What Is The Greatest Human Chess Game Ever Played?

[Robert Flesher]

For the fireworks thing, I very much like Kasparov-Topalov 1999 Wijk aan Zee. That rook sacrifice is fantastic to play at that level (24 Rxd4).

Here is the game:

[Event "Hoogovens A Tournament (cat. 18)"]
[Site "Wijk aan Zee NED"]
[Date "1999.01.20"]
[Round "04"]
[White "Kasparov, Garry (RUS)"]
[Black "Topalov, Veselin (BUL)"]
[Result "1-0"]
[WhiteElo "2812"]
[BlackElo "2700"]
[Opening "Pirc Defence"]
[Variation "Ufimtsev-Pytel"]
[ECO "B07"]
[PlyCount "87"]

1.e4 d6 2.d4 Nf6 3.Nc3 g6 4.Be3 Bg7 5.Qd2 c6
6.f3 b5 7.Nge2 Nbd7 8.Bh6 Bxh6 9.Qxh6 Bb7
10.a3 e5 11.0-0-0 Qe7 12.Kb1 a6 13.Nc1 0-0-0
14.Nb3 exd4! 15.Rxd4 c5 16.Rd1 Nb6! 17.g3 Kb8
18.Na5 Ba8 19.Bh3 d5 20.Qf4+ Ka7 21.Rhe1 d4
22.Nd5 Nbxd5 23.exd5 Qd6 24.Rxd4!! cxd4?!
25.Re7+! Kb6 26.Qxd4+ Kxa5 27.b4+ Ka4 28.Qc3
Qxd5 29.Ra7 Bb7 30.Rxb7 Qc4 31.Qxf6 Kxa3?
32.Qxa6+ Kxb4 33.c3+! Kxc3 34.Qa1+ Kd2
35.Qb2+ Kd1 36.Bf1! Rd2 37.Rd7! Rxd7! 38.Bxc4
bxc4 39.Qxh8 Rd3 40.Qa8 c3 41.Qa4+ Ke1 42.f4 f5
43.Kc1 Rd2 44.Qa7
1-0

Saw this game live, after looking at the game the next day. We found that Rxd4 does not work. If Topalov would have ignored the rook sac, the game would have been long forgotten. It took Topalovs blunder after Rxd4 to make Rxd4 work, so I do not consider this game one of the best ever.

I remember some people using the rook sac as a computer test position, but no computer program would play the rook sac....because even in those day the chess programs saw the sac was not winning.

I remember doing analysis and watching the game live. I had at least two engine back then that would play the sacrifice. I am not sure that the sack does not work, but I have not analyzed it in a long time. I am pretty sure me and a friend concluded it did not force a win, but was sound.
In the interest of time, Houdini 1.5 likes Rxd4.


