25 .. .ixd3 26 cxd3 b6 27 4 6

...d wold g ht m to crat thrat 8  f6 9 g4 28 i2

8 xc5 bxc5 wold  Black th bttr trctr. 28. . J8 29 J3?

ht choo to b pa  a bad dc on t wa ncar to compcat mattr wth 9 g4!? 29 ld7

Now th knght  comng to 5 wth a gan of tmpo. 30 4?

30 f5 g6 31 d4 wa bttr 30 . .l5 31 g3 d  32 �h2 h5 33 ig1 lg4+ 34 �h1  35 3 g6 36 g3? 2 37 i2?

ong on th pot. 37 ..  + 38 ig1 h3 0-1

ame 43 Traer-GFlear A 1998 1 4 5 2 l3 lc6 3 lc3 l6 4 d4 ib4 5 d5 l7 6 lx5 d6 7 ib5 �8!

Aftr th mo ht  n trobl bt th tabl wold t followng   c6 8 dxc6 0-0 and now a) ht wold fall too far bhind  d lopmnt aftr 9 cxb? xb 10  x4 1 1 0-0 ht o aftr 1 1 d xd 1  xd a5 13 d3 d5 14 0-0 xc3 15 bxc3 xc3) 11..xc3 1 bxc3 xc3 13 d3 xb5 14 g5 14 xb5 x 15 gx c8  impl bad for ht) 14g6 15 xb5 c8! 16  h6 1 4 1 h3 wold b mt wth 1...xg 1 8 xg g4 and Back wn) 1...8 18 b4 b8 9 g3 x1 0 x xc and Back  wnnng b) 9 d x4 9 . xd 10 cxd x4 11 d4 xc3 1 xb4 cd5 13 b3 ga ht a clar adantag  Tal-chmatlank Prag [mJ 1960) 10 0-0 n 10 xf8!? work; aftr 10xc3 11 bxc3 xc+ 1 d xa 13 cxb xb 14 xa xf8 15 00 ht  a lttl bttr) 10...xd 1 .. xc3?? 1 1 xf8 a5 1 cxb xb 1 3 d wold b winnng for Wht) 1 1 x4 bxc6 1 a3 a5 13 d3 13 xd6 cxb5 14 b c 15 xa5 f5 g Back good play) B  .d5 14 c5 and ht ha a p tonal adantag 8 l3

t do not kp th baanc aft 8 d3 xc 9 bxc3 x4 10 f3 f6 and now a) 11 c4? g4! 1 f4 c6! 13 dxc6 bxc6 8

The Fo Knghts

14 b g6 15 x6 a5 16 d e+ 17 e xe 18 xa c and lack ganed a aeal adanage  Hor Trinoic araeo 1964. b) 11 c4 g4 1 4 h 13 0-0 exd is plainly bad or Whe. c) 11 g? exd 1 xd? e7 is  iportan ite nesse, winnng a pawn or nothng. d) 11 4 (he bes) 11.g4 1 g3 (1 d3 c6 13 dxc6 bxc6 14 .a4 a 15 b3  looks ery dangerous or Whie) 1.. 13 d3 e7 14 �1 h4 15   16 d g6 17 e d7 18 xd7 xd7 and lack was beer Kunas-Dauo Gerany 1993

16 �d1 e7 17 e b7 18 c4 bxc4 19 xc4 d7 woud ilustrae the probles We has wth the d-pawn.  6. . ! 1  c4

1 7 e b7 1 8 c4 bxc4 19 xc4 d7 0 h3 exd 1 �g s us a pawn or notng. lack has soe probles wh hs kng, bu tha is  1  bxc4  8 xc4

Whe be a pawn down withou co pensaon aer 18 xc4 b 19 xb axb 0 d4 d7 1 c4 bxc4  xc4 exd5 1 8 . b 9 b1 xd5 20 xd5 xd5 21 c4 b6 22 3 c6 23 d2 bd 24 f3 8 25 �f2 5 26 d 4 h5 2 h 1 fd 28 �g  f6

8  x4 9 d4 ixc3

Tepng s 9 .xc3! 10 bxc3 c 11 "d xd (11.c6 1 dxc6 bxc6 13 d3 g4 14 g d 15 e4 is less clear) 1 "xd c6 13 d3 (13 xc6 bxc6 14 xc6 e7 15 1 b7 16 b a looks ery dangerous or Whe) 13...cxb 14 xb (14 00 a6 15 e3 .e6 does no gie Whie enough copensaon or he pawn) 14 e7 1 5 e xe+ 16 xe , when the two bishops and the ead in deel open is well suppeened by a beter pawn srucue 1 0 bxc3 f6  1 ig5

Also possible was 1 1 c4!?  aer the strange-lookng 11...g4 1 g  13 d3 "e+ 14 �1 x 1 5 gx3! Whe has a errible pawn scture bu also two bish ops and the g and bles or s rooks noe tha 1 5 x e4 16 4 xd is good or lack) 1 d1 e4 13 00 6 14 b d7 and lack has an edge   .  .f5 1 2 d3 h6 1 3 id2?

We has no rea copnsaon or the pawn and no rea chances or saing the gae 29 h3 �f 30 f2 c5 31 3 6 32 d5 xd5 33 cxd5 f4 34 Ja3 xd5 35 xa6 a8 36 xa8 xa8 3 b5 �6 38 b2 d3 39 c2 3b4 40 c4 xa2 41 b3 d3 0-1

ame 44  oorovDelev Bua 1995

Aer this oe i wl be ery hard or Whie o sae the d-pawn. Whie could hae kep the balance wth 13 x6 x6 14 00 e7   e  1 6 d a6 1 7  g 1 8 d4.

 4 5 2 f3 c6 3 c3 f6 4 d4 b4 5 x5 00

1 3  .a6 4 a4 b5 5 .b3  8 6 �f

Ths s raher rsky 5 .xe4  . e7 and  ... xc3! are considered laer  the chaper

90

   6 d3

te can queeze a mall advantage out of 6 Ixc6? dxc6 e accuate i the pave 6bxc6 7 d3 Ixe4, afte which te gain a mal advantage wh  00 xc3 9 bc3 xc3 10 xh7 h 11 d3 g 12 xc3) 7 gS 7 eS Ie4  a3 xc3 9 d2 bS 10 axb4 xd4 11 d3 dS 12 f4 fS 1 3 xfS IxfS 14 c3 hS and Blac i bete  Gutman)

and now a) 7 h6  xf6 xf6 Hee Blac ha compenaon, but Wite hould be able to eep the advantage, fo example 9 eS 'h4 (9f4 can be anweed b 10 d2 'xd 1 1 xd2 d 12 e3 cS 1 3 Ie2 cxd 14 Ixd4 c  S c3 e 16 f4 and \te eep the advantage) 10 a3 (afte 10 'd3 d  11 0-0-0 'xf2 1 2 'f3 'x 1 3 gxf3 e6 1 4 a3 xc3 S bxc3 cS Bac equaed  Lavo Cepo-Todeoevic, a Pama 1999) 0.xc3 11 bxc3 cS 12 d3 d 13 g3 g4 14 h3 (14 e2?) 14d S d a4 16 g2 'xa3 17 00 and ite poion eem to be pefeable b) The altenave 7 . e pobabl doent wo  d3 ( eS? i too geed too oon cS 9 bS c6 10 0-0 cxbS 11 exf6 cxd4 12 a3 c 13 Ie2 g4 14 e "dS and Blac ha a completel dominang poi on) xc3 9 bxc3 fS Accodng to Gutman ti poition i bette fo Blac, but afte 10   beleve that it  ve had fo

Blac to geneate countepla, fo example 10 h6 (10.xe4? 11 xd Ig3 12 d2 xd3 13 xd3 Ixh 14 h4 would leave Blac a piece down) 11 e3 and it eem clea at Blac doe not have enough com penaon fo the pawn. Blac need anothe move to eplace 1Q h6 fo 7  e to wo. 6 .. 8 7 .d2 xd4

Thi peudo-combnaon doent loo at tacve, but te altenave alo loo bad a) 7  xeS  deS xeS 9 000 e 10  (thi loo moe local than 10 e c6? 10 d6 would be a clea mpovement] 1 1 eS xc3 12 xc3 IdS 13 g3 Ie7 14 d3 and ite wa ceal bette  Bov azev, UR 1972) 1Q .d6 11 g4 and te ha good atacing chance. b) 7  dS doe not wo Afte  Ixc6 bxc6 9 eS cS 10 O-O-O Ig4 11 e1 g 12 b1 b 13 f4 \ite had a clea advantage in uaev-Kochiev, Funze 1979 The lne contnue 1 3fS 13  xf4? 14 xdS and \te wn 13h6 14 h3 cxd4 S xd4 cS 16 g Ie3 17 'xe3 d4  'g3 xc3 19 xc3 dxc3 20 'xc3 xf4 21 c4 give \te a clea advantage) 14 'f3 c6 S h3 Ih6 16 e2 xe 1 7 xe ih an obvou advantage fo \ite 8 xd4 c5 9 d3

9 'd6? xc3 10 xc3 Ixe4 11 "dS Ixc3 12 bxc3 f6 and Blac  9 . . x5  0 0-0-0

1 0  .d5? 91

The Fo Knghts

10...d6 11 f3 gves Whe a sd advan age bu hs s sl referable  he ex. 1 1 xd5

s s a precse reacn he aleaves ve Black gd pla: a) 11 g3 xc3 (1 1 .. .xe4? as lks gd) 12 xc3 x e4 13  e3 14 xf6 xf6 15 xd5 e6 gves Black he freer gae. b) 11 exd5 g4 12 g3 (12  f5 gives Black gd aacking chances) 12...xd5 13 xd5 'xd5 14 c4  xd+ (14...d4? 15 d3 xd1 1 6 xd 1 xd+  7 xd2 e4 18 .xe4 xe4 19 d3 gves Whe a sghy beer endgae) 15 xd2 �xd+ 16
Afer  1 .. .xd+ 12 xd2 xd5 (12...xe4 13 f6 'xf6 14 d+ and We aes) 1 3  c4 Whe has a clear exra pawn. e psn afer 3...xe4  4 xd5 d4 15 'e3 f5 16 d4 cxd4 17 xd4 c7 18 c3 des n prse real cpensa n.

c2 xe4 27 xe4 xe4 28 c3 s clse  g fr Whe whle 22 ... b3 23 a3 als des n acheve anyhn 23 �xc3 xc3 24 bxc3 Whe wud have very gd chances f wnnng he gae. 22 .. .b3?

Bh players sees  have prbles ndng e way. Beer was 22... a3 and Whie has prbles wh he hrea f . .. d4 fr exaple 23 e5 (23 b3 �d4 24 �c1 c6 25 e1 e8 and Back dnaes he shw) 23 ...c6 24 d1 d5 25 b3 �c3 26 c1 xe5 and Black s clse  g. 23 d1 ?

Here Whe shuld have played 23 a3 b5 24 c1 xd3 25 cxd3 d4 and al hugh Black has se cuner chances Whe s beer afer a ve ke 26 c 3. 23...bxa2 24 �xa2 a3?

Bad aackng echnque. Afer 24...b5 25 d2 b4 Whes kg s n grea danger. 25 b3 6 26 4 8 27 f6 g5

Beer was 12 f4 e8 (12 .. .g4 13 exd5 f5 14  e3 gves Whie a clear plus) 13 exd5 g4  4 f3 �f6 15 g3  f5 6 c4 and Black des n have enugh cpensa n fr he sacrced pawn.

Black has n defence a) 27...g6 28 6 f8 29 xf8+ xf8 30 xa3 and Whe wns. b) 27...gxf6 28 g+wns. c) 27...5 28 g7 b5 29 e6 e6 (29 ...e8 30 d6 and he dnaes he bard) 30 £ e8 31 �h6 e7 32 f+ xf8 33 f8 xf8 34 xe6+ g7 35 xc6 and Whe has a winnng enng.

1 2  .xb4

28 g4 8

1 2 .xb4?!

12...g4 13  cxb4 14 exd5  f5 15 d4 c7 16  d3 ves Whie he advanage. 1 3 f4 7 1 4 xd5 d7 1 5 h5 7 1 6 g3 Jxd1  1 7 xd1

Nw Black has se cpensan fr he pawn. 1 7.  ..6 1 8 �b1 a 5

Afer 18...a 5? 19 b3 d8 20 d3 c5 Back wuld have se drang chances. 1 9 .d3 a4 20 5  d7 21 d2 5 22 f4?

s leaves he dark squares n he ng sde unpreced. Afer 22 c3 bxc3 (22...c6 23 cxb4 d4 24 c3 d8 25 xd4 xd4 26 9

29 d5 1-0

  

ame 45 Nayranov-Sain Bihe 1993 1 4 S 2 f3 c 3 c3 f6 4 d4 b4 S xS! x4

This lne is ver rsky and  my opion not ver good.  g4

The crcal test.  �f? is not so good  ..f and now a) 7 g 0-0 8 g5 xd 9 0-0-0 f5  0 �f xc  xc ( �xc? e does not work)  . .e7  2 xf5 d   xf gxf   d7 xd7 5 f e ves Black a etter game though many compicaons sll re ma ) 7 e 00 8 d e8 9 xc dxc 0 0-0 g  g d7 2 h (2 g5 h5  h f  h fxg5 5 xg xg  hg f would leave Black wih the etter chances) 2..f5  ad? ( g5 d  f xd 5 xd d5 with equali was etter) xd  xd xc 5 xc e   ' f d5 and Back has a sight edge Adams-Howe Uoyds Bank 992

8 ...aS

t seems that 8 a5 is the ony good move. Fo 8 .. xd see Games -7 After 8..a 9 ax xc 0 xc dxc   xc hite has a clear advantage while he same goes for 8 .. h 9 xc d 9dxc 0 ax d5 [0g   g5   h and White wins) 0 xc dxc   d e7+ 2 d. 9 xc6 dxc 10 S 7 1 1 x7 �x7 1 2 d2 fS 1 3 bxc3!

A ice touch. Afer  xc xc  xc xc2  5 d2 g  h h  7 h c5 8 f cxd 9 e+ d7 20 cxd ae8 2 xe8 xe8 the posion was drawish  O A.Petrosian oscow 992. 1 3  ...xc2 14 c4 .xd2 1 S �xd2

 .. .xc3 7 xg7

Black comes out on top afer 7 a? f 8 ax x (8.. .xe5 9 g [9 dxe5 'xe 0 e d5 9   0 gx d5   a5 is ess clear) 9 xc (9 d d 0 g cd5    xd 2 cxd 'ews for Black) 9 ...xc 0 d xa   'g 0-0  2 c d   g5 'f5  d �e  5 c d5. 7 .. .f8

7...�f? 8 �xf e 9 c xf 0 xc xc  xc dxc 2 d gives White a clear advantage due to the o ish ops and the etter stcure. 8 a3

Less clear is 8 xc? dc 9 a 9 e e 0 a d would leave Black a piece p) 9  e7 0 e5 xe5   dxe5 d  ax x when Black has acve coun erplay

n his ending hite can connue to play for a wn forever without havng to take any risks at  1 5  ..g

 5    d xd  7 he+ f  8 xd ad8 9 c g8 20 g ge8 2 e ves White good chances to ght for a win in the rook endgame 1 h4! h

Or  .. .d  7 h5 f5 and now a) 8 d? xd 9 xd c5 20 h ad8 2 e fe8? afte the preferale 2..cxd 22 xd c 2 g xd+ 2 xd d8 25 c d7 2 g5 White wold retan oy some winning chances) 22 xe8 xe8 2 f e7 2 e d7 25 Mxe8 93

The Fo Knghts

xe8 26 c5 e7 27 e4 e6 28 g4 -0, stevez-Corjedo correspondence 988. b) 8 h4! ts ooks sronger) 8..g8 9 f4 g5 20 e and Black is under con siderable pressure. 1 7 h3 ad8  8 �c3 8 1 9 1  �8

9..d7!? is probably better  te king s needed in the cenre. 20 h3 x3 21 x3 8? 22 x8 �x8 23 d3?

This looks temping, giving a three versus to majort on e kgsde but it was bet ter to prepare s it 23 4! when te has some wnnng chances 23 ...xd3??

r 24..f5 25  e7 26 g4 f6 27 'e3 and now: a) 27 .. a5 28 'f4 b5 29 e3 g4 30 g4 �e6 3 d3 �f6 32 c3 �e6 33 d cxd5 34 cxb5 and Wte ws. b) 27 ..e6 28 g5 hxg5 29 hxg5 � 30 d3 �g6 3 f4  32 �c3 e6 33 b4  34 c5 �e7 35 d5 cxd5 36
c) 27 ..b5 28 d5! g4 29 g4 bxc4 30 c6
A bad mstake t was important to im prove the posion of the king before aow g Wte to mprove the posion of is king. After 23 ... 'd7! 24 g4 'e6 25 xg6 or 25 f4 xd3 26 'xd3 f5 27 g5 h5 28 c3 d6 29 'b4 b6 30 a4 a5 and White cannot w) 25..g6 26 d3 b5! 27  a Black draws.

25.. b5 26 cxb5 cxb5 27
24 �xd3

WlS

25 �4 �6

26 4 6

The endgame is also lost after: a) 26 ... b5 27 d cxd5 28 cxd5 e7 29 'd4 a 30
This endgame is wnnng for White. As pawn endngs have a tendenc to be an exact science we  have to be exact as we ome mght wonder what a pawn enng is dong n an opening book but is pawn endg s actuay ver close to the teorecal posion and could easly arse, so to me t seems ver natral to ncude it. 24 �7



Or 27...bS 28 cxb5 cxb5 29 d 'e7 30 �d4 a 3 �c5! but not 3 gS?? hxg5 32 g5 f5! and a of a sudden it is Black who comes out on top) 3...b4 32 axb4 a4 33 �c4 33 c6? a3 34 �xc7 a2 35 d6 < 36 d7 a  37 d8 c 38 d7 d 39 c8 a 40 �b8 b 4 a7 xb4 ony ves White good nng chances) 33...
   28 a4 �d6

g4 42 �b6  43 a

te also wins ate 28  b5 29 axb5 cxb5 30 cxb5 a4 31 'd3 a3 32 �c2 a2 33 �b2 �d5 34 g5 g5 35 g5 xg5 36 xg5 and 28.b6 29 d5 cxd 30 cxd5 �e7 31 �d4 � 32 �c4 �e7 33 g5 xg5 34 g5 g5 35 g5 ' 36 �b5 'g6 37 'c6.

33 axb5 cxb5 34 h5 a4

29 c5 �6

29 �e7 30 d5 cxd5 31 'xd5 ' 32  'e7 33 �c4 wit te sandard idea o b5) 33 ...c6 34 �d4 � 35 �e4 'g7 36 �4 � 37 g5 xg5 38 g e7 39 g �x6 40 �e4 and te wins

e also wns aer 34..d6 35 g5 and 34 .e7 35 g5 g5 36 g5 a4 37 gx6  38 d 35 g5 a3 36 �c2 a2

Or 36 ..g5 37 g5 xg5 38 d5 '6 39 d6  40 b2 g4 4 d7. 37 �b2 fxg5 38 fxg5 hxg5 39 d5 �f7 40 h6 g4 4 h7 �g7 42 d6 g3 43 d7 g2 44 h8   -0

ame 46 Yanemrov-Bezoov Smolens 1992  4 5 2 f3 f6 3 c3 c6 4 d4 b4 5 x5 x4 6 g4 xc3 7 xg7 f8 8 a3

30 �3??

A terible blunder. ite ins ater 30 5 �e7 31 d5 cxd5 32 'xd5 �! 32 .. c 33 e4 �e8 34 �4; 32 ... 'd7 33 c bxc6 34 'c5 �e7 35 �xc6 �d8 36 �b5) 33 �c4! c6 34 �d3 e7 35 �e3 � 36 �4 g7 37 g5. 30.. .b5??

305! 31 g5 5 32 d3 '7 33 �c3 �e7 woud ave drawn 3 cxb6 cxb6 32 �d3 b5

Now Black is ost again, as tese varia ons sow a) 32 .. .5 33 g 5 34 c4 �d6 3 g6 �e6 36 d5 cxd 37 �d4. b) 32..�e7 33 5 � 34 �c4 'e6 35 d5! cxd5 36 �d4 5 36 ...'d6 37 g5 g5 38 g5 'e7 39 gx6 8 40 �xd; 36 ...b5 37 axb5 a4 38 �c3) 37 gx5 �x5 38 �xd5 �x4 39 �c6 �g4 40 �xb6 x5 41 xa5

8 ...xd4?

is variation as some attracons be case it is qite compicated nortunate o Black it is also ver bad. ies king migt be walng around in te cente, but Black cannot acvate s pieces, and ite mmediael gans a strong iniave. 9 axb4 xc2  0 �d2 xa   �xc3 a5  2 c4! 7

12 axb4 is cnsideed in te nex game  3  ! d5

e alternaves also do no work a) 13  axb4 14 d2! d ite also wins ater 14 d5 5 b5 �d8 15 c6 95

he Fo Knghts

rnpoe o he ne gme 1 d e 17 d7 e8 18 g b3 19 e7 e7 20 f) I  d3+ d8 1 d 1 b3 1 7 f nd Blck reigned  mh-Fuh, corre pondence 1981 fer 1M1 1 8 c2 d 19 d2 f 20 g+ i l over. b) 13 b+ 1 d3 nd now b 1) 1.. .d I  nd Blck reigned  .CDiz-Am.Rodrigez, Cub 1981. Py could hve coniued 1. e 1 b+ c 17 d �d8 18 g+ nd ck i med b2) 1.d+ I 
Blck h o red 1  bc 1 7 c+ d7 18 d7 d7 19 e c+ 20
Whie, 198

Everchueze,

correpondence

1 7 b8! 1 0

Blck reigned becue of 17 . �e7 18 g d 1 9 f8 c7 2 0 d+ b8 2 d+ which led o me, or 17.d8 18 f+ c7 19 c b8 20 d 7 2 e+ nd Whie win.

ame 47 PolovoinRuman Lningd 1978 1 4 5 2 f3 c6 3 c3 f6 4 d4 b4 5 x5! x4 6 g4 xc3 7 xg7 f8 8 a3 xd4 9 axb4 xc2  0 �d2 xa 1 1 1 �xc3 a5  2  c4 axb4 1 3 �d2 d5

1 3 .e7 1 e d i conidered  h noe o Bck 13h move in he prevou gme. 4  b5 c6 1 5 1 !  piece join he ck

1 56

lck i ll ou of luck ) Ib+ 1
1 6 xc6! bxc6

Or 1  d 17 : e e 17...e 18 dn) 18  nd i gme over 1 7 Jx6 �d7  8 xc6

  

Black is an exchange up but the threats to his kig are impossible to meet

11 xg4 xg4 12 e1 �7 wit uncear pay

1 8. . .7

8 bxc3 xc6 9 00 xc3

After 18 :a5 Whte has the quet 1 9 e5 b+ (or 19b 20  �8 21 6 e8 22 c8 mate) 20 c "xb5 21 c7�8 22 e7 mate

After 9 5 10 b 00 11 a3 e4 12 xf8 xc3 13 "3 xb 14 xg7 �xg7 1 5 xb1 Whte is a litte etter  Nunn

1 9 �d1 ab8 20 f4 1 -0

tonger is 1 1 5!  hite shol prevent Back from paying 75 11a5 12 b5 a4 13 b2 6 14 xf6 gxf6 15 2 f5 6 b3 was angerous for Bac in agma suenBsger an 1 971

Black ha seen enough

ame 48 Skripenk auerAams Bundela 2001 1 4 S 2 f3 c6 3 c3 f6 4 d4 b4 S xS! 7

1 0 b1 00 1 1 b3

1 1 . . .c6 1 2 bS dS 1 3 d3 d6 1 4 c4 hS 1 S 2

Whte shou not exchange queens a pawn own of course After 15 xh5 xh5 16 e1 f6 17 g5 e8 18 bb xe1  9 xe 1 e6 te compensaon is te 1 S. . h4 1 6 h3

16 g3 ooks better 1 6. . .8 1 7 f3 hS 1 8 d3 d7

18  b6! 19 5! gves Whte some compensaon because of he strong bishops we 1 9 x �x 20 "f+ f6 21 xa8 xh3 woul be in Blacks favor 19 3 h6 6 xc6

Afer ths move Black has easy equaty For the crca 6 3 see the net game 6 .x4 7 2 xc3

Even stronger is 7  xc6 8 00 h4 an now a 9 e1 00 10 g3 3 1 1 f1 g4 12 f4 x 13 ax g4 14 c1 ac8 15 g5 5 16 x5 cx5 is competey equal  Gutman b 9  e6 1 0 e2 g4 1 1 g3 h5 1 2 f4 x 1 3 xh5 x 14 xf+ �e7 1 5 x1 �xf6 was agee rawn in D e a Via GarciaMellao, pain 200 c 9 g3 (vey risky 9  :h3 10 f4 g4 1Oh5 wou be too opmstic; after  1 e f8 1 2 f1 7 1 3 h4 hte i s better

Prophyaxis against 2 wth the threat of g5 foowe b "a5 attackg te queen sie 20 c2 ad8 21 d2!?

21 3 f5 22 a3 wt unclear pay was aso possible but Black shol avoi 22 f4

7

he Fo Knghts

23 d3 we te cme ut  tp. 21 .. .Cf6!

 Black were t a aleep wt a mve lke 2.e6 te te ammer wud al 22 6 g6 23 6 wt eru treat 22 d3 �f8

22.d? 23 cd c8 24 e wuld gve Wte te advatae a 24d? le t 2 6. 23 b3 b6

Ater 23..c8 24 d a6 2 4  26 d4 Black  uder eru preure Oe uld t uderemate te preure  te dark quare 24 f4 h5 25 d 1 g5 26 �h2

26 c2 wuld aw 26 ..3? 27 g4 g4 28 4 g4 wt d cmpea t r te quee

g4 f7 45 f1 h8 46 d1 c8 47 g4 8 48 d 1 �f8 49  1 g6 50 c2 �f7 51 �f2 7 52 �3 Cd7 53 �d4 5 54 Jc3 Cc5 55 d �f8 56 g4 i8 57 f1 h5 58 d1 .d7 59 h1 �f7 60 c2 h4 61 g4 �g6 62 f1 �h6 63 b1 8 64 b1 g6 65 f1 8 66 f 1 c8 67 nb1 �g5 68 d3 Cd7 69 f 1 C5 70 2 �h6 71 d1 8 72 c2 Cf7 73 d3 Cg5 74 b1 �h7 75 b2 �g8 76 b2 �f8 77 f2 �7 78 .c2 �d8 79 f2 7 80 f2 5 81 f2 8 82 d3 .d7 83 f2 �c8 84 c2 �b7 85 b2 8 86 c2 �a7 87 d3 c8 88 2 Cf7 89 b1 Cg5 90 b2 8 91 d3 �a8   

ame 49 Semek-Meani Bled 1998

26 . . ..c6 27 d5 d7 28 g3 g6 29 ic2 h5 30 h4 2 31 �g1 x3!

A aety acrce atr wc Black a little t ear He eve a te advatage but   ard t cme up wt a wig pla.  maybe t wa uecear but l t wa ard  ee w Black culd mprve  

1 4 5 2 Cf3 Cc6 3 Cc3 Cf6 4 d4 b4 5 Cx5! 7 6 d3

32 fx3 8 33 4 d2 34 d3 xd3 35 ixd3 c8 36 �g2 Cd7 37 hf4 C5

e crca tet. 6 ...Cx5

 perect blckade but w ca Black w ? 38 i2 �7 39 b1 a5 40 bf1 f8 41 1 id7 42 c1 f6 43 c3 8  98

A pble  6d4 7 d4 c ad w a Accrdg Lev Gua 8 d2? ve Wte a advatage but t  ard t ee w 8 .e 9 4 e7 9.d4? 0 e g4  d4 d4 2 b + 3 e2 b6 4 3 6  4 de eavy avur Wte) 0 e (0 d3 0-0   e2 d6 lk gd 

    

Black) 10 ...d6 looks better or Black. b) 8 d xeS 9 e 00 10 4 e7 11 000 e8 12 xc xc 1 h d6 14 e2 (14 g4? 2 woud expose the pawn and give Black good counterpla) 14 ... d7 15  c6 th an unclear middlegame ahead ZeicMikac Pula 2002. This might be the best line or Black  the whole sstem with 4..b4.

weakness o Whites pawns is less mportant. l Black can generate some counterpla ith ...d6h6 even though it is doubtl whether t be enough. b) 11...xe4 12 .x6 xd 1 xd gx6 14 e dS 1 5 e7 c6 16 de2 b8 17 c7 e6 (17...h6 18 ee7 ith a clear ad vantage  Gutman) 18 xh7 h8 1 9 d and White has the better structre.

7 dx5 x5 8 .d2 00 9 0-00 d6

1 0 4 7

Also possible is 9...xc 10 xc  (10.. xe4? 1 1 g g6 12 d  1 d2 hS 14 xc7 dS 15 b4 g4 16 .x8 xd 17 e7 g4 18 xb7 gave White a clear advantage n pielmann Bogoljubow tockhom 1919) 11 d2 (Whte comes out worse ater 1 1 b ? xe4 12 g xc 1 xc x 14 xc7 d6 15 .c4 S 16 b ac8 1 7 xd6 xc 18 xc2 xc+ 19 a h6)

10 ..aS 1  a would give White a serious advantage.

and now a) 1...dS? 12 exdS e8 1 b S (or 1 .. e4 14 d4 h6 1 5 bS c6 16 dxc6 bxc6 17 xc6 xd2 18 �b2 a6 19 xe8 c 20 bxc4 xe8 21 e and Whte ns) 14  e1 15 b2 gS? 16 h4 g6 17 d6 e4 1 8 dS e6 1 9 eS g4 20 d7 6 21 xb7 10. BellonLucacs Bucharest 1978. Better was I.x 16 gx d8 1 6 ...h 17 g2 makes no progress or Black) 17 x6 gx6 18 g2 though it does not oer Black enough compensaon or the pawn The bate il be n the quenside so the

1 1 1 d7 1 2 g3

White has seous attackng chances. The wa Black now tres to start an attack s too slow. e should have tried to pla wth his peces. 1 2. . .c6 1 3 d3 b5 1 4 h1 h5 1 5 3 h4 1 6 5 5 1 7 g3 g4 1 8 xg4 xg4 1 9 6 xc3 20 .xc3 c8 21 b4 n8 22 5

Whte has an overhemng posion. 22.. d5 23 .c3 a5 24 a3 b4 25 axb4 axb4 26 .xb4 6 27 c3 4 28 .x4 dx4 29 b3 g6 30 x4 h5 31 n5 gx5 32 x5 ix6 33 xh5 1-0

ame 50 Eorovulaev Aden 2002 1 4 5 2 c3 6 3 3 c6 4 d4 b4 5 x5! xc3!? 99

he Fo Knghts

A novel ea tat omes lose ng



equals

6 bx3 7 7 d3

7 x6 xe4 8 e2 x6 woul tans pose to Game 48

22 f4 d 7?

lak was ara o te treat o 5 but e a to brng te gt bak it 22.b6 even toug Wte as a terrible attak ater 23 g4 g6 24 4!

7  x5 8 dx5 x5 9 -

Wte as only a sigt avantage rom te openg 9 .d6

9 ..x3 10 b 0-0 11 b2 6 12 e5 woul ve Wte many attakng anes. 0 

nteresg s 10 4 x3 11 Mb 0-0 12 b2 "5 13  g4 4 e wen Wte as ompensaon or te pawn n s pressure on te ark squares 1 ...-   f4 h5 2 4 d7

12 .g4! wou be wasg me ter 13 3! (1 3 g3  14  5 s laks ea) 13.+ 14  6 15 g3 Wte wou ave goo attang anes.  3 .b2 5  4 g3

14 3 was nteresg.  4. . g4  5 3 a4?

Here te ngt s msplae Te bsop also look strange on a3 but t an move bak to te gse mu aster tan te kngt so te misplaement s less o a prob lem or Wte Ater 15. e8 16 Mae1 7 17 M "4 18 g3 6 19 " �8 te poson wou ave remane unlear.  6 ia3 J8  7 Jf3

Wte organses an assault on te kng. 17 5 x5 1 8 x5 x5 1 9 x5 wou onl be level.  7  .d7  8   i6  9 Jg3 6 20 5

Opening up or te bsops an limng te spae or te blak pees. Ater 20 4 6 21 5 x5 22 x5 x5 23 x5 8 lak keeps te balane 20 . . dx5 2 fx5 Jad8

lak s starng to eel te ol. 21..b6 22 g5! g6 23 3 5 24 "6  25 e6 g7 26 4 wit te treat o b2 s very angerous or lak. 100

23 �h ?

lopp Ate 23 6 g6 24 Mee3 b6 2 M3 6 26 ex6 5 27 b2 Wte as a inng attak. 23 . .g6?!

lak oul ave ougt bak wit 23 .. e6! planning 24 5 xe5 an lak wns. 24 g5 g6 25 4 remans unpleasant but t s s a gt 24 f 6 25 h4 b6 26 b2?

