Let "Nalimov" Speak

Время публикации: 01.06.2012 02:34 | Последнее обновление: 01.06.2012 09:23

(СN translation)
There was long-suffering in the third game of yesterday's World Championship tie-break, which Boris Gelfand could have won in the middlegame, but then lost his advantage. Viswanathan Anand started again getting into difficulties after about move 40 and then went into the rook endgame with his move 48, which turned out to be complex.

The Indian should have lost, having made further errors, but the Israeli grandmaster was the last to make a mistake.

Only a general outline of the events (as well as pointing out one of Anand's main misses 51...Kf5?was covered by our site's news and Twitter comments. Unfortunately, the moves of the final part of the game didn't appear on the official site of the match for dozens of minutes, while "The Week in Chess" site actually fell over due to a large number of visitors. And then the events of the third game were quickly superseded by the events of the decisive fourth game.

Let's briefly fill the gap with the aid of the Nalimov tablebase that knows all about six-piece endgames like this.

(Game 3 of the tie-break)

It makes sense to mention that in the game 9 of the PCA World Championship Match Kasparov-Short (1993), both opponents made gross mistakes in turn on move 46, in an identical endgame according to the material: Kasparov missed the win and Short immediately missed the draw with his next move. More persisting researchers can give additional historical analogies; if not from the sporting point of view, then relative to the position from the game Gelfand-Anand and its variations. There's a handful of studies.
50...Rc3! 51.Rc7

This variation: 51...Rc2+ 52.Kg3 Rc3+ 53.Kg2 Kf4!, just as this one: 51...Kf4 52.Rc8 Rc2+! are enough for survival at the end, but the relatively more direct way is (the variation, which was possible to calculate quite quickly with some luck): 51...Kh4! 52.c6 (after 52.Kg2 Black can implement the main threat 52...Rxc5!!, and taking of the rook leads to the stalemate) 52...Rc2+ 53.Kg1 Kxh3 (53...Kg3 leads to the more complicated draw) 54.Rc8 Kg3 55.Kf1 Kf3 56.Ke1, and now there's the sole, but very important, nuance: only 56...Re2+! 57.Kd1 Re7! saves, and after 58.c7 Ke4, with the following Ke5, Black takes the pawn.

It was correct to start with 52.Rc8! and after 52...Rc2+ to play only 53.Kg3 Rc3+ 54.Kh4!! with the win, although I, personally, can't call the further variations simple. Without the Nalimov tablebase, the program Stockfish starts to increase assessment of the position in White's favour only at 32-ply, but not at 30-ply.
And now 52...Rc2+! was saving. The king's walk to h4 doesn't win any more, since Rc8 turns out to be more useful in that position than c6. (So, what's with the rule about moving passed pawns in an endgame!).
53.Rc8! was also winning.
53...Kd6 54.Rc8 Ra3 55.Kg2 Re3 56.Kh2 Ra3 57.Kg2 Re3 58.h5 Re5 59.h6 Rh5 60.Rh8 Kxc6

This is an unbelievable mistake by Boris. Either 61.Kg3 or 61.Kf3 led towards the aim, as even did the slightly absurd 61.Kf2?! (because in case of 61...Kd6 the winning move is 62.h7).
For Black to be able to consolidate his defence and to draw, the king actually had to go to the first rank: 61.Kg1?? (or to f1) 61...Kb7 62.Kg2 Rg5+!, and so on. With his move, Gelfand also improved his opponent's position: Anand is now in time to just walk up to the pawn with the king.
61...Kd6 62.Kg3 Ke6 63.Kg4 Rh1 1/2-1/2

