Anatoli Karpov-Veselin Topalov ( 1-0 ) , Linares 1994

ΤΟ ΣΚΑΚΙ ΕΙΝΑΙ ΜΕΝ "ΠΑΙΧΝΙΔΙ" ΑΛΛΑ ΕΙΝΑΙ ΚΙ ΕΝΑΣ ΟΛΟΚΛΗΡΟΣ ΚΟΣΜΟΣ ΓΕΩΜΕΤΡΙΑΣ ΚΑΙ ΦΑΝΤΑΣΙΑΣ -- ΤΩΡΑ ΤΕΛΕΥΤΑΙΑ ΤΟ ΕΧΟΜΕ ΠΑΡΑΜΕΛΗΣΕΙ -- ΚΑΚΩΣ !!!  ΙΔΟΥ ΛΟΙΠΟΝ, ΑΣ ΨΥΧΑΓΩΓΗΘΩΜΕΝ ΜΕ ΤΗΝ ΠΛΕΟΝ, ΙΣΩΣ, ΕΝΤΥΠΩΣΙΑΚΗΝ ΠΑΡΤΙΔΑ ΤΟΥ ANATOLI KARPOV !!!

Anatoli Karpov-Veselin Topalov ( 1-0 ) , Linares 1994




One of Karpov's most brilliant games. The sacrifices in this game seem to be the work of
maybe Kasparov or Tal ... and not the stodgy player that Karpov is usually portrayed as.

1.d4 Nf6;  2.c4 c5!?; 
A very sharp move, but not an unsound one. 

This move follows the general principal that Black should strike
at the center as quickly as possible.

     [  More commonly played is:  2...e6!?;
        aiming for a Nimzo-Indian.  

        or even  2...g6;
        aiming for a King's Indian Defense or even a Gruenfeld. ]   

3.Nf3, ('!') 
Karpov develops a piece, controls the center, and even prepares
King-side castling.
(But it is more normal for White to play d5 in this position.) 

This line actually transposes out of the QP Openings, and makes
 the game a type of symmetrical English where White breaks
 with an early d4. 

(The common move order for this line is: 1.c4, c5;  2.Nf3!?,
  2...Nf6;  3.d4, etc.)  

Here the pawn structure is very similar to the Maroczy-Bind formation
in a Sicilian Defense. 

     [  Conventional opening theory says the best line is d5. But this is not
        set in stone. After the moves:
        3.d5 e6;  4.Nc3 exd5;  5.cxd5 d6;
        we have reached the pawn structure of a  Modern Benoni.
        (Black will almost always follow with a fianchetto of his King's Bishop.) 

        This line, while usually good for White, has never been proven to be
         a forced win.  ]   

Both sides continue to develop in a very reasonable manner.
3...cxd4!?;  4.Nxd4 e6;  

The standard method of development for this particular variation. 

     [  By playing the moves:  4...d6!?;  5.Nc3 g6;  6.e4 Nc6;
        7.Be2, "+/="  {Diagram?}
        {White has a VERY small edge - in this position.}
        we transpose to a Maroczy-Bind Sicilian type of set-up.

        (This formation could have come from just about any move order,
          which includes 1.c4, 1.d4, and of course 1.e4.)  ]   

5.g3!?,  (Maybe - '!') 
Some (older) books say the correct line here is Nc3, but this is highly
debatable.  

This is a good move here, White prepares a very hyper-modern
development of his King's Bishop. The slight drawback to this line
is that White's c-pawn can be seen as a little under-protected.  

     [  Another line is:  5.Nc3 Nc6;  6.g3 Qb6;  7.Nb3, 
         7...Ne5!?;  "~" 
         and Black has good counterplay.  

        My own database, and also ChessBase's on-line db says there
        has been nearly 350 games in this line. The earliest example was:
        GM M. Taimanov - GM E. Geller;  22nd USSR Championship;
        Moscow, (RUS); 1955.  (1-0, 44m)  

        The latest pertinent example is:
        GM G. Zaichik - S. Kriventsov;  R. Aronow Mem. Tourn; 2002.  (62)
        (White took a nice game, although it took over sixty moves to win.)  ]   

Both sides continue their normal plans of development.
5...Nc6; ('!?') 
This - a simple developing move - is perfectly playable.
(Sometimes in the modern main line, Black often develops
 this Knight to d7 ... which is sometimes seen as a more
 flexible square.)  