26 5 s more aurate (see below 26 .. .a4?

lak sou not go bak ere t 26Me7 27 a3 e7 28 5 x4 29 "x4 x4 30 e6 (30 x7 Mx7 31  xe5 32 4 s probab sigtl better o Wte but tis n o engame s ar t win) 30..4 31 ex �x 32 g4 �g 33 e7 85 34 + �6 35 x4 x an lak as goo rawng anes O ourse ts s not a ore lne but t oe sow tat lak as a tenabe poson 27 ia3 b6 28 h5 d7 29 b2 5 30 .f5! 7 3 hxg6 6

Or 31g6 32 xg6 an Wte rase toug 32 gxf7 �xf7  0

t s mate ater 33 xe xe6 34 5

   ummary

Afte 4  b4 thee is nly ne citical me because 5 d5 Ce7 des nt seem t e Whte a gd game at all Theefe 5 Cxe5 s abslutey the best me Then Black has a ae f chices. 5 ..Cxe4 6 g4! Cxc  xg7 f  a is ne f these Hee Blak is fced t g nt an uncmftable endgame with  a5 as afte Cxd4 White has a wnng attack staight fm the penig Back has als tied 5  0-0 but by etung the pawn and nshg s deelpment, Wte hlds a easnable plus. Ts leaes Back with 5 .. e7 and 5 .�xc3+, bth f which hae faied t equase cmpletey, but they d hae sme qualty In the nal analysis we can say that White is slghtly bette afte 4 �b4  4 5 2 f3 c6 3 c3 f6 4 d4 b4 5 x5

5 d5 Ce7 6 Cxe5 d6 7 C ()  Game 42; 7 �b Game 3 5 .. .tx4

5 .. �xc+ Game 0 5  0-0 -Game 44 5 . e7 6 d3  Game 9; 6 Cxc6 - Game 48 6 g4 xc3 7 xg7 f8 8 a3 xd4   �a5 () - Game  9 axb4 xc2  0 d2 xa    xc3 a5 1 2 c4 () axb4 - Game 47

12...e7  Game 46

7 f3

1  

101

CHAPTER EIGHT The Belgrade Gambit

1 4 5 2 3 c6 3 c3 f6 4 d4 xd4 5 d5

Th Blgrad ambit ariss aftr th abv squc This gambit is spciay ppular at cub v hr hm prpara  has t rachd th v f prfssia chss. Th ida sms  b t sacric a pa ad gt cmplicas  mattr hat. Back has fur mai rpis 5...xd5 (ams 51-52 5 ...b4 (ams 53-54 5 . 7 (ams 55-5 ad 5  x4 (ams 6065

ame 51 Sviler-Cerepkov Lningd 1990 1 4 5 2 f3 c6 3 c3 f6 4 d4 xd4 5 d5 xd5

 th psi afr 5 d5 Black has may altativs H has trid th txt mv 5 ..b4 5 . 7 ad 5 x4 Hr is a summary f lss mptat lis a 5 .d6 6 xd4 7 (6 .x4 7 b5 uld f curs b highy ufrtuat 7 b5 d7 ad ht ds t appar t hav a advatag. b 5 .. h6 (mayb Black sll has a dct psiti aftr is mv but it ds tg fr th dvlpmt ad i crats ak 10

sss 6 xd4 c5 7 b5! 0-0 ( ..b6  xb6 axb6  d3 givs t a sight dg  bxc7 x4  3 x3 10 3 b (r 10.  . 1 1 g3 xg3 12 hxg3 xh 13 xa ad it s 11 g3 d6 12 g2 f5 13 0-0 ad accrdig t Brc s it has a sma dg. s is a Amri ca mast ad grat spciaist  h Bl grad ambit. c 5 .b4+ 6 xb4 xb4 7 xd4 ( 5 7  2 g4 [ ..4  0-0 c6 10  h6 11 xd4 00 12 3 g5 13 dS gav it a clar advatag hmirBzski isla 12]  f4 d6! gvs Black a gd gam as 10 xd6 4 uld lad t imm dat disastr. d 5  c5 6 g5 7 (6  h6 7 xf6 gxf6  "d2 d6  b4 b6 10 b5 f 11 0-0 uld v it a larg advatag 7 x7 x7  d3 0-0 ( .. b4  d2 d6 10 0-0 6 1  f4 00-0 1 2 f5 d7 13 a3 "c5 14 b4 5 1 5   1 6 b5 5 1 7 xd4 gav \hit a clar advatag i th gam s-Chss Pr 3.5  0-0 d6 10 b5 d7 11 1 f 12 xc6 xc6 13 xd4 ad ht has a sght dg  s. 6 xd5 b4

6 ...b4 is csidrd i am 52. Black l t d ay rf i 6  7+ Th chck lks ctrary t a lgic as h

he Be/gde Gmbt

f8-bish is blced and the fbish is frced t devel  e2 b4 8 0-0 c5 9 xd4 xd5 1 0  c6 (r 10 . e 1 1 e d6 12 e3 a5 13 c4 c6 14 f4 and White is better i every cnceivable way) 1 1 e1+ d8 12 b4 c4 13 g5 f6 14 xd5 xd5 15 f4 xb4 16 c4 xc4 1 e6 e This was layed in Krnghy-vaic UR 194 Nw White has the cmbiain 18 g d8 19 d4!! winnig the qeen  Mnsn 7 id2 7

Blac is wse after the less natral  ..xd 8 xd2 e+ (r 8. e 9 xd4 0-0 10 0-0-0 d6 11 g4! b6 12 h4 and White has gd attacig chances in Fernandes ngh Dbai 1986) 9 e2 b4 10 xd4 (interesg is 10 00 0-0 [with the idea 10.xd5? 1 1 c4 c5 1 2 g5!!] 1 1 c4 b5) 10.xd5 1 1 f5 e5 12 xg+ xg 13 xd5 xb2 14 0-0 d6 (14..0-0 1 5 ab 1 g 16 b3 h8 1 g3 f6 18 h5 [18 c5!? d6 19 xc with advantage was cleaer 18d6 19 d3 gave Whte a srng attac i Fabregas-Magnei ai 1 992 Af ter 19 h6 White shld be awae f rics ie 20 e e6 21 g6 x? wh an even endn 1 5 ab g 16 b3.

0 ..7

Als iteresting is 10 b4 1 1 he (1 1 d6!? is interestin 110-0 (11xd5?! 12 c e 13 e2 d5 14 xd5 g4 15 ae1 0-0-0 16 xe xd5 1 x xf3 18 gxf3 d 19 ee xe 20 xe g6 ls cmftabe fr Blac althgh the end game is nt s easy t in fr Wite) 12 d6! c5 1 3 c4

8 2

Frced After 8 e2?! d3 9 cxd3 xd 10 xd2 b4 11 0-0 00 12  xd5 here is n real cmensatin fr the awn 8 .. .ixd2 9 xd2 x2

The best 9 .. b4?! 1 0 xe xe 1 1 c4 d6 1 2 he ! c5 1 3 b3 xd5 1 4 e c6 15 xd5 f6 1 6 f5 g6 1  f6 xd5 18 e gves a cear advantage accrding t Mnsn And 10 d6! ls even beter  Blac has n way t eqalse anymre. One idea is 10 cxd6? 11 b5! fllwed by e wining materia.  0 ix2

This sin is mre r less eqal te reticaly seaing bt Wte has a dangers iave Therefre 6 ..b cat be rec mmended fr Blac.

Nw Blac has ed the fllwig mves a) 13 c6?! 14 g5! d8 15 e gave Wite a big edge n Olszewsi-McKaig CCF 192 b) 13..a6 14 e?! (14 a3 c6 15 d5! gves Whte a clear advantage as Blac can nt easy free himself; the it is hat 1 5.b5 1 6 e ! b8 1 xc6 xc6 1 8 d b 1 9 b4! simly wis fr Whte) 14b5 15 a3 xc2! (1 5 ..c6 16 d5 ves a clear advantage accring t Pvoi b a can 103

The Fo Knghts

be seen bove Blck hs no defence t l) 1 'xc2 (1 d x1 17 x8 b 18 �c2 c4 gives Blck resonble counterply) 1 .bxc4 1 7 eS b7 18 xd7 8 19 xc xg2 nd Blck is in the gme. c) 13..bS 14 xbS b8 1S c4  1  eS xc4 17 xc4 gve White  cler dvntge in Polovodn-Tseitlin eningrd 1979 d) 1 3.g 14 e7 bS 1S xbS dS 1  eS b7 1 7 c4 f 1 8 e1 gives White  cler dvnge  Gutmn e) 13  b! 14 e7 (14 eS  1 S x x 1 xd7 fd8 17 e7 b4 18 eS xd ves Blck the better gme) 14 1S x x 1 xd7 (1  b8 17 e f mght give Whte  sight dvntge but it is doubtl) 1.. 8 1 7 xd8 (1 7 b7 xd with uncler ply ws plyed in rnovGusev UR 19S9) 17.x d8 18 e 1 �f8 19 e7 xd 20 x7 nd the endgme is microscopicly better for Whte 1 1 d6

Ths is he themc , but oher moves hve lso been nlysed ) 1 1 c4! nd now )  1 cS 12 dxc dxc 1 3 xd4 g4 14 e �d7 S e3 s better for White nd so is  1 .d 1 2 xd4 d8 1 he  1 4 f! when Blck hs problems getng ly mobilsed (for some reson onson only gives 1 4 xfS wih equlity)   1 ..bS! 12 b3 (White hs no dvn tge fter 12 xbS b7 1 xd4 xdS which gives  dynmic nd uncler posion.  computer mght prefer White but here re no weknesses to trget nd  he long rn the open b-le shoud give Blck excel lent counterply Tht is lso why fte 14 he Blck would ply 14d8!  order to brng both rooks into ply) 12..b7 13 xd4 xdS would gve  sfe pth to equl ity while fter 1.  14 he1 d8  S d! cxd 1 e2 White woud hve compens on for the pwn nd good resons to de cine  dw offer 104

b) ore mbious is 11 he1 d (for  1 ...xdS 1 2 c+ e7 1  e2 see the note to Blcks 10th move) 12 b5 'f8 (ctly Blck would be better off plyng 12d7! 1 xd4 �d8 14 c4 g S e3 wth  worse but defensible position) 1 xd4 (1 e4! fS 14 xd4  1S d xd3 1 Mxd �e8 17 e1 'd7 is equl) 1  . 14 e3 (wthout ths ide White would hve nohn 14 . cS (14xbS 1S e1 cS 1 xe7 e 17 c7 xdS 18 xbS nd White is close to winnin I S e 1 cxd4 1 xe7 e 17 c7! xdS 18 c4 xc4 19 xc4 hS 20 xd4 h nd Blck hs som chnces of sving the rook ending wih ctive ply 1 1 . .  c6 1 2 ad1

Both the lternves re more energetic ) 12 dxc7 d 1 bS e 14 xd4 'd7 IS xe e 1 he wih n dvn tge for te in White-Wiims Los nge les 1995 b) 1 2 he1 cxd 13  nd now b) 1 3 �f8 14 c4! (14 d dS!  S e2 d nd Blck is going to surive) 14 d8 (or 14eS 15 dS th  cler dvntge for White e4xd is coming nd so is -f4) IS e4 e 1 xd �e7 17 bS ih n dvntge for White b2) 1 3. .. 00! 14 e4 dS 1S d wih  distinct plus for White  Gutmn b) 1  . d8 14 f (14 b5  1 S e4 'e7 1 g+ e 17 f5+ 'd8 18 xd 'c7 19 fS b  gives Blck  good gme) 14..e 1S e4 �e7 1 g �f (1g 1 7 e2 'f 18 �d3 e8 1 9 xd4 gives White the dvntge) 17 d1 b8 nd White hs ood ply but wheer ths i enough to clim n dvntge is not certin  2. . .cxd6 1 3 �c1

13 he1! 'd8 14 �c1 e8 1S xd4 xd4 1 xd4 b would equse 1 3. . .b6?

This seems sound fter 1dS 14 xd4 xd4 1 S xd4 d 1 xdS White hs  much better endng but sl it requies very

The Be/gde Gmbt

ood ehnique o n uh a poiion 1 4 he  >d8 1 S c4

Or 1 5 xd xd 1 6 xd e8 17 xd6 a6 and a daw i ey ley

o 1 5 "d6 15 d d7 16 h5 h6 and hie doe no hae  ompenaion for he pawn 8 S!?

Or 8 "?! f6 9 5 "e7 10 e2 e5 1 1 xf6 �xf6 12 �xf6 xf6 13 0-0-0 6 and Ba i bee  Monon 8 . .e7 9 gS

1 S. . .b7 1 6 xf7 eS  7 xeS dxeS 1 8 xeS f8  9 dS  

A draw wa aeed probaby beaue of 9 xd5 20 xd5 xf2 21 1xd 7, when Whie adanae i impreie ony from an aehe poin of iew

ame 52 Bello opeamieso Wk aan Zee 1977 1 e4 eS 2 3 c6 3 c3 f6 4 d4 exd4 S dS xdS 6 exdS b4 7 xd4

Thi i more naura han 7 , whh ae y ranpoon  Game 5 7 . . .xdS I prefer he foowin ine 7 .. f6! hreaenn 8..."xd and hu on down Whie deeopmen 8 3 xd5 9  b6 hi i he naura moe Maybe he boo hae oeooed he enh of 7 .f6! beaue of he folown ame: 9.e 1 0 d2! "5 1 1 2 6 1 2 d3 f6 13 e1 d8 1 "h5 wih a on aa, zmean-uo, Mar de Paa 10 d3 e7 1 1 0-0 00 12 e1 d5 13 f3 e8! (13h6 1 2! wih he hea of d3 i en by Monon Here Whe mih be a ie beer een houh i anno be a

9 . 6

Thi alow for Whie o ar a ron a a GM Le Guman reommend 9 d5 1 0 xe7 xe7 1 1 x7+ f8 12 h5 d6 wih unear pay. Ao poibe i 12..b 13 3 e7+ 1 e2  15 f x2 16 xe2 d6 17 00 17 xd5?! e8! ae Ba an aa in Hoaeze Podebrady 1956 176 wh a dynamiay baaned poon  0 x6!

Th i naura he idea behnd Whe pay 1 0. . .gxf6   hS g6 1 2 0-00

Monon e he foloin ine: 12 ? 6! 13 0-0-0 d5 1 h 7 15 x6 hx6 16 �x6 f8 17 he1 h6 and Ba  wnnn 1 2. . d6

Ba anno open up hi in poion oo muh, bu i i hard no o Afer 12 .d5! 13 h! (1 3 xd5?! xd5 1 7+ xg7 15 xd5 6 16 e+ e7 17  f5 i no ear a al oy Mhai Ta ha eer y underood hw o ue he ueen aain an 10

The Fo Knghts

cea f mr pieces) 13...7 4 xdS e7 1 5 d3 c6 16 xg6+ hxg6 17 xg �f 1  d4 e6 9 d3! Whte has a very strg attack fr the pece ut maybe hs s the best ie lack has at hs dispsal.  3 h4 7  4 4?

Ths lks very atural but w lack has the chace t ee t the queesde wth hs kg whch s very pleasat as Whtes peces are al eher  the kgsde, r ected agast t. etter was 4 e   1 5  c4 wth the idea f xg6) S..dS 16 xdS g4 17 xg6!! xdS 17e6   xe6 hxg6 19 xg+
23 .  xb2

lack s pulg the emergecy crd. After 23xe6? 24 xd d7 25 xf6 bxc4 26  b6 27 xa7 Whte s much better because  lacks weak kg 24 xb2 bx4 25 3 7 2 d2 xe 27 xd .d7 28 xd7

Whte escapes wt ths perpetual check 28 .. xd7 29 x5 e7 3 5 d7 3 f5 e7 32 e4 d8 33 d5 e7 %  %

ame 53 Van HaaserGolo Dieren 1998  e4 e5 2 3  3 3 f 4 d4 exd4 5 d5 b4  4  4. . d7  5 he  d8   x hx  7 x  h  8 b 7  9 e4 b8?

lack culd have repulsed Whtes atack wth 19.. fS! 20 dS f6 21 e2 Jg7 22 c3 f4! whe lacks peces wil develp ad utumbe he small white army. 2 d3 f5 2 d5 f 22 a5 b5?

Ater 22 ...g7 23 c3 a6 Whte prbably cat make ay pgress. lack wuld  after smehg ke 24 e6 b6 25 a3 Jd7 26 Jxd7 \xd7. 23 e 106

Ths s the ly serus mve The alteratves are t prmsg a) Whte des t have eugh after 6 x x6 7 c4 d6  gS?! ths s reckless but  a3 c6 9 bS g6 wuld as gve lack the better game) dS 9 exdS JS 10 Jb5
The Be/gde Gmbt

c2 b and lack n 2...c  3 c2 b  wn c2  c2 S and ack ha a clea advantae. b 6 d? e (the tet 6 ...bdS 7 edS cS 8 e and accodn to Nunn Whte may be hy bette and now b  Afte 7 bS dS 8 dS e 7

  a new dea and a vey ton one. ntead 8 .. b (at leat  tn t deee a queon mak 9 c3 e7 0 c7 'd8  cb  2 e6 e6 3 b3 ve an unclea poon accodn to Monon analy. Acualy Whte mht be bette a ack  potenally vey weak on the dak quae. Gon back to 8 ... e7 we have he ol lon connuaon: b 9 c7 �d8 0  d6  0-00 'c7 and lack n becaue afte 2  S hte doe not have enouh compenaton o the pece. b2 9 d  he choce o Fz 7 but afte 9 ... dS 0  c3   'd2 bS 2 b5 c6 3 he  .e6 Whte doe not have any weaknee to attack wth h add onal development. lack  clealy bette. b3 9  (the bet 9 ..c 0 eS bS  bS c6 2 d3 d6 3 e7 e7  0-00 ..e6 any compenaon fo the pawn  an lluon. b2 Accodn to Monon 7 fS  mply bad. n my opnon t not th move that to blame but the ad mtake made on the

pevou move 7..c6 8 b b (8..aS a n Pe-Pak Pa 990  not o clea 9 c3 6

and now b2 0 3 c3   .d2 (  a3 e 2 e3 aS 3 b c7  3 e 5 e eS 6 d d 7 d 0-0 and ack  wnnn   ... 6  2 bc3 S 3 d c  d3 dS and Whte  to pawn down fo notn. b22 0 7 'd8    (  c2 c3 2 a3 aS 3 b e  e3 b6 and lack wn quckly ...c3 (a won del tte combnaon 2 bc3 7 3 b2 dS and Whte mut be lot  both ..e8 and ... ae theat. 6 .. bxd5 7 exd5

7 dS c6 wok n lack avou. 7 .. b4+ 8 id2

Whte ha ao epemented wth 8 ' 0-0 9 d h6 0 h d6  S c 2 f6 6 3 6 6  d3 e8, wth equalty n the ame Hecto-Kaoly Copen haen  985. 8 .. :e7+ 9 e2 xd2

 9 ...e 0 e2 cS   b b6 2 a a 3 baS aS  d whch wa ouy equal n Fahne-Wel Gaz 99.  0 �xd2 xe2+

0 ... e   d3 c 2 'd 2  3 e2 b6  he  00 5 b ve a clea advantae fo Whte  Monon.   �xe2 c5 107

he Fo Knghts

so perfety possbe s 11.d6 12 he (12 Cxd4 d7 1  he [or 1 Cb5 f8 14 C e8 1 5 d2 Ce4 16 Cxe4 xe4 wth a key draw] 1. . -- 14  he+ 15 f2 e5 and Bak s at east equa Grmberg ear, Begrade 1988) 12... 1 d �f5 14 xd4 fe8 (14.x2 15 e7 gves Whte good pay) 15 b e4 16 xe4 xe4 wth a probabe draw. Bak an also pay the more tesng 11.b5 12 xb5 (te should avod 12 b b7 1 Cxd4 Cxd5 14 Cxb5 [rsky s 14 he 1 f8 15 Cxb5 Cf 16 d2 a6 17 C Cxg2 18 Me5 Ch4 and Whte does not have  ompensaton for the pawn] 14.Cf4 15 d2 d8 and Whte s under more threat than Bak) 12...Cxd5 1 Cxd4 Cf (1.b8 treatenng ...7-5 s an nteresng alternave) 14  Ce6 15 he b8 16 Cxe6 e6 17 xe+ d8 18 e5 a6 19 �d xb2 and Bak has good pay 1 2 dxc6 bxc6! 1 3 xd4

Whte ould aso onsder 1 h e and now a) 1 . .. d5 14 f1 d8 (14..f8 15 d  5 16 b4 xb4 1 7 Cxd4 a5 1 8   woud gve Whte ompensaton  t s hard for Bak to nsh hs deveopment) 15 d 5 16 b4 xb4 17 Cxd4 a5 18  bx 19 a1 a6 2 �xa6 xa6 21 x e8 22 b wth ompensaon for the pawn. b 1    14 Cxd4 b8 15 d 2 d5 16 d 5 17 C6 xb2 and Bak s at east equa

Cxe8 Cxe8 2 bx4 �x4 4 h4 e6 wth an unpeasant poson or Bak) 2 axb �xb 21 Cxe8 xe+ 22 d2 �x2 2 x2 e+ 24 d xf2 25 xa7 h5  the endgame shoud be drawn 1 .. .ad!

Now Whte has probems keepng every thng together 19 he1

9 h1 d7 2 g Cg4 21 g2 Ce5 22 f d4 2 bx4 ed8 woud also eary faour Bak 1 9. . e7 20 c3

Or 2 xd5 xd5 21 �2 d2 and the rook nvason s very unpeasant. 20 de 21 ec 1 d4 22 a4 g4+ 23 g3 e5 24 3?!

s s a bad dea n a dfut poson now he seond rank s serousy weakened 24 h was neessary. fter 24..d7 (24 .�e2 25 �xe2 xe2 26 Cx5 woud b unfortnate) 25 Cb2 Ch5 26 h2 Cf4 27 d t s not so ear how Bak s gong to oet the  pont

1 3. . d5 14 d3 c5 1 5 b5 00

Bak s dentey not worse. 1 6 b3

Bak woud wn after 16 4 e6 17 C7 (17 b woud transpose bak to he game) 17.. dx4 1 8 2 Mab8 1 9 Cxe6 fe8 2 d2 xe6 21  e2 22 hf d8 2 ad d4 1 6. . .e6 1 7 c4 e 1  3?

Whte oud s have kept the balane wth 18 C7 dx4 19 2 xb (or 1 9..g4 2 f e+ 21 g ad8 22 108

24 .. d7 25 b2

Whte woud aso be serousy worse after 25 f2 xa4 26 bxa4 e 27 d Cd7 28 e a  the dbshop s reay weak and t s lkey that a knght versus bshop endgame be terrbe for \hte 25. . h5+ 26 f2 4 27 c2 xd3+?

Bak mstakeny tnks tha t s me to oet the frts of hs hard work It was

The Be/gde Gmbt

better t put extra pressure  \hte with 27 ..Me3 ad after sethig ke 28 d2 28..c6 t ctuusy itesi the pressure  \hites ps. 8 xd3 5 9 xe5 xc 30 e 6

This is what Back had bee cug  but te ca escape with e ve 3 c6!

The y ve. 3  . . .Ja8

31 ..�f8 32 e2 g6 33 xa7 xe 34
Ts s ufrtate After 32 e7 a 33 d7 a6 33  b 1 34 a3 a2 3 d gives te the advatage) 34 c7 \hte has a sghty better ps 3 . a4 33 b5 a6

e7+ 8 �d

\hte ca as  8 �f1 ad w a) 8..b 9 xb b7 10 xd4 xd 1 1 d2 0-00 ves a very essy sia where t s reay hard t say wh s better ad why b) 8 . c 9 xd4 Back s ceary better after 9 e e7 10 d2 b 11 xb b7; te ca reay fee the prbe f hs cetrased g fr exape 12 c4 dxc3 13 bxc3 xd 14  c6 1 c4 xc4 16 xc4 a6 17 xa6 xa6 ad Back has a paw re  the edgae) 9 .. xd4 10 xd4 c ad w b) 1 1 g? f6 1 2 b
urprisigy t turs ut that the kight is trapped 34 b �7 35 �e �e6 36 �d �d6 37 a3 g5 38 4 gxh4 39 b4 cxb4 40 xb4 axb5 4 cxb5 �c5 0-

ame 54 ola-S Polar Braz 99  e4 e5  3 c6 3 c3 6 4 d4 exd4 5 d5 b4 6 c4 xd5 7 exd5 109

he Fo Knghts

8 .. :6?

he can resgn

Probably too slow. The Queen can be at tacked from he g5-square. There are many alteaves hat should b ooked into a) 8..g6? 9 e ! (9 e?? h6+would be awkward) 9 ...b5 (9..c5?? 0 d6 xe  xe d8 2 e5 and ite ns) 0 xb5 b7  d xd5  2 g5 xe  3 xe e6 4 xd4 with a clear advantage for White. b) 8.�d6 9 e e7 0 'e2 was played n HankeHaussler Eppngen 988. Now .b5!  xb5 xd5 2 d b7 ooks vey promsing for Black One possible ne is 3 g5 f6 4 d2 c5 5  c4 g5! with a stong iniave. c) 8.. .c5!? looks very logcal 9 e e7 and now c) 0 e ? was unsuccessl after ...b5!  a3 ( b3 b 2 e5 f8! 3 f5 d6 4 a4 f6  5 axb5 xd5 6 xd5 d3! and ite has no defence while  xb5?? �xc2 s mate) ..xc2 2 b4 xb4 3 axb4 xb 4 d bxc4 wth a g posion for Black ankenau-Pedersen Hamburg 993. c2) 0 a3! xd5 or 0.d3!?  cxd3 xd5 2 b4 �d6 3 xd5 �xd5 4 b2 wth an unclear poson  Monson; here Back should probably play 4.d8 5 he  e8 when xg7 can be met by ...g)  xd5 xd5 2 e  �d6 (2. .c5!? 3 d d5 4 b4 d6  5 xd4 c6 6 g5 f6 7 f4 d8 8 c g5 9 d2 h5 20 c4 h7 is unclear and rougy level) 3 d c5 4 g5 f6 5 f4 'e6 6 �d3 �d5  7 �e2 e6 with equaty  Monson.

 0. . .b5

9 e2+ e7 1 0 e1 ?

A more accurat move order miht be 0 d! h6 ( 0b5?  5 and ite ns)   e  b5! transposing to the game. After c6 2 d6! xd6 3 d2! a 4 xd4 ite is on his way to victory. Back seems to b forced into 4 ..g6 5 c3 b5 6  f4 �c5 7 b3 'h5 8 g4 h4 9 d6 when 1 10

..c6  d! h6 2 d6 would only transpose to the prevous note. 1 1 b3

Not giving Black the chance to free hm self n the cene.  xb5 xd5 2 d b 3 g5 e6 4 �d3 g4 5 xd4 �xd4 6 xd4 is very close to equality even hough White is sghy better devel oped 1...h6

Black plays a prophylacc move drected aganst g5. After .a5? 2 d h6 3 a3 a6 4 xb5 0-0 5 xa5 Black does not have sufcient compensation wle after   .b8?! 2 d h6 3 g4 ite has a strong attack 1 2 �d 1 

ite is explong hs iitiave. fter 2 xb5?! 0-0 3 a 3 a6 4  a2! b8 Black has good counteplay.  2. . .�d8

2  d3!? seems better but Black s s in rouble 1 3 g4!

White is exploiing the moment. 1 3.  . c5

3..e8 4 a3 a6 5 g5 gives Wite a decisive attack.  4 g5 g6

4 .. f5?  5 xd4! would be a seous blow.

The Be/gde Gmbt

 5 e5?

This looks vy tmpng, but it was stong to play 1 gxh6 d3 (1  b7?  6 g 6 1 7  and Whit s) 1 6 cxd3 xd3 17 hxg7 g 1  g 6 1 9 d6 x 20 c1 cxd6 21 xg d 22 xd3 �xd3 23 b3 and Wht ns Evn 1 xd4? hxg 1 6 a3 a6 17 xb looks vy pos g o Whit.  5.5  6 a3

At st sght Whit smply ns a ook a t 6 d6, but thn you spot 16 cxd6 1 7 x c7 1  xh b7 and you ais th ook is tappd on a o th momnt.

xd3 xd3+  cxd3 x 3 e d4 4 b xb 5 xb h3! 6 c d6 7 e4 e6  that mains is ndgam tchniqu.

Th xchang is natualy no match o Blacks pawns 8 d 5 9 b7 c7 30 c6 �d7 3  �e b3 3 c3 c4 33 g e3+ 34 � 4 35 e4 e6 36 5 d5 37 dc �e 7 38 c8 �6 39 xe6 xe6 40 8 �e7 4 a8 xd3 4 xa7+ �6 43 c6 e4  aa6 e3+ 45 �e  �e5 0

ame 55 Sviler-Morozevi Moow 2002

 6 .a6  7 d6??

Wht als o th atal attacon o w ning an xchang. H could still w th 17 gxh6 gxh6  xh6, xpoig th bad posion o th Black king. 1. b7 (..xh6 19 c xc6 20  mat) 19 g+ c (19..6 20  c 21  x 22 x mat) 20 g6 xg 21 xh xd 22 x x 23 x dos not v Black nough compnsaon o th xchang.

 e4 e5  3 6 3 c3 c6 4 d4 exd4 5 d5 e7

This sa mov lads to quaty ow Whit has ou opons 6 4 (s and  th olowng gams), 6 c4 (Gam 9) and th to dscussd blow 6 4  7 cxd6  8 d5

t is going o th oth ook, but it is also hopss. 1  x c7 19 xh b7 20 4 xh would giv Black a stong attack o th xchang With Whits kng stuck n th cnt and th ook buid on a, h has littl hop o suia 8.dxe5 9 x a8 hxg5 0 b4 d3 

Th most popula mov. Whit has aso td a) 6 xd4? xd 7 xd xd4  xd4 00 is at last immdiat quaty o Black 9 2 6 10 �d? (10 d3, as playd  Tal-ain, ga 194, is pobably btt, though Black sl has no poblms) 1 d6 11 0-0  12 d3 �d7 13 c3  14 a4 111

The Fo Knghts

e4 wth a ne game for lack. b) 6 b5!? and now b) 6 .e4 7 - c5 (Maybe not he most natural, but the alteaves are not necessaly better 7..a6 S f4 d6 9 a4 - 1  e1 f5 1 1 d4 h4 12 g3 d4 1 3 'd4 g5 14 b3
Jf5 14 f5 c  15 c (or 15 d3 'f5 1 6 Jf5 feS 1 7 f6 gf6 wh a slght advantage for lack) 15acS 16 f6 gf6 and ack was slightly beter  Mrkov-Kosc, oa 1947. 6 .. .d6

The more risky 6 ! s nvesgated  Game 5S 7 xd4 O-O!?

lack can equalse smply with 7 .d5 S ed5 d4 9 'd4, for eample a) 9 ..J f6 1 'e (1 'd3?! 'e7 1 1 e2 - 12  J f5! 1 3 d2 e4 14 g4 g6 1 5 b 1 feS and lack has some na ve) 1...e7 1 1 e7+ <e7 12 -- f5 13  h5 14 b5 hdS wih equalty b) 9 .- 1 -- f6 1 1 �d2 'd7!? (1 1 c6!? ypcal for is ind of postion 1 2 dc6 'b6 1 3 c3 bc6 14 Jd6 dS 15 c4 a6! 16 b3 Je5 1 7 e5 d2  S d2 dS 19 e d2?? 2 c7!!) 12 d3 (12 c4?! b5 13 e2 a 14 g4 b4) 12'a4 13 b d7 14 he feS with equaly n aairGurech, I 1994.

8 b5 xd5 9 exd5 a6!?

This move s not vey popular, but t s quite good. 9 b4 and 9  e5 are nves gated  he two following games. 0 dxc6

Wte has o ateaves a) 1 d4? d4 1 1 'd4 f6 12 'b4 e& and Whtes king s under attack. b) 1 c3 e5 1 1 e2 and now:

he Be/gde Gmbt

b) 11.g5?!  g3 g6 3 0-0 f4 14 �d4 g5 (4.xg3  5 g3! would ve White good atacking chance along the open f-le and Black ome problem with pace)  5 ae1 f5?! ( 5..xg3 6 g3! d7 7 d3 e5 1 e4 �h6 9 c3 with a mal advantage for White wa better) 16 h4! h6 7 h5 e5 1 xe5 xe5 19 f4 g4 20 xg4 xg4 1 f5 xh5  e3! f6 3 h3 gave Black a long lit of problem, Van der Vaide-llenbroek The Netherland 1 995 b)   ..g6! (hi i an improvement)  e3 ( g3?! f5! 3 f4 c5! foowed by ..f6 and poibly .b7-b5 ve Black the better poion) 1. £5 (or  ...g5!?) n both cae wth complete euaity. 1 0    axb5 1 1 ixb5

12 xb7

Or: a) 1 c1 e 3  (3 e3? xe3 14 e3 g5! and hite i in eriou trouble; hi i the main reaon for playing .e and not ...�e+) 13. . xf4 4 xf4 e5 5 cxb7 (15 c4 b6 would cut off the b5-bihop from the acon) 5...xb7 16 d3 e7 with good compenation for Black. b)  g3 e+ i alo be ne for Back, a after 13 e bxc6 Whte ha nothing

Thi i the mo logcal After   f3?! Black ha two path that lead to a ight ad vantage a) 11.g5 1 xb5 f6 (but Black hould not fall for the trck 1 ...e! 13 e3 xe3 13. ..a5!? drecy might be bet ter 14 -!! a5 15 cxb7 xb5  6 bxc� xc 7 e3 "e6 1  b3, when White tand better becaue in the rook ending he ha a very dangerou paed pawn on the ale and hi rook i wel placed behnd ti pawn) 3 xg5 �xb 14 00 �xb5 1 5 cxb7 xb7 and Black ha good chance  a, c and g are a good colecon of weaknee b)  1 . bxc6   xb5 b 3 xc6 xb 14 e4 (or 14 0-0 xc and White doe not have ufcient compenaon for the pawn) 4...e6 5 0-0 xa 16 xa xa give Black realic wnning chance, a after 7 �h3 g6  h6 he ha the mportant nter mediate move 1...e6!

I thnk Black either here or on the next move houd have exchanged on f4 Here i the argment 3 .. xf4 14 xf4 xg  5 g e7 16 d (6 e3 �f6 17 c3 d5! and hite king i in trouble n the mddle while he alo ha plit pawn) 16..e4 17 f3 d5! (17...g6 1 ae �d 9 a4 i dffer ent  Black piece are not playng at al)   e4 4 19 c xb5 20 g5 g6 1 xd5 �e and Whte ha many problem to olve.

1 1    ig5

14 �d2

Alo poible wa ...f6 1 c3 ( -? xb 13 b "f6 4 cxb7 xb7 ve Black a good game White piece ae cat tered and miplaced) 1�e+  3 e3 bxc6 4 d3 g5 5 �f3 xe3 6 xe3 e6!? wth an even game. The exchange of ueen would of coure benet White.

White i ao n trouble after 14  f6! 5 c3 e4 when Black ha excelent attack ng chance

 2    ixb7

Black ha good play for the pawn 1 3 g4 e7

1 4    i6 1  he1 d 1 6 3 b

ndirecy attacking b 1 7 >2

Whte would be puhed after 17 a4? c6 1 13

The Fo Knghts

18 c4 c8 when the b2-pawn a 1 7. . .c8

17 .xg2 18  c6 19 h4 woul gie te ome e to generate countepla. 1 8 e2

ite houl not inert 8 h5? g6 be ore reeang wt 19 e2 becaue ... g7-g6 i uel or Black. 1 8. . .c6 9 d3 e6 20 d2

23. c3 24 a3 a3 25 a3 a8 26 e6 a 27 c6 a3 28 c4 d4 29 4 a5 30 �d2 c5 3 c5 c5 32 a4 �8 33 �d3 6 34 �e4 �e7 35 5 g5 36 �d5 g4 37 �c6 g1 38 h3 gh3 39 gh3 �6 40 d3 �g5 41 a5 �h4 42 a6 6 43 �7 �h3  a7 a7 45 �a7 �g4 46 �6 �4 47 �c5 �e5 % - %

Between plaer o uch a leel x6 i o coue a treat an not a trap.

ame 56 SkroekPinkas Waw 1987

20 .. g6

Black cou ao pla more acvel wth 20 ...b6? 21 �c1 a8 22 x6 8 2 4 a5 wit compenaon or te pawn. he rook on a i l not plang. 2 �c a5?

1 e4 e5 2 3 c6 3 c3 6 4 d4 ed4 5 d5 e7 6 4 d6 7 d4 00 8 5 d5 9 ed5 4!?

 i o coure er briant an heroic but not correct. 21  a8 22 c2 5 2 e5 c5 woul have been better when Black ha more than enough or the pawn. 22 d6 2

hough thi oe not promie euait it oe ve intereng pla. 1 0 c3

at a me 23 2?