[Event "WCh Rapid Tiebreak"] [Site "Moscow RUS"] [Date "2012.05.30"] [Round "3"] [White "Gelfand, Boris"] [Black "Anand, Viswanathan"] [Result "1/2-1/2"] [ECO "D12"] [WhiteElo "2739"] [BlackElo "2799"] [PlyCount "126"] [EventDate "2012.05.11"] 1. d4 d5 2. c4 c6 3. Nf3 Nf6 4. e3 Bf5 5. Nc3 e6 6. Nh4 Bg6 7. Nxg6 hxg6 8. Bd3 Nbd7 9. O-O Bd6 10. h3 O-O 11. Qc2 Qe7 12. Rd1 Rac8 13. c5 Bb8 14. f4 Ne8 15. b4 g5 16. Rb1 f5 17. b5 gxf4 18. exf4 Nef6 19. bxc6 bxc6 20. Ba6 Rc7 21. Be3 Ne4 22. Rb2 g5 23. Rdb1 gxf4 24. Bxf4 e5 25. Bxe5 Nxe5 26. Rxb8 Ng6 27. Nxe4 fxe4 28. Qf2 Qg7 29. Kh2 Rcf7 30. Qg3 Nf4 31. R8b3 Qxg3+ 32. Rxg3+ Kh7 33. Rd1 Ne6 34. Be2 Rf2 35. Bg4 Nf4 36. Rb1 Rf7 37. Rb8 Rxa2 38. Rc8 e3 39. Rxe3 Rxg2+ 40. Kh1 Rd2 41. Rxc6 Ne6 42. Rf3 Rxf3 43. Bxf3 Nxd4 44. Rc7+ Kh6 45. Bxd5 Rc2 46. Be4 Rc3 47. Kh2 Kg5 48. Rd7 Nf3+ 49. Bxf3 Rxf3 50. Rxa7 Rc3 51. Rc7 Kf5 52. c6 Ke6 53. h4 Kd6 54. Rc8 Ra3 55. Kg2 Re3 56. Kh2 Ra3 57. Kg2 Re3 58. h5 Re5 59. h6 Rh5 60. Rh8 Kxc6 61. Rh7 Kd6 62. Kg3 Ke6 63. Kg4 Rh1 1/2-1/2 [Event "PCA-World Championship"] [Site "London"] [Date "1993.09.25"] [Round "9"] [White "Kasparov, Garry"] [Black "Short, Nigel D"] [Result "1-0"] [ECO "E35"] [WhiteElo "2805"] [BlackElo "2655"] [PlyCount "103"] [EventDate "1993.09.??"] [EventType "match"] [EventRounds "20"] [EventCountry "ENG"] [Source "ChessBase"] [SourceDate "1993.10.01"] 1. d4 Nf6 2. c4 e6 3. Nc3 Bb4 4. Qc2 d5 5. cxd5 exd5 6. Bg5 h6 7. Bh4 c5 8. dxc5 g5 9. Bg3 Ne4 10. e3 Qa5 11. Nge2 Bf5 12. Be5 O-O 13. Nd4 Bg6 14. Nb3 Nxc3 15. Bxc3 Bxc2 16. Nxa5 Bxc3+ 17. bxc3 b6 18. Kd2 bxa5 19. Kxc2 Rc8 20. h4 Nd7 21. hxg5 Nxc5 22. gxh6 Ne4 23. c4 Nxf2 24. Rh4 f5 25. Rd4 dxc4 26. Bxc4+ Kh7 27. Rf1 Ng4 28. Kd2 Rab8 29. Rxf5 Rb2+ 30. Kd3 Rxg2 31. Be6 Rc7 32. Rxa5 Nf2+ 33. Ke2 Rh2 34. Kf3 Nh1 35. Rd7+ Rxd7 36. Bxd7 Kxh6 37. Rxa7 Kg5 38. Ra5+ Kf6 39. Bc6 Rc2 40. Rf5+ Ke7 41. Bd5 Kd6 42. Rh5 Rd2 43. Rxh1 Rxd5 44. a4 Ra5 45. Ra1 Ke5 46. e4 Ke6 47. Ke3 Kd6 48. Kd4 Kd7 49. Kc4 Kc6 50. Kb4 Re5 51. Rc1+ Kb6 52. Rc4 1-0 


Смотрите также...