(1) Kramnik,V (2807) - Deep Fritz (2741) [D27]
Man vs Machine Manama, Bahrain (2), 06.10.2002
[Karsten Mueller]

Vladimir Kramnik has found a good strategy with the white pieces against Deep Fritz. It is one he also has used in his regular games with good success. Once again in game two he exchanged queens quickly and set up a closed position with just two open files, both of which were easy to control. This minimised the danger of tactical surprises by the computer and retains the chance of winning the endgame with his fantastic technique. Bravo Vladimir!

1.d4 d5 2.c4 dxc4 3.Nf3 Nf6 4.e3 e6 5.Bxc4 c5 6.0-0 a6 7.dxc5 Qxd1 8.Rxd1
Looks unambitious, but Kramnik know how to keep increasing the +/= in such positions.

8...Bxc5 9.Kf1!?
Did Kramnik want to throw Fritz out of its openings book? [ Here are some of the games Kramnik has played against his opponents: 9.Nbd2 b6 10.Be2 Nbd7 11.Nc4 ( 11.Nb3 Be7 12.Nfd4 Bb7 13.f3 0-0 14.e4 Rfc8 15.Be3 Kf8 16.Nd2 Ne5 17.N4b3 Rc6 18.Rac1 Rac8 19.Rxc6 Rxc6 20.g4 h6 21.h4 Bc8 22.g5 hxg5 23.hxg5 Nfd7 24.f4 Ng6 25.Nf3 Rc2 26.Bxa6 Bxa6 27.Rxd7 Rxb2 28.Ra7 Bb5 29.f5 exf5 30.exf5 Re2 31.Nfd4 Re1+ 32.Kf2 Rf1+ 33.Kg2 Nh4+ 34.Kh3 Rh1+ 35.Kg4 Be8 36.Bf2 Ng2 37.Ra8 Rf1 38.Kf3 Nh4+ 39.Ke2 Rh1 40.Nb5 Bxg5 41.Nc7 Ke7 42.Nxe8 Nxf5 43.Bxb6 Kd7 44.a4 Rh3 45.Nc5+ Kc6 46.a5 Re3+ 47.Kd1 Re7 48.Rc8+ Kb5 49.Ne4 Rxe4 50.Rc5+ Ka6 51.Nc7+ Kb7 52.Rxf5 Be3 53.Bxe3 Rxe3 54.Rxf7 Re5 55.a6+ Kb6 56.Rxg7 Ra5 57.Kd2 Ra1 58.Kc2 Rh1 59.Kb2 Rh8 60.Kb3 Rc8 61.a7 Kxa7 62.Kb4 Kb6 63.Nd5+ Ka6 64.Rg6+ Kb7 65.Kb5 Rc1 66.Rg2 Kc8 67.Rg7 Kd8 68.Nf6 Rc7 69.Rg5 Rf7 70.Nd5 Kd7 71.Rg6 Rf1 72.Kc5 Rc1+ 73.Kd4 Rd1+ 74.Ke5 1/2-1/2 Kramnik,V-Kasparov,G/London ENG 2000/The Week in Chess 310 (74)) 11...Bb7 12.b3 Ke7?! ( 12...0-0 13.Bb2 Bd5 14.Rac1 Rfc8 15.Nfe5 b5 16.Nxd7 Nxd7 17.Nd2 Bb4 18.Nb1 Nb6 19.f3 Rxc1 20.Rxc1 Rc8 21.Rxc8+ Nxc8 22.Ba3 Bxa3 23.Nxa3 b4 24.Nc2 a5 25.Ba6 Nb6 26.e4 Bc6 27.Kf2 1/2-1/2 Kramnik,V-Lautier,J/Monte Carlo 1997/EXP 57 (27)) 13.Bb2 Rhd8 14.Ne1 b5 15.Na5 Be4 16.Bf3 Bxf3 17.gxf3 Rdc8 18.Nd3 Rab8 19.Nxc5 Rxc5 20.Rac1 Ke8 21.Bd4 Rxc1 22.Rxc1 Ra8 23.Kf1 e5 24.Bb2 Nd5 25.Rd1 Ne7 26.Nb7 Nc8 27.Rc1 f6 28.Ba3 a5 29.Rc7 b4 30.Bc1 Na7 31.Nd6+ Kd8 32.Rb7 Nc8 33.Nc4 a4 34.Bd2 axb3 35.axb3 Ra7 36.Rxa7 Nxa7 37.Bxb4 Kc7 38.Ke2 Kc6 39.Kd3 g6 40.Be7 Nc8 41.Bd8 Kb5 42.Kc3 Na7 43.Nd6+ Kc6 44.Ne4 f5 45.Ng5 h5 46.e4 Nb5+ 47.Kc4 Nd4 48.b4 Nb5 49.Be7 Nd4 50.f4 Nb5 51.exf5 gxf5 52.Nf7 exf4 1-0 Kramnik,V-Karpov,A/Frankfurt 1999/CBM 72/ (52)]

9...b5 10.Be2 Bb7 11.Nbd2 Nbd7 12.Nb3 Bf8?!
This move is not exactly convincing. In any case a human being would hardly have played it.

13.a4!
A typical strategy to fight for squares on the queenside and to create weaknesses in the opponent's position.

13...b4 14.Nfd2
Kramnik mobilises his knight to occupy the square c4, which he has just secured. From here it can keep a dangerous watch on the black position.