Wite take e ea wa out but there wa a wn 2 �xb2 8 24 g a8 (24...  25 �c2 a4 26 'c a 27 b2 a5 28 �2 woul ol help Wte to ecape to te knge 25 a b6 26 �c1 a5 27 xe6 xc 28 e xa 29 e8+ xe8 0 xe �g7 1 e7 c 2 c2 an Black ha run out o bullet. n te game Black manage to aw. 1 14

tonger i 10 a g4 1 1  (11 2 x5 12 x5 c6 1 2 cxb5 14 xb5 6  onl eual 1 1 ...h4 12 g 12 g? x 1  x xc 14 � xa 15 gxh4 6 i er me but mabe not o il 12..7 Brce Monon beieve at thi poion i much better or Black but i i not true. ter 1  axb4 xg 14 xg e 15 � xb5 16  Wite ha goo at tacking chance a well a much better e elopment. 1 0 . .a6?

hi i jut weak. ecear wa 10...a6

The Be/gde Gmbit

1 1 5! (1 1 e Jg5 s equal) 1 1.e 8 1 -- g6 1 3 � g5 14 d3 xf 15 xf4 e5 16 h4, when Wte has some hanes to geneate an attak but nothng onete

19 ...xd5 20 d2 6 21 Wg4 d7+ 22 Wf3 6 23 Wg4 d7+ 24 Wf3 6 %%

ame 57 IAmasiBezoov Batonbere 1996

1 1 a3 e8  2 .e2 i4

Afte 1Jg5 13 g 3 f6 14 f1! (14 -?? :xd5 would be what Blak s hopng fo) 14 f5 1 5 xb4 �xb 1 6 :4 xb4 17 f4 f6 18  Blak does not hae enough ompensaon fo he pee

 e4 e5 2 f3 6 3 3 f6 4 d4 exd4 5 d5 .e7 6 if4 d6 7 xd4 00 8 b5 xd5 9 exd5 e5

1 3 4

Afte ths Wte has pees hangng ee hee wh ges Blak a hane to su e the dangeous game he has been plang Wte had at least two bette moes hee a) 1 3 g3! f6 14 4! (14 xb4? xb 1 5 b xa3 16 b3 g4 17  d7 18 xa3 b5 19 d3 xd3  xd3 6 s less lea) 14 .xe (what else?) 1 5 xe (1 5 xe :xd5 16  s good fo Whte, but ths s bette) 15.�e7 16 f1 e4 17 :e3 :d3 1 8  and Whte wns. b) 13 1!? xe 14 xe �f6 15 g3 xg3 1 6 hxg3 g (1 6 ..e 17 f1 :xd5 18 4 looks e pomsng fo Whte; next omes �d3 and e 1) 17   1 8 f wth a lea adantage fo Whte

1 0 ie2

16 xb4?? g 17  �xf4 18 � xb4 and Blak wns

Wte an also onsde queensde as tlng but ts s pobabl not an opon head ng fo an adantage. 1 d 6! 1 1 3 :g6! (as we hae seen befoe ths s often a ke moe) 1 e3 5 13 e f5 14 f4 f6 Whte now gae up the dea of --, whh s pobabl wse Thee s no eas wa fo m to eate ountepla on the ngsde fast 15 :d  e8 16 - b5 1 7 J a6 1 8 Jf3 a7! and Blak was slghtl bette n Pe-Van de Wel, Cannes 199.

1 6. . .f6 7 4

1 0  .g6

1 7 xb4 g5! 1 8 xg5 xg5 1 9 �d4 xe3  e3 e5 es Blak good ountepla due to he e5squae, the weak wte pawns and a bette kng

Ts moe aan Blak wants to pla ..f5 and ..75. Blak also has good pla afte 1.6 11 :3 :g6 1 g3 f5! (ths s bette than 1.. 5 13  h4 14 d xg3 15 hxg3 e8, wh was equal n Bagn-okn, oh 199) 13 f4 f6 14 �d3 5 15 - a6 1 6 h4 h8! 1 7 h5  and Blak has the bette hanes as te has no eas wa to penetate on the ngsde.

1 3  .g4  4 e3 xe2 1 5 Wxe2

1 5 �xe? :xd5 1 5.  d7

Blak annot sae the nght on b4, so he eates ountehanes n ote was. 16 g3

 7  b5+ 1 8 Wf3 xe3+  9 xe3

Afte 19 e3 :xd5  3 �6 1 e4 :b6  ad e8 3 he :4 4  b5 Blak has good ompensaon on the lght squaes

1 1

he Fo Knghts

1 1 g3

o d5

\te should possiby cosider 11 e3 c5 12 0-0 12 Qc6 bc6 13 d4 c7 14 0-0 14 c6? c6 15 3 d5 16 d5 b4 17 c3 c3 18 bc3 c 19 �f1 a6+ 20 g Mad8 woud ve Black a wig at tack) 14 ..Mb8 Black is sighty better  o so) 12...f5 13 c4 f6 14 'b3 f4 15 d2  16 Mae wth a compicated game ahead, alhough it might be easie o pay Black. 1 1 . . .5! 1 2 f4

Forced After 12 f3? h4 13 00 g3 14 hg3 a6 5 c3 g5 16 e f4 17 g4 d7 18 d2 Mae8 Black would have a very good game because of the weakess of the dark squares i Whites camp 19 e4?! e5 would y make matters worse, as ..g4 is ow a treat. 1 2.  .f6

Black shoud hae played more eerge caly wth 12...c5! ad ow a) 13 c4 f6 14 d2 a6 15 c3 d4 ad Black has the beter game b) 13 dc6 is always rsky White has ot ished his developmet yet ad ca easily be caught  the midde 13...bc6 14 d4 6! ad Black is somewhat better 1 3 c3 e8 1 4 0-0 a6

Ths move is ot very ambious Better was 14 c6! 15 c6 bc6 16 d4 6 d6 %e2 ad Black is better) 16...%b8 with a advatage for Black 1 5 d4 xd4+!

20 c1!

Ths move seems a bit arcial, but after 20  e1 21 Me e8 Back will co tiue to be better i the edgame. So White shoud play 22 M! with some couter chaces 20 . c8 21 c4!

The bishop is ot wel placed here. Stroger is 21 f3 whe Black is ol sghtly beter ater 21.e1 22 e1 e 23 c4 b5 24 3 a 25 a4 ba4 26 c4 as the mssig paw is compesated by te mulple weakesses o Blacks queeside Sti White has some problems wth getg his g3-bishop to the game. 21 . ..e7 22 c2 xe1+ 23 xe1 xd5!

After this White suddely becomes acve teresg would have bee 23..b5! 24 b a whe Black cotiues his aressive prophylacc srategy

15 e7!? 16 %e 7 wth a eve game was also a idea, but ow Back due to his better paw strcre, has sme chaces to maoeuvre hs way i.

24 h4 %e8 25 %xe8+ xe8 26 id8 b5 27 b3 e6 28 h3

1 6 cxd4 d7 1 7 e1 !

28 . c6 29 �h2 h6

This move ves Black a free had to uickly double the rooks o the ee. Now or o the folowg move it was beter to play % to create some couteplay o the ope le.

Fritz 7 has a iteresg sueso here t wats to allow a edgame o opposite coloured bishops after 29.e4!? 30 e4 fe4 31 c7 e3 32 �g �. The argumet woud be that foowig 33 d5 cd5 34 d6 e6 35 e5 �f5 Blacks kg is ver acve ad might eve w te game. Better

1 7. . f6 1 8 d2 e4 1 9 ad1 f7

Ths hghights the weakess of the paw 1 16

28 c7 allows mate i oe with .'e1

he Be/gde Gmbt

woud b 33 xd with dcnt drawing chancs  th b3-bishop sll has som nc ions to prorm 30 a4!

dlgam whr hr ar muua chancs o atack, h numbr o pics is gnrall mor important compard to thir ndividual vau whl ths rvrss somwha in th ndgam For tis rason a rook might b strong nough  compt with to bishops in som ndgam whil it is no match or hm in th middlgam dpndng on othr actors o cours). 7 .. :e8

Black has missd th boa. 30 .. .. 3 d e2 32 ixd5 cxd5 33 'd2 c4 34 b4  35 d2 e8 36 h4 xa4 37 e d3 38 'e6+ �h8 39 i6 gx6 40 7 % - %

Game 58 ello LopezIzea Taarri Eoibar 1998  e4 e5 2 3 c6 3 c3 6 4 d4 exd4 5 d5 .e7 6 4

As wll as th sa  .d Black also has a wa o ghng or th iniativ. 6. . 00 7 ixc7  th obvious atrnavs a

a)  xd4 xd5 8 xd5 xd4 9 'xd4  givs Black a clar advantag  Nunn b)  c4?! d 8 xd4 5 9  9 x5 x5 10 x x 11 b3 ' s sghtl bttr or Black bcaus o th wo bishops  onson) 9 .xc4 1 0 'xc4 xd5 11 xd5 g5! 1 xg5 xg5 and Black has good chancs to cra an iav c)  xc h5 8 xa8 8 d5 x4 9 x4 b 10 d 8 1 1 d3 d5 1  5 and its cnr colapss) 8 x4 9 'd  10 c4 b 1 1 x dx 1  00-0 b with a clar advantag or Black In th mid

is is th crical posion or th lin starting with . 0-0 8 e2

t has th ollowng atavs a) 8 x x 9  9 d3 d5 10 5 x5 11 x5  and Black has th advantag) 9. 9 .  10 5  1 1 a3 b 1 000 b 13 b ac8 14 d xd 5 xd xd 1 '4 gav Whit lots o pa n Polovodin-rnr, Bl 199) 10 5 4 11 c3 dxc3 1 0-00 8 13 d and now a) 13..cxb 4 b 'a4! 15 d  1 d3 xd 1 xd b 18 c a5 19 '4 a 0 g5 b4 inciting taccs) 1 b3 (h onl mov 1 x h8  3 xg5 3 xb4 ac8 and Black shoud win du to his bttr kin 1h  x h  (Black thratns a winnng chck on d3) 3 h1 ? 4 x x with a craz situaon, but probabl also quit a promisng on or Black a) lso vabl s 13 ..'xb 14 xb 117

The Fo Knghts

cxb 1 5 b j 16 jc b6 ( 6  b5 1 7 xb5 xd6  8 xd6 b 7 9 d2 s slghy btt f Wht accdng t Mnsn smtg that culd b dscussd) 7 jc7 a5 1 8 b5 c5 and n my pn t ds nt hav  cmpnsan f th pawn b Pbably th bst mv f Wit s 8 jc4? d6 (8  xd5 9 xd5 d6 tanspss 9 xd4 xd5  0 xd5 d7   xc6 bxc6  2 ja5 a6 and nw b) 13 jc3? f8 1 h5 Accdng t Mnsn 1 'f g5  5 f5? gvs Wht th dg Hwv aft th smpl 1 5x 16 x 8 7  x4 8 fx4 6 19 g3  Black  ctanly w 14jf6 5  is was agd dawn b caus f tm pssu n Baung-Lma Bga 1989 Black shuld cnnu  5c5  6 f5 g6 17 h6+ f8 18 d5 jxc 19 bxc3 g7 20 g f5 21  ab8 wth btt chancs  Mnsn b2 13 ? (an mpvmnt v 3 c3?) 136 4 g4 jf6 H  hav bn yng wth a wd sm-fcd ln 5 0-0-0 xa2 16 c3 xc3  7 bxc3 g6 18 h4 jd3? 9 xd3 a1 20 d2 'xh1 21 g3 ab8 22 g5 b5 23 f6 xf5 24 xf5 8 25 3 x3 26 d+ g 7 27 f h6 28 x3 '1 29 d3 xf2 30 f h5 31 'x and Wt cnus t hav dcnt cuntplay

H Black culd play th sng 8  x4 9  f4 d8 ths s a nw da) 0 xd4 f6 11 c3 ja5 2 jd6 4 3 db5 xd6 1 xd6 5 and Black has wn th png ba 9 x6

Mnsn vs 9 xd xd5 10 xd5 xd4 1  xd4   2 c3 d7 13 a5 b6 4 b4 a 1 5 a3 b5 wth uncla pay 9   . x6  0 xd6 .e7   xe7 xe7  2 d2?!

Wht shuld nt wast m dfndg ths pawn Aft 1 2 00 x4  3  d 8 1 d2 d5 15 d3 Wht has a slght pul  2. . 5  3 ex5 e

Guan vs 1 3 xf5? 4 00 b4 5 c4 h8 6  wth an advantag f Wt

 4 g4?

Aft ths Wh s smply lst Ncssa was 1 4 f x  5 d3 whn Back has a bg na f t pawn but Wht has n waknd hmslf t much and can ght f a daw  4. . .b6

Black dcds t dvlp th bshp whch must b th cct da  5 W b7  6 . 3?

16 f3 was bt bu W s sl  a h bl mss aft 16h  6  e5  7 xb7 xb7   g  ad  . d6 1 18

Black s nw y mbisd wh Wit cannt gt any cdnan

The Be/gde Gmbit

1 9 g2

What ls? Afr 19 f4 6 20 g b4 21 g dS th waknss of th -suar is fl and Whi los. 1 9  d3 20 3 b5 21 b3 7!

Hadng for gS in ordr to aack th kng. 22 3 b4!

aking th d4suar away from th kngh

30 3 xd1 ?

0 . .f8! was uickr 31 ¥xd1 1 32 e2 7 33 1 xh2 34 e2 4 35 2 xb4 36 g2 3 37 e2 e3 38 g3 h3! 39 1 4+ 40 h2 d6 41 6 x6 42 5 x5 43 gxf5 e2  h1 3 45 2 e1 ! 01

Game 59 esieeGliori Mnreal 1998

23 xb4 e2

Back domnas a Whis pcs so who cars for pawns?

1 e4 e5 2 3 6 3 3 6 4 d4 exd4 5 d5 e7 6 4!?

his is an da from Mihajlo rajkovc and is hrfor cald th rajkovc Gamb 6 00

24 g1 2 25 g3 b6 26 e3 g5 27 e1 ¥6 28 d1 d2 29 ¥e2 1

Actay it is oy hr tha computrs un drsand tha Whit is los.  th way from mov 14 hy hav prfrrd Whit, but sug gsd no rasonabl improvmns o much for askng your compur how you playd!

h ony srous altrnav is 6 ...x4, as afr 6 xdS?  xdS b4 Whi has th powrl 8 c! dxc 9 00 7 tr is 9 ..aS 10 a4 xc4 11 xb4 d6 12 1 f8 1 bx wih a clar advanag for Whi) 10 d6 (o prvn Black from dvl opn 10 .xd6 11 g 0-0 2  h6 1 x x 4 x h8 1 1 bS 16 b b 1 S! and a thr is lft for Back o do is o rsign. 7 0-0

Black was sighy br afr 7 xd4? xdS 8 xdS 8 xdS? xd4 9 xd4 f6 10 d1 + 11 2  12 c 4 is h usual sory  Wh is in troubl 8 xd4 9 xd4 f6 10 d c6 11 b dS 12 0-0 4 1 x4 8 MorrisWdbrg, Nw York  991  7 .. .d6 8 xd4 xd4

A cauous mov h othr possbs ar a) 8 x4? 9  ! f6 1 0 bS fS 1 1 d and Black loss matral. b) 8 .S? 9 b xdS (Wh probably has a sigh advanag afr 9 cS 10  xfS 11 xfS c4 12 a4 b8 1 xf6+ xf6 14 dS b6 1  c d 1 6 bS bc8   ) 10 xdS f6 11 f4 d 12 c c6 1 19

he Fo Knghts

13 b3 S 14 2 e8 IS e 7 with an even game. ) 8 ...xS give Whe the hane o  9 exS (9 xS x4 10 x4 ranpoe o the ame) 9 ..x4 10 x4 f6 11 3 e8 12 3 g4 13 e3 athough h oen aheve any avanage 9 xd4 xd5 1 0 xd5 .6 1 1 d3 c6

Th i the naura move, even though i ha to be auae ha 6 oe no fa Bak oe no y equae afer 1 1 aS 12 a4 e7 13 3 6 14 b3 b6 (even wore i 14 e6 I S 2 g6 1 6 f4 g7 onon gve 16S 1 7 eS 7 18 fS fS 19 xfS xfS 20 xfS � 21 h fe8 22 h6 when Whie win afer 22  xh6 23 f g7 24 x7 h8 2S fS] 17 fS 8 18 e3 with a ear avanage for Whie, Trajkovi-eg, orreponene 1 968) I S e3 a6 16 4 x4 17 �x4 7 18 4 x4 19 x4 an ie  ray beer 1 1 ...e7 i a more nara wang move. fer 1 2 3 e8 1 3 f4 S (1 3. .eS 14 2 6 S b3 e6 16 f4 f6 17 2 gve ome irriang hrea) 14 g3 h4 he poiion  equa

Whe a ear ege aorng  onon bu ghty beer i more reai a Bak an pay IS...g4 16  e6 foowe by ..8 an ..S owever 14 xa4 ook more oa afer whh Whe reay i beter. 1 3 c3

13  e7 14  e6 equae beaue after 15 x6 a8 Bak win. On the ohe han 13 f4 e7 (1 3 xb2? 14 a f6 15 x6 e6 16 f4 give Whte the hane o reae a knge iniave) 14  S 15 exS xb2 16 a f6 17 fe fS ve Bak the beer pawn rtre an goo eveopment 13.e6

onon ve 13 .. e7 14 2 (14 2 e6 15 fe 1 a8 16 f4 xb3 17 axb3 a6  imiar to e game an even) 14 .g6 IS f4 S 1 6 exS f 1 7 2 x2 1 8 x2 xS when Whe mut b ighy beer afer 1 9 e3 14 e3 xb3 1 5 axb3 a6 1 6 3 e6

1 2 b3 e8

The mo natra. Th foowing move have ao been anaye a) 12 ...e6 13 3 xb3 14 axb3 e7 IS  fe8 16  a8 equae a afer 17 xa7 S 18 e2 xe4 19 x8 x8 20 e4 b6 21 e3 xb3 Bak i even be er b) 12..7 13 3 e6 14 2 an Whie i beer Ao poibe i 14 f4 b3 (14a8 IS f b6 16 e3 7 17 f4 wa a e beer for Whe in the ai game ononYoram  1995 wth a ifferen ime onro onon mgh have paye a inea of f) IS axb3 (IS x6 4!) S..eS 16 e3 S with uf ien ounerpay for Bak ) 12 bS 13 a4 bxa4 (13.b4 14 a i beer for Whe) 14 xa4 7  S  give 10

The aw i obvou The weaknee on 6 an b3 ane eah oher ou 17 ad1 c7 18 d4 ae8 19 2 h6 20 x6 x6 21 d4 e6 22 c4 a5 23 d2 g5 24 2 % -%

Game 60 KenworVan er Serren Ramgae 1981 1 e4 e 5 2 3 c6 3 c3 6 4 d4

The Be/gde Gmbt

exd4 5 d5 xe4

hi i Back mo pincipled epone. 6 c4 e7

hi poion cold alo have been eached via  . �e7 6 c xe 7 xd4 -

Black hold be cael afe 7 ...xd  xd f6 9 0-0 00 (9..xd 0 xg7 canno be ecommended) 10 xe7 xe7 11 �g e 12 h3 d6 13 ae ccodng o Lev Gman Whie ha a eo nave he. he lne cold contne 13. .e   xe xe1 1 h2 e 16 'c3 e6 17 xe6 xe6  xc7 b6 19 b7 ee 20 xf6 gxf6 2 g wih a clea we advan age  b5

o eevan i  0-0 c 9 b3 (9 b  conideed in he nex game) 9 ...x 10 xf2 xf2 11 'e2 h! 12 f (12 g3 h3+ 13 g2 d 1 xh3 e 1 g2 d6 folowed by �g gve Black a good game) 12 ..g 13 xc7 d 1 xd b 1  b �d7 (1 ..a wold aow a epe on) 16 xb xb wih an nclea poion

atacking chance Whe i e-knigh goig afe e 1  b) fe 9  c he whie pawn on c3 i vey el o him 10 0-0 xf2 1 1 h e  2 bxc7 d6 (12fg 13

Foced. 10 x h 11 g3 xc 12 bxc7 c+ 13 e3 b 1 �xc 'xc+ 1 g2 b6 wold ve Black a cea advan age de o the peoity of hi bhop 1 . . .x   cxb4 xb4??

 ly hoible move. ack had many bee move ne of them wa  1 . .e 12
  . .b4

 . c i conideed i he nex game. 9 c3

9  . xf2

hape and be. eavely a) 9 .�a 10  e   00 give hie decen compenation and vey good

and now a) 1  d6! 1 g �g 16 f e6 17 bxc7 c   xe e 1 9 b3 g 0 f i ghly be fo Whie accoding o Monon caly e oion look vey good b maybe 1.e6! wold have changed ha aemen b) 1 .g 1 xe xe 16 f6 'f (Whie wn afe 16...xc 17 < xd   xc7 d 19
The Fo Knghts

24 f+ 1    c b 19 f2 c4 20 a a White is somewhat better c 14..e4! IS 3 eS 16 eS eS 1 bc 6!? 1.h2   a g 19  h1 wt a raw is Mosos le   g g3 1 9 f4 (19 hg3? 4 20 h g4! was the iea beh I g3 19 ...b 20 hg3 gS 21 e4 wth chaces for both sies

1 2 g5!

Black must have misse ts oe, or maybe the 14t move  2  e8

Or 12 . e 1 3 e h 14 c  I S   16  3+ 1 2 eS I f a Wite ws. 1 3 �1 e5 4 e1 !! 1 -

e c+ 1 2, whe Black  lose s quee Game  KersenVoeer Cndence  1 e4 e5 2 3 c6 3 c3 6 4 d4 exd4 5 d5 xe4 6 c4 e  xd4 - 8 5 c5!? 9 - x2

Chess is certay a cotact sport! Also possible is a 9 + 1 0   1 1 e2 h4 12 g3!? 12 f4 g4 13 g3 gS 14 bc b IS e a accorg to Lev Gutma Wite stas much better i my opio Wite is better, but oly sightly 12h3+ 13 g2 ! 14 bc b S h3 + 16 g2 eS 1  b3 g4 I f2 c 19 bS f+ 20 g S 21 S aS 22 a4 g4 a Black has some play for the ma teral b 9 ...h4 10 e3 6 ot 10 ..? 11 f2 c4 12 bc wch is clearly better for Wte  Nu 1 1 bc b 12 3 e3 13 e3 geerates a very slght pul for Wte  0 'h5!

10 ? + 1 1  h a Black is a echage up. is theme comes back coualy  t ie 1 0 . e5

Elmiatig te efeer Black ow re sige ue to 14e1 IS e f 16 1

s is best Also teste have bee: a 10 ..e 1 1 h1 b 12 bc3 + 13 f2 f2, but tis ki of me loss caot go upuishe, a after 14 gS Black resige i the game Kraft-calvazo correspoece 1990. After 14.. e I S f+ f6 16 f6 Black wil get swly mate b 10 6 11  eS 12 gS g4 13 h4 f+ 14   IS 4! ae 16 h3! c4 1 f+ gf6 I f6 eS 19 hg4 2 20 eS e 21 e3 eS 22 fS a Wite keeps a clear avatage  Moso.

The Be/gde Gmbt

x ae 22 a g6 23 6+ x+ a Black ws) 20  a 2 x x 22 x � 2 cxb6 axb6 24 e6 g6 a Wite has some caces o srg 1  W1 

Tere is o salao to be o  2xc4  bxc c6 4 gx6  6 e    x6  6 cx6 g 7 b2  I  x a Wie s) 6 gxh e  7 g!! oso 7fxg  x7 x 9 x7 h 20 xe 1 1 b4

Moso es the olog log a ererackg aalss   bxc7!? 6  2 f2 g4  3 gS xg  4 xg xc4   xa xa 6 b x+ 7 �x2 e  e7 c 9 x6 xc 20 �g e2 2 c3 e+ 22 'f4 xc3 2 Jxe c 24 4  2 xa7 t a ra

1  Lx 14 5 6 15 '6 l4 1 6 .xd8 lx6 1 7 6 lS 1 8 bxc7 b8

1 1 . . .l+?

Black mistakel tiks tere is a perpet al or  bt is is ot the case Beter as 6 2 bxc! g4  g + 4 gx h+   xg 6 g4 c6 7 x6 cx I  Jx 1 9 4? 1 0

Back resige  m opo this was too earl, bt o ae to bear  m hat ths as a correspoece game Whites last me was ot the strogest Beter s 9 ! b (94 20 x!) 20 Jb3 b7 2 x c 22 e 'f 23 f6 6 24 xh7 �e7 2 f+ �7 26 xb e 27 c a Wite s easil Ater  9 g4 g7 20 b e 2  e  22 xe xe 23 e a 24 6 �h 2 e White has a clear aatage althogh Black ca gt   m opo White oes ot hae eogh compesatio for he qee I b6  9 xg xg 20 c4! 20  2! [20 xg4 2 x 4 22 + x+ 23 x ax 24 x x 2 cxb6 axb6 wih a aw s Mosos sggeso ere] 2

Game 62 Gunikov-Tal Lningrad 1951 1 4 S   lc6  c l6 4 d4 13

he Fo Knghts

exd4 5 d5 xe4 6 e2 5

1 d 2 15 S d 16 2 b d Blc wis  s) Nw 1 06 11 6 d6 12 �g g6 13 g2  1 �h3 c6 15 00 givs hit sm cm ps s), but Bc c ls py th crzy-lig 1O..c6?

7 g4?

 g d  g r h y srus rs fr hit i this psi.  f? is sdid i th t gm, whi hr r sm thr trvs: )  g?  S h S   9  � 10 d d is just  ugh) S .g 9 hg d6 with  clr blc dv tg. b)  d2? d3  cmm thm   f th pws is rtrd) S d3 S cd3 d 9 �d f6 0 f f6 11 2 �6 d Bc wis) S..b 9 b b  0 c3  ftr 1O .. d2 1 1 d2 � 12 2 c 13 0-0 00 hit s hs  tugh jb prvig ht h hs uh cmps fr his pw)   cb 1 1   is crly btr fr Blc ccrdg t Krs, whi ftr 1  2 c 12 00 d6 Blc hs  tr pw) 11.g3 12 2 h 13 f1 d hit hs sm chcs, v thugh h is sl wrs ftr 13..0-0 7   7

his s vry strg, but Bc s hs  rspcbl ltrtiv   .. ? S f S f d 9  � 10 '  1 1 d f6 is clry btr fr Blc) S.d S... g 9 d f6? 1 h3 gh3 1 1 h3 wuld blg i h 19th ctury, whr  fts wr ccptd wth  quss sd) 9 d  10 f6 t sw is 10 g2 c 1 1 d �d 12 b c3 1 3 c dS 14

This is thr rsg s id.  cu is 11 dS d 12 d3 f3+ 13 2 c6 1 c6 bc6 15 g2 f which, ccrdig t s, is  trbl mss l, h is bviusly righ Friz  qut musigly sssss this psiti s 003 ftr  fw mmts brth d6? is bd S h3 Blc wis mus igly fr S gf? f 9 h3 h3 10 � d 11 g S) S ..g6 9 gf f 10 d with  clr dvtg fr h  4?

Afr this mv Blc is dig vry w but th rm sus is S gf, fr which hry his Blc is slighy bttr. Prbbly h is v mr th ths ftr    d6 9 �c b Bt rs c sid tht yu shuld ccpt lmst l mtril  ffr, if it is t sig dircy  yu c lwys rtur it l hs is s th phils phy f chss cmpurs d  imprtt prt f mdr chss, which sms t b gtg wildr.  cuti is 10 b 10 'b3 c 11 g5 �g 12 g b3 13 b3 b d Blc is prbbly just wig) 10  S 1 1 2  with  w ig iiv fr Blc

he Be/gde Gmbt

Theory ormaly prefers 8c5? bu Whe has some eresg acics ow 9 c7 b+ 9  c7 10 e e7 1 1 f looks good for Whe) 10 d f 1 1 �f2 c7 12 c+ f8 13 d d6 ad maybe Whe s acaly somewha beer here Hs developme obvousy looks be er ad afer 1 c3 Black has o obvous way o proceed because 1d 15 �d �d 16 cd e7 17 e3 mus be much beer for Whe

Game 63 HaroTmman euwarden 1978 1 e4 e5 2 tc3 tf6 3 tf3 tc6 4 d4 exd4 5 td5 txe4 6 Ve2 f5 7 f4?!

Ths dea does o   wel The move d7 -d6 proves o be moe useful for Back ha f s for Whe 7 .. .d6

8 ... d6?

Looks aural bu ow Black has prob lems wh he deveopme of hs kgsde Beer was 8  d6 whe Back s smply beer Oe le goes 9 d6 d6 10 0-0-0 e8 1 1 �d3 g wh a clear advaage 9 O-O-?

Whe could have kep he balace here wh 9 gf5 f5 10 0-0-0 �d7 11 d d 12 d e8 13 �f3 whe he pos o mgh o be comforable bu for whom? 9 ... e6 1 0 c4 te7! 8 0-0-0

Now Whe s shd 1  xc7 xc4 1 2 xc4+ d5 1 3 te5+ �g8 14 txd5 xd5!?

Ths works wel d5 1 5 d ef6 16 d5 d5 17 d d6 18 d+ ' s he maeriass gude o a uck . 1 5 xd4 Vxc4 1 6 xc4 tf6  7 h4 fxg4 1 8 h5 ed5 1 9 g5 t6 20 f4 td5 21 f5 e7 22 h6 g6 0-1

Or 8 g5 ad ow a) 8d3 9 cd3 d 1 0 �h g6 11 h c6 12 de cd5 13 ed5 g 1 'h3 s ucear accordg o Moso Ths le has so may possbes for mprovemes ha I do o kow where o sar Maybe wh 9 �d3 whe Whe avods a lo of possble mprovemes laer o  he varao b) 8  e5? 9 e e 10 e c6 11 b f6 12 0-00 f5 wh very easy pay for Black c) I recommed he ured 8e7 9 e fe 10 e c6 1 1 7 (1 1 b b6 12 d3 f5 s horribe for Whe) 11�7 12 e7 e7 13 0-0-0 0-0   g3 g+ 15 b 1 f 16 d g3  7 g3 d5 wh a ood dgame for Black 8 ...te7!

 classica souo o a smple probem The kgh o d5 was aoyg, so  s em aed 1

The Fo Knghts

9 xe7 xe7 0 g4?

10 xd4 0-0 11  gS 12 xgS xg 13 b 1 c 1 4  d7 1 5 c4  8 16 4 6 17  1 a8 and Back as a pawn for vray nong Van Oostrom immrman corrspondnc 1986  0  c5

Wit as compltly lost t opnng battl. H as no ral compnsaon for s o pawns.

wr t s too dirct. 7  d3 8 cxd3

Wit as l coic Aftr 8 "xd3? b4 9 �2 xdS 1 0 x4 7 (1  4? 1 1 " g6 12 "5+ �7 1 3 �x8 favors Wt 1 1 d2 b6 (1 1 .0-0?! 1 2 "c4 c6 13 c3 aS 14 0-0-0 qit nncssary gvs Wit  initiav 12 g3 0-0 Black s a pawn p t no worris 8  d4 9 h5+

Wit mst av missd tat on ftr 12gx3? 13 g dS 14 S Wit as som nav

Aftr 9 x4?? x2 10 gS Black cam p wit t stnning novty 10.f4 11 xf4 (11 xd8 xdS 12 g 4 13 4 b4 and Black wns  Lmaci 1 1.4 12 xc7 7 13 xa8 b4 14 d2 xd 15 xd2 a 16 d1 dS and Wt r sgnd ImpriaLmac corrspondnc 1994

 3 e5 OO

9  g6  0 h4 c6

  ih3?!

ftr 1 1 gxfS xfS 12 g1 �aS Wt  also com ndr atack bt t txt is vn wors.    fxg4  2 xe4 d5

is mov is vry accrat Aftr 10.c 1 1 d xa1 12 x4 c6 13 x! cxdS 14 d4 g8 1 5 f6 Wit ns  Krs.

ow t is a ovr.  4 hg if6  5 h5 gxf3  6 g3 h4 7 xc8 xc8 8 g5 ixg5 9 xg5 e8 20 h6 f7 2 dg cc7 22 xd5 ce7 23 d e5 24 xe5 xe5 0-

Game 64 Monson-Mla ndene 1996  e4 e5 2 f3 c6 3 c3 f6 4 d4 exd4 5 d5 xe4 6 e2 f5 7 g5

is is t drct attmpt  qson is 16

  dxe4

Wit as litl coic. Aftr 1 1 c3 xc3 12 bxc3 c (12..6 givs Black a clar advantag  Monson 1 3 d xa Wt as no ral trats for s rook and  soon los t gam    cxd5  2 exd5

Mayb Wit s not so badly off aftr 12 xfS? xfS 13 3 b4 14 d 7 15 d3 00 16 a3 wn t king is strangly saf on d and  pics wi dvlop qit

The Be/gde Gmbit

easy. Bu Blak has he muh sronger 1 2.g7 akng over he iniave and eaenng ..h7-h6 amongs oher hngs

siuaon afer moves ike 14e7.

1 2  :a5+

even highly risky ones sarng wi 14.2+? Ths should probably be nvesi gaed more han i has been so far 15 d1 xa1+ (1 5 e 16 2 f4 17 ' b5 17  4+ 1 8 e1 xf 19 '3 gves We a lear advanage  onson 1 8 ;e 1 'xb2 19 : xf 20 'e e7 21 x and Whie has a srong and probably deisive aak) 16 b3

12 ..b7 is isussed in he nex game 1 3 id2!

13 d1? que amusingly leads rely o a fored draw afer 13 d5 14 4 'x4 15 :e 1+ e7 1 6 xe7+ ;xe7 17  and now a) 17..e6 1 8 'f d5 19 3 ;5 20 b4 (20 e3 8 21 b4+ 'xb4 22  :xe3 23 e3 6 and Blak wins  u man) 20..xb4 21 d ;a5 (21 ..;x3? 22 a3 b3 23 b [bu no 23  b4 24 d 5 25 axb3 f1 26 2 xf2 27 'e5;6 28 d3 and Whe has a wnng aak onson alhough 28 . b6 29 :1+ 5 30 x bx5 does no seem o be ompleely lear 23..d3 24  and Whe wins as afer 24 . 7 25 b4 x 26 d4 Blak is horribly maed) 22 �a b6 23 d a5 (236? 24 d a6 25 7+ a5 26 d2 a4 27 :b 1  gives he a srong aak for example 27.g 28 f3 d4 29 d5 and 3 omes nex) 24 a 1/2_1/2 Varady-arsson orrespondene 1961 b) Blak an also play 17.g5 18 xg (1 8 'xg e6 19 'f d5 20 3 5 21 b4+ xb4 would lead o noe a bu whou Blaks g-pawn I anno see ha his  make any diffeene) 18e6 19  e5 20 d xe4 21 f3 x (21d3 22 a b3 23   gives Whie a win ning aak) 22 gxf3 ;x 23 'g ;e4 24 f d5 25 xf5 d6 (bu no 25...6? 26  'x1+ 27 x e8 28  d5 29 d ;5 30 e 6 31  d6 32  ;e5 33 d4+ ;f4 34 + and Whie wns) 26 f4+ e7 27 e5 7 28  g8 29 'g4 and again is a perpeual hek. 1 3. . :a4

 3..'xd5 14 0-0-0 g7? 15 3 woud ose bu Blak would aso be in an awward

1  g3 f! Al lines seems o lead o a draw hee

and now a) Bak s s afer 16. d4 1 7 4 xb3 (17..5 18 e1 f8 19  and Whie wns 17e7 18 e1 f8 19 f3 g7 20 g5 would also give Whie a win nng aak) 18 e1+ e7 1 9 xb3 f8 (19d6 20  a1+ 21 ;2 f4 22 xf4 'g7 2 3 f8 24 'xd6 f 25 b2 gves Whie a winnng aak  he rooks are of no value o Blaks kin 20 xe7 x7 21  'a1+ 22 ;2 d6 23 3 and Whie is winning vensson-uglof orrespon dene 1969 b) Blak needs o seek surival wih e exhange of queens 16 .g 17 'xg4 fxg4 18 b5? (18 3 g8 19 4 d6 20 e1+ d8 21 f 7 22 xa gves Whie good ompensaon bu Blak an also deend) 18..e 7 19 e 1 f8 20 d6 6 21 e4 g7 22 g5 and Blak resigned n eier-Kreshar orespndene 1964 17

The Fo Knghts

Mab Black should ha instigatd 22...! whn I do no s an win fo hit. Th is pon with 3 h g   g but tha sms to b it ht dos win how aft 22 .. d 3 'f6!. ft 1 c3 g 16 c1 OO! 1 b3 xa2 18 d6 + 19 �d1 '6 20 c Monson bls that ht has a good gam. But t us tak th anasis a bit th 20..b! 21 xb 'xg 22 xg xb3+ 23 c b 1   b (h Black coud tak a daw but wh not  fo mo?)  c �h8!? 26 c 6 2 h (2 xf a6 28 3 b1 9  b would aso g Black a fantasc gam) 2 .. b 28 h  and Back has a  powl attack.  nn ln is 1 xf?? 'c+ 16 �d1 (16 �2 xf 1 xf 'xa is just winnin 16... '} and Black ns.

