(3106) Anand,Viswanathan (2790) - Kramnik,Vladimir (2788) [C42]
Corus A (12), 30.01.2010

1.e4 e5 2.Nf3 Nf6 3.Nxe5 d6 4.Nf3 Nxe4 5.d4 d5 6.Bd3 Nc6 7.0-0 Be7 8.c4 Nb4 9.Be2 0-0 10.Nc3 Bf5 11.a3 Nxc3 12.bxc3 Nc6 13.Re1 Re8 14.cxd5 Qxd5 15.Bf4 Rac8 16.h3 Be4 17.Qc1N
A new move in a very well-known position.

A natural rejoinder, threatening the fork but also to park the knight on the great c4 square. Now 18.Nd2 looks obvious, when the obvious rejoinder 18...Bxg2 leads to very sharp play that seems to favor White. Anand found a different idea.

[18.Nd2 Bxg2 (The solid 18...Bd6 looks better. 19.Bxd6 cxd6 20.Nxe4 Rxe4 21.Bf3 Rxe1+ 22.Qxe1 Qb3 and White doesn't seem to have anything. So perhaps this was why Anand chose 18.Qe3.) 19.Bg4 f5 (19...Bf6!? Calling White's bluff. 20.Bxc8 (20.Rb1 ; 20.f3 Ra8 21.Kxg2 h5 and now it's White's bishop that gets trapped. 22.Bxc7 hxg4 23.hxg4+/= ) 20...Rxc8 21.Re3 c5 again Black has some compensation, but it's not obviously enough.) 20.Re5 fxg4 21.Rxd5 Bxd5 22.hxg4 c5 obviously favors White. Black might have enough activity to draw, but it's not going to be quick or easy.]

[18...Bf6?? is a blunder that shows the point behind Kramnik's move. 19.Bd3 and the bishop can't be taken because 20.Qxe8+ forces mate next move.; 18...Bd6 is possible though, as is the immediate retreat; 18...Bf5 , with a likely transposition with 18...Bd6 after 19.Ne5 Bd6 We might see this position get tested in the future, when one possible continuation is (19...c5 looks possible too.) 20.Qg3 f6 21.Ng4 Bg6 22.Ne3 Bxf4 23.Qxf4 Qd6 with a position that still seems on the traditional border between White's being slightly better and Black equalizing.]

19.c4 Qd8 20.Ne5 Bf5 21.Qc3 b6
White is looking very comfortable here.

[22.c5 may improve, so that on 22...f6 White has 23.Nc4 , with a superior position following 23...Nxc4 24.Bxc4+ Kh8 25.Ba6 Ra8 26.Rxe8 Qxe8 27.c6+/- (27.Bxc7?! Qc6 ) ]

[22...f6+/= isn't appealing, but it might be best anyway.]

[23.c5+/- Bxh3?! 24.Qg3 Bf5 25.Bg5 Qe6 26.Bh5 Bg6 27.Bg4+/- Bf5? 28.d5+- ]

Very natural, trying to bring the knight back with the rest of the army, though it was performing a function hitting the c-pawn.

24.Ng4 Qg6 25.d5
[25.Bxc7 h5 (25...Bxg4 26.Bxg4 Rxe1+ 27.Rxe1 f5 28.Bf3 Nxd4 29.Bd5++/- White's strong Bd5 gives him a serious edge.) 26.Ne3 Rxe3 27.fxe3 Rxc7 28.Qxc7 Bd6 Was a line Anand mentioned in the post-game press conference (at least the snippet I caught), but he was concerned that Black's counterplay would be serious. In fact, Black is losing after 29.Bxh5! (29.Qxc6 Bh2+ 30.Kxh2 Qxc6 was given by Anand as a possibility, and he felt, rightly, that White is better here. It's true, but he's not yet winning or even close to winning.) 29...Bxc7 30.Bxg6 Bxg6 as White has too many pawns for them to all be safely blockaded.]

[25...Bc2 26.Rd2 (26.Rc1 Nd4= (26...Na5? 27.Bxc7+/- is the game.) ) 26...Na5 27.Rxc2 (27.Bxc7 Nxc4= ) 27...Qxc2 28.Nh6+ Kh8 29.Nxf7+ Kg8 is nowhere near as favorable for White as the game version - that's because Black still has his c-pawn and there's no passed d-pawn to worry about.]

26.Bxc7 Bc2?!
[26...Bc5+/= keeps the game alive.]

27.Rc1 Nb3? 28.Rxc2+-
White is winning now, thanks in large part to his passed d-pawn.

28...Qxc2 29.Nh6+ Kh8 30.Nxf7+ Kg8 31.Nh6+ Kh8 32.Nf7+
Just repeating moves to gain time on the clock.

32...Kg8 33.Nh6+
This is not a threefold repetition of the position, as some online kibitzers and possibly even Kramnik thought for a moment, because the first time around Black still had a pawn on f7.

33...Kh8 34.Be5! Qg6
[34...Qd2 35.Nf7+ Kg8 36.Bc3+- holds everything.]

[35.Qxg6 hxg6 36.d6 Rxe5 (36...Nc5 37.Bg4 ) 37.Nf7+ Kg8 38.Nxe5 Bxd6 39.Nxg6 should be winning, but there's still some work, as White's queenside pawns are weak, there's no kingside passer and won't be for some time, and there are opposite-colored bishops.; 35.Qf4 is another possibility, going for a middlegame decision rather than a technical ending.]

[35...Rcd8 was the last chance to fight. White is much better after 36.Qxb3 (36.Nf5 might be even better.) 36...Qxh6 37.Be6 , but it's not yet trivial.]

36.Qxb3 Rxe5 37.Rxe5 Rc1+ 38.Kh2 Bd6 39.f4
[39.Qe3 does the job too.]

39...Bxe5 40.fxe5 gxh6
[40...Qxh6 41.Qf3 is crushing; but 40...Qe4 offers a moment's excitement before grim reality sets back in. 41.Nf7+ Kg8 42.e6 Qf4+ 43.Qg3 Rh1+ There's the excitement... 44.Kxh1 Qxg3 45.d6 ...and here comes the grim reality. Black has nothing in his power to stop the d-pawn's advance, and fishing for perpetual doesn't work either: 45...Qxa3 46.d7 Qc1+ 47.Kh2 Qf4+ 48.Kg1 Qe3+ 49.Kf1 Qf4+ 50.Bf3 Qc4+ 51.Kf2 Qh4+ 52.Kg1 Qd4+ 53.Kh2 Qf4+ 54.g3 Qd2+ 55.Bg2+- and done.]

That'll do it. Recentralizing the queen takes f4 under control, so there won't be any dreams of a perpetual (or more), hits the rook and supports the advance of the e-pawn. Black makes a few more moves, but in vain.

41...Qb1 42.d6 Rh1+ 43.Kg3 Re1 44.Qf4 Rf1 45.Bf3
The "attack" is over, and now White can savor the position and all his threats: Qf8#, d7, e6, Qf6+ followed by Bd5+, etc. 1-0