FEN: b2r3r/k4p1p/p2q1np1/NppP4/3p1Q2/P4PPB/1PP4P/1K1RR3 w - - 1 24

Houdini_15_x64:
4/15 00:00 2,604 0 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Rd6xc6 Re1e7+ Ka7b6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Re8e1+ Kb1a2 Rc6e6 Bg2h3 Re6d6 Bh3c8 Nf6d5 Bc8b7 Re1e6 f3f4 Rd6d8 Bb7xd5 Rd8xd5 b2b4 a6a5 Ka2b2 c5xb4 a3xb4 a5xb4
5/33 00:00 3,975 0 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Re8e1+ Kb1a2 Rc6e6 Bg2h3 Re6d6 Bh3c8 Nf6d5 Bc8b7 Re1e6 f3f4 Rd6d8 Bb7xd5 Rd8xd5 b2b4 a6a5 Ka2b2 c5xb4 a3xb4 a5xb4
6/33 00:00 5,724 0 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Rd6xc6 Re1e7+ Ka7b6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Re8e1+ Kb1a2 Rc6e6 Bg2h3 Re6d6 Bh3c8 Nf6d5 Bc8b7 Re1e6 f3f4 Rd6d8 Bb7xd5 Rd8xd5 b2b4 a6a5 Ka2b2 c5xb4 a3xb4 a5xb4
7/33+ 00:00 7,534 0 -0.17 Na5c6+
7/33 00:00 8,466 0 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Rd6xc6 Re1e7+ Ka7b6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Re8e1+ Kb1a2 Rc6e6 Bg2h3 Re6d6 Bh3c8 Nf6d5 Bc8b7 Re1e6 f3f4 Rd6d8 Bb7xd5 Rd8xd5 b2b4 a6a5 Ka2b2 c5xb4 a3xb4 a5xb4
8/33 00:00 12,012 0 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Rd6xc6 Re1e7+ Ka7b6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Re8e1+ Kb1a2 Rc6e6 Bg2h3 Re6d6 Bh3c8 Nf6d5 Bc8b7 Re1e6 f3f4 Rd6d8 Bb7xd5 Rd8xd5 b2b4 a6a5 Ka2b2 c5xb4 a3xb4 a5xb4
9/33 00:00 20,069 557,000 -0.21 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 c6c7 Kb6xc7 Re1e7+ Kc7b6 Re7xf7 Nf6d5 f3f4 h7h6 Rd1e1 Nd5e3
10/33 00:00 28,182 640,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Re8e1+ Kb1a2 Rc6e6 Bg2h3 Re6d6 Bh3c8 Nf6d5 Bc8b7 Re1e6 f3f4 Rd6d8 Bb7xd5 Rd8xd5 b2b4 a6a5 Ka2b2 c5xb4 a3xb4 a5xb4
10/33 00:00 47,790 955,000 -0.23 Rd1xd4 Ka7b6 b2b4 Qd6xf4 Rd4xf4 Nf6xd5 Rf4xf7 c5xb4 a3xb4 Nd5xb4 Na5b3 Ba8d5 Rf7f6+ Nb4c6 f3f4 Rh8f8 Rf6xf8 Rd8xf8 Bh3d7 Rf8d8 Bd7xc6 Bd5xc6 Re1e7 a6a5 Re7xh7 a5a4 Nb3c1 Rd8d2 Nc1d3 Bc6e4 Kb1c1 Rd2e2
11/33 00:00 75,562 956,000 -0.28 Rd1xd4 Ka7b6 b2b4 Qd6xf4 Rd4xf4 Nf6xd5 Rf4xf7 c5xb4 a3xb4 Nd5xb4 Na5b3 Ba8d5 Rf7f6+ Nb4c6 f3f4 a6a5 Nb3d4 Rh8f8 Rf6xf8 Rd8xf8 Nd4xc6 Bd5xc6 Re1e7 h7h5 Re7e6 Rf8g8 Kb1b2 a5a4 Re6d6 Kb6c5
11/33 00:00 80,150 1,001,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 Rf7f4 g6g5 Rf4f5 d4d3+ Rd2xd3 Re1e2+ Kc2d1
12/33 00:00 104,712 1,203,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 Rf7f4 g6g5 Rf4f5 d4d3+ Rd2xd3 Re1e2+ Kc2d1
13/33 00:00 163,981 1,012,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 Rf7f4 g6g5 Rf4f5 d4d3+ Rd2xd3 Re1e2+ Kc2d1
14/33 00:00 272,890 1,421,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 Rf7f4 g6g5 Rf4f5 d4d3+ Rd2xd3 Re1e2+ Kc2d1
15/33 00:00 458,011 1,982,000 -0.21 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Rd6xc6 Re1e7+ Ka7b6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 f3f4 Re1e3 c3xd4 Rd8xd4 Rd2xd4 c5xd4
16/33 00:00 804,437 2,812,000 -0.21 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 f3f4 Re1e3 c3xd4 Rd8xd4 Rd2xd4 c5xd4
17/33 00:00 1,452,674 3,853,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 Rf7f4 g6g5 Rf4f5 d4d3+ Rd2xd3 Re1e2+ Kc2d1
18/35 00:00 2,746,309 5,011,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 Rf7f4 g6g5 Rf4f5 d4d3+ Rd2xd3 Re1e2+ Kc2d1
19/38 00:00 4,908,460 6,022,000 -0.22 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 f3f4 Nf6g4 Rf7b7+ Kb6a5 b2b4+ c5xb4 a3xb4+ Ka5xb4 Rd2xd4+ Kb4a5 h2h3 Ng4e3 Bg2f3 Ne3c4
20/40 00:01 10,344,418 7,134,000 -0.18 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Kb6xc6 Re7xf7 Rh8e8 Kb1c1 h7h5 f3f4 Kc6b6 Bh3g2 Re8e2 Rd1d2 Re2e1+ Rd2d1 Re1xd1+ Kc1xd1 Nf6g4 Kd1d2 Ng4e3 Rf7b7+ Kb6a5 Bg2h3 Ne3c4+ Kd2c1 Rd6c6 Bh3d7 Rc6f6 Bd7h3 Nc4e3 Kc1d2 Rf6d6 b2b3 Rd6d8
21/42 00:02 20,698,104 7,976,000 -0.19 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Kb6xc6 Re7xf7 Rh8e8 Kb1c1 h7h5 f3f4 Kc6b6 Bh3g2 Re8e2 Rd1d2 Re2e1+ Rd2d1 Re1xd1+ Kc1xd1 Nf6g4 Kd1d2 Ng4e3 Rf7b7+ Kb6a5 Bg2h3 Ne3c4+ Kd2c1 Rd6c6 Bh3d7 Rc6f6 Bd7h3 Nc4e3 Kc1d2 Rf6d6 b2b3 Rd6d8
22/45 00:05 42,379,863 8,225,000 -0.21 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Kb6xc6 Re7xf7 Rh8e8 Kb1c1 h7h5 f3f4 Kc6b6 Bh3g2 Re8e2 Rd1d2 Re2e1+ Rd2d1 Re1xd1+ Kc1xd1 Nf6g4 Kd1d2 Ng4e3 Rf7b7+ Kb6a5 Bg2h3 Ne3c4+ Kd2c1 Rd6c6 Bh3d7 Rc6f6 Bd7h3 Nc4e3 Kc1d2 Rf6d6 b2b3 Rd6d8
23/54 00:16 132,706,888 8,215,000 -0.18 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Kb6xc6 Re7xf7 Rh8e8 Kb1c1 h7h5 f3f4 Kc6b6 Bh3g2 Re8e2 Rd1d2 Re2e1+ Rd2d1 Re1xd1+ Kc1xd1 Nf6g4 Kd1d2 Ng4e3 Rf7b7+ Kb6a5 Bg2h3 Ne3c4+ Kd2c1 Rd6c6 Bh3d7 Rc6f6 Bd7h3 Nc4e3 Kc1d2 Ne3c4+ Kd2c1 Nc4e3
24/54- 00:29 256,124,787 8,548,000 -0.25 Na5c6+ Ba8xc6
24/61 01:11 629,868,144 8,862,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Kb6xc6 f3f4 Rh8f8 Bh3g2+ Nf6d5 c2c3 d4xc3 b2xc3 c5c4 Kb1b2 Kc6c5 Bg2xd5 Rd6xd5 Rd1xd5+ Kc5xd5 a3a4 Kd5c6 Kb2a3 Kc6b6 Ka3b4 a6a5+ Kb4a3 b5xa4 Re7d7 Kb6c6 Rd7e7 Kc6b5 Re7e5+ Kb5b6 Ka3xa4 Rf8d8 Re5b5+
25/61- 01:18 700,574,044 8,885,000 -0.29 Na5c6+ Ba8xc6
25/68 02:47 1,534,780,264 9,180,000 -0.37 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Kb6xc6 f3f4 Kc6b6 Re7xf7 Rh8e8 Kb1c1 c5c4 f4f5 Re8e2 f5xg6 h7xg6 Rd1d2 d4d3 c2xd3 c4xd3 Kc1d1 Re2e3 b2b4 Nf6e4 Bh3g2 Ne4c3+ Kd1c1 a6a5 b4xa5+ Kb6xa5 Bg2f3 Re3e1+ Kc1b2 Nc3a4+ Kb2a2 Na4c5 Ka2b2 Rd6e6 Rf7a7+ Ka5b6
25/68 03:31 1,932,497,852 9,148,000 -0.29 Rd1xd4 Ka7b6 b2b4 Qd6xf4 Rd4xf4 Nf6xd5 Rf4xf7 c5xb4 a3xb4 Nd5xb4 Na5b3 Ba8d5 Rf7f6+ Nb4c6 f3f4 Rh8f8 Rf6xf8 Rd8xf8 Bh3d7 Nc6b4 Re1e7 Rf8f7 Re7xf7 Bd5xf7 Kb1b2 Nb4d5 Nb3d4 Nd5e3 Bd7e6 Bf7e8 Kb2c3 Ne3f1 Nd4f3 Kb6c5 Kc3b2 Nf1e3
26/68 04:57 2,771,423,701 9,314,000 -0.30 Rd1xd4 Ka7b6 b2b4 Qd6xf4 Rd4xf4 Nf6xd5 Rf4xf7 c5xb4 a3xb4 Nd5xb4 Na5b3 Ba8d5 Rf7f6+ Nb4c6 f3f4 Rh8f8 Rf6xf8 Rd8xf8 Bh3d7 Nc6b4 Re1e5 Rf8d8 Re5e7 Bd5g8 Re7g7 Nb4d5 f4f5 Nd5f6 Bd7e6 Bg8xe6 f5xe6 Rd8e8 e6e7 Nf6d5 Rg7xh7 Nd5xe7 Nb3d4 Ne7d5 Kb1b2 Nd5e3 Rh7f7 Ne3c4+