     [  MCO  gives the following continuation as the main line here:
        5...Qc7;  6.Nc3 a6;  7.Bg5 Be7;  8.Rc1 d6;  9.Bg2,
        9...Nbd7;  {Diagram?}
        The normal move. (And the end of the column.) 

         (Of course not: 9...Qxc4??;  10.Nd5, "+/-")     

        10.0-0 h6;  11.Be3 0-0;  "="
         MCO calls this position even, I would say that White
         has a tiny edge here. ("+/=")  

         GM A. Yusupov - GM L. Psakhis;  Yerevan, 1982.  

         [ See  MCO-14; page # 702; columns # 103 through # 105,
           (mainly column # 105 here);  and see also (mainly) note # (j.). ]  ]   

6.Bg2 Bc5;  7.Nb3 Be7;
Standard development.  

     [  7...Bb4+!?; "~" - GM Z. Ribli  ]  

8.Nc3 0-0;  9.0-0 d6;  10.Bf4! Nh5!?;
Black decides to remove the troublesome White Bishop.
(This does not seem to a grossly unreasonable plan, nor is there
 any dramatic change in the evaluations of most programs.)  

     [  If Black plays ...Na5!?; Karpov had prepared a nice Queen
        sacrifice, viz:  10...Na5!?;  11.Nxa5 Qxa5;  12.Bxd6 Rd8;
        13.Bxe7 Rxd1;  14.Raxd1,  "+/="  (Maybe - '±')  {Diagram?}
        with a nice positional superiority.  - GM A. Karpov.  

        Or Black could try: = 10...e5!?;  11.Bg5, "+/=" {A.J.G.}  

        Maybe: 10...Ng4!?; "~"  ]  

11.e3!,  (TN?)  
A nice move, and rather unusual too.
(The natural move is to retreat the Bishop and safe-guard
  White's control of the dark-squares.)  

'!?'  -  GM John Emms.   '!'  -  GM Anatoly Karpov. 

This looks to be a new idea, I can't find any games with e3 here
 prior to this one in any database. {A.J.G.} 

     [  The standard move for White  -here, in this position -  is:
         11.Be3, "+/="  {Diagram?}
         and White maintains a modest advantage in this line.
         - GM John Emms.  ]   

11...Nxf4;  
Black captures, there is no reason for him not to.
(His Knight was being attacked by the Queen.) 

     [   11...Nf6?!;  12.Qe2, "+/=  ]  

12.exf4!, 
White captures AWAY from the center - a violation of most
conventional wisdom. 

I am sure many players would be at least tempted to play gxf4,
and try to use the g-file for the attack. 

Karpov sees that the open lines give him a huge space advantage.
The remarkable thing about this game is how quickly he manages
to convert this into tangible assets.  

     [ Interesting was:  12.gxf4!? ]   

12...Bd7;  
Black develops, he will have to mobilize his Queen-side sooner
or later. 

     [  Another move was:  12...Bf6!?; "~" 
        with the positional threat of doubling White's c-pawns.  

        According to one author, a much better move was: 
        >/=  12...Na5!?; -  I. Damsky. 
        (This move fails to impress me  much, White retains
         a large advantage after Qd3.  {A.J.G.} )  ]   

13.Qd2, (Maybe - '!')  White patiently prepares the possible doubling of the heavy
pieces on the d-file.  

      [ 13.Ne4!? ]  

13...Qb8!?;
Black must defend d6 somehow.
(Of course ...Qc7? immediately, runs into Nb5.)

"A classic Hedgehog move."
  - GM John Emms.
.
(Seirawan questions this move, but analysis does not justify his conclusions.)