1 5.  b4+ 1 6 d c!

This is pobabl th smplst lin. Aso possibl a a) 16 ... xd+ 1 �xd2 c+ 1 8 �1 00 19 d3 xb 20 b 1  1 �f1 ' 22 gxf3 xd3+ 23 g2 xd  bd1 and Black has compnsaon fo  sacicd pawn  Monson b)  6.00!? 1 d3 xd+ 18 �xd2  19 �d1 "a 20 �d b wh a daw. 1 7 d 1 e3+ 1 8 e2!

A winning attmpt 18 �1 xd+ 18

(18 'c wih a daw s of cous possibl 19 xd2 d 20 d3 'c 21  'xb is appan good fo Black, but aft 22 2 00 23 'xh! xf+ 2 xf xd+ 2 1 c 26 \1 �xh 2 ! hit has good chancs of gnan an inia Actuall Black onl has on dfnc 2 . 8 d6 b+ 29 �g1 b6!, pinnng h un. Now aft 30 3 �g! th gam  nd  pptual wh 31 h!? 'c 32  �g8!, as 32�h6?? 33 g! x+ 3 \xf g 3 �g3 'xd6 36 h! ss th black kg tappd  a mang nt stating with 3 f!. 18 ..."b5+ 19 xe3 xd+ 20 xd2 "xb+ 1 e3 "xa1  d3 f6

22.."xh1?? 3 + gs a mang at tack 3 xh7 xh7 4 xg6+ f7 5 c1 !

H w acual la th gam, which connud  x? �x 6  d6 2 c �f8 8 f!? f! with a cla adantag fo Black. Instad w folow th poson to its logca conclusion. 25 .. .b6

An incoct winning attmpt. Bst is 2 . :b 26 2 +  3 b+ wth a daw. 6 c4 a6 7 xf7 + xf7 8 e4+ d8 29 e5 c8 30 e8+! x e8 31 xe8+ b7 32 xa8 xa8 33 h4

ow hit wns asil aft 33 .  .c4 34 h5 xd5 35 h6 g8 36 f4 b7 37 f5 c6 38 f6 ad 39 f7!

he Be/gde Gmbt

Game 65 Van Der ee-Timmermans Sa Van Gen 1996

15 d8 �d8 16 gf d6

O 16 d4 17 h4 e7 1 8 2 6 19 g 1 9 e3 xe3 20 e3 7 21 4 d6 wod gie Blak a peasat edgame 197 20 e3 xe3 th a kely daw

1 e4 e5 2 f 6   f6 4 d4 ed4 5 d5 e4 6 e2 f5 7 g5 d 8 d d4 9 h5 g6 1 0 h4 6 1 1 de4 d5 1 2 ed5 g7! ?

1 7 f4?!

This seems to e ey heahy As White moes his kg ayway, why ge a hek? Why ot s ish deelopmet st? 1  �d 1 !

hee ae of ose some alteaes, ut oe of them is ay good a 13 4? h6 14 g3 5 afte 14hxg5 15 xg5 f6 16 xf6 xf6 17 0-0-0 5 18 e3 d8 19 f1 h4 20 f4 6 21 dx6 6+ 22  7 23 d3 xg2 24 e5 Whte had a ea adatage  Moso hess Maste 4000, 1994 15 h3 x4 16 'xg6+ f8 17 e3  18 e2 xe3 19 fxe3 f6 ad Blak wis  Moso  13 g3? 0-0 14 d6 14 d3 a 15 f1 6 16 3 h6 17  xd5 gae \hite a aw posio i Nikoo Ydoh, UR 1949 4 �a5 15 d1 d5 16 d3 5 17 e1 6 18 'f3 xd6 19 f4 d4 ad Bak is leay ette  13 h3 h6 14 3  15 d1 xa1 16 g3 f6 ad Whte is a ook dow 1 . . .h6 1 4 f f!

imple ad good

Payed with he idea of soppig g6-g5 But i wod hae e ette to meet his head o wth 17 g 1 g5 18 h4 g4 1 9 e2 19 g4? g4 2 e2 d4 21 e h5 gae Blak a ea adatage i Wieiga-Va de ak, Haalem 1976 19g8 1 9  d4 20 g2 h5 21 g5+ 7 22 1 8 23 4 x2 24 e 7 a3 25 g4 g4 is a te ette fo White  Moso; aally he is lose o wig with 26 g3 h7 27 f6 5 28 x5 dx5 29 e 7 30 f3 5 31 d6 d7 32 fxg4 ad Blak is ompletely doated 20  1 6 21 f4 d4 22 e1 e8 23 f1 gx 24 x a6 with a ee edgame 1 7  .d7 1 8 4?!

qally ad is  8 g1 a 9 e 1 e8 wth a ea adatage fo Bak 1 8. . 8 19 b �e7! 20 e1 �f7 21 e2 he8 22 e g5?

Afte ths mstake White ges ogised Blak would hae ee mh ete afte 225 23 d2 a 24 a4 4 25 x4 x4 26 2 4  hite has a ad stu e ad o otpay 2 1 1  24 1 e2 25 e2 gf4 26 f4 e5 Y   19

The Fo Knghts ua

Th bt  for Black  robably 5 .. 7 wh h  boud to obta qualt  mor or l vry . I rcommd that Black avod 5  4 too may comlcato ad oo much forcg thory to rmmbr) ad 5 d5 6 d5 b whr th dg ook bttr for Wht. For t  ca oly rcommd that you lay omthg l. Th gambt  o th whol a bt dodgy ad hould b mtd to blz gam ad oth tuao whr th oot do ot ha a chac to thk thg through 1 e4 e5 2 f3 c6 3 c3 f6 4 d4 exd4 5 d5 xe4

5 .d5 6 d5 6 .b4 Gam 51 6 .b4  Gam 52 5 ...b4 6 c4 ) 6 ..bd5  Gam 53; 6 ...d5  Gam 5 5 ..7 6 c4  Gam 59 6 f4 6  0-0  Gam 58 6  d6 7 d4 0-0 8 b5 d5  d5  ...a6  Gam 55 9 ...b4  Gam 5  ..5  Gam 57 6 e2

6 c4 7 7 d4 0-0 8 b5 8  b4  Gam 0 8  c5  Gam 1 6 . .f5 () 7 g5 7 f4  Gam 3 7 g4  Gam 2 7 . .d3 8 cxd3 d4 9 5 g6 1 0 4 c6 1 1 dxe4 cxd5 1 2 exd5 () g7  Gam 5 12 . .a  Gam 

6 jc

130

6 .  f

1 exd

CHAR NINE The Glek System wth 4 c5   

 e4 eS 2 3 c6 3 c3 6 4 g3 cS

4 g3 has bee amed the Glek System by eeroe wrg about t, ad wh good reaso As you sha see  ths ad he com g chapters, the Russa Gradmaster Igor Glek, at oe tme umber 12  he world, s a geat champo o ths system. No player  moder es has created a system  such a  way ad made t hs ow as Glek has doe  wth 4 g3 he closest sbg to he Glek System occurs  he Vea Opeg, whe after 1 e4 e5 2 c3 f6 the moe 3 g3 s sometmes played he key dea of he Glek System s to play the poso closed raher ha ope,  though Whte does ofte put pressure o the Black cetre wh d2-d4 or wh f24, ofte  combao wh h2h3 ad g3-g4. I ths  way he Glek System mght seem a ttle modest to stat wth but as the opeg de elops t becomes appaet that Whte ca orgase a ae ad hat Black should be cael ad wel prepared. 4..c5 s usurprsgly the most popular choce aer 4 g3 rst of al, Black does ot ope he log dagoa for Whtes bshop as he does wth ..d7-d5 Secodly, the bshop s ery aturally placed o c5, amg at the  weakest whte pot o f2

Whte has o real choce but to poceed  wh 5 g2 ad slow deelopmet Black the shoud play 5d6 ad later a7-a6 to orgase a safe heae or he bshop o a7, otherse Whte   ga he bshop par  wh a tmely a4 ad clam a sght ada tage. he oerall prospects for a adatage or Whte ater 4  c5 are sm the posos ae close to eua or smply eua

ame 66 inski-ezi Czetchwa 1998  e4 eS 2 3 c6 3 c3 6 4 g3 cs S g2

Not 5 xe5? xe5 6 d4 xd4 ad Black ws S  d6

Some poor souls hae tred 5a6 6 xe5! ow ths works 6xe5 7 d4 xd4 after 7.d6 8 xe5 xe5 9 f4 xc3 10 bxc3 d6 11 e5 e7 12 a3 �e6 13 -! Whte aeady has a  g adatage 13xe5 14 5 g4 15 d 6 16 d4 xd4 17 cxd4 h6 18 ab f5 19 xb7 e3 20 c 1-0 Glek-Aagaard, Co pehage 1996 8 xd4 d6 9 f4 c6 10 d2! 0-0 11 0-0 e8 12 b3 ad Whte was 131

The Fo Knghts

clealy better Gle-onan Haarle 1995 6 d3 0-0

Blac i tyg to play without the prophy lacc ove aa6 However exchang the c3-ight or the c5-bihop  proe Whte oe avantage 6 e6 can alo be anwere by  a4 an now a)  b6 8 xb6 8 g) 8..axb6 9 00 0-0 10 b3  11 ex x 12 b2 6 13 4 ex4 14 x4 x4 1 x4 c6 16 e1 e8  a3 b 18 e  19 ae1 with a clear avantage or White in Gle wnel, Zwoe 00 b)  ... h6 8 c3  9 0-0 h3 10 4 xg2 11 xg2 b6 1 e 0-0 Hector Chrianen, Reyavi 998. Now 13  e 14 b4 wth the iea c-c4 ve Whte oe avanage

or White) 1.xg3  x8 x8 3  b5 an Bac ha a lot o copenaon or the acrice aterial. Both .xb2 an  g5 are looing. 1  .  f5 1 8 f3?!

Better wa 18  g 19 e3 b6 wth an unclear poion. 1 8  f6 1 9 'd2 ae8 2 ae 1 ?!

20 h , prevenng ..h wa better. 2 .. .h5 21 ixh6?

 a4 ie6

 ... b6 8 xb6 axb6 9 0-0 i alo a tte better or White 8 0-0 'd 9 ig5! g4 10 xc5 dxc5 1 1 h3 h6

Force. Ater 11..6 12 x6 gx6 13 h2 Blac pawn trucre i terrible. 1 2 id2!

12 h4 6 13 hxg4 xg4 14 c3 g i a nice tap  Blac i better here. 1 2  f6 1 3 ic3?

T exchange i a tae I houl have playe 1 3 h wih oe avantage eg 136 14 c3 4 15 x4 cx4 16 d2  1 4 an White ha the initiave. 1 3.  ixh3! 1 4 xe5 xe5 1 5 ixe5 xg2 1 6 �xg2 g4

I ie thi ove when I playe 13 c3 1  if4

At t I thought that I ha thi all cobaon 1 xc But ater 1 5 1x i alo playable) Blac ha a very trong attac or exaple 18 4 18 ex x5 19 4 5 20  x+ 21 x h+ an Blac hou win 18.. 6 19  h5 0 e3 4 1 xc5 21 x4 x4 2 gx4 x4 23 h 1 a8 i uncoortabe 13

Pure blu I have alway ae Ta play Thereore I never i a chance to play ie y hero 1 e3 6 ve Blac a ght ege 21 .. .gxh6 22 'xh6 g

Blac correctly retain the piece Ate 2  e4 23 xh ex3+ 4 x3 xe1  g g 26 +   + 8 28 g h8 29 h5 �g8 Whte ha ae a aculou ecape 23 h1 'd6?

But thi i a gave itae that ony help White to accelerae hi attac Blac coul have retaine a clear avantage with 23  24 4 6 5 e 4 Now White can  a) 6 e4 xb  e6+ g8 8 �4 xc+ 9 1 x+ 30 � + 31 � 1 32 e1 x3+ 33 g1 h5 e eai et way to top the counterplay) 34 xh xh5 3 xh5 6 an Blac  win th engae b) 6 e6 tring to copcat 26.xe6 6xe6  xc) 2 x

The Glek ystem wth     

27 h7
Maybe Bak issed his one? Ater 2 h6? h8 Blak is learl beter 25 . ..e6?!

Better was 2 ..e 26 4 e6 27 x 6 28 x6 �x6 29 e 6 30 g1 e8 31  though White should s . 2..e 26 x x 27 ex e1 28 g6 g8 29 6 also wins or White. 26 xf5 e8 27 g5

5 5 g2 d 6 6 d 3 g4 7 h3

Aso possible is 7 a4!? and now a) 7..�d7 8 h3 h 9 g x 10 x3 d4? 1 o..e6 1 1 3 with ony a sight edge or White was moe natural) 1 1 x  12 x6 gx6 13 g4 gave Whie a lear edge in Vitolins-Agzamov Mosow 1967. b) 7 .. b6 8 xb6 axb6 9 h3 is sightl better or Wht. ) 7 ..d7 8 h3 h 9 3 x 10 x2 b 11 b4 bxa4 12 xa4 e7 13 h4 00 14  and Whte had a sght advantage in Averbakh-passky UR 1963 7   e6

7. x 8 x3 d4 9 g2 b 10 0-0 0-0 1 1
Aer 10.0-0 11 g d7 12 4 h6 13  e8 14 g4 Whie organised a menag oensive in ovaev-epp Germany 199 1 1 d4

White has three pawns or the piee and now the knght is apped 27 ...4 28 xg7 h6 29 e7 d8 30 d7 e8 31 e7 d8 32 d7 e8 33 xd6 xh7 34 Jd4 fh8 35 f4 h2 36 f3 xd3 37 xd3 x2 38 3 xb2 39 x7 hh2 40 �g4 b6 41 f5 �d8 42 g7 b4 43 �f4 a5  a3 b3 45 e5 f2 46 �g4 1-0

Game 67 Seer-Ko Dmund 00 1 1 e4 e5 2 f3 6 3 3 f6 4 g3

1 1 . . . .d7

A bad mistake woud be 1 1 .xa2?? A ter 1 2 d e7 13 d2! g6 14 b3 a6 1  b2 �a8 16 4 b 17 e3 Blak does not neessarly los he bishop saightaway but hal hs army is used up hg not to lose it Probaby th bishop  be lost sooner or later anyway. 1 2 b3 00

12  8 13 h2 would lead to a sh 133

The Fo Knghts

advantae for White after 13..0-0 while Black would et ito trouble with 1 3 xe4? 14 dxeS dxeS 15 b2 aS 16 b4 xb4? 16 ..a4 17 xeS xb4 18 d3 xb2 19 xb2 and Whte is clearly better but this is what Black should learn o ve with) 17 xeS f6 18 xd7 �xd7 19 e1 d8 20 xf xf6 21 'f3 and White has a winin attack. 1 3 2

13 dS? 1 3  .e8

Bette here was 13exd4 14 xd4 e8 1 5 e 1 xd4  6 xd4 c6 17 a3 e6 1 8 ad �e7 with countelay aanst the e4awn. Wite retins a small ede, as he has the two bishos but it is nothn secia 14 e1

Here White should lay 14 dS fter he black rook has one to 8 tis move is even more sensible. 14..e7 15 d2 b5 16 f4 6 1 7 fS ves White a asn inave. 1 4. . .8?

Black should si oen the ositon. fter 14.exd4 15 xd4 xd4 16 �xd4 c6 17 f4 White s somewhat beter 1 5 �2

1 5 dS. 1 5. . .h7?

Neither layer really cares about the centre it seems. fter S.exd4 16 xd4 xd4 17 xd4 e6 18 f4 c6 19  e7 20 xf6 xf6 21 4 7 22 4 h8 Black is only sihy worse 16 d2

Now after 16 dS e7 17 c4 6 18 �d3 fS 19 exfS xfS 20 �e2 f6 21 d2 White s oly very shy better 1 6. . .6

Black can no loner exchane on d4 as af ter 16. exd4 17 xd4 xd4 Black loses a iece after 17..S? 18 f4 xh3? 19 c3 xd4 20 �xd4 f6 21 �c4 h8 22 �1  the kiht is traed) 18 �xd4 f6 19 �d2 e6 20 f4 te is clearly better f4fS is  obous threat) . 134

1 7 a4 g5

17 ... xd4 18 xd4 exd4 19 �xd4  20 f4  21 c4 shows White domnaon. 18 g5 g5 19 de5 de5

Or 19 .xeS 20 f4  21 ad c6 22  wt an advantae for Whte 20 ad 1 e6 21 4 g4 22 g4 e4 23 4 a5

Black is also in trouble after 23... 24 eS eS 25 xeS xeS 26 xeS xe5 2 xeS. He cannot realy free imself because after 27..aS? 28 �e7 White ws. 24 g1

lso ossible was 24 e3 S 25 3 x3 26 �x3 eS 27 h1 wth a lear advantae e idea is to use the -le for a drect attack 24 .. .h5 25 1 7? 26 6 e7 27 3

Now Black is lost. 27 ...g4 28 d5 7 29 g5 2+ 30 Jg2 3 3 Jg7+ �8 32 3+ 3 33 g8+ 1 -0

Game 68 Smalz-Romanisin Franen 001  e4 e5 2 3 6 3 3 6 4 g3 5 5 g2 d6 6 d3 d4?

n unusual but erfecy viable road to equaliy One should always remember that in the Glek ystem Wite is not realy tryi

The Glek  ys tem wth 4    i c 5

to gt an opning advantag. H is ying to cat a al batl nstad o th usual long thocal lns, wh th pics gt choppd o staight om th opning

kngsid atack at th psnt momnt. 1 4  .b 1 S g2

O 15 ad1 c6 16 c3 ad 17 "3 b5 1  4 6 ith quaity 1 S . .6 1 6 4 6 1 7 2 J 1  1 bS 1 9 h1 4 2 S

At th connuaon 20 5?! 5 21 x x 22 x "x Black would vn b a ittl btt  Wit has mo waknsss 2 .. .6 21 4!

Wht tis to ght o h nav somhow but this mov lavs a tibl waknss on d3. At 21 a1 th chancs would hav bn lv 2 1 . . .6

Not 21.xa2?? 22 xd4!. 7 4

22 .1 b4!

Also possibl is 7 xd4 xd4 and now a  00 g4 9 1 "d7 10 3 x3 11 "x3 h3 12 d4 .xg2 1 3 �xg2 xd4 14 xd4 0-0 th quality in ingo-, Gmany 199. b)  2!? might b mo ambious.  .. .b6 9 0-0 c6 10 h3 6 11 �h2 h5 12 d4! (whn attackd on th ank, ply in th cnt; not 12 3? h4 13 g4 x3 14 3 xg4! 12 ... c4 13 d5 cxd5 14 xd5 4 1 5 b3 xd5 16 b2 ith compnsaon o h pawn in Kovaov-Alksandov, Mnsk 2000

Togth wth th nxt mov ths is a pit all that Black alls into. At 22 . 4?! 23 g4 d 24 "g3 d7 25 g5 Wit woud hav bn ab to sa an attack on th kingsid, but 22 .. d 23 3! xd3 24 .xc5 d 25 g4 bxc4 26 g5 wth compcaons was btt. Ts loss but at 23..c6 24 a3 a6 25 g4 Ud 26 d2 d6 27 d1 Wt l soon play g4g5. Th msplacd kght on a6 is a poblm o Black.

7 . ig4

24 2! bx4 2S J 1

Btt than 7x  x3 h6 9 xc5 dxc5 10 3 d6 11 2 (11 g2? 11  c4 12 0-0 .h3 13 1 0-00 14  6 1 5 d2 �b 16 "c3 cxd3 17 cxd3 and Whit had som ntiaiv n punskyGlado Gacia, Wik aan  1996

O 25 bxc4? xc4 26 d1 d and Black suivs

 xS xS 9 h3 ix3 1 ix3 x3 1 1 x3 6 1 2 - - 1 3 b3 7!

A tpical manouv. Th knight is had ng o th d4-squa 14 ib2

14 h5 b 15 b2 c6 16 4 6 also dos not pos any gat that to Black. Th a not nough mno pics on th boad o Wt to hop o a succssl

23 i3! x3

2S .. . 26 2 6

At 26d6 27 bxc4 "xc4 2 xd3 x+ 29 g2 compnsaon o th pics is had to spot 27 bx4 x4 2 g1 x4 29 Jx3 4 3 x x2 31 ixS hS 32 x h7 33 J1 

At tis Black is ab to cat count play wh th -pawn oct was 33 dd! 1 34 �g2  35   36 �g1 x 37  a 3 d4 x3 39 x4 and it is vy lkly that Wit   th ndgam. 1 35

he Fo Knghts

xf6 16 gS e7 17 h4 gies White a king side initiae) 14 e2 with chances for both sides

33...d3 34 f2 e3 35 e1 e2 36 f4 b1 37 Wf2 b6 38 �xe2 b5 39 Wf3 d5 40 We2 b5 41 Wf3 % -% 9  .g6

Game 69 Heo-Barkaen Swedih Championh, Sara 2002

nother idea is 9  g4? 10 0-0 d7 1 1 d2 g6 wih the plan of h7hS and Black has good counterplay

1 e4 e5 2 lf3 lc6 3 lc3 f6 4 g3 c5 5 g2 d6 6 d3 6 7 e3

fter 1 c7 12 0-0-0 d7 13 g4 h6 14 hf 0-0-0 15 c4 White has some adan tage due to the open le and hs better placed queen

his moe often leads to a ransposition to the positions arising after 7 00 but  this game te tries somethng different 7 xe3

lso possible is 7  d4 8 xd4 exd4 9 e2 g4 0 d2 x3 11 x dS and Black should be ne 8 fxe3 e7

Black should not part wih hs bishop too easily 8g4 9 h3 (9 0-0 e7 10 h3 x 11 xf3 c6 was okay for Black  George lterman Groningen 1997) 9 xf3 10 x e7 11 g4 and now Black needs to casle  the same drecon as te but Whte can improe hs posion whout cas tlng, while ts is a hard task for Black to achee 9 d4

lso interesng is 9 d2? g6 0 00-0 c6 11 h3 aS 12 �b bS 13 g4 (13 dS xd2 14 xf6 �e7 15 xd2 �xf6 was leel in Vasiesiu-isipeanu Bucharest 1998) 13 b4 (13e6 14 dS d8 15 xf6 136

1 0 3 c6 1 1 d3 0-0

1 2 0-0-0 5 1 3 g4 b5

fter ths moe Black has reasonable counterplay 14 g5 d7 1 5 Wb1 lb6

lso interesing was IS exd4 16 exd4 (16 xd4 b4 17 e2 deS 18 f4 with equality was also possible but not 18 xd6? e6 and Black has won ita time in the race for the kings wrong is 16 xd4? b4 and gS fas  6f4 17 d2 xg2 1 8 xg2 b4 19 e2 b6 wih unclear play 1 6 d2 exd4!

fter this Blacks queen is sadly msplaced and trapped by its own night Beter was 1 6a4 17 xa4 (or 17 e2 S wt threats) 17xa4 18 h4 exd4 19 xd4 wt an unclear posion, whle after 9 hS eS 20 xd4 e6 21 "xa4 bxa4 Black is a litte better 1 7 b3 b4 1 8 exd4 f4 1 9 g3 xg2 20 3! c4 21 5 e6

he Glek ystem wth       

70-0 8 3 x3 9 £3  10 x x 1 1 x x 12 00 6 1 3 � 7 14  a8 1 �f2 was play i ams-Va r trr, Wik aa Z 1998 rig t ams aftr 1 'h6 th psi t wul b qual 8 h4?!

8 3 x3 9 fx3 6 (9 .7 10 '2 g6 11 � 6 2 2 00 3 g4! was ular n Galz-Prz Cigs 1998) 10 4! wth quaty 8 e6 9 f5?! 22 6!?

Hr Htr hss t play a ntrst ng psial sari ftr 22 xg2 7 23  x 24 x x 2 x Wt wul hav kpt a slight g

9 3! th a lvl psi was bttr Nw lak gts a spy vlpmt 9 . f5 1 0 ef5 d7   4

Whit has  mpsa fr th x hag a a stabl avatag

 1 4 �xf 1 2 x  1 3 x6 bx6 14 3 wt mpsai was i at by Glk. ut I thik that aftr 14.4 lak is bttr Fr xampl  g4 � 16 x4 x4 17 0-0 0-0 a  paw is t as mprtat as th minag blak qu n th tr

26 . .b7 27 d4 ae8 28 f1 

1 1  . .dS!

22.. .4 23 d5 "d7 24 "2 e3 25 3 d 26 d1

Vry am ftr 28 h4  29 h 8 30 g6 h6 31  'x7 lak is kay! 28 .. .h8 29 f4 "e7 30 d 8 31 f2 "a7 32 3 S 33 f3 8 34 h4 ne5??  -0

lak was straight it 3 a pba bly lst  m bfr Wit ul xut   fr 34.g6 3S f4 �4 36 3 � 37 h4 Wht ha g attakg has Game  Gek-Mkasn Dmund 1998  e4 e5 2 f3 f6 3 3 6 4 3 5 5 2 d6 6 d3 a6 7 h3

akig l f th 4-squar 1  000? 12 4 a t is kay.  2 e2

his s th start f a pla aimg t ga  tw bishps ut as w shal s is pla is vry m sumig a mr r lss igrs al thr s i th psi 7 h3 is n itslf t a ba mv, but th flw up is vry ubius

ftr 12 f4 xf4 13 xf4 0-0-0 14 2 8 1 3  16 4 h8 lak has a lar avatag his is th lgal utm f f2-f4 as lak is bttr vlp a thrfr prpar fr a pig f th psi

7 . . .h6

1 2. . .0-0-0 137

The Fo Knghts

Black is also btr aftr 12h5!? 13 g3 13 5 xf5 14 f4 h7 15 g3 d7 16 xh5  17 g3 xf4 18 xf4 0-0-0 s Black a strong attack 13 .h4 1 3 0-0

hr is no dfnc aftr hs mo. 2 xh4 xh4 28 5 d6 29 e8 a 30 c8 x6 31 a8 b6 32 b4 x2 33 1 g3 34 e2 2 35 d1 5 0-1

ftr 13 c3 d4! 14 c4 4! 5 0-0 h8 16 f4 3 Wht is in crsis 13 g3 h8! 14 d2 4 also gs Black a srong nia

Game  lek-Zeier Baden-Baden 2002

1 3. . .5 14 g5

Wit ould los aftr 14 g3 hxg4 15 hxg4 g6 16 g5 7 17 �2 d8 18 c3 h5! 19 gxh5 xg5 20   21 c4 h5 22 xd5 h4. his is ust on iustraon of th prs Wht is confronting

1 e4 e5 2 C3 Cc6 3 Cc3 C6 4 g3 c5 5 g2 d6 6 d3 a6  0-0

s is h main mo and h onl on hat rall maks an sns

1 4. . .Ch 1 5 c3 x5 1 6 Cg3

 ..e6

Or 16 b4 a7 17 g3 �g6 18 xh5 xd3 and Wit is dsorgasd and undr drct atack Of cours h sll has th o bishops (!) but it has bcom clar hat Wt took too man brs in h opng  ordr to gt hm Of cours, somms hn ou  somthg n, it gos rong But mor oftn than not Glk has bn r succssl h s impromnts in this, hs on opng.

Black knos hat th ght ll not b on th dark but h lght squars, and h quckl crats a st-up rad for ths h chancs  s l should probabl b about qual, but hr is a lot of pa and possibls for ni idual originali and n plans hs is probabl hat has attractd such cra  grandmastrs as Hctor and Glk to ths sstm

1 6. . :e6 1  d4 exd4 1 8 Cxh5

In h long r Wht canot l h th bishop on c5 and nds to lmiat it soonr or latr. But a ita quson s hhr o not Wt nds to pla h2-h3. As  shal s in hakdLko (Gam 73), it can somms b crucial, but oftn it should b possibl to da ts mo

Or 18  g6 19 �xh5 xh5 20 xh5 7 21 xg7 xg5 22 f5 f6 and h black na and xra pan lasts nto h ndgam. 1 8  .b8 19 e1 d 20 4 8 21 4 Cg6 22 g3 5 23 gx6 gx6 24 c1 d3 25 xd3 Cxh4 26 Cx6 !

138

8 e3

8  .xe3 9 xe3 Ce 10 C4

The Glek ystem wth   .  i  

To exploit the open e is vey naaL te has also ed 1 0 a4 g6 1 1 a "d7! bt not 11...0-0?! 12 "d2 h6 13 b4 with advantage o White in Romeo Holmes Koeev Gan Canaia 1997 12 a4 hS! 13 h4 (moe o less oced. Now Black shold have payed 13...0-0-0! wth good chances o opening the kngside beoe the eenside is opened nstead o 13..h3?! 1 4 c4 xg2 1  �xg2 O-O? (1 ...0-00 16 h2 whch was good o White n Pnskpsiewcz Ka kow 1998.

1 8 a4 g5 1 9 xg6 xg6 20 b6 d8 21 3 4?

21 ..:hg8 22 �h2 cS was bette bt Black is bldng p a macho mage.

1 0. . c6

Black can aso  10 ..."d7? 1 1  (maybe 1 1 d4? is a moe ambios attempt 1 1 ...0-0-0 1 2 d4 h3 13 xh3 "xh3 14 E xS 1 "x+ "xS 16 xS and the posi on appeas to be eveL This shows hat Whte cannot always dela h2-h3 o eve and tha is why Whte oten plays it as ealy as move eight (see Games 74-7. The es on o when h2h3 shold be plaed is one o the geat nances and nansweed es ons o ths openng. 1 1 5

22 g4 xg4??

This sacce is headess. Ate 22 . :hg8 23 a 'e6 24 12 h7 2 ' 6 26 c4 White is bette bt that is as a as it goes. 23 xg4 xg4 24 2 g8 25 i3 g3 26 c4 6 27 a1 g6 28 d6 5 29 xb7 e8 30 .d7 1-0

Ate 11 d4 "c7 (1..g6? 12 E xS 13 ex e7 14 e4 wold gve Whte sometn 12 a4 dS 13 exdS exdS 14 xdS xdS 1 "d2 e7 Black was ne in Hecto-Aleksandov Kge 1997. 1 1  . ..x5 1 2 ex5 d5 1 3 e2 d6 1 4 a4 5 1 5 3 OOO?!

An naccracy. Black shold have gone o S..d4! 16 e4 xe4 17 xe4 000 18 a  h6!? with nclea play. 1 6 a5 dg8

n 16 ... d4 White  now have the e soce 17 a4! when the knght has he opon o gong to b6 and then c4 we White can smpl conne with Ra(b and b2b4 wh an aack. 1 7 e4 d4?

Bette was 1 7.'c+ 1 8 �h2 d4 19 a4 (19 b g6 20 g6 xg6 is in Blacks nte est 19. xaS 20 c3 c7 and the posion can best be descibed as a mess.

Game 72 Her-ISklv Malm' 1997 1 e4 e5 2 3 c6 3 c3 6 4 g3 ic5 5 ig2 d6 6 d3 a6 7 00 ie6 8 ie3 .xe3 9 xe3 d7 1 0 d4 g4 1 1 d3

I pee White hee. Black has managed to pevent h2-h3 bt maybe his is too ealy becase White also has othe was to place hs pieces. 1 1 . . .0-0

Alteavely a 1 1 ...0-00 1 2 a3 hS 1 3 h4 e7 14 b4! and I have a eeing hat the White attack is stonge. b Ate 11...x3 12 :x 00 White has 1 3 :a wh a sight edge o 13 x6?! gx6 14 dS �g7  S  S! 1 6 exS 6 17 'd1 :ae8 18 c3 with compensaon bt notng 139

The Fo Knghts

moe. 1 2 th4 te7

1 2 .b4 13 2  14  a 1  ves White the ee. The battle zoe is o the kisie 1 3 tf5 xf5 1 4 exf5 exd4 1 5 exd4 6 1 6 f3?

This elets he most impotat obje ve o the poso ilu he ook o a1 i some ay. Best as 6 4 to push the pas a vaate  o the ook 16 .. e8 17 4 h6 18 h4  9 a1 a Whte has a pomisi attak.

22...x3 23 bx3 b6 24 e3 8 oul also ve Blak the bette haes. 23 b3 a5 24 f4?

Hee the ook is very e Bette as 24 e3 b4 2  2 a4? xa4 26 bxa4 b8 27 h3 a White is a little ose) 2 ..a4 26 bxa4 xa4 27 3 8 ith a sliht blak ee. 24 . ..b4 25 e1 h6!

A so pophyla move. Blak pe vets both 4 a  eas o bak ak mates. 26 g4 a4 27 bxa4 4

Bette as 27a8 28 a1 axa4 29 xa4 xa4 ith lea omatio 28 d3 a5?

A mistake v White a hae to eate some outeplay. 28..a8, ilui  piees  the attak o the queesie, as iht. 29 e2?

29 a 1  is the best hae.

1 6. . .ted5 1 7 txd5 xd5 

Ate ths Blak shoul be okay. He ob tas some eak pas, but also otrol o e4 a he ope e. 17 .x 18 4 6 19 4  20  e8 ih ulea play as also possible 1 8 b3 6 1 9 f3 b5 20 a4?!

White shoul ot ope es o he quee sie; this is hee hs eakesses ae 20 3 b7 21 e3 e8 22 ae1 xe3 23 xe3 8 24 3 e8 oul ve a eve e ame, hee White sl has to ee a itle bit.

29 . . xa4 30 h4 f8 31 g5 th5 32 g4 hxg5 33 xd5?

20. . ab8

33.. .x2 34 hxg5 xe2 01

Ao to va okolov Bak is al eay sihtly bette. 21 axb5 xb5

O ouse ot 2.axb? 22 3, he Whte is ave a Blak passive 22 3 b6 140

33 hx 3 34 1 3 ith a lea blak avatae as the sa alteative. ame 73 Sake-eko Tilbu 1997 1 e4 e5 2 g3 t6 3 tf3 tf6 4 3

he Glek ystem wth       

c5 5 g2 d6 6 d3 6 7 0-0 00 8 e3

As explaed earer, hs moe s he mos aural Whe has aso red a 8 d? xd 9 exd e7 10 d4 exd4 1 1 xd4 ad ow a1 11f 12 b3 also possble s 12 f3? e8 1 3 g f6 14 f4 d7 1  c3 wh a sgh edge 12  a7 13 'd3 'f6 14 d2 d7 1 c3 d4 16 h f 17 ac1 8 18 e3 xb3 19 'xf6 f6 20 cxb3 ad Whe has some adaage, Dree achuk Frace 1984 a2 1..g6 headg for e) 12 c3 �f6 13 e3 e8 14 c2 xe3 1 xe3 d7 wh equa Georga-Garo, Moscow 1992 b 8 g h6 9 e3 d4 10 xd4 10 h3 shoud be a le beer for Whe accordg o Glek 10 exd4 1 1 e2 g4 12 d2 d 13 exd x 14 x xd h a ee game  Ah-Geee, Charlero 1994 8 .. .xe3

Aso eresg s 8.d4 ad ow a 9 h3? a7 9c6? 10 xd4 xd4 11 xd4 exd4 1 2 e2 'b6 1 3 'c1 e8 1 4 c3 dc3 1 'xc3 d was leel, Beko-Kuzm, Russa 1997 10 h2 c6 11 'd2 e8 12 ae f 1 3 f4 e4 14 dxe4 �a 1 f2 h ee chaces, Zhag ZhogZhu Che Cha 1997 b 9 xd4 exd4 10 e2 g4 1 1 h3 xf3 12 x d 13 exd xd 14 c4 dxc3 1 bc3 c6 16 d4 a7 17 c4 Brodsky Byhosky, Rsho e Zo 1997) 17f6 1 8 3 'd7 ad Black keeps he balace.