I'm not sure what your point is, It is clear Houdini does not think the sac is winning. That was my point. Great game, but not the best ever.

You stated the sac does not work, but infact it does work as it's playable, and perhaps the best try for a win. The fact that Rxd4 is playable was my only point. I never said it was winning, but was impressed to see Houdini play it. But I agree, nice game, but hardly the best.

[oudheusa]

I have always liked this one.

[Event "02, Wijk aan Zee"]
[Site "02, Wijk aan Zee"]
[Date "1985.??.??"]
[EventDate "?"]
[Round "?"]
[Result "0-1"]
[White "Alexander Beliavsky"]
[Black "John Nunn"]
[ECO "E81"]
[WhiteElo "?"]
[BlackElo "?"]
[PlyCount "54"]

1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.f3 O-O 6.Be3 Nbd7 7.Qd2
c5 8.d5 Ne5 9.h3 Nh5 10.Bf2 f5 11.exf5 Rxf5 12.g4 Rxf3 13.gxh5
Qf8 14.Ne4 Bh6 15.Qc2 Qf4 16.Ne2 Rxf2 17.Nxf2 Nf3+ 18.Kd1 Qh4
19.Nd3 Bf5 20.Nec1 Nd2 21.hxg6 hxg6 22.Bg2 Nxc4 23.Qf2 Ne3+
24.Ke2 Qc4 25.Bf3 Rf8 26.Rg1 Nc2 27.Kd1 Bxd3 0-1

[OliverUwira]

Speaking of positional queen sacrifices, anyone remember this one?