14...Bd5 15.f3 Bd6 16.g3 e5 17.e4 Be6 18.Nc4 Bc7 19.Be3 a5
This kind of move is never pleasant, because the square b5 and the pawn on a5 become very weak. But there was no real alternative. White has achieved a nicely dominating position, the kind that is very comfortable against a computer because Black has hardly any counterplay. Black would like to complete his development by castling and then mobilise his kingside rook. It is for this reason that Kramnik executes the following manoeuvre. He is determined to keep the initiative.

20.Nc5
[ 20.Rac1!? ]

20...Nxc5 21.Bxc5 Nd7 22.Nd6+ Kf8
[ 22...Ke7? 23.Nf5+ Kf6 24.Be7+ Kg6 25.Rac1+/- ]

23.Bf2 Bxd6
Deep Fritz gives up the pair of bishops in order to release some of the pressure. But Kramnik gets a permanent advantage. [ 23...Ke7 24.Nf5++/= ]

24.Rxd6 Ke7 25.Rad1 Rhc8 26.Bb5 Nc5
looks like a tactical blunder. But we must remember that Black is a computer that will never overlook something as shallow as this.

27.Bc6 Bc4+!
[Kramnik confessed later that he had completely overlooked this move and was horrified because he though he was lost. "But after a few seconds I saw a way to force the draw with perpetual check, and so I calmed down and started to look for a way to win." He found it in a very subtle rook endgame.] [ 27...Ra6?? 28.Bxc5 Raxc6 29.Rxc6++- with check; 27...Rxc6? 28.Rxc6 Nxa4 29.Rd2+/- ]

28.Ke1 Nd3+ 29.R1xd3 Bxd3 30.Bc5!
[ 30.Rxd3?? Rxc6-+ ]

30...Bc4
[ 30...Bc2 31.Kd2 Bxa4?? is refuted by 32.Rd5+ Kf6 33.Bxa4+- ]

31.Rd4+ Kf6 32.Rxc4 Rxc6 33.Be7+ Kxe7 34.Rxc6
The resulting rook endgame is better for White because the black rook is tied to the passive defence of the a-pawn. One of the main principles of rook endings is activity. The rook is useless if it is engaged in passive defensive tasks.

34...Kd7
[ 34...Rd8 35.Ke2 doesn't help]

35.Rc5 f6?
After this Black is condemned to passivity. Maxim Notkin proved in Chess Today No. 699 that [ 35...Rc8 is better because after 36.Rxa5 Rc2 36.Rxc8 is a draw. The point, however, is so deep that even Deep Fritz would have difficulty finding it. This is the main line continuation of Notkin's analysis: 36.Rxc8 Kxc8 37.Kd2 Kd7 38.Kd3 Kd6 39.Kc4 Kc6 Black has the opposition and White cannot win the tempo battle: 40.h4 g6 41.g4 h5 42.g5 Kd6 43.Kb5 f6 44.gxf6 g5 45.hxg5 Ke6 46.Kxa5 h4= and the final queen endgame is drawn.]

36.Kd2 Kd6 37.Rd5+ Kc6 38.Kd3 g6?!
In the next phase Deep Fritz only weakens its own position. [ 38...Ra7 seems to me to be tougher.]

39.Kc4 g5 40.h3 h6 41.h4 gxh4 42.gxh4 Ra7 43.h5
Kramnik has all the time in the world to improve his position before resorting to Rd8. In such positions it is wise to restrain oneself and first maximise the strength of one's position.

43...Ra8
Now Kramnik embarks on a manoeuvre in order to force Fritz to move. He could achieve this with b3, but that gives up a possible waiting move later on. Kramnik really plays this endgame superbly. Rubenstein or Capablanca would have loved to watch it!

44.Rc5+
[ 44.Kd3 with the plan Ke3 and f4 was also a possibility.]

44...Kb6 45.Rb5+
[ Naturally not 45.Kd5?? Rd8+ 46.Kc4 Rd4+-+ ]

45...Kc6 46.Rd5
Zugzwang

46...Kc7?!
[ 46...Ra7 47.Rd8 Rc7 48.Rf8 Kd6+ sollte zwar auch verloren sein, ist aber aktiver und daher unangenehmer. should also be lost for Black, but it is more active and less pleasant for Kramnik.]

47.Kb5 b3 48.Rd3 Ra7 49.Rxb3 Rb7+ 50.Kc4 Ra7
[ The pawn ending after 50...Rxb3 51.Kxb3 Kb6 52.Kc4 Kc6 53.b4+- is lost because of the distant white passed pawn]

51.Rb5 Ra8 52.Kd5 Ra6 53.Rc5+ Kd7 54.b3 Rd6+
[ 54...Ra7 55.Rb5 Ke7 56.Kc6 Ra8 57.Kb7 Rd8 58.Rxa5+- ]

55.Kc4 Rd4+ 56.Kc3 Rd1 57.Rd5+
[ 57.Rd5+ the transition into the pawn ending is the simplest way to victory, for example 57...Rxd5 58.exd5 Kd6 59.b4 axb4+ 60.Kxb4 Kxd5 61.Kb5 Kd6 ( 61...f5 62.a5 e4 63.fxe4+ fxe4 64.a6 e3 65.a7 e2 66.a8Q++- ) 62.a5 f5 63.a6 Kc7 64.Kc5 e4 65.fxe4 fxe4 66.Kd4 Kb6 67.Kxe4 Kxa6 68.Kf5 Kb6 69.Kg6 Kc7 70.Kxh6 Kd7 71.Kg7+- Thus far Kramnik has succeeded in keeping the match in calm waters. We will have to wait and see if Fritz will be able to stir up a storm in its next white game.] 1-0