Here I hae a ew dea 13 h3 'c7 14 g4 f6 1 ad1 folowed by f ad Whe should be beer.  3. . .g4

coordag he whe forces  Black has equased  4 3 e6  5 e2 b5

Thraeg c4  6 b3 5

Black s a le beer here  7 d2 c7  8 h d8  9 g2 6 20 4 b4 2 b 5 22 e2 d5 23 d2 e8

Black has a aleae pla  23 .c? 24 hf3 24 dxc? 'xc 2 exd xd 6 xd xd+ 27 e4 b7 28 c4 xd1 29 xd1 f ad Whes poso colapses 24..cxd4 2 exd4 c6 26 dxe eS 27 exd xd 28 e4 fS wh adaage for Black 24 h3 7

24exd4 2 exd4 dxe4 26 xe4 ef 2 'd3 d ad Black s a e beer

9 xe3 e7

25 2 c8 26 e?!

Ths s he aural moe Afer 9  h6 10 h4 h7 1 1 �d2 e6 1 2 f2 e7 13 af c6 14 h3 fg8 1 d4 Whe had he ae  Arbako-Agecheko, Moscow 198 9 e6 10 �e2 'd7 1 1 h4 4 12 'd2 e7 1 3  c6 14 af was also a e be er for Whe  VorokoPolood, R 1982.

Whe could hae kep s dsadaage o a mum f he had plaed 26 xe xe 27 dxe �xeS 28 exd cd 29 f3 �c3 30 d4 a6 31 fe g6 32 'd2, whe Black eeds o pa 32�cS o keep some pressure.

 0 h4 c6   d4 g4  2 d2 h6  3 d?

26  .g6 27 dxe5

Or 27 exd cxd 28 f b7 29 c1 d6 ad Whe s slowly beg groud dow 11

he Fo Knghts

27 .. .dx4 28 lx4 gx5 29 ld4

9 xd d 30 d4  31  g4 3 3 e 33 xg4  34 4 e8 3 e xe 36 d  and the endgame is ery dlt or ite , this is proba bly the best hane he had

29  .c5 30 lb5 7 3 h3 b7 32 ld2 g5 33 4 d7 34 h4 lgf7 35 lf1 c4 36 l3 cxb3 37 lf5?!

White wold also lose ater the relaely sronger 37 d xd 3 exd dd 39 h3 (39 xb3? d3 39 b7 40 g g4 41  s; Wtes king ll not b able to sre he teats 37 . d8 38 xd7 xd7 39 cxb3 g6 40 l3 ld3 41 d2 x4 42 �h2 xg2 43 lxg2 x  lx1 lf5 0-1

Game  Gnuly-As Pardubie  1 4 5 2 lf3 lc6 3 lc3 lf6 4 g3 c5 5 g2 d6 6 d3 a6 7 0-0 0-0 8 h3 h6

Or a b 9 a3 b 7 10 h4 d4 1 1 e e6 1 b4 b6 13 b with a sight wite adantage in BelaenetsAkhmiloska UR 1979 b h8? (with the idea 6-g-e7 9 h4? (9 e3 wit the idea d3-d4 9 g 10 d ge7 11 e3 e6 1 3 d7 13 14

h d 14  d4 and Bak was worse  lek-Blenen t ngbert 1991 Bt o orse ite did not play well  8e8 9 h4 d4 1 0 d 6 1 1 e3 e6 1 3 g6 13 h 8g7 14 b6 1 d4 �6 16 e3 and ite was a little better in lek-Arbako Mosow (rapid 1988 d 8 e6 9 h (9 h4 d4 10 e3 6 11 a4 a7 1 3 b 13 xa7 a7 only ges eqaliy 9 h6 (neessary as White teatened g and 4 10 h4 d4 (1 0d 1 1 exd xd 1 xd xd 1 3  g 14 g4 6 1  xg d4 1 6 h6+ h7 17 e+ g6   xg6+ g6 19 g4  0 b ae8 1 d wth a slght edge or White 1 �d7 1 1 4 ex4 1  4 �e7 13  with the niae 11 d d ( 1 6? 1  x6 x 6 1 3 3 b 1 4 4 is nomortabe or Blak 1 exd d7 13 3  14  x 1 4 and White is a litle better 9 lh4

lso possible is 9 e3? xe3 10 e3 e7 11 h4 6 1  g6 13  and ite is probably sightly better 9 ...l7

An interesng alternae is 9 d4? and now a Aording to lek 10 e3? es te the adantage Howeer ater 1g 1 1  (1 1 3? 3+ 1 � g4  1 1 ex x 1 3 x  Blak is bettr b 1 0 e e 1 1 d4 d4 1   6 13  x 14  d 1 e 1 e6 wt a balaned posion in zn-Makarhe Mosow 1986 1 0 3 .x3 1 1 fx3 c6  2 d4!

is is stronger than 1 6? althogh the sare is ey interesng Bt White has no real reason to sarie an exhange here besides the pre pleasre o doing so 16 13 h g7 (Blak needs to be arel; White wins ate 13b6? 14 �xh6 �xb 1 f x3 16 �6 ' 17 g+ h7 18 6 g8 19  x7 0 g6

The Glek ystem with       

h8 2 ) 4 n e6 5 e2 and Whte ha compenaon fo he echange but nohng moe I l beeve  Black eouce hee

1 9. . x5 20 x5 a8 Al Back pece ae now wokg

21 g4

Whte ha a mal advantage hee whle 3 g6 g6  4 d3 e6 would be level

Black  alo lghtly bette afte 2 f4 b5 22 d4 c4 23 c c 24 e4 e3 25  c5 26 bc5 g2 2 g2 c5 28 d5 e4.

1 3  8?

215

1 2. . .g6 1 3 5

Now te manage to w a pawn ec eay wa 3..f5 4 ef5 e 5 d3 wth a ght edge fo Whte 1 4 dx5 x5

 a) 4 de5 5 h6 gh6 6 d8 d8  f6 e6 8 a4 d2 9 c5 f8 20  ad8 2 d3 f2 22  w a clea eta pawn. b)  4. e5  5 d6 (afte  5 d6 c 6 f5 e4  e4  Black euale becaue folowng 8 h2  9  te  no bette) 5.f5 6 d8 d8  e e  8 e4 d2  9 ac and Black ha no compenaon fo the pawn. 1 5 xd6 7 1 6 d4 6 1 7 ad1 ? 

Whte hould pay moe attenon to the poblem wth h ueen Afte   f5 f5 8 ef5 Whte woud have a clea ad vantage 1 7. . .c7 1 8 4

Afte  8 f5 d 9 b4 he opm c accal attack ag wth 9 d5  epeed by 9cd5 20 ed5 b 2 h6 gh6 22 f6 g6 23 g4 ad8 24 c4 e5 and Black poon  pefeable) 9 c5 20 d Jd 2 b3 Whte woud be a pawn up but h ueen  mplaced and he ha double pawn n the cente He  bette but nothng moe 1 8. . .d7 1 9 5?

Afte th Black  eceve counteplay down the e-le. Coect wa 9 a4 b8 20 c3 e8 2 f5 f5 22 ef5 when Whte keep h eta pawn and pevent al th countepay. Now Black take ove the na ve

By ecug the c4-uae Black  ceate counteplay on he dak uae He  pobably aeady bete depte the pawn dect. 22 4

22 c5 c4 23 e4 d 24 d4 a 25 a4 wh ony a ghtly woe poon wa teeng 22 ...c4 23 xc7 xc7 24 1 x3! 25 4 x1 + 26 x1 d5 27 c5 x4 28 a3 xa3 29 8+ h7 30 4! c4

30f6 3 e6  32 f4 followed by g6 and h8 mate

31 6 g6 32 7 x7 33 x7 d6 34 7??

Whte ha defended wel and could have made a daw afte 34 c3 d5 35 d5 cd5 36 a6 e8 3 b4 Now he  cloe o beg lot 34 .. .8 35 d8 d5 36 xc6 dc7 37 8? a5 38 c6 a4 39 a6 a3 40 4 g7 41 2 6 42 3 x7 43 d3 g7  c4 a6 45 c2 4 46 c5 3 01 14 3

he Fo Knghts

Game 75 Gleknkiov P San Gii 2001 1 e4 eS 2 f3 c6 3 c3 f6 4 g3 cS S .g2 d6 6 d3 a6 7 00 0-0  h3 h6 9 h4 e6 1 0 .e3 xe3

There are no serious aternaves to his move a) After 10 .. h?! 1 1  e 2 xc xc 1 3 e3 d 1 �h2 ad8 1  h f6 16 f White has a strong niave ek Kuzmi Begrade 1989. b) 10 ..e?! 1 1 xc xc 12 f c6 1 3 f c8 1  e8 1 g and Back was under atack i Kuzm-Vamrov Tash kent 198.

1 xd xd 18 xd xd 19 xd c 20 d6! xc+ 20 . fd8 21 ad1 and Whe retains a strong night on d6) 21 g1 and Whie dominates. 1 7. . .xfS

After 1 .c6 18 h! 1 8 xh h 19 h g 20 xg g+ and Back wis  a reay surprising and origna acc in a knd of posion seen hunreds of mes before) 1 8xf 19 xf d 20 exd xd 21 xd xd 22  White has better pieces ess obvous weaesses and a sgh ead n deveopment. A sght but very stabe advan tage 1  xfS c6 1 9 f4 e 7

1 1 fxe3 e7 1 2 �h2 c6 1 3 d4 g6 1 4 fS dS 1 S exdS xdS

After ths White takes contro of the dark squares but he coud aso have attacked the dpawn A n a Back woud have been much better off had he payed 1xd 16 xd xd. Now 1 g g! 1.xg2 18 xg2 d 1 9 e b 20 h! woud gve ite a strong attack) 18 xg hxg 19 xd cxd 20 xe ae8 21 ad1 xe eads to a eve endgame.

20 xf6!

This ind of exchange sacce is auto maic for a payer of eks caibre. 20 .. .gxf6 21 xh6 c 22 xdS

This combaon wins. Anoher way was 22  c6 23 xd! xd 2  and after h Back is mated 22 . ..xdS 23 e4 f5 24 .xfS f6 2S d1 c6

r 2..e 26 g h8 2 d an Wte wins.

1 6 dxeS xeS 1 7 d4

White probaby had a stronger option i

144

26 gS+ �h 27 xd xd 2 d3 dS 29 f4 �g7 30 e4 d4 31 eS d7 32 fS Jdc7 33 eS d7 34 gS+ �f 3S d+ �g7 36 e6 f6 37 e7  3 g4 10

he Glek  ystem wth      c 

ummary

his system leads to posions wth complex equaity Black usuay plays a7-a6 to keep his dark squared bishop, afer which White has no real choice but to pay e3 He  then be able to creae some play aon the halfopen fe, while Black prpares to counter ith   dS n the cenre 1 e4 e5 2 f3 c6 3 c3 f6 4 g3 c5 5 g2 d6 6 d3 a6 6  d4 (  Gam 68; 64  Gam 6 60-0  Gam 66 7 0-0 7 h3  Gam 0 7 e3 (  Gam 69 7  0-0

7 e6 8 e3 xe3 9 e3 9 d7  Gam ; 9  /e7  Gam 1 8 h3

8 e3  Gam 3 8    h6 9 h4 e6  Gam 5 9  e7 (  Gam 4

6 .  d

7 e3

9. . . e 7

1

CHAPTER N The Glek System wih 4 d5 . . .

 e4 e5 2 3 6 3 3 6 4 g3 d5

Together wth 4..c5 his is Blacks main choice aganst the Glek ystem. Black sikes  the centre and obtans an open game ith easy development. In retrn Wite can open the long agona for is ight-sqared bishop After the obvios 5 ed5 d5 6 g2 Black has to make a choice: to potect the knght on d5 wth 6  e6 or to echange on c We stdy 6  e6  Games 7779, while 6 c, the more poplar choice, is the sbject of Games 888 After 7 bc Black can play 7..c5 (Games 81 -8), 7d6 (Games 84-86) and naly 7 ...e7 (Games 87-88).

Game 76 imaSanos Bl2000 1 e4 e5 2 3 6 3 3 6 4 g3 d5 5 exd5

Whte really has no decent alternaves 5 ...xd5 6 g 2

Again oher moves wold make ttle sense 6. . .b6

After tis move Whte shold have easy play, bt it all comes down to govois 146

7..c5! Not very crcal is 6.f6 7 0-0 d6 8 d4 ed4 (8 .00 9 h h6 10 e1 gave White a small edge in iTseitn-Kostakiev, oa 1987) 9 d4 d4 10 d4 0-0 11 b5! e7 12 d8 d8 1 f4 and Wite has a clear advantage Glek chlehoefer, Bad Zwesten 2000 conned 1  ..d7 14 c4 a6  5 c7 c8  6 d5 c4 17 e b4 (otherse b7 ns a pawn) 18 d6 and Wite won the echange for a pawn, vng hm a close-townning posion 7 00 e

Black has some alteaves here: a) 7d6 8 d4 0-0 9 b5! (9 e5 e5 10 e e5 1 1 h5 c 1 2 bc c6 13 e1 e6 is not so clear) 9 e4   g5 e7 1 1 e4 a6 2 c7! (2 bc d4 wold only be eqal) 12c7  d5 d5 (Black loses after   . d8  4 f4! e5 1 5 d6 d6 16 d6 bc4 17 c4! d1 1 8 ad1 f6 19 fe �c4 20 d&  2 e5 e5 22 d5 e6 2 c4 d8 24 e6 e6 25 e2, when there is no com pensation for the pawn decit) 4 �d5 b4 15 5 �c2 16 d2 and Wte is beter, Glek-Korneev, L Toqet 199. b) 7 c5! is a fresh idea from Granas ter govoi. 8 d4! (note c) is the real test; Wite has not eperienced any sccess wth

he Glek ystem wth     d

th altavs: a   0-0 9 d3  10 h3 fS 1 1 a3 a6 2 3 was qual n udakova-Lugovo, t Ptsbug 2002. b  b? 0-0 9 b4 d6 10 bS? (10 a3 was btt 10  d4 1 1 xd4 xd4 12 4  13 4 a6 and accodng to Lugovo Black s slghtly btt, oovjov-Lugovo, t Ptrsbug 2002 c  d4

and now c ..xd4 9 xS 0-0 10 3 6 1 1 xc xc 1 2 h s plasant for Wht c2 xd4? 9 xd4 wh a rthr splt: c21 9 xd4 10 1 6 11 bS 0-0 12 xc6 bxc6 13 xd4 dS 14 g4 gvs Wht a smal advantag  Lugovo. c22 9 xd4 10 bS d  11 xd+ xd 12 d d 13 3 c 14 a4 g4 15 d3 c4 (or S..d 16 a c4 1 xd+ xd 1 a6 c 19 c wth h nav 1 6 cS and Wht has good com pnsaon. Now Black should avod 1 6..xb2? 1 b3 c4  xc6 bxc6 19 xa d 20 b4 6 21 d1  22 xc6 wth a cla advantag for t c23 9 ..xd4 10 f4 (too low s 10 1? f6 11 f4 0-0 12 S S, whn Black has an dg du to bttr dvopmnt; thn 13 xb ? xc2 14 xa xa lavs th ght squas tmnally wak 10 .fS (th acv and local rply 1 1 xb and now c231 1..xc2? 12 c6  13 xa

�d (or 13.xa 14 S xa 15 xfS and \ht wns 1 4 h xa  5 S and Wht has a wnnng advantag aft S..xd 16 xd   1 b b4 1 g5 6 19 bc c232 11..xc2 12 hS 0-0 13 xa "xa 4 S and Black dos not hav su cnt compnsaon. c233 11.b 12 4 x4 13 x4 fS (13.dS 14   looks btr o Wht 14 gS 4 15 c3 c6 16 b3 dS  6, whn Wt mght b a ny bt bttr aftr 1xb3  axb3  19 xc hd 20 3 but Black dos hav som acv pay for th pawn c3  xd4 9 1 6 10 g 0-0 11 h  1 2 4 dxc3  3 xfS h6 (n ess nant 85 Lugovo only vs 13 . d? 14 g4 and Wht wns, but ths s bttr 14 h h 15 x x 16 x d4 (th only mov 16 ..f? 1 g6 'x 1 x d4 19 + x 20 x xc2 21 b 1 gvs Wt a wnnng poson, wl atr 16..xh 1 f5+ th bshop on cS hangs 1 fS cxb2 1 xh6 � ( ..x 19 x2 bxa20 xa '6 21 xf6 gx6 22  1 S and Wht has vry good wnnng chancs 19 xg xg 20 h5+ g 21 ad? vs a vry doubl dgd poson wh many tats on Blacks kng, w 21 + ads to a draw. o Lugovos  .. cS has yt to b rfutd, but m sur mo dscovs  b found n hs ln 8 e1

Th obvous mov. Wt has no chanc for an advantag at  a3 0-0 9 b4 6 10 1 f6 11 d3 a 1 2 bS d4 13 b2 a4 14 xd4 C-1/ ZVagaGolubovc, Opata 2002. 8   6

ftr  .g4 9 h3 x 10 xf3 f6 1  xc6 bxc6 12 �g4 Wht has a btt scur and th nav. 9 d4

Evn btt s 9 a4 (n od to oc 17

he Fo Knghts

Bak to giv up th ontro of th bquar) 9 ..a (nar bau hit wa thratn ing to ra th op of th gbihop with a4aa) 1 4 x4 1 1 b  (1 1 . 1 fx4 gv t a ight g) 1 x  1 � h8 14 , whih look bttr for hit

Xx 18 x  19 x bx  4 wth a ong poion 1 7. . .xd4   .xd4 d 1 9 %ad 1

Or 19 xb xb (19�xb  xb xb 1 7 giv a wning nin  4 8 1 a1 From a trat pot of w hit ha a ompltl ing poion

9 .. .exd4 10 S d3 1 1 cxd3

Not th on poibl aptur Aftr 11 x! x 1 x 8 1 f4 x4 14 x4, with th ia , hit tan ghtl btr. 1  . . .a 1 2 c3 00 1 3 d4 �?

Too ow Bttr wa 1 ..g4 14 � h8 1 S  8 1  h 1  S aS 17  g 18  x 19 x x  h4 7 1 x+ hx  a1   4 v Wht a rong attak)    a 17  b4 (17..a4 1 8 xb xb 19 4 look a itt btr for t; th  quar  a ni outpot fr  rook aftr  an x) 18 f4 '7 with an un lar poition Blak i ra to ght for th ntral quar.

1 9. . .fS 20 a4 as 21 x cx 22 e4 xe4 23 xe4 e 24 e xe 2S dxe c7 2 d7 c1 + 27 d1 xd1 + 2 xd 1 1 -0

Blak rign u to th follong : 8 . 8 9 7! g8  +h8 1 

Game 77 Mowani-Cisensen Cpenhagen 1991 1 e4 eS 2 3 c 3 c3 f 4  3 d S S exdS xdS  2 e 7 00

Thi poition wa atal rah va a ffrn mov orr n i gam, but hr w folow th uual rout 7 .. .xc3 14 dS a7?

Not a vr mprv kght; aftr th mov hit tan arl bttr Bttr wa 14b4! 1 h4  1 a gxh4 17 axb4 hxg 18 hxg xb4 19 'h, whn hit ha ompnaon for th pawn but nog mo. 1 S d4 4 1  3 cs 1 7 e3

t oul alo hav pla 17 ! 18

hit woul gan th av aftr 7 . 8 4! Now Blak mut hoo b tn: a) 8  9 x x (9..x 1 x bx 11 bx lo a pawn, an 9  x 1 x  lar, on  wor irumtan) 1  x x 1 1 bx x 1 xb7 b8 1 g o not gv Blak muh ompnaon for th pawn Tru, 

he Glek ystem wth 4    d 

ext pwn is dobed bt sl t contros sqes Blck  nd most conterpy te terr 1 3 . xd xd 1  :xd xd g bt hs prob d5 lems persist ter 15 :! bt not 15 :d5 + 16 h e! wth good drng chnces. b  ... exd 9 xd xc3 0 bxc3 xd 11 'xd! 0-0 12 xb with  cler dvn tge, tge, seitlinBr seitlinBrnov, nov, oscow 1 9 c   xc3 9 bxc3 nd now 0-0 1 1 d2! 6 6 1 2 c1 9 c 10 e 0-0 bxc6 1 3 dxe5 dxe5 e e  1  e nd nd there there s xc6 bxc6 not enogh compenstion or the pwn Khoov-Zhhovitsky oscow 1969 c2 Bette Betterr is 9 .. exd! exd! 10 xd xd  xd xd 1 1 cxd 11 'xd! 0-0 12 b c5 13 d3 :b b  1 g2  g2 e5 wold ve Blck very cve ply; here the ext pwn conts or lttle lttle 1 1 ....cc66 nd White White ss oy very very sigh sightly tly better, i better t l a bxc3 e4

his bckres t ws better to ply d6 9 d nd  tht wy trnspose to the  pevos note 9 le 1 !

9 d! d! xd xd 1 0 cxd 'xd 'xd 1 1 b 1 0-0-0 12 d3 wold give some compenson or the pwn. Bt it wold lso be com pletey nnecessry nneces sry

poson is somewht ncomortble or Blck e is more or less orced to cste qeenside beore it is too lte nd wil hve some problems with the b-le. l White needs to cvte the ngh on e1 nd Blck is certily  the me 1 0 d3 ex exd3 d3 1 1 xd3 xd3!!

t is bsotely essentil to bng the ight bck into ply Bcks min problem here is his bd development, nd tht cn only be expoited th swt deveopment 11 cxd3 d5 12  e 13 :b b  :b b  1 c e6 15 g5 xg5 16 :xb xb 00 1  xg5 xg5 'xg5  s xc6 xb xb 19 xb :d gve Bck resonbl conteply or the pwn in BenkoVgnin, BenkoVgnin, Vjck Bnj 191 1 1  ..  d5

There is no lternve to stop the pres sre down the long digonl. 11..e 12 g! xd3 Blck cn lso  1 2...e6 ...e6 13 xg 6 1 'h6 xc3 xc3 15 :b b bt then there re no good moves let, or exmpe 1 5..d ..d 16 b d6 1   e  s h x x 1 9 x x c 20 xd6 xd6 cxd6 2 1 :c c x 22 x 0-0 23 d5 nd Blck cnot sve the endg endgme me 1 3 xg xg x or 1 3. .6 1  xc6+ bxc6 15 e1 e1 d 16 'g nd Whte ins 1 xc6+ bxc6 15 'xh d 16 'd c 1 x gives White  cose-towinnng endgme 1 2 e1 e1 +  e7  3 xd5  xd5 xd5 xd5  4 a a 3

9 ..  c4

O corse it is not ice to py . . .-5, . -5, bt it seems seem s to be necessry neces sry Ater 9  5 10 d3 d5 1 1 e2 e e 12 c  13 b the

his posion White

IS

vrtly wiing or 149

he Fo Knghts

1 4 . . d7

Tere is no slaon in 14... 14... d8 15 g4 8 1 6 4 4 d 1  e e  e e 1 8 ad ad c8 1 9 d d8 20 �3! 5 21 b Pobably Pobably e only cance is to play 14   8 staigt away, as Black be oced to do so soon anway Tis way e mgt ae e idea o ..ae8 ..ae8 in some sitaons. 15 e e 16 g4 promses ite some inia e e Black rook is pretty ineece on 8, bt sl tere is no concrete retaon o Blacks play 1 5  xe7 x e7 xe7 1 e2 d 17 ad1 d8 1 8 f4 5  9 Jd5 Jxd5 Jxd5 20 xd5 xd5 d7 21 xe7 xe7 1 -0

Besides being a piece down, Blacks king is also in great troble

Game 78 Marinkovi-uneev Lningd 1989 1 e4 e5 2 3 6 3 3 d5 4 exd5 xd5 5 2 e  3  7 0-0 5

ere it mst be Black. c) Ineresng is 9 e5!? e5 10 e5  11 , wen te idea o 12 d4 pre sents problems or te 2-bisop and gt e te some cnces or an adantage 8  0-0 9 d3

More ambios is 9 e5!? d4 10 'd b4 (10 . e8 w compensa compensaon on is also a able opon) and now: a) Not 1 1 d3? d3 d3 12 cd3 cd3 d, d, wen Black as a perect posion. Glak Glako o Rkimo, Taskent 1993 b) 11 b bc2 12 b 3 13 e4  14  c6 15 c6 c6 16 c6 c8 1 e4 d4 w cances or bot sides. c) 11 e4 d6 12 c3 dc2 13 d6 d6 d6 14 d4 a 15 cb4 and Black is orced into 15 15a2 a2 16 a4 d5 Blacks posio pos ionn is ragi ragile le ater ater 16 .. �e6?! �e6 ?! 1 b) 1 a a g2 18 g2 d4 d4 19 �a5, wen Black as good conerplay wit a rook and pawn aganst e ncoordinated two mnor pieces Te posion is probably abot eqal 9 .. ..  10 a3 ?!

Tis is an easy way or Black o eqise I do no know wy i is played so rarely

Alternaely: a) 1 . . d4?! 1 1 d4 d4 12 b5! nd ite as ideas le 12   .b6 13 c4 e 14 b b8 15 e4 Tog Black as com pensaon, it is no completely cler weter its enog enog or a pawn. b) Mc saer is 10 . . .d! 11 e4 d4 12 d4 d4 13 'd b6 nd ite migt be appy wt e two bisops, bt mor importantly te c1bisop canno d a good sqare, and Black as a sigt space adan tage Te cances are probably abot een, bt te sold be carel.

8 e2

1 1 a a4 d de7  2  4  4 1 3 e e4

Or 8 e 1 00 and now: a) ite can win a pawn wi 9 a4?! d6 10 c4 d db4 b4 11 1 1 c5 e 12 e5 e5 e5 e5 1 3 e5  e5 d3 bt i is certainly certainly not wort wort it b) 9 d3 c3 9.. .. 6!? wi eqaity is also good) 10 bc3 6 and i anyone is beter

I t is no obios at te qeen is wel placed ee. More precise was 13 b6! ab6 14 b2 d4 15 d4 ed4 16 e4 3 1  and ie is beter de o te speoriy o is bisop to Blacks kg

10

 3 . . .8? .8?

The Glek ystem wth     d 5

Tis seems somewt rcil More rl ws 13  d7 14 xb6 xb6 d tere re o ppret defects coected to te qees positio

2 1   gg4 22 d3! d3! f5 f5 23 c3 e4 e4 24 d2 "d7 25 4 gives Wite  strog posio Al Blcks mior pieces re misplced

1 4 xb xb6 6 axb6 axb6 1 5 b2

22 22 d7 d7 23 d xe 24 xe does ot solve y of Blcks problems problems

1 5 b5!? b5! ? keepng keepng te pressre pressre log log te   8 igol, ws tempg It l depeds o te ie 15 15x x 16 xf3 d4 17 g4 "d8 (fter 17 17f5 f5 18 "xe5 d8 19 d Blck  regi e pw bt oce te bisop comes to b2 Wte  ve  very strog pressre 18 b2, we it seems tt Wite wi oce g gi te dvtge of bisop verss kngt 1 5 . . . . 5 1 6 e2 e2 e8?

Too pssive ow Wite s me to g perfect coordio Blck wod be worse fter 16 16b5?! b5?! 1 7 d4! d4! exd4 exd4 18 1 8 xd xd44 xd4 xd4 1 9 xd4 bt stroger is 16 163!, 3!, excgig Wites best piece Eve so Wite is better if e rect rectss wt 1 7 x3 "x3 1 8 b5 7 7 19 d4 �g4 2 4, voidig  ndesirble pw strctr o te qeeside, s well s mg ng to pt Blcks cetre der pressre 1 7  e1 g6 1 8 1 !

22 xd4 e5 23 c3

troger ws 23 xf5! �xf5 24 xe5 xe5 25 xe5 xe5 xe5 25  e5 26 �e2 wold leve Blck wit  bd srcre d bd pieces 26 e c3 26�d6 27 �d3! 'xd3 28 cxd3 c6 29 4 d8 3 e7 ves Blck eormos problems i e edgme 27 e7 �xc2 (cess resembles wr; wt good  it d y if yo ve te lrgest rmy, bt cot brng yor roops to te frot? 27  x3 28 "e2! is lso very dger os for Blck 28 �e1 f8 29 d7 �c5 3 �b4! Tis igigts Blcks problems, eg 3�c1  �c1 31 2 g8 32 d5 wi  wi ig ttck, or 3 3�xb4 �xb4 3 xb4 c4 32 xc7 d6 33 d7 xb5 34 xb7 d Wite s  nig edgme 23 .. ..g4?!

ecessry ws 23 23d7!? d7!? i order to g coterply gist te b-pw fter 24 f4 ec4 25 �d3 c5 25 Je1  Je1 26 xe x3 27 e7 looks iviti 26 bxc6 bxc6 27 g6 xe1 xe1 28 xe e8 29 �d3 7! Wite stds mc better, bt Blck s mged to keep is lt e of defece d is sill i te gme 24 4 ec4 25 d3 .5

Blck is so i trobl troblee fter fter 25   d7 26 'g6 d6 27 f! (wit te ret of  d3 s wel s protecg te b-pw fter 27 27 f5 f5 28 xf5 �x 29 �xf5 xf5 3 f2 te edgme is ro for Blck

A very crypc move tt ppers very strog Wite s te ide of 2-4-5 to crck ope Blcks kngside s wel s b4-b5 d d3-d4 Blck, o te oter d is com pletely witot coterply 1 8   �8  �8 19 4 6

Uplest, bt wt is Blck to do? 20 b5 a5 21 d4 exd4

26 d5 3 27 e6 ad8?

t is  pity tt Blck lost is cool ere  we c ever ve eog ftsc combi os Te best move ws 27 27 c6! d ow: ow:  It is wortwile nvesgig e posi to fter 28 xf6?! Tis is  nccrte move, bt te posios re so nteresig t it is fc fctt to c t  mstke! 28   gxf6 151

The Fo Knghts

29 fS!! x 30 e7 (ths certaily looks ike the end, but .) 30 .. ! (3�xg3! 31 �xg3 gS 32 xgS xS 33 xg8 xgS 34 c7 cxb5 35 e and hanks to he bad posiion of Back's king, hite has good wining chances) 31 x (31 x Jx 32 x g4! and Black is a piece up) 31 ...e3!! 32 �xe3 x! 33 x gS 34  � 35 f1 h 1+ 36   ith a daw by repeon (36 . �xaI ?? would ose in an nstant to 3 f5. b 2S g2! g4 29 f5 ih a clear advan tae On 29 .. g6 White has 30 x7! but also 30 e4! xf5 31 xg7! is a threat so Bc is ou of good moves.

00 3 c3 b6 14 xd5 xd5 1 5 f6 gxf6 1 6 xd5 gves hte a superor endgame, nyderheteff, correspondence 19) 1 1 x xd4 1 2 xb bS 1 3 g2 00 14 c3, when the two bishops and beter pawn srucre give te a clear edge, nyder Openrieder, correspondence 1 97. 9 ..0-0 10 d3 e7 1 1 a3

28 xaS CxaS 29 xd8 Jxd8 30 .g2 xd3 31 xh3 xg3 32 �h2 J3 33 e7 2 34 Wg3 xc2 3S 5 nc3 36 Wh2 Cb3 37 d1 1 -0

Gm 9 SmsloPoluaesk Bu 1961 1 e4 eS 2 c3 C6 3 g3 dS 4 exd5 CxdS S .g2 e6 6 C3 Cc6 7 0-0 e7 8 e1 6

The alternaves ar not so ood: a) S .. d6?! 9 e4 d7 10 d4 exd4 1 1 xd4 xd4 1 2 �xd4 0-0 1 3 g5 f6 14 c3! c6 15 f4 ith a clear wte edge Korolevauermann 197 b) S .xc3 9 bxc3 f6 (\hite won after 9  .f6? 10 d4 00 1 1 �e2 d [1 1 ..d6  2 e5 fxe5 13 5 d7 14 xe6 xe6 15 5 with a clear advantage to ite  ys lov]  2 dxe5 aeS 1 3 exf6 xf6 14 a3 xa2 1 5 f!, GufldGarcia Manez, Caagey 1974  Black " lose the ex change) 10 a3! and Back  come under serous atack Voronikovher UR 196 c) S.. .f6 9 d4! xc3 10 bxc3 00 11 e2 ansoses to note b. 9 e4

Here hte should pefer 9 xd5! xd5 10 d4 xf3 (10.e4  1 d2 xd4 1 2 xe4 1

Preparing an expansion on the queenside. 11..Cb6

f Black tries to prevent b2b4 with 1 1.. .a5, hite connues ith 12 ed2!, redi recng the night in order o attack the e pawn 12f6 13 c4 g4 14 h3 x 15 x3 d4 16 �d and Black has serious poblems For example, 16..eS 17 c3 c6  S d4 de 19 d5 b5 20 e3 bS 21 �e2 and ite is cose to being srategically win ning. 1 2 b4 g4 1 3 c3 a6 14 h3  hS 1 S cS

ite has a cea advantage Polugaevsk was a great ghte and a great defensive player But myslov was a reat naura player wh a song sense of posiona play. ome mes it is ike s you can ght  you ike but if the posion is bad, the result w be bad. 1 S . .b8 16 b2 d6 17 b3 Cd7 18 d4 e8 1 9 e3

Preparing the doubing of he rooks on the ele and hereby forcing Black to ake uncofortable exchanges 1 9. . .exd4 20 cxd4 xe3 21 xe3 as

The Glek ystem wth     d

Ao poibe wa 21xcS 22 cS �f8 23 g4 g6 24 eS xeS 2S �xeS f6 26 g3 wth the idea of f4 when Whte ha a cear cear edge edge  myov 22 b5 e7 23 2 3 e d5 24 b3 56 56

24..xcS .. xcS oe by force 2S dxcS xc 26 MeS x 27 xf3 c6 28 �xdS cxdS 29 MxdS c8 (or29 ..b6 b6 30 fS! Mf8 31 c3 and White win a piec) 30 xg7! and Wte have a nnng endgame. 25 d3 b6 26 e5 a4

correct, becau it happen to be m own idea I payed it in few ame, but when I faced 8 d4 m ove for .de7 .de7 perhed. 8 d4 wa brought to my atenon by the trong Poih  (now G Pawe Behm After After hi game gam e againt againt G G  arcn ar cn ini, which aw 8 d3 not 8 d4 thi move ha become quite popuar. The at e  tried 6 ..de7 ..de7 (in 200) my teenae opponent, Poh mater Gregor Gajewki payed 8 d4 a we Oh what a miery creative ife i!