Karpov - Kasparov (WCC, Lyon, 1990)

[D]rnb2k2/pppq1pbp/6p1/2nNp3/2P5/5N2/PPQ1BPPP/3RK2R b K - 0 14

14...Nc6!? 15. 0-0 Ne6 16. Nb6 axb6 17. Rxd7 Bxd7

(too bad that Karpov returned the queen soon after to get rid of the black pressure)

Code: Select all

[Event "World Championship 35th-KK5"]
[Site "Lyon/New York"]
[Date "1990.10.15"]
[Round "3"]
[White "Karpov, Anatoly"]
[Black "Kasparov, Garry"]
[Result "1/2-1/2"]
[ECO "E92"]
[WhiteElo "2730"]
[BlackElo "2800"]
[PlyCount "105"]
[EventDate "1990.10.??"]
[EventType "match"]
[EventRounds "24"]
[EventCountry "USA"]
[Source "ChessBase"]
[SourceDate "1991.04.01"]

1. d4 Nf6 2. c4 g6 3. Nc3 Bg7 4. e4 d6 5. Nf3 O-O 6. Be2 e5 7. Be3 Qe7 8. dxe5
dxe5 9. Nd5 Qd8 10. Bc5 Nxe4 11. Be7 Qd7 12. Bxf8 Kxf8 13. Qc2 Nc5 14. Rd1 Nc6
15. O-O Ne6 16. Nb6 axb6 17. Rxd7 Bxd7 18. Qd2 Be8 19. b3 e4 20. Ne1 f5 21. Bd1
Ne5 22. Nc2 Rxa2 23. Qd5 Ke7 24. Nb4 c6 25. Qxe6+ Kxe6 26. Nxa2 Nf7 27. Be2 Nd6
28. Nb4 Bc3 29. Nc2 f4 30. Rd1 h5 31. f3 e3 32. g3 g5 33. Bd3 h4 34. Kf1 c5 35.
Ke2 b5 36. cxb5 Nxb5 37. Bc4+ Ke7 38. Rd5 Bf6 39. Rxc5 Nc3+ 40. Kf1 Bg6 41. Ne1
Kd6 42. Ra5 fxg3 43. hxg3 hxg3 44. Ng2 b5 45. Ra6+ Ke7 46. Ra7+ Ke8 47. Ra8+
Bd8 48. Nxe3 bxc4 49. Nxc4 g4 50. Kg2 Ne2 51. Ne5 gxf3+ 52. Kxf3 g2 53. Rxd8+
1/2-1/2

[Taner Altinsoy]

Speaking of positional queen sacrifices, anyone remember this one?


Karpov - Kasparov (WCC, Lyon, 1990)

[D]rnb2k2/pppq1pbp/6p1/2nNp3/2P5/5N2/PPQ1BPPP/3RK2R b K - 0 14

14...Nc6!? 15. 0-0 Ne6 16. Nb6 axb6 17. Rxd7 Bxd7

(too bad that Karpov returned the queen soon after to get rid of the black pressure)

Code: Select all

[Event "World Championship 35th-KK5"]
[Site "Lyon/New York"]
[Date "1990.10.15"]
[Round "3"]
[White "Karpov, Anatoly"]
[Black "Kasparov, Garry"]
[Result "1/2-1/2"]
[ECO "E92"]
[WhiteElo "2730"]
[BlackElo "2800"]
[PlyCount "105"]
[EventDate "1990.10.??"]
[EventType "match"]
[EventRounds "24"]
[EventCountry "USA"]
[Source "ChessBase"]
[SourceDate "1991.04.01"]

1. d4 Nf6 2. c4 g6 3. Nc3 Bg7 4. e4 d6 5. Nf3 O-O 6. Be2 e5 7. Be3 Qe7 8. dxe5
dxe5 9. Nd5 Qd8 10. Bc5 Nxe4 11. Be7 Qd7 12. Bxf8 Kxf8 13. Qc2 Nc5 14. Rd1 Nc6
15. O-O Ne6 16. Nb6 axb6 17. Rxd7 Bxd7 18. Qd2 Be8 19. b3 e4 20. Ne1 f5 21. Bd1
Ne5 22. Nc2 Rxa2 23. Qd5 Ke7 24. Nb4 c6 25. Qxe6+ Kxe6 26. Nxa2 Nf7 27. Be2 Nd6
28. Nb4 Bc3 29. Nc2 f4 30. Rd1 h5 31. f3 e3 32. g3 g5 33. Bd3 h4 34. Kf1 c5 35.
Ke2 b5 36. cxb5 Nxb5 37. Bc4+ Ke7 38. Rd5 Bf6 39. Rxc5 Nc3+ 40. Kf1 Bg6 41. Ne1
Kd6 42. Ra5 fxg3 43. hxg3 hxg3 44. Ng2 b5 45. Ra6+ Ke7 46. Ra7+ Ke8 47. Ra8+
Bd8 48. Nxe3 bxc4 49. Nxc4 g4 50. Kg2 Ne2 51. Ne5 gxf3+ 52. Kxf3 g2 53. Rxd8+
1/2-1/2

I remember this game very well. It was home preparation for from Kasparov and it would have been incredible had he pulled a win out of it. But Stubborn defence from Karpov saved the game. Nice game indeed.

[javimm]

For the fireworks thing, I very much like Kasparov-Topalov 1999 Wijk aan Zee. That rook sacrifice is fantastic to play at that level (24 Rxd4).