     [  Maybe a TINY improvement would have been:  13...a5!?;  "~" 
        but White probably maintains a very small advantage ... 
        no matter what Black plays here.  ]   

14.Rfe1!,  (Maybe - '!!')  {Diagram?} 
This is easily one of the best moves (and deepest!) of the game.
 It is not at all clear even what future - if any! - the Rook will have here.

You could reasonably argue that this is a natural move ... to place a
heavy piece on a half-open file, and you would be correct. But in modern
GM practice such generalities, (without a concrete follow-up or plan); are
rarely good enough to defeat the very best players in the world! 'Nuff said? 

     [  Fritz 7.0  computes for nearly an hour and plays:  14.Rad1, "+/=" 
         and White has a solid edge. ]  

Black's next has been condemned as inferior, but is a common plan,
{in these types of positions}; to re-deploy the dark-squared defender to
the g7-square ... where it is usually much more effective than e7. 
14...g6!?;  {Diagram?}
Black's lack of space makes it difficult to come up with a really effective
plan in this position. 

'?!' - GM Yasser Seirawan.  

     [  Maybe a little better would have been the try:  14...a5!?;  
        but White seems to keep a fairly substantial edge no matter
        what Black tries here.  ]  

White immediately sets about gaining more territory ...
and also weakens Black's King-side.

15.h4!,
The best move, and also given an exclam by several other annotators here.  

     [ 15.Rad1!?, "+/=" ]  

15...a6; 
This could be best here, ...h5 (?!) looks like it weakens Black's King-side
far too much.

     [   15...h5!?;  16.Rad1 Rd8;  17.f5! gxf5[];  18.Qh6, '±'  {Diag?}
                - GM A. Karpov.  ] 

16.h5!,
An excellent move ... several authors have said that Nd5 here might have
been better here. (Its definitely not!)  

     [ 16.Nd5!? exd5;  17.cxd5 Bf6;  18.dxc6 bxc6; "=" ]

***

Black now makes a rather desperate bid for counterplay.  

16...b5!? ; (Maybe - '?!') 
Nunn labels this as an error.
(Ftacnik calls in dubious. GM Y. Seirawan gives it no mark at all.)  

'?'  -  GM John Emms.  

Part of Topalov's style is fierce resistance and often a disregard
for material considerations. I think that Black thought he would get
adequate play by this move ...
and a subsequent sacrifice of an exchange. But Karpov crosses
him up ... and does not take the bait! 

     [  >/=  16...Rc8!?;  17.Rad1, "+/="  

***

        >/=  16...Ra7;  17.Qe2,  "+/="  

          (Or 17.h6 b5;  18.Nd4, "+/=")    ]   

17.hxg6 hxg6;  18.Nc5!, 
A nice move ... and a surprise for the Bulgarian super-star.
 (He was probably expecting White to capture on b5, but this line
  would give Black a lot of counterplay on the light squares.)  

'!'  -  GM Anatoly Karpov.   '!'  -  GM Yasser Seirawan.
'!'  -  GM John Emms.  

     [  Very tempting for White had to have been the continuation:
        18.cxb5 axb5;  19.Nd4!?, "+/="  {Diagram?}
        when the first player seems to have a very clear advantage.
        (But Black had planned to sacrifice  the exchange to generate
         much-needed counterplay.)  ]   

18...dxc5!?;
Topalov said after the game that he felt this was forced.  

**********

     [  Black could have tried:    18...Qa7;  19.Nxd7 Qxd7;  20.cxb5,
        20...axb5;  21.Nxb5 d5;  22.Rec1,  '±'  
        but White is clearly better here, and Black's compensation for
        the pawn is highly questionable.   

***

          18...Be8?!;  19.Nxa6!,
        I think this is best. 

          ( 19.Nxe6!? fxe6; 20.Rxe6 Rf6!?; 21.Rae1!? Rxe6!?; 22.Rxe6 Bf7?!; (?) 
             23.Qe3!? Qc8!?; 24.Bxc6, '±'  (Maybe  "+/-")  {Diagram?}    
             - GM A. Karpov. 
             {Some of these moves are not clearly forced or best.} )      

        19...Rxa6;  {Diagram?}
         What choice does Black have?  