Back' potion i beyond redempion redempion for for exampe: exampe: 26  xeS 27 dxeS a4 a4 28 28 c3 fdS fdS 29 d4 d4 with compete donaon or 26 fd7 fd7 27 f4 a4 28 c2 with a very proming poion

7 00 g6 8 d4! d4!

27 c2 g6 28 xg6 hxg6 29 c5!

Highighng the probem of the ight quare 29. 29 .  .bd5 .bd5 30 xa4 b6 3 d3 8 32 e5 d6 33 c6 e8 34 e5 e7 35 c4 5 36 xe8 xe8 37 e5 e7 38 x7! e 39 . xg3 40 e5 �7 4 xg3 xg3 xg3 42 �h  -0

Game 80 SaiyazanvBlau Caua 200  e4 e5 2 3 6 3 c3 c6 4 g3 d5 5 exd5 xd5 6 g2 de7?!

t i hard for me to accept that the idea of de7 de7 foowed by ...g7-g6 ...g7-g6 and ..g7 ..g7 i not

Or: a) 8 d3 g7 9 �e3 0-0 10 d2 fS   cs cs Me8 Me8 1 2 h4 �h6 �h6  3 dl dl b6 1 4 �a3 �a3 b7 IS h2 and the payer agreed a draw in th rather eve poion, Kamink-Behm, Kozain 1998. b) y rt preion of 8 b4 wa that White coud etabh a trong iniative A time howed, however, not  preion can be trana tranated ted nto fact fact 8 .. a6 9 a4 �g7 1 0 a3 a3 00 00 1  bS (1 1 Mel Mel g4 12 1 2 h3 xf xf33 1 3 x b8! b 8! ith ith the idea of .. . fS ve Back ood chance chance  1 . .d4 2 Mel xf} xf} (thi eem ighty unnatura; probaby better wa 2. 2. 4? 13 h3 x 14 x x axbS axbS  S axbS c6 16 bxc6 xc6 7 �2 fS, whn Back retain the tron niht on d4 and the chance ar eve) eve) 13 x3 and White ha an ave, Pinki-Grabarcyk, Waraw 1 998 99 8  Thi game how the effect effect of practi ca che pychoogy. To me it wa very dif13

The Fo Kngh Kngh ts

cult t play aganst ne f my wn pet deas   lst the game. 8 ... exd4 9 tb5 7!

ess gd gd s 9 .. . .a6? 10 bxd4 xd4 11 'xd4 'xd4 12 xd4 7 7 1 b 00 14 c and Whte has a clea adantag At st t mght nt lk e much, but hs peces ae sghtly bette placed and Blac has pblems deelpg the queensde. Beee me, ths psn s athe uncmfabe One pss ble ne s 14 .. .. a5 15 c5 a7 16 g5 f6 17 fe 7 18 e b6 19 f1 g 20 c4 g6 21 e6 xe6 22 xe6 and Whte kept a dsnct adantage GGajewsk-nsk  anca d 2000. anks  the hep f m ppnent  managed t daw but ths ps tn has n alue f Black.

1 8 e1 e1 h6 1 9 t te4

Afte 19  xf+ 20 x c6 21 xc6 bxc6 22 xc6 the psn s eel eel 1 9  . .tec6 .tec6

Bette than 19  ef5 20 xf5 xf5 20 20 .x . x 21 d6) 21 d6 xd6 22 xd6 e8 2 xe+ �xe8 24 c7, c7, whch ges Whte a stng nae f the pawn Blac shuld pbably etu the pawn, and maybe ge up anthe ne n de t get hs peces ut f the ce.

1 0 4 4 0-0 0-0 1 1 txc7 txc7 b8 b8 1 2 b5 b5 a8 a8

20 tc5 te6

20.  .a7 was me ambtus 21 txe6 xe6 22 txb7 xd1 23 cxd1 td4 td4 24 e5?!

24 xh6 xh6 25 Jxd4 xa2 26 a4 s sghty bette f Whte 24 .. . tc2 tc2 25 x7 Wx7 Wx7 26 e2 % - % 1 3 c3?! c3?!

Ths s the stat f a bad plan Whte had t pefeable atenates 1 c7 th a daw, as  the gam he s sn wse, but best s 1 e I  a6 14 d6 f5 15 xc8 xc8 16 d2 6 17 a and Whte has a sght plus plus because f the t bshps 1 3 dx  dxc3 c3 1 4 bxc3? bxc3?

The nmal 14 xc was clealy me sund 1 4  . .a6 . a6 1 5 td6 td6 xc xc33 1 6 c1 c1 7 1 7 t5 td4?

Bette Bette was wa s 17 . h6 18 ge4 ge4 �a5 and and hte des nt hae  cmpensan f the pawn. 14

Game 81 Glek-Marelin Glek -Marelin Gea 00 1 1 e4 e5 2 t3 tc6 3 tc3 t6 4 3 d5 5 exd5 txd5 6 2 txc3

Ths s the mst ppula me 7 bxc3 c5

7 ...d6 ...d6 s cnsdeed  Games 8486, whl whlee 7   e7 e 7 s cnsdeed cnsdeed n Games 87-88. 87- 88. 8 00 00 9 e1

As pssbe s 9 d f6 9.. 9..g4?) g4?) 10 e b6 10 10xe xe 11 fxe e4 12 d4 seems pleasant f f te) 1 1 d2 'e7 12 b

The Glek ystem wth     d5

e 13 c a tis is a pawn sacce in the long n because of the theats own the b e but his is bette than givng up the ak squaes 1 b cb 15 C2 fS 1 b "c7 Hee White shoul have playe 17 f wth to play on two lanks nstea the  game Reefscage-anka Gemany 199 conn connue ue 17 e   S b2 f5 19 1 1 e 20 b ab8 when Black ha goo go o coun teplay fo the pawn ecit. 9 .  :6!?

he man iea of 7.. 7..c5 c5 i s to t o take take cono conoll o the squae. 9...f10 3 g  g 11 h3 h5 12 g  13 C2 b 1 a 5 15 Ce f5 1 gf5 f5 17 e3 was sighty bette fo White in Glek-Fo ay Open 1995 whe 9.e8 .e8 is nvesgate in the net game.  0 d3 d3 h6?!

his is not necessa. Wite cannot use the g5squae fo vey much. Black shoul be equal afte 10 .b. Now No w play coul coul  con nue 11 e3 e 3 g 12 h3 h5 h 5 13 b1 e 1 e "c3 15 3 "f "f an Whte W hte shou be cael ue to is weakene pawn stuctue If he stats to play too casuay he might qckly en up n touble

also a tle loose so White is not moe than an inch bette) 15 C2 this highlights the awka posion of the knght on a 15. 15.g g 1 1 aeS 17 C f5 S b ab 19 Ce3 h5 20 C5 " 21  "S 22 h White has an ovehelming avantage Glek-patchev Ba Zwesten 2002 3 b3 e6 4 c4!

his is the most aessive move focng Black to make concessions so possible  was 1 Cc5 c5 15 c5 7 1 "h5 an hanks to the w bishops Wite has a sma ege.  4 . . .x . xe3 e3  5 xe3 xe3

Again the most natal move. Afte 15 fe3 abS 1 Cc5 "g 17 Ce wong  wol be 17 Cb7 g 18 "1 e 19 b5 c8 20 e e 21 e b 7 an Black is close to wnin 17.. 17..e e 18 f though Whte has some avantage Black can ght on.  5 . . .ad8 . ad8

  e3 b6  2 d2 d2

 6 c  ?

beieve that th at White Whit e is sighty bette hee  beieve  2 . . e8 e8

fte 12. 12. g 13 Cc f 1 a Ca5 1...e ...e 15 a e3 1 Ce3 a es Wte some pessue against b7 but a is

Wite hee isses he chance to uin Blacks pawn stcte Pehaps he mis uge the compensaton Black might get on the light squaes. squa es. 1c c bc17 "e2! 17 e h3 18 a5 also ooks bette fo Wite but the queen shoul not go astay  ike this an ve Black hope of counteplay on the kingsie 17. 17. 77 18 18 e e an Wite  wns a pawn. He is cealy bette afte 18.. 18..e e 155

The Fo Knghts

19 Mxe4 Mxe4 20 dxe4

2 4

1  . .8 1  b2  1 8 a1 &d8 1 9 

Whte immedatel spots the weakness of d6

White would have had better chances to get an advantage with 19 f4? exf4 20 xf6 gxf6 21 Mxe7 Mxe7 22 Mxe7 xe7 23 gxf4 g6 24 a c6 25 f e7 26 f2  Glek. His man idea is 26 . xf 27 h3! Going back a bit, 19.b6!? might be better 20  (20 xc6? xc6 21 J:xe b7 and Black has a strong attack) 20 .. g6 21 e4 d6 with unclear play.

2.. .d8 2 ! b 2 d bx 28 x8 x8 29 f4?

1 9. . d

Black could equalise wih 19...d4 20 c3 f 21 M3e2 d6.

Simple and best was 29 xe xe (29 ...Mce8 30 Mxe7 xe7 31 b1 is very uncomfortable for Black) 30 xe b7 31 f4 with a clear advantage 29 .. .8 3 x b8 31 4?

his does not prevent Blacks counterplay. Better was 31 c3 b6 32 f2 b7 and even hough Black has improved hs posion problems si reman 3 1 . . 4 32 d4?

21 xb4? xb4 22 3e2 g4 23 d2 f would aow Black to take over he nave.

White is drifng After 32 d4 b2 33 J:d e8 34 d7 J:e7 Black has no more than a repetion. Now Black takes over

21 . .a

32 . .b2! 33 1 2 xa2 34 f2 b2

21a4? 22 c3! "a3 23 d d7 24 f4 " 25 h1 woud play into Whites hands

34b7 35 e6 f6 wt very promisng prospects was better

22 3

Glek tnks White s preferable here Per haps he is right.

ore precise was 3..c3! 36 e3 b4 37 d3 e8 and Black keeps his extra pawn wih very few problems

22. .�f8?

3 f3 ?

fter this Black has an unpleasant end game. Better was 22  b6? 23 a3 wth only a sight white edge Note at 23 d2 4 24 e4 6 would no achieve anythng for te.

First Whites pla fe apart now Blacks game disintegrates as well. After 36  e8 Black is s better. Now Whte has the pos sibiity of 37 d wit an initiave.

2 4 b4 21 a1

3 1 b

23 xa xa 24 3

Black has problems wth his knight on a I t does not really do anthng out there, and  has no easy way o become usel. 24 . .. ?

hs does itle to solve the probem of the knigh on a. nstead of improvig his wors piece, as many theoreticans recommend Black is actualy making the suaion for the knight even worse. t was better to play 24  .f6 25 h4 b6!, with the idea of ..b7-c fter 26 d J 27 f4 c6 28 c3 Mfe7 29 fxe xe 30 xe xe 31 xe fxe 32 f2 White has better chances in he endgame 16

3 a3? b??

Very bad 37 ..cxd4 38 xd4 exe and Black is close to wnning

he Glek System with  .   d5

38 a7 td6??

It des nt imprve After 38 cxd4 39 xc4 a5 40 xa5 xa5 41 d5 d7 42 xd4 c6 43 d (43 e6 d6 acieves it e) 43..xe5 te endgame is a daw. 39 a8 0

A eal cmedy f errrs

Game 8 HarikrisnaCvek Pardubice 00

lmasi Biel 1996) 15 e4 e7 16 c4 c6 17 a3 xe4 18 x4 b8 19 xe5 xf2+ 20  and Wit is better de t te  bisps, Glek-Almasi, Germany 1996. 4 ...te7  5 a4

Wite needs t  smeting t distrb e Black set-p 15  d7 16 b ad8 leads t an nclear psin  5. . .c6  6 g5 6  7 d td5

 e4 e5  t3 tc6 3 tc3 t6 4 g3 d5 5 ed5 td5 6 g tc3 7 b c3 c5 8 0-0 00 9 Je e8  0 d3 g4   h3 h5   g4 g6  3 tg5

Black as eqaised.  8 c4?!

After ts Wit as sme weakesses 18 " was mre naal  8 . t4  9 i4 e4 0 d

is psitin as been played a nmber f times Te cnclsin seems t be at Wite as te beter cances  3. . .b6

After 20 c3 f5 21 x5 xf5 22 d4 a5 we see w wel he t bisps wrk t geer i an pen psiin 0 .. .ie4?

Tis is raher safe Black sld play fr an edg wit 20 f5 21 gxf5 xf5 22 "xf4 xe4 23 xe4 f8 24 4 xf+ and ites kig \ always b a ittle ncm rable, even tg draw is te mst ikely reslt

Black as as tried a) 3..6 14 e4 b6 5 f3 a5 16 b1 xe4 1 7 x4 xc3  8 c4 d4 19 �xb7 it a slig pls fr te, Glek Lane, Belm 1996. b) 1 3. . .f6 14 e4 b6 1 5 4 7 16 5 6 17 3 b8  8 e4 e7 19 �3 wi a small edge fr Wite, Rsanv-Smikvski Rssia 2000

Black is cntining wit iaccrate mves After 24 c7 25 d4 b6 26 e1 e6 te psitin is cmpletely qal

 4 te4

5 c3 g5

e best mve is 1 4 b1  6 (1 4f6 15  �x 1 6 x ad8 1 7 g2 f6 1 8 xc6 gave Wte a mall pls in Glek-

25 �c5 26 e1 xe1 27 xe1 7 wi a ikely draw.

 e4 e4  e4 d6 3 3 e8 4 g e5

6 e e 7 e 157

he Fo Knghts

Game 83 Solovjov-Gavrienkov Tufa 1999  e4 e5 2 3 c6 3 c3 6 4 g3 d5 5 exd5 xd5 6 ig2 xc3 7 bxc3 ic5 8 0-0 0-0 9 e1 e8 1 0 d4

27 .. 7?

Black has complete os his sense of danger. s move ooks inncen, bu i is  fact a terrble mstake 27...�e5 28 d2 c5 would sl keep the blance. Here we see the importance of paing at  concentaton even in du posions as i is never eas to foresee when a slght inaccurac change the evaluation. 28 d4 h5

Ts does litte good but there is no defence aganst c4c5 and �b 1 with hreats against h7 and b7 29 c5 ic7 30 b1  hxg4 31 hxg4 b6 32 e4?

ore precise is 32 a b5 (32bxc5 33 b7 and hite wis) 33 a f8 (33...e7 34 e d7 35 xc6 xc6 36 e6 b7 37 a and Whte wis) 34 e6 f5 35 f1 fxg4 36 xc6 d8 37 xg4 xa5 38 xf4 f6 39 "8 �d8 40 xb5 and Whte should wn. 32.  bxc5?

Bad defence. Whte woud sl have a pomising posion after 32�d5 33 h7 (33 xd5 cxd5 34 xd e7 35 c6 nd Back has good drawng chances 33 .. e6 34 h e7 35 d5 cxd5 36 xd5 but ts is stil much bettr than the text. Now it is  over. 33 xc6 .b6 34 d7 \g8 35 id5 \h7 36 ie4 \h6 37 c8 id8 38 5 3 39 \x3 -0

8

Ths move should give Black eas pla. The poblem is that White is not read to support a big cente with his pieces, and Black quck develops we to attack it. 1 0. . .exd4 1 1 xe8 "xe8 1 2 cxd4 b6 1 3 c4

Or 13 d3 g4! 14 g5? (14 c3 with equalt s better) 14..e1 1 5 f1 g6 1 6 c3 (even worse was 16 b2? e7 17 h4 h6 18 e4 b4 19 a3 f5! wth a clear advantage for Back, DelchevAlmasi Odoheiu 1995) 16..:e8 and Black s better. 1 3. . .ig4 1 4 c5 d8?

Ths sacce is ver ambious, but unfounatel awed uch better is 1 4 a5 and a) 15 d2 :d 8 1 6 d5 e5 (1 6 d7! 17 c1 �xd5 18 e5 d1 19 xd Jd1 20 xd xd 21 xc6 bxc6 22 xc6 d2 23 xa7 a4 would ve an endgame where Black has enough compensation to daw 17 xe5 xf3 1 8 xf3 �xe5 1 9 b 1 b6 20 c6 and Whte is a litte better acordng to Glek although m not so sure. b) 15 e3 :d8 16 a4 c3 17 :c1 x 18 x xd4? (a gave taccal erro;

The Glek ystem wth     d5

after 8x4 9 xc6 xc6 20 xc6 bxc6 2  f6 22 b f8 23 b7 4 24 x4 x4 25 xa7 c4 te rook en ing i cometey awn)  9 xe8 xe8 20  xe3 2 e3 2 22 c4 xe 23 g2 e6 24 c6 wt goo winning cance for Wite, Bekov-V.Ivanov, ocow 996  5 cxb6

e roof of te ung i in e eang. After  5 g5! x4 6 �c x3 7 x3 a5 8 e3 8 9 g2 �e6 Back wa ju a awn u in Gek-Van er Serren Ger many 995

Game 84 Heor-Timosenko Bled Ompiad 2002  e4 e5 2 3 c6 3 c3 f6 4 g3 5 5 ex5 x5 6 g2 xc3 7 bxc3 .6 8 0-0 00 9 nb !

 5. . .x4

Generay Wie ou icue ti move i i arena in te 6 .. xc3 ine, a Back  ave to make e convenient move to ni  eveoment 9 .. .b8  6 x4!!

A fanac conce  6. . .x4

 6 xf3  7 x8 x8  8 x give Wite a wining engame

Back a ao rie to o wtou ti move 9 ..e8 0 4 6   e (more aive i  e3 b8 2 2 a5 3 5 b6 4 fe b7 wit uaiy Gek-Hector, Co enagen  995)   ..ex4 2 xe8 xe8 3 cx4 b6! (3. b8! ou be aye)

 7 x4 e    8 f axb6  9 b3

te oiton i a tecnica win, but  it nee to be aye w care an attenon; Hector ot ti otion in a game in 997  9 . .c6 20 e3 b4 2 3 f3?

2.. e6  moe reen 22 2 5 23 b e7 24 a4 f5 25 xb6 g5 26 a5 f4 27 gx4 gxf4 28 x4 g7 29 g3 c3 30 nxb7 x2 3 a6 a2 32 a7 c5 33 7 a4 34 x5 xa7 3 5 .6 g7 36 g2 a  37 �2 c 38 xc5 2 39 .g3 �g7 40 c7 �6 4 e4 �g5 42 nf7 5 43 f5 0

4b 3! b  5 e3 t woner ook 159

he Fo Knghts

w i omeh ha houl be emem be) 1...'f8 Geko Gemay 1994). Now Whe houl play h4 ih ome itiatve acco o Glek.

 6 h b6  7 b1

1 0 d 

1 ... h6 i coiee i Game 85 whle 1 0...4 i he ubjec of Game 86. Black ha alo tie a) 10 ...Me8 1 1 Me 1 1 1 2! f5 12 c4 '7 i complex) 1...ex4 12 x4 12 Mx 'xe8 13 cx4 wa maybe moe atu al) 12...x4 3 c4 b6 14 e3 f5 a Black ha equale compleely NaybullNikolic Yeeva 1996. Noe ha Black ma tae ha bee o emove al obec of aack fo the oe piece  Whie camp he 2bihop. b) 10...ex4 1 1 cx4 4 2 h3 h 1 3 Mb! 6 14 h4 xc2! bette i 14...a6 15 Mb 1 'f6 16 6 wh a lh plu fo Whie bu ohi moe) 15 "xc2 x4 16 'c4 xb 1  "xb5 c6 18 "a4 b6 19 f5 'b4 20 'c2 a Whie ha oo atacki chace Glekaka Gemay 1994. 1 1 h d7 1 2 ?

I believe Whie coul pla moe roly hee wih 12 Me ! e8 13 e4! "f6 1 4  4   'x ...x 16 '2 'e6 17 x look oo fo Whe) 16 x x 17 Mbx h6 18 3 a Black ha poblem hol h ce. 1 2. . ex 1 3 x x 1  x e7

Black houl avoi 14..  15 5 "f6 16 M xf 17 e4 e 7 18 xf5 xf 19 Mb whe Whie  have a vey to aack aai he ow weakee black ki. 1  2 b?

Thouh trctually ute thi i imply oo low.  wa bee o eutale Whie cao wh 1..c6! 16 xc6 16 h x2 1 x2 ' woul eve ve Black he avaae a l he weakee i hie camp  cou i he eame) 16...bxc6 17 Mxb8 "xb8 18 "e2 e8 a Black ha fou acv coueplay a equaty. 160

uely al of Whie piece ae eay o aack Black k. 1 7. . .g6

Black coul alo hav e 17 ...e6! 18 f3 18 e4) 18...c4 19 Me  20 e5 whe hie ha ome offeive poibltie bu Black ca efe wih 20..f6 foowe by .... 18 c .c6 19 e 1 xg2 20 xg2 d 21 e?

Th i o he h way o oaie he ook. Whie woul have bee bette afe 2 Mfe2 '6 22 Me5 f6 23 f4. 2 1 .  .6 22 e1 d?

Bee wa 22..6! wih equal. 23 e2 d6 2 e3 a6

O 24...xe3 2 xe3 a6 26 Me7 7 27 M7e with a vey pomi poio fo Whe. 2 g?

v Black o ake ove the iitiave. 2 x Mx 26 �c6 M6 27 " woul have ecue Whie a mall avaae. 2 . . 26 2 h

 i achepal fo Black o he weake Whe kiie i hi mae.  ake a empo bu he weake i pema e. 27 e hxg3 28 hxg3 6

Now Black i bee ue o a afe k. 29 a3 g7 30 c1  h8 31  h 32 1 ?

he Glek ystem th  .   d

Whte is drifig t was necessary to ex change the knights with 32 e4! xe4 (32 d5 33 �d2 c6 34 c5 Mxe5 35 Mxe5 would ve Whte counterpay, after which the game is far from cear) 33 xh5 gxh5 34 Mxe4 f6 and ough Back is et ter here a draw is stil a very ikey outcome

the ngsde and takes me a for the right to pay 'f6 The costs are too heavy

32. . .d5

  e

33 d2

Aandoning he ing  not a good idea. Better was 33 'f3 f6 (33�h8?? 34 e4 would e nny ut only for White) 34 'd3 Mxe5 35 dxe5 f5 when Back is etter, ut that is  33 ..8 34 e2 f6 35 xh5

After is there is no way ack 35 .. xh5 36 xh5 gx5 37 e x3 38 e5 xa2 39 5 f3 40 d3 xg3 4 f2 g4 42 e3 4 43 x7 3 44 xf7 xf7 45 e5 e6 46 xg4 d5 47 d3 b4 48 3 a5 49 f6 e6 50 g4 f5 5 h2 f4 52 b3 d5 53 4 a4 54 a3 bx4 55 xa4 3 56 b3 g3 0-

Game  Hor-oannssn Mam 00  e4 e5 2 3 f6 3 f3 6 4 g3 d5 5 exd5 xd5 6 g2 x3 7 bx3 d6 8 0-0 00 9 b   b8  0 d4 6

I dont reay ike ths move t weakens

White has aso had success    d2!? f5 (  a5? to prevent c4 woud not work due to 2 dx5 e5 3 5 xc3 4 d5 'e  5 e4 c6  6 h5! and White  start a strong attack Back has more pro ems than  can care to mention) 2 c4 �f6 3 e3 Ovious White does not want to exchange the strong nght for he ishop on d6. He woud ike to have he to ishops ut th atte w e on he ight squares and herfore th knght is more usel on e3. 3..h ( 3 xd4 4 xf5! 'xf5  5 cxd4 gves Whte a standard edge)  4 2 fd8 5 d5 a5 6 c4 6  e e8  8 c3 'd8  9 'd2  20 g4 wth an attack, Gek-Zonierowicz Bern 990   . . f6  2 d2 f5  3 e4 g6  4 d5 a5  5 h4

This is too optimisc and weakens the ngsde fter the moe reaxed  5 'e2 fe8  6 f3 \Xit wold e a ttle etter  5 . .xe4

Ths move proa ame as a surpse to Whte. He might have eckoned on some thng ike 5c4?! 6 h5! h  'e2 5 8 a4 a6 9 ax5 ax5 20 g4 d 2 g5 wth good attacng chances  6 xe4 f5  7 5 f6

 . 'e8?  8 e ( 8 h4?! e!)  8.a4! woud aso have ven Back very 161

h e Fo Kngh Kngh ts

good chances Whte s punished for s recklessness and can now ony ook at the ruins of his posiion 1 8 a4

Trying to gain some me o organise defence. Of course the rook is also somewhat strangely paced out here but when you defend, often you ave to mae this nd of compromse with your posion. 1 8 . . .b6 . b6 1 9 2 2 b b8 8 20 20 id2 id2 4

20 .. .f4!? 21 gxf4 exf4 22 � 'f would aso be better for Black, but the tex move blocng out the g2bishop, s more principled.

Black does not have a good feeng for the fray ray of is dark squaes squae s After 2 . . .b8 26 ;b3 Wte is worse, but can sl bite After 26 ;xb& xb8 27 a3 'e6 and 26 d b6 Back as good wnnng chances. 26 c4+ ff7

26 .. e7 27 %d6 'xd6 'x d6 28 xc 'd 29 �xb4 ves hite a cear edge but now e ins. 27 xd6! xd6! xd6 28 ixc5 c6 29 xb4 c7 c 7 30 b8+ Wh 7 31 b5 g6 32 id4 gxh5 33 h8+ �g6 34 g8+ �h7 35 5!  5! 10

Game 86 Glek-Kroeze The Neheands 1996

2 3

After 21 c4 e3 22 e3 (22 xe3 f4 23 f4 xf xf44 is very uncomfortab uncomfortabe e 22  . .�g Black has a very promising attack. 21 .. 7 7

Ts s a tte low Black should have connued to put immedate pressure on hites centre with 21  .  22 c4 22 d4 e 23 dd xc3 is a pawn 22... 22 ...�d7 �d7 te awardness of te rook becomes transparent 23 b a6 24 ;bxa bxa 2 ;xa f4 and Black soud wn.

1 4 5 2 tf3 tc6 3 tc3 tf6 tf6 4 g3 d5 5 xd5 txd5 6 ig2 txc3 7 bxc3 .d6 8 0-0 0-0 00 00 9 b1 ! nb8 1 0 d4 g4

Tis move does not promise equat. As Black does not want to ve up the bisop it is acuay a tle siy. The idea is of course to provoke Wite into weakeng his kingside wit 2h3, but this is not such a big weakness, and e bishop is not we placed on d7. 1 1 h3

Better than 11 d3 �d7 12 e3 exd4 13 cxd4 b4 14 'd2 d 1 fe xe3 16 xe3 xe3 when Black is ne, PopovicPytel rstenik 1979  1 . . .d7 .d7

22 c4 c5 23 dxc6 txc6 24 c5

Wte has suddenly decided to ght back

Te alternat alternative ivess are worse a) 1 1 ..xf . .xf33 12 xf3 xf3 �f6 13 e3 a a 14 14 g2 b6 1 4 was better for hite in W.WatsonZ.Polgar Brussels 1988. b 1 1 ..h? ..h? 2 g4 g6 13 dxe dxe xe 14 xeS xe 1 'xd8 d8 16 f4 xc2 17 :b2 b2 d3 18 eS xf 19 xf ves White a tecncally tecncally nning nning position position  Glek.

24 . . bxc5 25 a6

 2 d 3

2 �c4+ 8 26 xc xc 27 �xc was maybe better for Black but sl te lesser evi

Or 12 e ;e8 13 g �c8 14 dxe xe xe 1  xe xe xe 16 f  f4 g6 g6 1 7 ;xe ;xe xe8 18 e3 b6 19 h4 h6 20 c4 with adantage for White, DrazicGodena Bratto 1999.

25 .. . tb4 16

The Glek ystem wth     d 

1 2   exd4  exd4

12 ..h6 ..h6 13 e e 14 dS e7? (14... (14...S S s sronge sronger) r) I S xeS g6 1 6 3 3 gve gve lck lck no rel comenson or he wn, ForserUncker, Zch 199. 1 3 g5 g5 g6 14 cx cxd4 b4 1 5 d2!?

hs s led wh  combinonl de, wch shold onl e  drw Objecvel sronger ws IS b3 S 16 e4 wh  sgh edge 1 5 xa2  xa2

We s  s no n o r rdd o o  king king sks   drw drw s obvosl no n oon in hs mnd. Now lck shows h s oonens nssence on connng o gh ws  wise decson 21  bS?  odd mske becse lck ws orced

o sss s kng. Aer 21.'6! 22 d3 e (lck needs breng sce 22 ...'g? ...'g? 23 h6 �6 24 h7 nd 22 .'h .'h 23 h6 e7 24 be1 6 6 2S eS  eS!! re boh nnn nnngg or Whe) 23 h7+ h7+  24 h e7 2S be 1 d! d ! 2S e6 leds o  dw er 26 g7 'g 27 h6 g [276 w reeon s lso ossble] 2 xe6 xe6 29 ' e7 30 g  31 h6) 26 h4 nd ccordng o Glek Whe hs he ck.  eel h lck shod now l 26 . . !, nendng nendng 27 g g 6, 6, when whe n lck mgh mgh js js  weher he sorm Now Whe s wns 22 g5 e7 23 g6+ �h8 24 f6+ ixf6 25 h6+ �g8 26 d3 Ie2+ 27 �h2 1-0

16 Ixh7! �xh7 17 h6+ �g8 18 id5 Ic3

Rsk  ws robbl ser s o l  ... . S! S! 1 9 x2 6! 6! wh n ncler oson.  no 19... 19...e e 20 bS 'd7 21 dS, whch wold ve e  srong ck. For exmle 21c6? 22 gS 'e6 23 xS! xS x S 24 2 4 g4 g4 nd Whe ns ns  1 9 xg6+ �h8 20 h5+ �g8 �g8 21 ic4? ic4?

Game 87 Kovalevskaya-Xie un New Delhi Delhi 000 000 1 e4 e5 2 If3 Ic6 3 Ic3 If6 4 g3 dS 5 exd exd55 Ixd5 6 ig2 xc3 xc3 7 bxc3 e7 e7 8 00 00 0-0 9 e1 

Whe s imovng her orces grdl 9 d4 does no mke  lo o sense when he eS wn cn be hreened wih emo  sl  mgh no be so es or lck 9 d4 .6 10 3 e 11 dS S 12 d2 e7 13 b4 cS! 14 xS xS 1S d6 (Glekmn, Een 1993) S S xd6 xd6 1 6 c4 �6 s eql ccording o Glek he bes lne seems o be 17 xd6 d I  xd6 19 'xeS e6! (ll oher moves led o  whe dvnge) 20 xb7 (20 'xcS? d olowed b d2 d2 wold ve ck ver cve l or he wn) wn) 20. .xb7 . xb7 21 �xd6 h3 22  x1 23 x1 c4 wh comension or he wn Whe hs lso red 9 d3 d 3 g4 (9    S!?  163

he Fo Knghts

Glk) 10 b b8 11  d6 12 g5 d7 13 �3 5 and lak anagd to rat annoyng trats V.MaxnkoPt t Ptrsburg 1993

x6 x6 6 1 6 �x2 �x2 5 17 4? 4? (1 7 3 3 6 1 8 d4 g4 19 3  20 x x x x 21 �5 �4 22 22 �x4 �x4 x4 x4 23 :xb7 xb7 5 24 7 xd4 25 x4 d3 26 :4 4 d8 was draws  awMannon dnburg 1999) 17 5 5 1 8  4 (  8  3 a a 19 d4 :a6 a6 20 �d g4 21 3 8 22 4 g4 g4 s  s unlar wl wl 22. .4? 23 \1 woud b good or Wt) 18.. 18..a5 a5 (or 18 186 6 19 g2 wt an nav or ) 19 x5 �x5 20 xb7 xb7 5 21 �d2 xb 22 xa8 and Wt s rosopally bttr ) 11..b8 ( bst) 12 d2 (12 g5 x5 13 x6 :6 6 gvs t no advan tag) 12. 12.6 6 13 a4 (13 4? d7 14 b2 d4 s plasan p lasantt or or lak) 13 . \d7 14 a3 w unar pla

9 .. .i 10 b1

1 1 . . .ig . ig4 4 1 2 h3

s s t natural ov bu Wt also as anor pa to an advantag 10 d3 4 I I x4  x3 12 b 6 and now a) 1 3 g4?  7 1 4 3 5 1 5 gx5 gx5 x5 x5 16 a4 w an unar poson Krasnkowandza ngrad 1986 b) 1 3 d2 prov provs s  kng kngts ts poson and vs t qun a an to prov ts loaon loaon 13  .. b8 b8 (13  5 14    :b8 b 8 15 4 d4 16 b5 b5 s sowat bttr or Wt Wt) ) 1 4  a3 :8 8 1 5 x8 x8 x x8 8 16 �5 and Wt s bttr

2 d5? 4 13 dx6 �xd1 14 :xd xd bx6 would gv lak  advantag

1 0 . .e8 1 1 d4

Or 11 d3 and now a) 1 1 ..g . .g44 1 2 3 s bttr bttr or Wt Wt s n otr ls ls t t  bsop as no good squar squar spaly wt  pawn on b 7 anng. b)  1 4 12 d2 d2 (12 dx dx44 x3 x3 1 3 d2 xd2 xd2 14 xd xd22 b6 1 5 5 b7 b7 1 6 4 4 b8 b8 would oly v qual 12 12xd3 xd3 (or 1 2... ... x3 13 x4 wt a strong tav) 13 x ss lar s 13 4 dx2 13 13: :x4? x4? 14:x4 x4 5  5 xd3 x4 16 x4 s vr vr good or or Wt Wt  4 �x2 �x2 5 1 5 xb7 xb7 6 6 1 6 x6 x6 x6 x6 17 x6 x6 x6 and lak as good ountrpay ountrpay)) 13  . .x8 x8 14 4 dx2 (14. (14.7 7 15 xd3 s sowat bttr or t as lak as probls dvop 15 164

1 2 . . .ix3 .ix3

12  5? 5? 13 g4 g6 14 g5 7 1 5 x5 x5 and W wns a pawn pawn 1 3 ix ix3

Hr Wt ould av playd or strong strongly ly 1 3 x3 and now a) ad s 13 13xd4 xd4 14 :x x x8 15 xb7 xb7 �1 (1 5   5  6 �4 �4 dx dx33 1 7 x7 d8 18 :xa7 xa7 gvs t a lar advanag) 16 2 5 17 4 4 �x1 (17..�x2? 18 xd4 and W W wns) wns) 1 8 :b   g g 19 1 9 xg4 xg4 \6 20  d8 21 xd4 wt a lar ad vantag or Wt b) 13. 13.: :b8 b8 14 d5 a5 15 \4 b6 (or 1 5 b5  b5?? 1 6    a6 16. 16 .d7? d7? 17 b4 dos not work work 17 a4 wt wt  a lar advanag or or t lak sould not  to ontst t lgt squars on  qunsd as  s bound o os os su a gt) gt) 16 1 6 a4 a4 b7 1 7 a3 a3 (17 (1 7 \xa7 5  5 18 a3 a3 a8 a8 19 1 9 \b4 \b4 :xa2 xa2 s lss la lar) r) 17  . .a5 18 4 w a sgt advantag or Wt. 1 3 . . .exd4 .exd4

Equaly possbl possbl was 13  a5? 14 dx5 x5 (but not 14 ..xd? 15 :xd xd x5 [15 .: .:x5 x5 16 4 :5  5 17 xb7 xb7 18

he Glek ystem wth     d

b is obvious good for White 6 d c4  Jb with  white dvntge   Jb c ( d?! 6 Jd c  c b4 8 Jc Jc8 9 Jc8 Jc8 20 d woud gve White compete domnon d is ver unpesnt 6 e �e8  Jc d 8 8  b4 9 Jc d6 with good counterp

20 &xc3 d 1 + 21 f1 g5 22 xe5 .xe5 23 &c5 f6 24 g2 d2 25 xa6 d4

Bc does not f for the tric 2..Jf2? 26 f2 d4+ 2  c 28 e4, when he endgme is ost 26 xc7 xf2 27 h1 h5

1 4 x7

4 Je e8  b woud be un cer, but White cnnot p  cd4? d4  6 b Jd8, fter whch Bc hs l the chnces 1 4 . .e5 1 5 g2

After his White is ghng to eep the bnce Better ws  e4 dc 6 f4 �d  Jd g6 8 g hg6 9  wih  e dw 1 5. . .dxc3 16 f4

White is  pwn up, but Bc hs  ver cve roo nd bishop, nd  the white pwns re we Bc shoud be o. 2 c4+ f 29 d3 f3 30 h7 h4 31 gxh4 x3+ 32 g2 g4

Bc is ting  ot of iss wih this move 2..h4 ws  sfe drw, but Bc obvious si hd some ghng spiit eft

1 6. . a6

This is rther sow in  shrp posion. t seems more sensib to ght for the initive wth 6g!   ( e Je 8 d J d8 9  e 20  Jd+ 2 h2 Jd2 woud gve White serous prob ems with the cpwn nd his ngside c6 8 f4 gf4 9 f4 with compic tions. 1 7 3

   g 8 e (8 e c4 fvours Bc 8.J e 9 e e oos prefer be for Bc 1 7   :Vxd1 1  xd1 ad 1 9 xd xd

33 a4 .e5 34 a5 h2+ 35 g1 g3 36 d7 f4 37 e4 .e3+ 3 f1 f2+ 39 e1 f5 40 d3 f4 41 a6 g2 42 xg2 xg2 43 &a3 g3  f1 g7 45 e2 e 46 f3 a7 47 a5 d 4 h5 c 49 h6 7 50 a7 xa7 51 xf4 7 52 h5 %  %

ame 88 Glek-Nikoli W!k aan Zee 199 1 e4 e5 2 f3 c6 3 c3 f6 4 g3 d5 5 exd5 xd5 6 g2 xc3 7 xc3 e7  00 00 9 e 1 .f6 1 0  1  1 1 d3 e6

This is btter thn   S becuse White ow hs o pend tim defending 2 This 16

he Fo Knghts

wa Whe does no have tme or manoeuvng wih a3 and d2

wih the idea o 9 2. 1 5 c3

1 d2 g 16 c3, wh he dea o b and 2 was a possbe mprovemen. 1 5. . .b6 1 6 b

16 a3 �d6 17 d6 cd6  s d2 d4 19 d4 ed4 would have been even.