Here is the game:

[Event "Hoogovens A Tournament (cat. 18)"]
[Site "Wijk aan Zee NED"]
[Date "1999.01.20"]
[Round "04"]
[White "Kasparov, Garry (RUS)"]
[Black "Topalov, Veselin (BUL)"]
[Result "1-0"]
[WhiteElo "2812"]
[BlackElo "2700"]
[Opening "Pirc Defence"]
[Variation "Ufimtsev-Pytel"]
[ECO "B07"]
[PlyCount "87"]

1.e4 d6 2.d4 Nf6 3.Nc3 g6 4.Be3 Bg7 5.Qd2 c6
6.f3 b5 7.Nge2 Nbd7 8.Bh6 Bxh6 9.Qxh6 Bb7
10.a3 e5 11.0-0-0 Qe7 12.Kb1 a6 13.Nc1 0-0-0
14.Nb3 exd4! 15.Rxd4 c5 16.Rd1 Nb6! 17.g3 Kb8
18.Na5 Ba8 19.Bh3 d5 20.Qf4+ Ka7 21.Rhe1 d4
22.Nd5 Nbxd5 23.exd5 Qd6 24.Rxd4!! cxd4?!
25.Re7+! Kb6 26.Qxd4+ Kxa5 27.b4+ Ka4 28.Qc3
Qxd5 29.Ra7 Bb7 30.Rxb7 Qc4 31.Qxf6 Kxa3?
32.Qxa6+ Kxb4 33.c3+! Kxc3 34.Qa1+ Kd2
35.Qb2+ Kd1 36.Bf1! Rd2 37.Rd7! Rxd7! 38.Bxc4
bxc4 39.Qxh8 Rd3 40.Qa8 c3 41.Qa4+ Ke1 42.f4 f5
43.Kc1 Rd2 44.Qa7
1-0

Saw this game live, after looking at the game the next day. We found that Rxd4 does not work. If Topalov would have ignored the rook sac, the game would have been long forgotten. It took Topalovs blunder after Rxd4 to make Rxd4 work, so I do not consider this game one of the best ever.

I remember some people using the rook sac as a computer test position, but no computer program would play the rook sac....because even in those day the chess programs saw the sac was not winning.

I remember doing analysis and watching the game live. I had at least two engine back then that would play the sacrifice. I am not sure that the sack does not work, but I have not analyzed it in a long time. I am pretty sure me and a friend concluded it did not force a win, but was sound.
In the interest of time, Houdini 1.5 likes Rxd4.