         20.cxb5 Rb6;  21.bxc6 Bxc6!; 22.Bxc6 Rxc6;  23.a4, '±'  
         White is better ... and Black has almost zero counterplay here.  ]   

19.Qxd7 Rc8;  {See the diagram just below.}
This looks to be forced.  

     [  One fan suggested a Rook move, but it does not work out:
         I.e. 19...Re8??;  20.Bxc6,  "+/-"
         with a decisive edge for White.  ]   

***

***

White has quite an obvious edge in this position. But I think it can be
CLEARLY argued that Karpov's play has been incredibly non-routine
and VERY inventive.  (!!!)  

While White is clearly better here, the final denouement looks to be
a VERY long way off.  

"Karpov now crowns some powerful positional play with a devastating
 sacrificial sequence."  - FM Graham Burgess.  

***

 20.Rxe6!!,   (Maybe - '!!!')  
An extremely brilliant and well thought out sacrifice that is far from
being obvious.
(Several Masters, that I showed this game to, did NOT seriously
 consider this sacrifice. NOTE: AFTER posting this game, I got
 like 7 e-mails, & they all said basically the same thing:
 "White wins a free piece on c6."  WRONG!!)  

While White has sacrificed on e6 before, the combination of
(other) elements here is probably unique. 

Seirawan  calls this both brilliant and charming. 

 (Every annotator worth his salt gives this move two exclams.)    

     [  White gains a fairly decent advantage after continuation:
        20.cxb5 axb5;  21.Bxc6 Ra7;  22.Qd3 Rxc6;  23.Nxb5, "+/=" 

 *** 

         One reader wrote in and asked: "Why not just grab the free piece 
         on c6 here?"  Well ...  (I offer just one line, based on the moves that  
         this particular chess fan suggested.)    

         20.Bxc6!? ('?!/?')  20...Ra7; ('!')  21.Qd3 Rxc6;  22.cxb5 c4!;  23.Qf3!?,
         23...Rc8;  24.bxa6!? Qxb2;  25.Rab1 Qa3;  "=/+"   26.Re3?!,  ('?')
         26...Bc5;  "/+" 
         and both  ChessMaster8000  and  Fritz  agree Black is clearly better.
         (The pawn on a6 is doomed, there is no rush to capture it.)  ]  

20...Ra7!; 
This is clearly is the best defense. 

'!' - Damsky and Burgess.   

     [    20...fxe6;  21.Bxc6 Ra7;  22.Qxe6+ Kg7;
        23.Be4, '±'  (Maybe  "+/-")  {Diagram?}
        White is clearly better.  

***

          20...Rc7?!;  21.Qd5 fxe6;  22.Qxe6+ Kg7;
        23.Bxc6, "±"  (Maybe  "+/-")  ]    

21.Rxg6+!,  (Maybe - '!!') 
This is clearly the best ... and is also clearly planned in advance by Karpov.  

 (desperado)   '!!' - GM John Emms. 

     [ 21.Qd5!?, "+/=" ]  

21...fxg6;  
This is forced.
(Other moves look like they will allow Black to be mated.)  

     [    21...Kh7?;  22.Qh3+ Kxg6;  23.Be4+, "+/-"  

        or    21...Kf8?;  22.Qh3,  "+/-"  ]  

22.Qe6+ Kg7!?;  23.Bxc6 Rd8;  24.cxb5 Bf6;
This could be forced as well.
 (Karpov gives it a "box" symbol.)   

     [  Two continuations that were MUCH worse than the game were:  

          24...Qd6?;  25.Qxd6 Bxd6;  26.b6, "+/-" 
        (White is winning.)  

        Or    24...Rd6?;  25.Qe5+ Bf6;  26.Qxc5, "+/-"
        and, again ... White is winning; and rather easily too.  

        And    24...axb5?;  25.Nxb5, "+/-" 
         - FM G. Burgess.  ]  

The smoke has cleared, and White has a material advantage.
 ---->  But Topalov is always dangerous!  