1  c4 d7 1 3 b

 3 g! g4 14 d2 d4   e4 e7 woud be raher award or Whe. Al his ieces are srange placed. 1 3. . .g4 1 4 c1

Ths is beer han 14 'd2 because Whie eaves he d2square or he knigh 1 4  e8

Black develops his pieces ino pla smoohl.  he sars acion oo soon wih 14 .. �!   d2 d4 he will end  trou be 16 d4 ed4 17 £ and Whe wns a pece. The more normal ...h3 16 e4 d7  7 1  eein d and  17 ..d4 1 e3 c6  9 d1 aours Whie and so does  ...g 16 e4 d 17 c3 d  b,

166

1 6.  .d4 17 xd4 exd4 18 h4 nxe1+ 1 9 xe1 ne8 0 xe8+ xe8 1 c1 h5

21..c, in order o preven he weaken ing o he hpawn was probab more eac. Aer 22 4 c 23 s d7 Whe would be a ie worse bu the mos lkel oucome would be a draw  4 c5 3 g5 xg5 4 hxg5 g6 5 e4 e7 6 3 5 7 � xe4 8 xe4 d7 9 b8+ �g7 30 e5+ % -%

 Gk   4 .   d 5 ummary

4dS  probably the mot exble way fo Black to meet the Glek ytem. It eem that after 4 .dS Black can equale with almot any idea except for my own 6 .de7. Lugovoi 6..b6 folowed by 7  cS eem to ve Black good play whle the man line after S . xc3 alo offe Black good counterplay Whte  natualy  to create ome theat along the b- and e-e or  to puh wth d2d4 but t hould never be realy dangerou for Black 1 e4 e5 2 3 c6 3 c3 6 4 g3 d5 5 exd5 xd5 6 ig2 xc3

6  b6  Game 76 6 de7  Game 80 6 ...e6 (  0-0   e7  Game 79  . cS  Game 78   xc3  Game 77 7 bxc3 c5

7  d6 8 0-0 0-0 9 b b 8 10 d4 ( 10g4  Game 86 1  h6  Game 85 10 S  Game 8 .e7 8 0-0 00 9 e f6 1 0 Ub 10 .b8  Game 88 1 .e8  Game  8 0-0 0-0 9 e 1 e8

9.6  Game 81 1 0 d4  Game 3 10 d3 ()  Game 82

6 .  i 6

1 0 d4

10 d3

167

CHAPR EN he Glek System: Sidelines 1 e4 e5 2 3 c6 3 c3 6 4 g3

5 .g2 d6 6 h3

In hs chaper we ll dscss rare and less vs lnes a Blacks dspsal: 4..e7 4 ..g, 4.. and 4. .d4 Pray Black shld lk fr eqaly n he an lnes and n n he sdelnes  here are a sl any new dscveres wang  e nd n he Glek yse

 s hard  nd her gd sqares fr he c8shp han g4 and as Whe fen has deas wh g3-g4 laer and as he psn s relavey clsed he lss f e s ly jsed.

Game 89 Glek-Vu Zilleal 1993 1 e4 e5 2 3 c6 3 c3 6 4 g3 e

6    0-0  0-0 e6 8 d4 d

Black cann have fel cfrae a akng hs ve  hs chces were e een he npleasan and he ncfrale. Afer 8 .ed4 9 d4 d4 1 'd4 Whe s ny a le eer,  e psn s very pleasan and  s vey hard fr Back  nd a gd plan 9 e1 e8 0 b3

Gek decdes n sw develpen Here  was pray ly sed  clse he psn and ry  esash a kngsde na ve edaely wh 1 dS? 8 1 1 h, wh he dea f g and ff4 wh a srng nave. 1 0 . .a5!

e dwnsde  1 3 s shwn ed aely. Black sddenly has a hk fr hs cnerpay. 1 1 b2

fer hs ve Whe ans an ecel len versn f he hdr Defence 168

Whe cld sll change plans Afer 11 dS 8 1 a3 a 13 e3  14 d c 1 5 f4 he cld s have creaed a srng kngsde ffensve.

The Glek System: Sidelines

1 1 . . .f8 1 2 d2

12 S? Ce7 13 C2 Cg6 14 h2 looks beer or Whie oo 1 2. . .h6?

Black is rying o aus o being wihou space, bu his can never be he roa o happiness Here he cou have rie he pawn sacrice 12a4? 13 Cxa4 ex4 14 Cx4 Cx4 15 x4 c 16 h2 xa4 17 bxa4 C7 wih some counerplay, even hough Whie reains he beer chances 1 3 ad1

exchange on S Whie wl ominae he ligh squares compleely an be cose o winning 1 8. . .xd5?

Aer his Black is orever in rouble on he ligh squares  was beer o pay a move ike  .a4 19 xd5

19 exS? Cc 20 g4, aking he Ssquare rom he Black queen, gives ie a clear avanage ill  is logica o keep he pawn cenre u an le he lighsquare bishop dream o he long iagonal 1 9. . .tb6 20 d3 a4 21 b4?

is unnecesary weakenng o he c4 squae leas o mplicaons imply 21 c4 wih a clear plus looks beer 2 1 . . e6

21  a3? changing he naure o he game was eniely also possible 22 a3 a2!?

Black nees o creae some counepay 22.c6 23 4 C7 24 3 gives Whie a grea avanage 23 d4 5?!

According o Glek Whie sans much beer I is har o  couneplay or Black 1 3. . .exd4

This shows ha 7 is probably no such a goo iea er 13  c 14 h2 Back has harly improve his posiion 14 txd4 txd4 15 xd4 8 16 �h2 6 1 7 f4!

I you have a space avanage you shoul no exchange pieces Aer 17 CS? CxS  exS 7 19 c4 S he whie avanage wou have she. Back woul be hre o have some breahing space or his remag peces. 1 7. . .d7?

I is no obvious ha he knigh  be beer pace anwhere else 1.bS? wih he iea o . 7 looks much beer 1 8 td5!

Now e siuaon has change Aer an

Bu his is no so goo  Back is now losing by orce Beer were moves like 23.Mac or 23.c4, when Black is much worse bu can sll gh 24 g 1 a8 25 bx5 dx5 26 e5 ed8 27 xd8!

imple chess The pin is lehal 27. . xd8 28 xd8 xa3 29 e6 fxe6 30 xe6 a2 31 ee8 x2 32 xf8 �h7 33 f5 10

Game 90 Heor-Harman P E 1996 1 e4 e5 2 t3 f6 3 tf3 t6 4 g3 g6

Anoher way o he Pirc Deence However, is no a goo version because Blacks knigh on c6 is mispace 5 d4!

onger han he auomac 5 g2 g7 6 00 00 7 3 7 4? ex4  Cx4 Cxe4 an 169

he Fo Knights

Black has wo a paw for vr itle compe sao) 7  d6  h3 h6 whe there is o ap paret reaso why Black should be si catly worse or eve sihtly so 5 ...exd4 6 txd4 g7 7 tde2

s move is local but it does ivolve a sliht loss of me Equaly iteresti are a) 7 xc6!? bxc6  2 a6!? - 9 0-0 b looks ver soud too) 9 f4! 9 eS �e7 10 f4 0-0 is far from clear Whtes ki is qute ucomfotable i the cee) 9  dS 9 .. 0-0 looks better 10 e5 d7 1 1  00 12 e e 13 e f6? a bad istake; after 13..e7 Black would oly be slihtly worse) 14 xd5! fxe5 15 b4 exf4 16 f4 b7 17 xc6 White was cleay better i tpusky apetaovic oroto 199. b) 7 e3!? 0-0  2 d6 9 h3 d7?! 9 . e 10 0-0 d7 is mre atral after which hit s somewhat better; this is act ay a drect trasposio to the 3 system of the Pirc) 10 f4 b6 1 1 b3 d5 12 !? e? sply bad after 12 dxe4 1 3 xc6 bxc6 14 xe4 xa 15 a hite would have a eat itiave to compesate for the ex chae but sl this is uclear) 13 xc6 bxc6 14 d4! ad Wte has a reat posioal advatae ikel-uzi Budapest 1999

fd7! would ive Black ood couteplay aaist Whites cetre 1 0 00

Or 10 5!? h6 1 1 e3 e5 1 2 b3 a 1 3 0-0 ed7 14 'd2 c5 1 5  h7 16 ad wt a smal advatae for Whte in Glek Ibramov, ussia 1995 10 .. .d7 1 1 f4

his alows Black to creae couterplay troer are a) 11 4! b 11..b5? 12 S! h5 13 xb5 'c 14 h2 ad Whte is much bet ter  there is o rea compesaio for he paw) 12 a4!? with the idea of 3 ad -f4 with advatae. b) 11 h2 b5! 12 a3 a5 13 f4 e5 14 cd5 c4 15 xf6 xf6 16 c3 with chaces for both players i Rashkovsky Othma Biel 1999 11...a5

Black misses his chace. fter 1 1.b5! 12 a3 a he would have aned ood couer play. 1 2 g4 h6  3 e3 a4

7 .. .0-0 8 g2 d6

Or   e 9 0-0 b 10 a4 ths move is ot ecessary Glek writes that White should play 10 f4 d6 1 1 h3 d7 1 2 4 b5 1 3 3 b4 14 ce2 or 10 f4 d6 11 h3 b5 12 fd5 wt some advatae i both cases) 10...a5 1 1 f4?! 1 1 h3 would s leave Whte better) 11.d6 12 d2 e6 13 d5 xd5 14 exd5 b4 15 fe e4. White has playd the ope less tha wel ad Black has a pref erable posio, rasekow-Glek Moscow 195. 9 h3 e8

his is vey atural. 9  a5 is cosidered  the foow ame. 9 ..d7 1 0 5 h6 1 1 3 e 12 0-0 is a little better for Wite while 12 4 e5 13 f4 c4 14 c1 b6 15 00 c6 16 3  70

4 d2!

Wte correcly decides to let Black creae some weakesses o the queeside wih the paw push. He knows that it costs Black time ad tha after 14 a3 'e7 1 5 d2 Black would be able to use the weaken of the lht squares to his advaae wi 1 5...a5!, leavi Whte awkwardly placed 1 4 . .a3 15 b3 b5 16 tg3 b4 17 tce2

The Glek System: Sidelines

h7

20 xeS?

Not 17xg4? 18 hxg4 �xa 1 9 xa xg4 20  h4 21 c3 bxc3 22 xc3 bcau o pc ar much ongr than a rook hr whr th wakn of th dark quar  pcaly mporan

Black rturn th faour Ncary wa 20 h4 21 f2 afr 21 c5? x5 22 xa8? xa8 Black would ha mor than nough compnaton for  xchang) 217 and Black ha managd o kp h pc ogthr Th poon would thn ha bn ry compcatd and who  b tr and why  a ry dfcult quon to anwr Afr a mo lk 22 d4 Black mut choo btn 22  x5 and 22  a6 wth a racaly compl poon

 8 eS! dxeS?

Afr th Wht ha a ry dangrou na Bttr wa 187?  9 ad f8 20 d4 and Wht  ony a l btr  9 d 1 8

Forcd Aftr 19xf4? 20 xf4 (or 20 xd7? xd7 21 xd7 5 22 xc7 3 23 xa8 xa8 24 4 wth good chanc to in th ndgam) 20   2 xh6 b8 22 xg7 xg7 23 f4 Wht ha a trong at ack  am r doubtl whthr Black  abl o hold th poon togthr at all for xamp 235 24 h5 gxh5 25 xh+ g6 25  h7 26 f6 g7 27 g+ fg6 28 xd7 xd7 29 h+ g8 30 'h6 and Wht n) 26  xf6 27 4 f5 28 �xf xf5 29 gxf xh5 30 'f4 and h hat of d 1-d4  forc Black to part th at lat anothr pc

21 xh6 xh6?

Anor carl mo Aftr 21 b8 22 4 Wit ha th chanc but  th wa prfrabl 22 xh6 S

Black can no longr a hmlf 23 4 h 7

Now th wdih hampon nih th gam n tyl

20 xeS?

In h gam t work ou wll But th alta wr worh ngang a) 20 c5? xf4 206?? 21 'xd7 xd7 22 xd7 xc5 23 xc7 and Wht n h xchang a lat) 21 'xf4 (21 xf8? xf8 22 'xd7 'xd7 23 xd7 g3 24 xc6 x2 do no work out c+  a horribl that) 215 22 xa8 xa8 23 xf8 xf8 (23xf8? 24 xd7) 24 'xb4 c6 (24f3+ 25 xf3 xf3 26 xd7 wn for Wht) 25  f3+ 26  g5 27 f4 and Black ha om praccal compnaon for th xchang ftr 275 t  not obou that h  wor b) Vry trong  20 f5 brnng th f rook nto play wil multanouly kpng th Black forc ou 20  a6 (20   21 xf5 xf5 22 xc6 'xd2 23 xd2 g Wht a wnnng adanag) 21 g6 g6 22 xh6 and Whit ha a clar adanag

24 hS gxhS 2S xhS gS 26 6+

Th pon bhnd th combnaon 26 x6 27 x6 x6 28 x6 cS 29 dS xc2? 30 xeS xeS 31 xe5 x2 32 gS+ �8 33 cS+ �e8 34 xb4 1-0

Game 91 Blkov-Zasv Mocow 1996 Th annoaon n th gam ar bad on 171

he Fo Knight

thos by Vlad Bkov in New n he Yearbook 42.

1 e4 e5 2 3 c6 3 c3 6 4 g3 g6 5 d4 exd4 6 xd4 d6

A sight divrsion rom 6 ..g7 which opns th way or som harassmnt with g But basicay it maks litt dirnc. 7 h3

Aso possibl was 7 xc6 bxc6 8 g2 b7 9 0-0 g7 10 b3? 0-0 (10..x? 11 x  xa 1 2 g5 6 1 3 xa dos no work) 1 1 b2 c5 1 2  8 13 d3 wih a smal dg or Whit 7 . g7 8 de2 00 9 g2 a5

1 2. . e8 1 3 b3

Whit now vs Black a choic. To b considrd was 13 d2? 5 (1 3 .. h7 1  givs Whit a sma advantag 1 xh6 xh3 15 xg7 xg7 16  c 17 d xg2 1 8 xg2 xb2 19 h1 , whn Wit sms o hav som compnsaon. On in is 19  c5 20 3 g 21 3 c 22 d5 and Black is in oubl Or 20 b5 21 5 and Whit holds a strong initiav. ayb 19..c5 is wrong but in any cas Whit sms to hav som dangrous opons 13 a3 5 woud v Black good coun play and should not b considrd. 1 3. . axb3 1 4 cxb3 a5!?

Black dvlops th rook via th ourth ank This is obviously qui doubldgd, as  ook can th caus a o o damag or com into troubl isl. A normal mov ik 1 h7 would hav l h posiion rathr balancd. 1 5 d2 ae5 1 6 ae1 !

This tim tis plan coms without .. 8. 1 0 g5?

10 0-0 coud quit asly hav transposd to th pvious gam. With his mov Bikov provoks Black o play h7 h6 Whthr Black would hav playd this mov voluntar ily as Hartman did, or i i is a waknng o th kingsid, is vy dicult to dcid. That two strong gandmastrs cam to a dirnt conclusion is a vry good indicator o his 1 0. . .h6 1 1 e3 a4

1 1 8 1 2 0-0 would b a ittl bttr or Wit 1 2 0-0

Also possibl was 12 d2 h7 13 a3 5 1 d c5 (1   c 15 d3 6 [15 xb2? 16 "b5 drops a pic] 16  woud also giv Whit a sma advanag) 1 5 x5 dx5 16 0-0-0 and i is bttr. 1 7

A sound posiona dcision Th rook is activatd and vacad rom th unplasan prssur on th long diagonal ss aracv was 16 xh6? xh6 17 xh6 xh3 (a vry nic tacc ha can asly b missd) 18 xh3 h5 9  xh3 20 g2 h5 (20 ..h8 21 h g7 would b vn 21 h1 5 22 xh5 xh5 and Black has som nitiav. 1 6. . h7 1 7 4 b6?!

Too passiv. Bttr was h acv 17..b5

Te Glek System Sidelines

in order to create weaknesses on e queen side. On the other hand Black could not play 17..xe due to 18 xe xe 19 xe xe 20  e8 2 g2 and the compensa tion for the exchange is hardly enough 1 8 f3 .6 1 9 f2 th5?

Black mus feel under pressure to play such a move The rook  be very badly placed on hS the knight was a ittle bit in the way on f and also the f6-knight was keep ig a good eye on the e-pawn t was better to accept a sighly infeior position with 19. d7 20 fdS xdS 21 xdS e7 22 b 22 f xdS 23 eS xe3 2 xe3 xeS is e for Black) 22 ..�b7 23 �d e6 2 xg7 xg7 2S  and Wite might be a le better. 20 txh5 xh5 21 td5 te7

21heS 22 'e �b7 23 b would give Black a headache 22 g?

This is a risky decision and probably a it tle unnecessary. White would keep a signi cant plus after 22 f exposig te awkward posiion of the rook on hS 22...txd5

The exchange sacrice is forced. fter 22  h 23 �f xdS 2 xdS �c8 2S g3 e6 26 d3 ere is no counterplay whatsoever. 23 gxh5

  g8 28 xg xg7 29 �d f6 30 e �b 7 3 f th posiion favours no one 23  txe3 2 xe3 gxh5

Beikov beieved 2.dS was a serious al ternave. s w shal se this is an ilusion: 2S d2 d 26 f e7 and in my opinion Black does not have enough compensation for the exchane i tis position After 27 eS cS 28 e6 �c8 29 f Black is under havy pressure for example 29�xe6 30 hxg6+ g6 31 �c d7 32 de2 d3 33 xe6 xe6 3 xe6 d2 3S d  xe6 36 xd2 when Black must choose between this hopeless posiion or losig the endgame after 36.xh3 37 " xg 38 g. 25 f f5 26 Wh2 fxe?

 mistake. There is no reason to let Whites bishop come from g2 to e where it endangers Blacks king White woud also be better after 26 ..h 27 d2 but no 27 g3 xg3 28 xg3 d 29 c2 cS when the active bishop gives Black good compen saon for the exchange 27 xe+ Wh8 28 g 1 

Exposing th problems for Blacks king 28  . e7?

This loses by force. 29 fg2 g8 30 f3 1-0

There is no way to meet dS.

Game 92 Gek-Graaryk Geheim 2002 1 e e5 2 3 tc6 3 tc3 tf6  g3 .b 5 ig2

The local coniuaon After 23 exdS heS 2 f Se7 2S fS MeS 26 g6+ g6 27

s siply mus be the ght move oth erise the g3 system is too slow to be justi ed t s out at Black has a reasonable game after S d3 dS 6 exdS and now: a) 6 ..xdS 7 �d2 00 after 7 ...xc3 8 bxc3 �d6 9 �2 0-0 10 0-0 White has a tempo extra compared to ines starng with . dS the plan is simply to play c3-c and c3 o put maximum pressure on the S 1 73

he Fo Knight

pawn) 8 �g2 �xc 9 bxc �g 10 h �h5 1 1 00 f6 12 b1 b8 1 b5 and Wit has th to bishops and i mor acv Laktic-Laic Kladovo 1995 b) 6 .d5! 7 �g2 �g 8 h �xf 9 �xf ! Black nds  us hs lad in d vlopmnt bfor it disappars) 10 �g2 �xc 11 bxc 0-0-0 12 00 h8 1 f f5 1 c1 ith prospcts for both sids Barczay-cs Budapst 1 995 5 ...x3

5 . 0-0 and 5.. .d6 ar both covrd in th foowing gam. 6 bx3

s s th natral rcapr but far from th ony opon. so intrstng is 6 dxc and now a 6..0-0 7 00 d6 8 h �6 9 f xf 1 0 xf g 1 1 f5 g5 12 g f6 1  h 'd7 1 b a8 15 �g Gk-Rontais Bad Godsbrg 199 and Wit is bttr  Glk b) 6  x 7 5 x5 8 �x 0-0 9 00 d6 10 f g 11 f5 f6 12 g2 8 1 g d5 1 �g5 c6 1 5 h1 d6 16 d and hit is a ittl bttr Glk-Grabarczyk atowc 1992. c) 6 .d6 7 g5! (7 h h6! 8 f5 �xf5 9 xf5 'd7 10 g 0-0-0 wold giv an un clar poston hit has waknd his king sid and is o bishops hav yt to crat sious scop) 7 ..h6 8 �xf6 xf6 9 d2! h5 10 c �6 11  0-0-0 12 d2 and Whit is a ittl bttr accrding to Glk Al of this nds to b tstd in pracc bfor any crtain rl valuaon can b prsntd

1a5! to stop th a-pawns march coms into consdraton  4 a5 d7  5 �h2 7  6 e3 g5?

h Poish G opts for somtng intr stng This mov was Glks own rcom mndation from an 68 1 6d6! 17 b 1 ! b5 1 8 b2 ac8 1 9 b 1 7 20 b gav Wht a clar advntag in th gam orchnoi-Glk Wijk aan Z 1997  7 h5 h6

17  18 h 6 was possibl l though Wt s a ittl bttr.  8 ab ?

Not th bst Aftr 18 f! �7 19 2 6 20 f2 Whit woud hav kpt a sight dg Now Black dcids to captu th a pawn  8. . .7  9 f4 xa5 20 xb7 7

Black could also tak th cpawn straight away 20  xc! 21 h � (21..7! with th ida of 22 �h f5 2 xf5 f6 with unclr play is aso posbl) 22 g �6 (22.. 6 2 xh6 looks vry dangrous) 2 'd1 and Whit has compnsation for th pawn but no mor 2 d

21 g 'xc 22 h transposs to t pr vious not 2 .. .x3 22 d2 a3?

6 ...xe4 7 xe5 xe5 8 xe4 d5 9 g2 00  0 00

Whit is a bt bttr  h has th o bishops and a xibl cntr.  h posi ton is not compltly clar hit dos hav a awd pawn structr and th knight looks good in t cntr  0. . e8   h3

 prcauon. 1 1 d als loks natal.   . .6  2 d3 f6  3 a4 e6 1 74

h qun gos astray for no apparnt rason Black wants to com hom and d fnd g7 wth h qun but ths is way too passiv. It was bttr to kp th qun cn

The Glek System Sdelnes

tralised and read for acon. 22.�d4! 2 e a4! (keepng an ee on the g4sare) 24 d2 Meb8 25 Mb xb7 26 Mxb7 a and the posion is ver ncear. \hite has some potena threats against Backs king bt Back has a passed pawn spported b the rook. Al in l the chances ook prett balanced. 23 f4 e6

Back wold also be in trobe after 2 ...a6 24 g5! �xb7 25 gxf6 g6 (or 25 .e6?! 26 �hS 7 27 �h4 a 28 g7 xg7 29 xh6). Now both 26 c and 26 g4!? give \hite seos threats aganst Backs kng. Even so there is nothing con crete and Back can sl hope for the apawn or the exchange to be a decisive factor ater on. 24 f5 g5?

This eaves th knight  a terribe sita tion Better was 24...d8 25 d7 when te  contine wit g4 Back shod avoid the folowing rap 2S ..Me7?? 26 ! and the een has nowhere to go (b4 and cS  be taken from im wth tempog pawn moves). 25 4 7 26 g4 5 27 f4 f8

40 c3 g5 41 ie7 d7 42 xd7 xd7 43 xg5 fxg5  ixg5 6 45 W3 c4 46 if4 Wg7 47 W4 Wf6 48 i3 ie8 49 ie5 1 0

Gek retrns one pawn in order to gan conro of the dark sares for the passed pawn. Ts bishop ening is easi winning te  pa gS f6  and g6 and Back can do nothng abot it.

Game 93 Glek-ells Oend 1993 1 e4 e5 2 f3 c6 3 c3 f6 4 g3 i4 5 ig2 00

Or S .. d6 6 d h6 7  e7 8 d4! (tis is oca as Back liits s conro of d4) 8 .. xc 9 bxc g6 1  a (1 eS!? dxe5 11 xd8+ xd8 12 e wold slghtl favor \hite  Gek) 1...d7 (GekOnischk Wijk aan Zee 1997). Now Onischk recommends that te shod conne with 11 �! b6 12 Md e6 1  eS dxeS 14 d4!? exd4 1 5 cxd4 with compensation. The natral connaon for Back is S ..c4 and now \hte needs to prove his iniave or at east increase the pressre. Aer 16 eS dS 1 7  a4 1 8 c4 g4 19 cxdS xd 2 xc7! \hite gains a g iiave. Bt ths lne is of corse far from forced. Back shod be both ess greed and more caos 6 00 d6

If o want to know who is better and who is worse ask the knight on h7! 28 i4! c5 29 ia3 ac8 30 d4 e2 31 f2 xf2 3 2 xf2 d8 3 3 ixc5 c7 34 xa7 xa7 35 ixa7 c7 36 ic5 f8 37 e3 d8 38 3 d7 39 7 f8

Aso possibe is 6 ..e8 7 d (7 e!?) and now a) 7 ...xc 8 bxc dS 9 exdS xdS 1 gS h6 11 ! (11 e4 fS 12 c4 f6 was paed in GekZatsev oscow 1991 after 1  xf xf6 14 b2 te wod keep a slight ps) 1..hxgS 12 xdS xdS 1 xd5 e7 1 4 g2 f6 1 5 Mb ! with advantage to ite b) Less impressive is 7 ... h6 8 dS! f8 9 h (9 c4!? d6 1 h also ooks beter for \ite) 9 ..d6 1 xf6+ (better is 1 c4! with 17

he Fo nhts

some adantage; the man dea s that f Blak takes on d Whte  reaptre wth the  pawn and reate pressure down the -le wth he rooks, as wel as n the ente wh the push d3-d4) .. f6 1 1 h2 'g6 12 g4 f 13 ef (not my nest moment; better was 13 e3 e4 14 e4    g2 and maybe Wte has a smal adantage) 13  f 14 e3 e6 1  d   6 4 d4 17 f4 d7  h2 wth an unlear poson, Pnskagomedo, appele la Grande 1997.

hs pawn sae s not jused. Blak would hae been better off playng 1 ab. Often a rook that s not er wel plaed on a an be equay nae on b, eept that t protets a pawn. f t urrently has no better plae to eerse ts powers t s usual for t to be deployed n ths way. Probably the hanes are balaned after 1 ..ab!.

7 d3 d7

1 6 xb7 e5 1 7 h3

Blak has tred oher moes here: a) 7 .g4!  h3 d7 (hs s smply a lost tempo beause Whte often plays h2h3 freely) 9 h4 d4 10 f4  1 1 h2 6 12 a4 and Whte was learly better n Glek onson, Essen  99 b) 7 . e  h4! (yet another Glek dea) ... 3 9 b3 d 10 g de4 1 1 e4 h6 12 f6 f6 3 b and Wtes pos on seems preferable. ) 7 .. h6  d2 e6 (...a6 9 a3 a 0 b4 b6 1 e3 4 2 h3 3 13 f3 d4 14 g2 was a ittle better for Wte n Glek-Wedberg, Bel 1993) 9 h3 a  0 a3  1 e3 e3 12 e3 d 3 ed d 14 d 2 3   3 d6 16 d2 f 1 7 4 4   4 h epelleer-Ani, Nar bonne 1999). Now Whte an seure a better sruture wih 19 6 6 20 '6 b6 21 a4 wth some adantage. Blak shoud probably play 19 ...b6 and hope to reate more ounterplay wth the queens remanng on he board

Whte s arel, but ths s a ittle slow tronger was 17 f6! 'f6  e4 de4 9 d4 wth some edge One eample s 1 9f+ 20 3 e 21 f6 gf6 22 h3 and Whte has good hanes n the endgame.

leae Blak wth serous holes n hs kngsde. 13 xg5 xc3 14 bxc3 d6 15 b1 e4?

1 7 ..ed7 1  d2 c6 1 9 b4 e5 20 4 ae?

Blak should hae been more prophyla n hs lne of thinkin to preent Whtes net moe After 20 ...a! 2 bb1 (21 f6 f6 22 4 d6 23 a4 h should not e Blak problems; the rook on a4 s probably the worst-plaed pee on the board) 2...h 22 'd2 f 23 fe f6 24 e3 e Blak knghts are naly playng. he poson s probably n a dynam balane, but Blak must feel a bt more omfotable

 h4

s s a standard manoeure n ts stru ture.  ... h6 9 5 x5 1 0 ex5 d5 1 1 g4 e

Whte would hae the adantage after ..d7 2 f4 ef4 13 f4 d6 14 d2, manly beause of the to bshops. 1 2 g5 hxg5

 2... 3 1 3 f6! d4 14 g7 would 1 76

21 c4!

Now the entre s broken up and Whtes bshops omes nto he own.

he Glek System Sdelies

2 . ..d6 22 bb dx4 23 dxe4 t5?

23 x4 ws bsoluty ncssry t is curious to s th niffrnc Blck rsnts o his own mtl in this gm Aftr 24 f c6 2 4 xg 26 xg f6 27 xf6 xf6 28 xc4 Wht s bttr, but th gm is still in rogrss 24 d 1 ! a6 25 3 xa2?

A fst th is bttr thn  slow tortur 26 x6 gx6 27 h6 td7 28 �h1 1 0

Game 94 GlekKlovans Willingen 2001 1 e4 e5 2 3 t6 3 t3 t6 4 g3 td4!?

s mov is insir by slr mnou vrs in th Englsh oning whr both  c4  2 f3 c6 3 c3 4 n  c4  2 c3 c6 3  f6 4 g3 4 hv bn tri Hr it givs Blck rsonbl chncs to ght for qut, but ts too soon to jug whthr it  bcom  mn con nuon

5 g2

s looks k h ogcl mov hr but ccing th wn scric is not ncssrly comtly holss  x n now: )   7 looks ntrl, but still th con clusions r not so sml to rw 6 f4 6 7 f3 g4 (7 ... x4 8  xf3 9 xf3 g 0 3 x3+   x3 6 2 f

woul lv Blck sriousl woun) 8 g2 (8 2 xf3 9 x x4 0 x4 x   x   2  2 3 f 3 2 4 4 x4 4  3 00-0 6 0-0-0 with  clr vntg, oorovic-Kio, Bust 2002) 8 .. h (Blck ns to sk ctiv countrly 8 x4 9  8 0 c3 x   xf3 x3 2 xf3 c 3 00 n  woul giv Whit  clr vn tg) 9 3 h4 0 h4     c6 with . h to com. Blck hs scric two wns but hs  surior strctur n goo chncs in th milgm Prcticl tsts of cours show if ths is nonsns or not b)   (th ctiv ry) n now b) 6 x 6 7 c4 (7 f4 x 8 fx g4 9 2 x2 0 x2 x   0-0 x n Blck hs bttr chncs in th milgm u to his sfr king n bttr vlomnt) 7 g4 8 2 x2 9 x2 0-0 0 0-0 x with goo comnson for th wn in this ynmic milgm. b2) 6 3 (Wht shoul  to stbils his cntr) 6  6 7 f3 g4 8 g2. hr is no oubt tht \hit is unr unlsnt rssur hr but h i stl  wn  hv not bn bl t  nything rlly srong for Blck, so myb ths is rly worth  rcc tst. 8   (8 . b4 9 0-0 0-0 9 .. 4 0 x4 n Whit is ust  wn u] 0 x 8  3 n Blck hs noth in 9 00 x4 0 x4 x4   x4 f6 2 f4 c6  g x 4 xg x   xb2 6 b f6 7 f3 n Wht looks btt in th ngm u to  wknsss on th light squrs Evn 7 xb7 xg  8   8  9 xc7 b8 20 x7 (wth is ik 20 0-0 2 f4) looks  goo  for th ont 5 ... tx3 6 x3

6 x3 wi b sn in th nxt gm 6 ...5

h nrl vlomnt Blc hs lso tri 6 b4 7 0-0 6 8 4 0-0 (th nturl mov; ftr 8xc3 9 bxc3 0-0 0 g2 8   3 b6 2 4 Whit ws bttr n 177

The Fo Kights

GlkFr Elsta 1995 9 J2 xd4 10 'xd4 S 1 1 'd a6 a hl fr th bishp in as f Ca4 12 J th bishp has n bttr squars 12. Jx 1 "x 8 and th psin is mr r lss baland

a 19 ad8 20 f Cd4 21  CbS 22 a4 Cd6 2 4 and Wht has an av b 19.f 20 xfS Cd4 21 xd4 xd4 22 S xS 2 S  and th psin s rathr unlar

7 0-0

7 d 6 8 0-0 d6 mayb 8..0-0 9 J2 bS with unlar nsquns is bttr 9 2 4 10  d7 1 a6 is mr naua 1 1 Ca4 av Whit a smal advanta n Gk-Rizuk Bil 199S 7  d6

7 .6 8 d 0-0 9 J2 bS wh ular play s th ht way  a4

his is a standard manuvr Nw Whit ains h tw bishps and thrby a small ndurn d. l Blak has sm dvl pmnt as mpnsatn and wh is bttr and why has nt yt bn didd.   .h3 9 e1 d7 10 xc5

20 g4

Nw Wit has d attakin pssibi s baus f th wak dark squars

N nd fr any day. If 10  x 1 1 xf2 bS Blak has d untrhans

20 .. .g7 21 g2 e6 22 d2

1 0. . dxc5 1 1 d3 00 1 2 g2 fe 1 3 g5 xg2 1 4 �xg2 c6 1 5 h 3 d7

22 . .ae 23 c3 f6 24 a4 b6

ftr 1S .h6 16 d2 4 17  ad8 1 8   xd 1 9 xd "d6 20 ad1 S 2 1  Whit has th advanta baus Blak ds nt hav m fr th ln manuvr Cf6 d7-6d4 Wht shuld mt his da by prparin f2f4. 1 6 e3 f

ls wrth nsidrin was 16. 4 1 7  xd 18 xd 'd6 fllwd by .7-S and  b7-b6 with a mplx middlam ahad 1 7 �g1 e6

1 7.. 4 1 8 hS ad8 shuld iv Blak d untrplay 1 8..xd  9 xd ad8 20 a1 wuld b a ittl btr fr Whit. 1  f4! exf4 1 9 gxf4 g6

his is a hrribl mv  passiv and wakn at th sam m. Rmmbr that Blak shud b tryin t jusi th xhan f bishp fr niht ith avty h atr navs ar: 1 78

h bishp is in t h ln dianal. ain Blak is vy passiv. It was ns sary t  t rat sm untplay bfr Whit ts his idal st-up fr th atak Bttr was a mv ik 24.4 25  6e7 26 f5 h5 27 �h1 g7 2 f3 f 29 d2 e7 30 fxg6 hxg6 31 g1 h7

1..h7 wud mt ih h flwn mbinan 2 x6 x6  1  4 xh5 6 5 f  6 Jh6 a8 7  6 8 W  9 J7 7 40 xf and Blak sn hav t giv in 32 xg6+ �h 33 h6 g7 34 ag1 f7 35 g4 f5 36 xg7+ fxg7 37 xf5 10

Game 95 GlekOnisuk Biel 1996 1 e4 e5 2 f3 c6 3 c3 f6 4 g3 d4 5 g2 xf3+ 6 xf3

The Glek System Sidelines

ous reat. tl his was the lesser el. 1 1 xd4 xa4 1 2 dxe5 dxe5 1 3 3 0-0-0 14 f3 e6 15 xe5 d3 16 3 d7 1 7 f2 e8 1 8 5 6 1 9 2 Jd8 20 f1

s looks more ogcal than 6 x. The bishop belongs to g2 and he queen does not belong on d but rather e2 i it had to choose rght now. With 6 x3 ite leaves the dreams o d2d behind , at least or the m being s sacrice is a necessity i he wants to play or an advantage 6 5

6 d6  d3 e 8 e2 00 9 00 c6 10 h3 b5 11  d 12 h2 a 3 d gave White good attacking prospects in Glek man ap dAgde 1995. 7 O-O!

ritica Also worth considerng are: a  h3 d6 8 a b6 9 xb6 axb6 10 00 e6 and Black does not look worse b)  d3 00 8 00 d6 9 h3 d 10 a a 1 1 xc5 xc5 1 2 e3 e6 1 3 ae with the idea o 'h5 ollowed b  and White is better BeikovZatsev oscow 996. 7 ...d6 8 a4 d4?!

s seems a bit dubious 8b6 9 xb6 axb6 0 'c3 with the idea o d2d despite what was said above) gves Whte a slight advantage accordng to Glek. 9 3 g4?!