FEN: b2r3r/k4p1p/p2q1np1/NppP4/3p1Q2/P4PPB/1PP4P/1K1RR3 w - - 1 24

Houdini_15_x64:
4/15 00:00 2,604 0 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Rd6xc6 Re1e7+ Ka7b6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Re8e1+ Kb1a2 Rc6e6 Bg2h3 Re6d6 Bh3c8 Nf6d5 Bc8b7 Re1e6 f3f4 Rd6d8 Bb7xd5 Rd8xd5 b2b4 a6a5 Ka2b2 c5xb4 a3xb4 a5xb4
5/33 00:00 3,975 0 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Re8e1+ Kb1a2 Rc6e6 Bg2h3 Re6d6 Bh3c8 Nf6d5 Bc8b7 Re1e6 f3f4 Rd6d8 Bb7xd5 Rd8xd5 b2b4 a6a5 Ka2b2 c5xb4 a3xb4 a5xb4
6/33 00:00 5,724 0 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Rd6xc6 Re1e7+ Ka7b6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Re8e1+ Kb1a2 Rc6e6 Bg2h3 Re6d6 Bh3c8 Nf6d5 Bc8b7 Re1e6 f3f4 Rd6d8 Bb7xd5 Rd8xd5 b2b4 a6a5 Ka2b2 c5xb4 a3xb4 a5xb4
7/33+ 00:00 7,534 0 -0.17 Na5c6+
7/33 00:00 8,466 0 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Rd6xc6 Re1e7+ Ka7b6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Re8e1+ Kb1a2 Rc6e6 Bg2h3 Re6d6 Bh3c8 Nf6d5 Bc8b7 Re1e6 f3f4 Rd6d8 Bb7xd5 Rd8xd5 b2b4 a6a5 Ka2b2 c5xb4 a3xb4 a5xb4
8/33 00:00 12,012 0 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Rd6xc6 Re1e7+ Ka7b6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Re8e1+ Kb1a2 Rc6e6 Bg2h3 Re6d6 Bh3c8 Nf6d5 Bc8b7 Re1e6 f3f4 Rd6d8 Bb7xd5 Rd8xd5 b2b4 a6a5 Ka2b2 c5xb4 a3xb4 a5xb4
9/33 00:00 20,069 557,000 -0.21 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 c6c7 Kb6xc7 Re1e7+ Kc7b6 Re7xf7 Nf6d5 f3f4 h7h6 Rd1e1 Nd5e3
10/33 00:00 28,182 640,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Re8e1+ Kb1a2 Rc6e6 Bg2h3 Re6d6 Bh3c8 Nf6d5 Bc8b7 Re1e6 f3f4 Rd6d8 Bb7xd5 Rd8xd5 b2b4 a6a5 Ka2b2 c5xb4 a3xb4 a5xb4
10/33 00:00 47,790 955,000 -0.23 Rd1xd4 Ka7b6 b2b4 Qd6xf4 Rd4xf4 Nf6xd5 Rf4xf7 c5xb4 a3xb4 Nd5xb4 Na5b3 Ba8d5 Rf7f6+ Nb4c6 f3f4 Rh8f8 Rf6xf8 Rd8xf8 Bh3d7 Rf8d8 Bd7xc6 Bd5xc6 Re1e7 a6a5 Re7xh7 a5a4 Nb3c1 Rd8d2 Nc1d3 Bc6e4 Kb1c1 Rd2e2
11/33 00:00 75,562 956,000 -0.28 Rd1xd4 Ka7b6 b2b4 Qd6xf4 Rd4xf4 Nf6xd5 Rf4xf7 c5xb4 a3xb4 Nd5xb4 Na5b3 Ba8d5 Rf7f6+ Nb4c6 f3f4 a6a5 Nb3d4 Rh8f8 Rf6xf8 Rd8xf8 Nd4xc6 Bd5xc6 Re1e7 h7h5 Re7e6 Rf8g8 Kb1b2 a5a4 Re6d6 Kb6c5
11/33 00:00 80,150 1,001,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 Rf7f4 g6g5 Rf4f5 d4d3+ Rd2xd3 Re1e2+ Kc2d1
12/33 00:00 104,712 1,203,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 Rf7f4 g6g5 Rf4f5 d4d3+ Rd2xd3 Re1e2+ Kc2d1
13/33 00:00 163,981 1,012,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 Rf7f4 g6g5 Rf4f5 d4d3+ Rd2xd3 Re1e2+ Kc2d1
14/33 00:00 272,890 1,421,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 Rf7f4 g6g5 Rf4f5 d4d3+ Rd2xd3 Re1e2+ Kc2d1
15/33 00:00 458,011 1,982,000 -0.21 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Rd6xc6 Re1e7+ Ka7b6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 f3f4 Re1e3 c3xd4 Rd8xd4 Rd2xd4 c5xd4
16/33 00:00 804,437 2,812,000 -0.21 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 f3f4 Re1e3 c3xd4 Rd8xd4 Rd2xd4 c5xd4
17/33 00:00 1,452,674 3,853,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 Rf7f4 g6g5 Rf4f5 d4d3+ Rd2xd3 Re1e2+ Kc2d1
18/35 00:00 2,746,309 5,011,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 Rf7f4 g6g5 Rf4f5 d4d3+ Rd2xd3 Re1e2+ Kc2d1
19/38 00:00 4,908,460 6,022,000 -0.22 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 f3f4 Nf6g4 Rf7b7+ Kb6a5 b2b4+ c5xb4 a3xb4+ Ka5xb4 Rd2xd4+ Kb4a5 h2h3 Ng4e3 Bg2f3 Ne3c4
20/40 00:01 10,344,418 7,134,000 -0.18 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Kb6xc6 Re7xf7 Rh8e8 Kb1c1 h7h5 f3f4 Kc6b6 Bh3g2 Re8e2 Rd1d2 Re2e1+ Rd2d1 Re1xd1+ Kc1xd1 Nf6g4 Kd1d2 Ng4e3 Rf7b7+ Kb6a5 Bg2h3 Ne3c4+ Kd2c1 Rd6c6 Bh3d7 Rc6f6 Bd7h3 Nc4e3 Kc1d2 Rf6d6 b2b3 Rd6d8
21/42 00:02 20,698,104 7,976,000 -0.19 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Kb6xc6 Re7xf7 Rh8e8 Kb1c1 h7h5 f3f4 Kc6b6 Bh3g2 Re8e2 Rd1d2 Re2e1+ Rd2d1 Re1xd1+ Kc1xd1 Nf6g4 Kd1d2 Ng4e3 Rf7b7+ Kb6a5 Bg2h3 Ne3c4+ Kd2c1 Rd6c6 Bh3d7 Rc6f6 Bd7h3 Nc4e3 Kc1d2 Rf6d6 b2b3 Rd6d8
22/45 00:05 42,379,863 8,225,000 -0.21 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Kb6xc6 Re7xf7 Rh8e8 Kb1c1 h7h5 f3f4 Kc6b6 Bh3g2 Re8e2 Rd1d2 Re2e1+ Rd2d1 Re1xd1+ Kc1xd1 Nf6g4 Kd1d2 Ng4e3 Rf7b7+ Kb6a5 Bg2h3 Ne3c4+ Kd2c1 Rd6c6 Bh3d7 Rc6f6 Bd7h3 Nc4e3 Kc1d2 Rf6d6 b2b3 Rd6d8
23/54 00:16 132,706,888 8,215,000 -0.18 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Kb6xc6 Re7xf7 Rh8e8 Kb1c1 h7h5 f3f4 Kc6b6 Bh3g2 Re8e2 Rd1d2 Re2e1+ Rd2d1 Re1xd1+ Kc1xd1 Nf6g4 Kd1d2 Ng4e3 Rf7b7+ Kb6a5 Bg2h3 Ne3c4+ Kd2c1 Rd6c6 Bh3d7 Rc6f6 Bd7h3 Nc4e3 Kc1d2 Ne3c4+ Kd2c1 Nc4e3
24/54- 00:29 256,124,787 8,548,000 -0.25 Na5c6+ Ba8xc6
24/61 01:11 629,868,144 8,862,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Kb6xc6 f3f4 Rh8f8 Bh3g2+ Nf6d5 c2c3 d4xc3 b2xc3 c5c4 Kb1b2 Kc6c5 Bg2xd5 Rd6xd5 Rd1xd5+ Kc5xd5 a3a4 Kd5c6 Kb2a3 Kc6b6 Ka3b4 a6a5+ Kb4a3 b5xa4 Re7d7 Kb6c6 Rd7e7 Kc6b5 Re7e5+ Kb5b6 Ka3xa4 Rf8d8 Re5b5+
25/61- 01:18 700,574,044 8,885,000 -0.29 Na5c6+ Ba8xc6
25/68 02:47 1,534,780,264 9,180,000 -0.37 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Kb6xc6 f3f4 Kc6b6 Re7xf7 Rh8e8 Kb1c1 c5c4 f4f5 Re8e2 f5xg6 h7xg6 Rd1d2 d4d3 c2xd3 c4xd3 Kc1d1 Re2e3 b2b4 Nf6e4 Bh3g2 Ne4c3+ Kd1c1 a6a5 b4xa5+ Kb6xa5 Bg2f3 Re3e1+ Kc1b2 Nc3a4+ Kb2a2 Na4c5 Ka2b2 Rd6e6 Rf7a7+ Ka5b6
25/68 03:31 1,932,497,852 9,148,000 -0.29 Rd1xd4 Ka7b6 b2b4 Qd6xf4 Rd4xf4 Nf6xd5 Rf4xf7 c5xb4 a3xb4 Nd5xb4 Na5b3 Ba8d5 Rf7f6+ Nb4c6 f3f4 Rh8f8 Rf6xf8 Rd8xf8 Bh3d7 Nc6b4 Re1e7 Rf8f7 Re7xf7 Bd5xf7 Kb1b2 Nb4d5 Nb3d4 Nd5e3 Bd7e6 Bf7e8 Kb2c3 Ne3f1 Nd4f3 Kb6c5 Kc3b2 Nf1e3
26/68 04:57 2,771,423,701 9,314,000 -0.30 Rd1xd4 Ka7b6 b2b4 Qd6xf4 Rd4xf4 Nf6xd5 Rf4xf7 c5xb4 a3xb4 Nd5xb4 Na5b3 Ba8d5 Rf7f6+ Nb4c6 f3f4 Rh8f8 Rf6xf8 Rd8xf8 Bh3d7 Nc6b4 Re1e5 Rf8d8 Re5e7 Bd5g8 Re7g7 Nb4d5 f4f5 Nd5f6 Bd7e6 Bg8xe6 f5xe6 Rd8e8 e6e7 Nf6d5 Rg7xh7 Nd5xe7 Nb3d4 Ne7d5 Kb1b2 Nd5e3 Rh7f7 Ne3c4+