25.Ne4!,
This centralized Knight is worth its weight in gold.
(And is probably better than both of Black's Rooks put together!!) 

     [ 25.Qe3, '±' ]  

25...Bd4;
This looks like Black's only chance in this position.  

     [    25...Bxb2!?;  26.Rb1 Bd4;  27.b6 Rf7;  28.Ng5, "+/-" ]   

26.bxa6,
This is very good ... and of course, it wins.  

Damsky claims that pawn to f5 may have won more quickly, and
was a better continuation  ... but Karpov has something VERY
specific in mind here.  

     [  The move  26.f5!?, '±'  ("+/" maybe  "+/-")
         was/is {also} worth a look.  

***

        Karpov gives a long line, that begins with the move of:
        26.Qg4!?, "+/=" 
        and seems to lead to a win for White as well.  ]  

26...Qb6!?;  {Diagram?}
This is close to being forced for Black ...
 the capture on b2 was probably too risky.  

     [  Probably too dangerous - mainly because it gets the Queen
         out of play - was the variation:
         26...Qxb2!?; ('?!')  27.Re1, '±'
         (I think I have worked out a pretty much forced win, but it is
          very long and banal - - - so I will skip it here.)   

***

        Black could not allow Qe7+ ... for example:
        26...Rxa6?;  27.Qe7+ Kh8;  28.Ng5 Ra7;  29.Nf7+ Kg7;
        30.Qxd8 Qxb2;  31.Qh8+ Kxf7;  32.Bd5+ Ke7;  33.Re1+ Kd6;
        34.Qd8+ Rd7;  35.Re6+ Kxd5;  36.Qxd7+,  ("+/-") 
         ... "and White finally wins."  - GM John Emms.  ]   

27.Rd1 Qxa6;  {Diagram?}
Black defends as best he can.  

     [  Maybe just as bad was:
        27...Qxb2!?;  28.Bb7, '±'  (Maybe  "+/-")

        {And then again, maybe it is worse!!}  ]    

White's next move elevates this game into chess immortality.
28.Rxd4!!,
Super-brilliant, especially if you stop to consider there were much
simpler ways to proceed from this position.
(The third Rook offer in only nine moves of this game. White has
 also managed to sacrifice TWO!! exchanges, another rare motif
 for high-level, GM chess.)  

The dark-squared Bishop is the only glue that holds the tatters of
Black's position together, so White naturally removes it. 

***

<< As the ad men say, "It doesn't get any better than this!"  We've
     seen two Exchange sacrifices in the same game, and an
     interweaving of the initiative with incremental positional pluses.
     Classic Karpov. >>   -  GM Yasser Seirawan.
     (Who gave White's 28th move only one exclamation point.) 

     [  Also winning for White ... and much simpler to calculate
        over-the-board; was the continuation:
        28.Nxc5! Qb6;  29.Rxd4 Rxd4;  30.Qe5+ Kh7;  31.Qxd4 Qxc6;
        32.Ne4! Ra8;  33.a3, "+/-"  {Diagram?}
        and White has FOUR pawns; a good, centralized Knight;
        {for a Rook}; and a vastly superior (winning) game.  ]  

28...Rxd4;
This (also) appears forced.  

     [  28...cxd4;  29.Qf6+ Kh6;  30.Qh4+ Kg7;  31.Qxd8,
        31...Qxc6;  32.Qxd4+,  ("+/-")  ]   

29.Qf6+ Kg8; 
This was forced to avoid a quick mate. (Damsky) 

     [  Bad is:  29...Kh6?;  30.f5,  "+/-"  

         Or  29...Kh7?;  30.Ng5+,  "+/-"  ]  

30.Qxg6+ Kf8[];  31.Qe8+ Kg7;  32.Qe5+!, 
The best line ... among many different and attractive moves.  

     [  Also winning was:  32.Nxc5 Rd1+;  33.Kh2 Qf1;  34.Qe5+ Kh6;
        35.Qg5+ Kh7;  36.Be4+ Kh8;  37.Qe5+! Kg8;  38.Qb8+ Kg7;
        39.Qxa7+,  ("+/-")  {Diagram?}  and White wins.  