9 .. b6 1 0 d g 1 1 d3 00 12 xb6 axb6 1 3 3 h5 1 e3 with an advantage was more prdent 1 0 d3 d7?

10 b6 1 1 xb6 axb6 12  would ve Black some problems as now 13 5 is a seri

20.. :d4

s does not work but Black is short o alteaves: a) 20x 21 x d 22 e3 xa 23 b2 xa2 2 b (wth the idea o a3 rappng he quen) 2d 25 a3 'e6 26 xa c6 2 a b8 28 b a 29 xc ves White a clearly wnng endgame b) 20 . d 21 a a 22 b5 d6 23 e5! 'xe5 2 b2 also wns. 21 xd3 xa1 22 a6 �8 23 2 x2 24 x2 e8 25 d4 6 26 2 7 27 e2 �7 28 g4 a5 29 f4 f6 30 f5 f7 31 �f2 a4 32 4 x4 33 x4 a3 34 xa3 xd4 35 �e3 5 36 2 d1 37 g5 e8 38 gxf6 gxf6 39 e5 �6 40 f2 e 1  41 �d2 e4 42 �d3 4 43 exf6 3  �2 d6 45 e2 f3 46 e6 �d7 47 e7 �d8 48 e5 xf5 49 x7 �e8 50 �d2 �f8 51 7 a3 52 7 �f7 53 d8 �g6 54 f7 �g7 55 x6 xa2 56 �3 6 5 7 f8  �xf8 58 x5 �e8 59 e7 �d8 60 4 h2 61 h7 f5 62 5 �8 63 6 g3 64 6 e4 65 �4 f6 66 8 �7 67 d4 d7 68 7 4 69 �5 5 70 5 �7 71 �4 h4 72 �d3 3 73 �e2 h5 74 �f3 1-0  79

The Fo Knghts uar

lack hould not look for equait n the ideine of the Glek tem 4.. �e7 eem to be too paie to make real ene wle 4 .. ie ite a pleaant poion if he pla  d4 and hould be aoided b lack. 4 .. b4 i more reean but after ..�xc3, whic mut be the main idea, White  ain ome adantae due to the wo bihop. 4..d4 i interetin, but White apparentl ha e eral ood opon at hi dipoal, one of them be he eed  xe 1 e4 e5 2 f3 c6 3 c3 f6 4 g3 d4

4 .b4  2 () 5  0-0  Gam 93;  .xc3  Gam 92 4.  d4 exd4  xd4 d 7 h3 7 8 de2 0-0 9 2 ) 9 ..e8  Gam 90; 9 ..a  Gam 91 4e7  Gam 89 5 2 xf3+ 6 xf3

Gam 95  xf3 (  Gam 94

5 g

180



9 g

6 xf3

CHAPTER  Unusual Fourth Moves for White  e4 e5 2 3 c6 3 c3 f6

n this nal chapte we look at ess eqent outh oves hte. ies sch as  CxeS?  a3  e and also he notnate  c?!. oally people play the Fo Knghts wh te in ode to avoid asses o theoy and copute analysis hs is aleady aceved at ove thee so to  to get the away o the books is haly a good idea. O al these oly  a3!? akes soe sense And this is aso whee we  stat.

Game 9 BlemS Warsaw 00  e4 e5 2 f3 c6 3 c3 6 4 a3?

Robet Fische said that n the Dagon Vaiation o the ician a playe with an Elo ang o 000 can occasonaly can beat a andaste n the sae way  a3 is not so supid as it looks at st sight hte says '1 want see you pans. Many welknown pay es have used this ove once o tice. A nny swne line is  CxeS? is ove called the Me and chlte abit only woks in a biz gae whee Black does not have tie to coe p with a deence Pay contnes with ".CxeS S d and now a S.Cg6 6 eS Cg8  c and te does have soe initiative o the piece Ate "6?! 8 ' dS 9 exd6 C6 10 'e d   .x hite is nnng accoing to soe od anaysis by al chechte. b S.Cc6 (tis is oe accuate 6 dS b hs is the geat hing about acceptng gabits. oeties you can etn the ateal and be bette o o it Ate atenatives the gabit stats to ake sense 6CeS   Cg6 8 eS Cg8 9 d6 with soe copensation o 6".Cb8  eS Cg8 8 d6 again wth play o ite  dxc6 Cxe 8 'd 'e 9 'xg Cxc} 0 e3 Cd 1 c3   cxb Cxe3 13 e3 'xb+ and Black is vey close to wning. 4.d5!? 181

he Fo Knights

According to te od masters you should play in the centre when your opponent plays on he anks. Thogh many old ideas have been revised, s one stays strong in modern chess strategy l Black has tried a lot of different opons here: a) 4 ..g6 is a reversed 4 g3 system. But now White has some interesg oppotniies to ght for an advantage The pawn on a3 is vey usel here a) 5 g3 g7 6 g 0-0 7 00 d6 8 d3 is level (a-a3 is hardy usel). a) 5 c4 g7  d3 d 7 h3 e6  e3 h 9 xe6 e 10 d4 exd4 1 1 xd4 d7 1 xc6 xc6 wih a eve game in Ham douci-Adams, Cannes (rapid) 00 a3) 5 xeS? makes more sense as Black cannot play ..b4 any longer. t is nterest ing hat tis is the only move at ves Wite some chances of ghng for te ad vantage. S ..xeS 6 d4 c 7 dS

and now Black has to possibies a31) 7 ..eS S f4 d6 S..hS 9 g3 and Wite regains te piece t good play; no tice that 9 eS? 4+ would be far too dan gerous) 9 eS dxeS 10 b5 d7 1 1 xd7 xd7 1 0-0 c 1 3 h h4 with chances for both sides. Praccal tests are needed (or lots of fee me and noing bet ter to do) a3) 7  b8  eS g8 9 d6 (also interest ig are 9 c4? and 9 '?) 9.cxd6 (9c6 10 'f3 g7 11 c4  1 h4 aS 13 f4 18

looks very dangerous) 10 exd 'f6 1 1 b5 a 1 c4 h (1  'e 1 3 e3 h6 14 00 xe3 15 e gives Wite a srong attack) 13 'e f 14 e3 xe3 15 e3 and Wte had a srong atack n Carlsten Nyys, Helsiki 00 a4) 5 d4 exd4 6 xd4 g7 7 de (ts s some kd of Prc Defence that should not be dangerous for Black) 7..0-0 S g3 d6 9 g d7? ess local was 9 .h8 10 h3 g 1 1 g4 fS 1 ex gx 13 gS and Wite was a little beer in BlehmKaufman Wash igton, 00) 10 00 c 1 1 e h3 and Black is hardly worse. b) 4e7 (In my opiion s move is too passive to equalse lly. l, e empirical evidence is against me) 5 d4 d6 6 bS exd4 7 xd4 d7  0-0 00 9 e xd4 10 xd4 xbS 1 1 xbS d7 lack is equal) 1  a4 e 13 a f6 14 b4 cS 15 d c 16 c3 b6 17 axb6 and a draw was ageed n Wals-Piket Nov ad 1990. c) 4 .. cS (a provocaon) 5 xeS? xeS 6 d4 d6 and now: c1) 7 f4 c6 (7 ..g6  eS e7 9 fS h4 10 exf6 xf6 11 e ves Black many problems wih he knight on h4)  eS e7 (now we see hat move a3 is qute usel  Black does not have .b4) 9 dS xdS 10 xdS (10 'xdS d 11 bS dxeS 1 eS 0-0 is unclear) 10 . d6 1 1 exd 'xd6   xe7 'xe7+ 1 3  (or 1 3 'e  e6 ith a good game for Black) 1 3 . 0-0 14 d3 fS wth equaity in Wolf-Lesege, Monreal 1993. c) 7 deS xeS 8 d3 (also nteresng is  e c6 9 f4 c7 10 g3 h unclear pay but 10 eS is very risky for example 10...g4 1 1 h3 h 1  e3 f6 when per haps Black s better) .dS 9 exdS x c3 10 bxc3 'xdS 11 0-0 0-0 and White is probably very sighy beter here 5 b5 d4

The only move. After S ..dxe4? 6 xe5 �d6 7 d4 exd3 (not 7a6? as n Bhend-Flear, Brocco 1991 when Wite is close to winning after  f4 axbS 9 xbS d 10 xc6

Un  l Fo th Mo ve fo Whit e

xc6 1 1 xc7+ 7 12 xa8 a5 13 4 xa8 14 0-0  Backs kng s awkwar an te apawn s potenay vey angerous) 8 0-0 e6  xc6 xc6 10 f4 '7 1 1 x3 6 1 2 '2 an tere s no copen saon for te a pawn sctue

7 g3 6 8  d2 de8 9 4 dx3

s was necessary. fter 1 ...e5 20 'e2 g8 21 4 Wte e ae to organ se a ajor attack on te queense  20 bx3

6 e2

fter 6 xe5? �6 (6..c3? 7 xc6 xc6 8 xc 7  xa8 xa8 10 3 cx2 11 x2 e7 12 0-0 0-0 13  s goo for Wte) 7 c4 c5 8 5 x5  ex5 'x5  0 'e e6 1 1 0-0 0-0-0 Back as e etter gae 6 ..d7

n nterestg aternave s 6 ..xe4? 7 3 (7 xe5 '5 8 xc6 xc6   c5 ooks goo for Back) 7 ..6 8 xc xc6  xe5 f5 10 0-0 6 11 xc6 (1 f4 xe5 12 xe5 0-0 wou e equa) 1...x2 12 x2  13 f4 xc6 1 4 e1+  1 5 2 7 w uncear pay 7 d3 d6 8 00 6

Its aso posse to case e sop an gan space, ke  te Ruy opez fter 8 ..a6?  a4 0-0 10 e 5 11 3 a5 1 2 a2 c5 1 3 f4 c6 ot payers ave goo pay 9 e  e 7

t s not y juse to pay  . g5? wen Wte  use s ea  eveopent to reak open e cente 1 0 c3 c3 1 1 xc3 a6 12 a4 'e7 13 c2 5 14 3 a5 15 a2 c 5 1 6 e3 8 17 g3. Wte  etter ue to s contro of te gt squares. s s of course just a fantasy varaton ut t oes sow te tenences n s poson.  0 xd7  xd7   4 ex4  2 3 6

Back cou aso ave equase t 12..g4? 13 xf4 (13 ex4 0-0-0 14 3 f5 wou e very angerous for Wte) 13 . .00 14 c3 c5.  3 x4 e5 4 �  x3  5 x3 000  6 5 g4?

Unnecessary copcaons fter 1 6..x5 17 x5 f6 18 'a5 8 1 c3 e5 Back s no worse.

Wte s etter. He a contro of te cen tre an posses for creang treats on te -e. 20 .. .e5 2 e2 7 22 d4 4 23 xg7 xg7 24 x4 xe4 25 nab 6 26 d3 e8 27 d5 5

27..cx5 28 �x5 8e6 2 x6 us tates Backs troues 28 3 e2 29 5 �7 30 4 e7 3 4 d8 32 xd6?!

More angerous was 32 a4 8 33 a t an atack. 32 . . xd6 33 4 e5 34 2 Je2 35 5 e5 36 g4 e3 37 b a6 38 g6?

38 a4 wt soe pressure was etter 38 . ..xa3

Now t s te. fter 38 .e ? 3 xe 'xe+ 40 g2 'e 41 3 5 42 'f5 foowe y e4 Wte coes out on top. 39 e4 d7 40 e ?

Fro ere on Wte oes not pay s est Wt 40 x6 f5 41 e6 xe6 42 xe6 xe6 43 xe6 e3  xf5 6 45  6 46 a7 Wte cou ave kept te aance an ae a raw 40 .. .�b8??

e troue aness. fter 40 ... f5 41 183

he Fo Knighs

Me6 Mxe6 42 xe6 'xe6 43 dxe6 d8 Blck hs seos wining chnces in he rook ending 4 e7 4

No 41'8 42 Me Md8 43 Mxd8 'xd 4 Me nd Whie wins 42 b ??

Rering he fvor fer 42 h7 Mb3 (or 42 3 43 Mb nd b7  fll) 43 Me 7 44 c7 'd7 45 'b b6 46 M8e7 e wins 42 ..nb6 43 xb6??

icide isccon 43  d  44 g2 2 4S 3   46 g4 hS!

I is vey likely h his is he ove Whie overooked fer 47 xh5 d1 Blck wins he een 47 �xhS h3 me (-)

 erible ge!

Ga 97 GoluoviSzao udapet 1995  e4 eS 2 3 c6 3 c3 6 4 3 dS S exdS xdS 6 bS

fer 7 �xc bxc6 8 bc3 �d6 he o bishos nd he iive re sfcien co enson for he bd wn srcre fer 9 00 0-0 10 d3 g4 he osion is sighly ncoforble for Whie fo exmle: ) 1 1 h3 h5 12 g4 (hs looks errible b here is no leve) 12g6 1 3 e wih  sighly worse osion for Whie b) 11 e? f5 12 h3 �h5 13 e3? e ws in oble b his is  bd bln der fer 13 e2 e8 14 e3 e4 15 e4 Mxe4 1 6 d2 x 1 7  Mxe1 1 8 xe 'f6 Blck hs  srong ck) 13  e4 14 dxe4 e4 1 5 xe4 �h2 0-1 iche-ck corresondence 1980 7  . .d6 8 d4 exd4 9 cxd4    - g4   e3

This cn be cored h he 1 0  �e6 line in he coch or Knighs (see Ges 29-31)   .. e7

1 1 'f6 12 e2 Me8 13 'd2 e 14 h3 �h5 15 g5 e6 16 fe c8 17 c4 c6 18 �f4 ws  ile beer for Whie  Tr kower-Grnfed Vienn 1928  2 3 5  3 d3

r 13 Me d 14 d2 wih ey Brening-enni Bodensee 1999  3. . .dS 4 c4 xe3

14  f4 sees ore locl B fer he cricl 1 5 c5 xf3 (1 5 xd3 16 cxd6 g 17 e5 xe5 18 dxe5 e8 19 c1! gives Whe soe ressre) 16 'xf3 (or 16 xh7? xh7 17 x d5 1 8 cxd6 xd6 1 9 b1 b6) 16xd3  cxd6 xd6 18 M b 1 Whie does kee n dvnge  he igh is no rey coforble Blck cn s gh wih 8c5? 9 dxc5 e5 wih soe conerly, b he longer rob les on he eenside s e Whie eners  coch For Kighs re versed where he hs he exr ove 23 This gives him he ossibiy o ly n oh eise dbios ie s he noevre c6-b4 hs been evened 6 ...xc3 7 bxc3 184

 5 xe3 c5

fer 15  e7 16 e4 x3 17 e5 (17 c5? b2 18 M4 x 19 xf3 Mfd8 nd Blck wins) 17�b4 1 8 c2 g 19 xg hxg6 20 'e4 Whie hs good coenson for he wn wih his gre cenre nd chnces

Un l Fo h Mo ve fo White

of eatig a ingide atta t  tae a ong ime befoe te a-pawn get oe to pomotion

4 . .e4!

1 6 c ig6

 nie ombinaiona teme 5 e4

od Wid i 16  x 17 x7 (17 xf 4 i peaant fo Ba 17 1 x g6 19 xg6 g6 20 xg6 xf 21  g7 22 x e7 wit ane fo bot ide 1 7 ig6 hg6 1 8 d5 c7 1 9 ae1 ae8 0 e Je7!

e pan of doubig te oo on te e e ve Ba a afe daw 1 d3 fe8  e4 f4 3 h1 d6 4 fe1 ig3 5 d1 if4 6 de1 ig3 7 d1 if4 8 a4 b6 9 de1 .g3 30 d1 f4 Y  

Wite annot mae ea poge n te foowng ai batte beeen to of te bet paye of ei age Emanue Lae butay augte iegbert aa i a ine tat a been own to be dubiou fo ite eve ie

Game 98 Taas-Laske Wod Ch. (ame 3) ein 1 916 1 e4 e5  f3 c6 3 c3 f6 4 c4?!

i i a famou begnne move Unfo unatey, afte ti inoent-ooig inau ay Wite aeady a to gt fo euaity

5 x x 6 xe4 d5 ve Ba a eiou advantage due to te to biop and e tong ente 5 ...d5 6 d3 de4

ee i no eaon to mae ing ompi atd fte 6 . f5 7  e4  �b5 Wite wi end up wi an advantage a d4! 9 xe4 4 10 e2 (Ba a pobem wit te epawn 10  f5 (1d5 11 4 f5 12 d e5 1 xe4 xf 4 xf and Wite a a ea pawn moe 11 d b4+ 12 d2 xd 1 xd2 e  14 6 bx6 1 5 e and Ba  truggig to nd ountepay tat juie te pawn deit b  ex 9 x e6 10 0-0 d7 1 1 e2! d6 1 2 d4 0-0 1  f4 xf4 1 4 xf4 and Wte a a poiona advantage, ooviagai No Bay 1996 e e5quae i a good re outpot fo a of Wte piee  epeiay te nigt 7 e4 d6

: a 7 e7 (au Kee idea   f5 9 �2 g6 (9e4 10 g5 6 1 1  e6 12  oo good fo Wite 10 d4 e4 1 1 e2 d6 12 0-0 wi unea pay i  a to be teted b 7  5  0-0 f6 9  00 (Gob exande Hang 947/4) and now 185

The Fo Knights

White should play  c2! ith soe advan tage.

 2 g3?

After 9 . xd4  xd4 -   e3 e 2 -- e8 3 he e6 4 dS! the position is equal  Gigoic.

Or: a) 2 xc6? b 3 d4 hS! 4 c4 fe8 and Wites posion is terble for exaple: S d4 (S  a! 6 g g and Black ns a piece) S...xg2 6 g dS  "c3 eS! th a clear advantage for Black. But not xh2?  8 xg+
0 i3

 2 . h3  3 2 c5

8 d4?

Not recoended After 8 -  9 c3 h6  e e8  h3 d te posion as equal i DjuricMalaniuk, Foia 99. 8 .. d4 9 ld4

Or 9 xc bxc6  xd4 -  - and no: a)   ...e8! 2 b3 cS 3 �c3 e6 4 b2 g6 S fe b ith an attack. ozi binoich Mosco 924. b) Aso relevan is the natral ooking ...cS! 2 c3 b. he authors of En opaeda f he Openn beieve this to be equal, but in y opnion Back has a serious intiave. 3 b3 d 4 b2 f6 S ad fS 6 c4+ h8  h4 h  8 h3 atakoerBogoljubo Piestany 922) and no 8... fS! ould have given Black the ad vantage  the knight on h4 is very eak. 9 .. .0-0!

The alteaves are orse  - xd4   xd4 xh  xc6 h4!   xh xh  2 d4 d8  3 e2 cS and accoring to Keres Black has ore than enough co pensaon for te pan  I agee.  0 .. h4   ixc6 bxc6

ere several oves are possible.

4 lb3

The alternaves also ve ttle cofot: a) 4 f3? g4  S gS g2  6  ! and Wite is copletely defenceless. b) 4 bS a6 S - b8!! (a great posional ove!) 6 c4 xbS  cxbS a6! 8 bxa6 �e6  9 b3 (after  9 b eS 2 c 186

Uns l Fo th Mo ves fo White

xa6 White has no defene; one line goes 21 b3 xa2 22 �xa2 g+ 23 2 xb3 and Blak ns 19... f8 20 he xa6 and Blak shold n. For example 21 f4 4 22 xe6 e6 23 b 1 xb3 24 axb3 xb3+ 25 2 £ 26 e3 d6 and Bak has a lear pawn more and an ae posion. 1 4  jg4 1  f1 h!

Blak wold also be better after 1 5.. xf1 1 6 xf ab8 bt there is no reason to exhange qeens when o are attakng.

openng experiment wth 4 e2 is not rewarded and neither shold it be. 4 2

he problem wth ths moe is that the extra tempo is seless and the bishop is not er ael plaed on e2 so probabl it  hae to moe later. as White st gies Blak the adantage of the rst moe for free.

1 6 td2 f8 1 7 g1

s moe looks terl senseless, bt the alteates are not at lbeter a 17 4 £ 1 8 g xh2 and Blak is winning. After 19 d2 g4 20 h1 xe3 it is loer. b 17 3 ab8 (1 7 ...e7!? is also a good moe 1 8 b3 e 7 1 9 �g2 be8 and there is no defene against ...xe3. 20 00 h3 21 £ �xf3 22 x x 23 xf annot be alled a real alteae. 1 7. . .ab8  8 t4 j

ra his bishop was doing little and now it is the king If White exhanges it he has no defene on the dark sqares  he has aread lost the battle for the light sqares. 19 h3

Desperation is he ommon trait of desperate men. Blak has a nie wn in the fol lowing ne 1 9 b 1 3 20 d2 ed8 2  xd2 22 .xd2 e+ 23 d xf 24 1 xd 25 xd2 h6 26 3 �f 27 d2 'd4 28 �d3 e+ 29 1 xg1 30 d e 3+ 31 �d e mate! 1 9. . .jxh3 20 2 g4 2 d3 Jbd8 22 tx xd3 23 txd3 x3+! 0-1

Gae 99 Van e iel-Timman Wk aan Zee 1985 1 4  2 tf3 t6 3 3 f6

his game is bween te two strongest Dth plaers in te 1980s. Van der Wiels

4 ...d  xd txd 6 0-0

6 b5 wold ge the oth For Knights ith reersed olors. his is probab not so sill when ompared with the atal game. 6 ...x3 7 bx3 jd6

Also nteresting was 7 ... e4? 8 d4 (s gambit is learl Whites idea otherse after 8 e Bak is ertainl not worse 8 ...xd4 9 xd4 xd4 1 0 3 d5 1 1 d3 d6 1 2 dxe4 �e5 (12..xd1 3 xd1 00 14 .e3 e6 15 a4 gae White some iniatie in Van der WielYspo Rekjaik 1985 13 g3 00 and the posion is eqal aording to Nnn. 8 d4 00 9 b1

9 b5 exd4 10 xd4 wold again transpose to the main e of the oth For Knights wit reersed olors. 9 . .h6 1 0 dx tx 1 1 tx jx 1 2 ja3?!

No a er good moe, bt Blak also has good prospets after 12 f3 h4 (12 ... b8 13 a3 d6 14 xd6 xd6 15 xd6 xd6 16 fd d8 eads to eqalit aording to 187

The Fo Knighs

mman, but 12 b seems a bt arial) 13 g3 f6 14 �e2 6 15 4 e and Blak has gd prspets n the ddlegme A ke mve uld be e7 flwed b h3

17ab 1  g2 (1  3 b7 19 b7 6 20   21 b h3 22 d2  and hte has n dene aganst e2) 1 g2 19 g2  a5 20 a a2 2 b b and the eta pawn unt

1 2. . .e8 1 3 f3

1 8 xb2 ad8 1 9 g2 xg2 20 xg2 e2 2 c3 de8 22 d2?

 3 �d d 14 4 b 1 5  d3 e6 wud gve Bak a sd advantage n the endgame hte has weaknesses; Blak desnt

Ater 22 g 6 23 e2 e2 24 a3 a6 hte has sme drawng hanes even thugh Blak s learl better paed

1 3. . 4  4 g3 'a4

22..a5 23 'b2?

he penng s ver and te has aumuated a great amunt  prblems wth hs aral pla

23 g1 6 24 d1 eaves \hte wth mre hanes r the daw

1 5 'c1 xc3 1 6 xb7 3 1 7 b2

Again naurate deene Better was 25 e2 e2 26 b h7 27 b 1 g6 2 b3 b3 29 ab3 b2 30 b4 ab4 31 b4 b4 32 d1 5 33 d7 and \hte has gd praal hanes fr a draw , Blak s nt fred t hse ts endgame f urse

ter 17 a a 1 �e3 (1  d ? �e4 and Blak wns) 1a3 19   20 b3 �a2 21 d   6 22 g4 2 23 bd3 g4 24 g4 e Blak has three pawns fr the ehange and gd winnng hanes

1 7. . .xb2?

Here Blak ul have ganed a g pstn wth the fwing seuene

188

23 . . .'c6+ 24 g1 c4 25 a3?

25 . .  e 1 !

After ts ere s n defene 26 xe1 xe+ 27 g2 '1+ 28 3  1 + 29 g4 5+ 30 3 ' + 0-1

Un s l o th Mo vs fo Whit 

ummay

4 a3  an ntrng m fr a par kng fr a gam wth rrd cur and m adantagu diffrnc. Prbab th gam   aftr 4..d5! 4 x5  a ng bt n with tt rca mprtanc 4 c4  t bad an aftr 4x4! Back arad ha th nta and qit ik a th adantag 4  i harml and prbab jut c t a t tmp. h bt ptin fr ht m t b ntng th tch Fur nght rrd.  in al ht huld prbab k fr 4 g3 f h want t pa ntrng din 1 e4 e5 2 f3 c6 3 c3 6 4 a3 4  ( - Gam 99; 4 c4  Gam 98 4. . .d5 ( 5 exd5  Gam 97 5 b5 ()  Gam 96

4 

4   d5

5 b5

189

The Fo Knights

I NEX OF COMPl GAMES

Acs-aian Hoogeveen 2002 26 AdamsKorchnoi Mad 1996          1 1 ekineLasker Moscow 191          61 masi.IBezgodov, Balatonbe 1996 1 15 rkpovZaitsev Moscow 1992          53 BekovZaitsev Moscow 1996          17 1 Beon Lopez-Izeta Txabarri Eoibar 1998       1 17 Belon LopezJamieson W aan Zee 1977                   105 Berg.KBronstein Tstp 1990                         10 BergSokolov Malm 200 1                                                                                                   BezmanVaravin Pe 1997          7 Blehm-Socko Warsaw 2002          181 BondarevsBronstein Moscow 195 76 BykhovskyHowe pe/le la Grande 1995                  0 ChrisansenGefand Munich 1992                       9 Egorov-Iudachev Aden 2002         99 ForsterHjararson  1996                                            7 GaagherCooper Bsh ague 2001                     36 Ganguy-Acs Pardubice 2002 12 GekGrabarczyk Gesheim 2002                        1 73 GekInov Poo San Goio 2001                       1 Gekovans Wiingen 2001                          1 77 GekKroeze The Netheands 1996 162 GekMarcein Gea'y 2001          15 GekMikhachisin Domund 1998          137 GekNikoc W aan Zee 1997                        165

190

Index of Complete Gmes GlekOnischk Biel 996.......... .......... ........... ........... ........... .......... ........... ........... .......... ...... 1 8 Glek-Vcic Zleal 1993 ........ ........... .......... ........... .......... ........... .......... ........... ........... ......... 168 Glek-Wes Ostend 1993         15 GlekZeier BadenBaden 2002 ................................................................................................ 138 GodenaBeavs Reio Emilia 199596............................................................................ 2 GolbevMalanik Alushta 1994 .......................................................................................... 55 GolbovicSzabo Budest 1995.......................................................................................... 184 Gtnikov-Tal Leningrad 1951 ............................................................................................... 123 HarikrishnaCvek Pardubie 2002.... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... 15 Haroch-Tman Leeuwarden 198 .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... 125 HectorBarkhagen Swedish haionsh Ska 2002 ..................................................... 136 HectorHarman Po En 1996 ........... .......... ........... ........... .......... ........... .......... ........... .... 169 Hector-Johannessen Malm· 2002 ........... .......... ........... .......... ........... .......... ........... .......... .. 161 HectorSokolovI almJ 199 ............................................................................................. 139 Hector-Toshenko Bled Oiad 2002 ....... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... ... 159 Hnydik-Kiselev Zabanski Wesien 1994 ........... ... ...... ........... ... ...... ... ...... ... ... ... ... ... ... ... .. 64 Idachevomov zakhstan 1994 .................................................................................. 14 IvanchkSvidler nas 1999.......... .......... .......... ........... .......... ......... ........... .......... .......... .. 20 KamskyAdams Domund 1992............................................................................................. 23 Kenworhy-Van der Sterren Ramsgate 1981 .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .. 120 KerstenVoeter oondene 1993.... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... 122 KobajaSevostianov Mosow 1994 ......... ........... ........... ........... ........... ........... ........... ........... . 83 KontzVan den Doel Dei:sau 2000 ................................................................................... 42 Kovalevskaya-Xie Jn ew Delhi 2000 .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .. 163 KceraOral Prev 1995 .................. ........... .......... .......... .............. ........... ............ ........... ......... 63 Laier-SokolovI oze 1992........ ............ .......... ........... .......... ........... ........... .......... ........... 5 1 LesiegeGigoric Montal 1998 ....... ........... ........... ........... ........... .............. .......... ........... ..... 1 19 LimaSantos Bra 2000 ......... ........... ........... ........... ........... ........... .......... ........... ........... ....... 146 LtzYspov Munih 1992 ............. ........... ......... ............. ........... ............ ........... ........... ........ 48 MalakhovPinter Balatonbere 1995 .... .... .... .... .... .... .... .... .... .... .... .... ... .... .... .... .... ... .... .... .... . 54 MaakhovYemen Ekatenbu 1996 ................................................................................. 61 Marinkovicneev Leningrad 1989 .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... ... 150 MonsonMilat oeondene 1996 .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... ... .... .... .... .... 126 MotwanChristensen openhagen 1991 .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... . 148 MotylevGrschk Mosow 001 ........... ........... ........... ........... ........... ........... .......... ........... .... 24 Movsesian-Iescas Cordoba Elista Oiad 1998 ........... ...... . .... .. ..... ...... ... ...... .......... 29 Nadyrhanov-San Bishkek 1993 .... .... .... .... .... .... .... .... .... .... .... .... .... ... ... .... .... ... .... .... .... .... ... 93 NikolenkoAleksandrov Smolensk 2000 ............................................................................ 19 NnnSlss Mosow Ompiad 1994 ... .... .... ..... .... ..... .... .. .... ... ... ... ... .. .... ... .... .... .... .... ... . 43

191

Th e Fo Kngh ts PavasovicBeiavsky Pooroz 1999             6 PcoaPolgar.S Bra 199  109 Pedersenrusciov Bld Oad 2002...... .  ...............   56 Pins-Pedzic Czohowa 1998  131 PolovodnRutman nngrad 198  96 Reefat Bn SattarHebden Dhaka 1995 81 RozentasAdams K 199 9 Santo RomanHector Manla Olad 1992               86 ScmatzRomanisin rankn 2001..........   ..... .......  13 SegerKoc Domund 2001 .   . .....            . .   . . .  . .  . . 133 SermekMedancic Bld 1998                   98 Sakedeko Tbu 199 10 Saryazdanov-Bauert Caua 2002       ...... .... .....    153 SirovKarpov cow 1992              66 Skripcenko autier-Adams Bundla 2001      . .           . .   . . . .  . . .  . .   . .             9 SkrobekPinkas Waw 198          1 1 Smyslov-Pougaevsky Baku 1961  152 SoovjovGavrtenkov ula 1999               158 SpangenbergYemein Blava 1993  59 SpeelmanKarpov nar 1992  12 Spiemann-Rubnstein BadnBadn 1925            1  SviderCerepkov nngd 1990 ..     .  . . . . . .   . .   . . .    .  . .   . .   .       .              .  .  . .   . . .  . .  102 SviderMorozeic Mow 2002              .    . . .  . .  . .   . . . . .    . .   .  .            .    . . .  . . . . . .  1 1 1 Tarrasc-asker Wod Ch am 3) Bn 1916 185 TodorovDelcev Buaa 1995           90 Trabert-Fear.G A 1998    ..........    . ... .. 89 Van Der WeideTimmermans  Van Gn 1996                    129 Van der Wie-Tman W aan Z 1985                    18 Van Haastert-Good Dn 1998             106 WinsnesHector b 1998              3 YandemrovAleksandrov  Prbu 1996  69 Yandemrov-Bezgodov molnk 1992                          95 Yemeinaov ocow 2002.   . .    .  . .  . .   .   . . .  . . .  . .  . . .       .                   . . .   . . . . .  . .   . . .  . . .          3 YudasinSagalck Kmvo 1995    32

19