I'm not sure what your point is, It is clear Houdini does not think the sac is winning. That was my point. Great game, but not the best ever.

You stated the sac does not work, but infact it does work as it's playable, and perhaps the best try for a win. The fact that Rxd4 is playable was my only point. I never said it was winning, but was impressed to see Houdini play it. But I agree, nice game, but hardly the best.

The game I posted isn't certainly the best from a perfect play point of view.

I remember an interview with Kasparov about that, and he said that he went into deep thinking about the sacrifice (about 15-20 ply deep!!) and he couldn't predict if there was going to work on the end because it was so complex.

The greatness of that move is the K played even then he knew it may could not work and a refutation could be found, be he knew it was going to give him great chances of success because the refutation is very hard to find: ie not accepting the sacrifice is not an easy move to play. Seing the winning line and the refutation one over the board in this position for a human in tournament conditions and having the guts to play that against a 2700+ player is what makes this game a great one. Not the best?. No game is.

As has been said, engines refute the sac, but we're talking about "Greatest Human Chess Game Ever Played". Even then, I don't see how taking the sac is a bad move per se, as refuting it is not a good move either (although a better move than accepting it).

[Don]

For the fireworks thing, I very much like Kasparov-Topalov 1999 Wijk aan Zee. That rook sacrifice is fantastic to play at that level (24 Rxd4).

Here is the game:

[Event "Hoogovens A Tournament (cat. 18)"]
[Site "Wijk aan Zee NED"]
[Date "1999.01.20"]
[Round "04"]
[White "Kasparov, Garry (RUS)"]
[Black "Topalov, Veselin (BUL)"]
[Result "1-0"]
[WhiteElo "2812"]
[BlackElo "2700"]
[Opening "Pirc Defence"]
[Variation "Ufimtsev-Pytel"]
[ECO "B07"]
[PlyCount "87"]

1.e4 d6 2.d4 Nf6 3.Nc3 g6 4.Be3 Bg7 5.Qd2 c6
6.f3 b5 7.Nge2 Nbd7 8.Bh6 Bxh6 9.Qxh6 Bb7
10.a3 e5 11.0-0-0 Qe7 12.Kb1 a6 13.Nc1 0-0-0
14.Nb3 exd4! 15.Rxd4 c5 16.Rd1 Nb6! 17.g3 Kb8
18.Na5 Ba8 19.Bh3 d5 20.Qf4+ Ka7 21.Rhe1 d4
22.Nd5 Nbxd5 23.exd5 Qd6 24.Rxd4!! cxd4?!
25.Re7+! Kb6 26.Qxd4+ Kxa5 27.b4+ Ka4 28.Qc3
Qxd5 29.Ra7 Bb7 30.Rxb7 Qc4 31.Qxf6 Kxa3?
32.Qxa6+ Kxb4 33.c3+! Kxc3 34.Qa1+ Kd2
35.Qb2+ Kd1 36.Bf1! Rd2 37.Rd7! Rxd7! 38.Bxc4
bxc4 39.Qxh8 Rd3 40.Qa8 c3 41.Qa4+ Ke1 42.f4 f5
43.Kc1 Rd2 44.Qa7
1-0