        (Line by - Yakov Damsky.)
        {White gives mate ... and ALL his moves are with check!  ...
         in like nine or ten moves.}  ]  

22...Kg8; 
This seems to be forced.
(Virtually all the annotators agree on this.)  

     [  Much worse was:    32...Kh7?;  33.Nf6+ Kg6[];
        34.Be8+ Kg7;  {Diagram?}  This has got to be forced. 

           (34...Kh6??;  35.Qg5#)   

        35.Nd7+ Kg8;  36.Qg5+,  ("+/-")
        and White will mate in like 10 - 12 moves from this position ...
        and win just about all of Black's pieces along the way. 

***

        Also bad was:    32...Kg6?;  33.Be8+! Kh7[];  (Box.)
        (This appears to be forced.) 

          ( Even worse was: 33...Kh6?; 34.Qg5+ Kh7; 35.Nf6+,  ("+/-")  {Diag?}  
            and Black loses his Queen ...  or gets mated. )   

        34.Nf6+ Kh8;  35.Nd7+ Kg8;  36.Qg5+,  "+/-"  {Diagram?}
        transposing to the above line.  ]  

33.Nf6+!?, ('!') 
This is/was (very) good, but there may have been other ways to win.  

  And while there may have been other ways to win, I find Karpov's  
  method of deciding the game to be the most attractive.     

     [  White is probably winning with:  33.Qg5+,  {Diagram?}  

        or    33.Bd5+, 

        or   33.Qe6+,  {Diagram?}  

        which all give White a decisive attack. ("+/-")  ] 

33...Kf7;  34.Be8+ Kf8;
Once again ... Black's choices are very limited.

     [  Slightly worse - if that was even possible! - was:
         34...Kg7?!; ('?')  35.Nd7+ Kg8;  36.Qg5+ Kh8;  37.Qh5+ Kg7;
         38.Qf7+ Kh6;  39.Qf8+ Kh7;  40.Nf6+ Qxf6;  41.Qxf6, "+/-"  {Diag?}
         and White is winning.  - GM John Emms.  ]  

35.Qxc5+ Qd6;  36.Qxa7!,
White has no fear here. 

     [ Also good was:  36.Qf5!?, '±'  ]   

36...Qxf6;
Maybe the only move.

      [  An interesting line was:  36...Rd1+!?;  37.Kg2 Rg1+;
         38.Kh3,  {Diagram?}  The correct move.  

          (38.Kxg1?? Qd1+; 39.Kg2 Qh1+!; 40.Kxh1, stalemate.)      

         38...Rh1+;  39.Kg4 Rh4+!?;  40.Kg5!,  "+/-"  {Diagram?}
         and White is winning.  

          ( White cannot play: 40.gxh4?! Qxf4+!!;  41.Kh3 Qg3+!; 
            42.fxg3, ("=")  {Diagram?}  and Black is STALEMATED. 
            (By the rules of chess, its only a draw.) )    ]  

You could (perhaps) give all of White's remaining moves an exclamation point.
37.Bh5 Rd2;  38.b3 Rb2;  39.Kg2, ("+/-")  {Diagram?}
Black Resigns.  (1-0) 

(There is nothing left for the second player to play for here.)

*******

A star of the very first magnitude.
(It is also a very sharp and scintillating game ... with as many sacrifices
 and difficult, complex positions as one could ask for.) 

This was picked as the game of the year by several different chess magazines.
It was also the top game - picked by a distinguished panel of judges -
for the Informant.

This is definitely one of Karpov's better games. The icing on the cake is that
it came during one of the very best tournament performances that he ...
(or anyone!!) ever had.
(He scored 11/13 ... vastly out-distancing a field that included all the best players,
 including some guy named Garry Kasparov! It was also the highest category
 tourney {18} ever held up to this point in history.)  

***

"A very beautiful and impressive victory by Anatoly. Bravo!"
   - GM Yasser Seirawan.  
http://www.lifemasteraj.com/old_af-dl/karp-topa_lin94.html