Saw this game live, after looking at the game the next day. We found that Rxd4 does not work. If Topalov would have ignored the rook sac, the game would have been long forgotten. It took Topalovs blunder after Rxd4 to make Rxd4 work, so I do not consider this game one of the best ever.

I remember some people using the rook sac as a computer test position, but no computer program would play the rook sac....because even in those day the chess programs saw the sac was not winning.

Would you care to show one of the greatest games ever where neither side made a mistake, please? I'd really want to know...

I think perfect games are very rare. More than likely a perfect games would be a draw, but I suspect that most draws have imperfect moves in them.

I define a perfect move as a move which does not change the game theoretic result - it does not have to necessarily be a good move from our point of view. For example sometimes we have a won game but play a move that makes the win more "difficult" for ourself, but it's still a win.

There is evidence that humans play several hundred ELO below perfect play and I'm estimating that the majority of top grandmaster games have these errors in them. Another way to say this is that a perfect player would rarely not win against a top grandmaster.

[Robert Flesher]

For the fireworks thing, I very much like Kasparov-Topalov 1999 Wijk aan Zee. That rook sacrifice is fantastic to play at that level (24 Rxd4).

Here is the game:

[Event "Hoogovens A Tournament (cat. 18)"]
[Site "Wijk aan Zee NED"]
[Date "1999.01.20"]
[Round "04"]
[White "Kasparov, Garry (RUS)"]
[Black "Topalov, Veselin (BUL)"]
[Result "1-0"]
[WhiteElo "2812"]
[BlackElo "2700"]
[Opening "Pirc Defence"]
[Variation "Ufimtsev-Pytel"]
[ECO "B07"]
[PlyCount "87"]

1.e4 d6 2.d4 Nf6 3.Nc3 g6 4.Be3 Bg7 5.Qd2 c6
6.f3 b5 7.Nge2 Nbd7 8.Bh6 Bxh6 9.Qxh6 Bb7
10.a3 e5 11.0-0-0 Qe7 12.Kb1 a6 13.Nc1 0-0-0
14.Nb3 exd4! 15.Rxd4 c5 16.Rd1 Nb6! 17.g3 Kb8
18.Na5 Ba8 19.Bh3 d5 20.Qf4+ Ka7 21.Rhe1 d4
22.Nd5 Nbxd5 23.exd5 Qd6 24.Rxd4!! cxd4?!
25.Re7+! Kb6 26.Qxd4+ Kxa5 27.b4+ Ka4 28.Qc3
Qxd5 29.Ra7 Bb7 30.Rxb7 Qc4 31.Qxf6 Kxa3?
32.Qxa6+ Kxb4 33.c3+! Kxc3 34.Qa1+ Kd2
35.Qb2+ Kd1 36.Bf1! Rd2 37.Rd7! Rxd7! 38.Bxc4
bxc4 39.Qxh8 Rd3 40.Qa8 c3 41.Qa4+ Ke1 42.f4 f5
43.Kc1 Rd2 44.Qa7
1-0

Saw this game live, after looking at the game the next day. We found that Rxd4 does not work. If Topalov would have ignored the rook sac, the game would have been long forgotten. It took Topalovs blunder after Rxd4 to make Rxd4 work, so I do not consider this game one of the best ever.

I remember some people using the rook sac as a computer test position, but no computer program would play the rook sac....because even in those day the chess programs saw the sac was not winning.

I remember doing analysis and watching the game live. I had at least two engine back then that would play the sacrifice. I am not sure that the sack does not work, but I have not analyzed it in a long time. I am pretty sure me and a friend concluded it did not force a win, but was sound.
In the interest of time, Houdini 1.5 likes Rxd4.


FEN: b2r3r/k4p1p/p2q1np1/NppP4/3p1Q2/P4PPB/1PP4P/1K1RR3 w - - 1 24

Houdini_15_x64:
4/15 00:00 2,604 0 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Rd6xc6 Re1e7+ Ka7b6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Re8e1+ Kb1a2 Rc6e6 Bg2h3 Re6d6 Bh3c8 Nf6d5 Bc8b7 Re1e6 f3f4 Rd6d8 Bb7xd5 Rd8xd5 b2b4 a6a5 Ka2b2 c5xb4 a3xb4 a5xb4
5/33 00:00 3,975 0 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Re8e1+ Kb1a2 Rc6e6 Bg2h3 Re6d6 Bh3c8 Nf6d5 Bc8b7 Re1e6 f3f4 Rd6d8 Bb7xd5 Rd8xd5 b2b4 a6a5 Ka2b2 c5xb4 a3xb4 a5xb4
6/33 00:00 5,724 0 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Rd6xc6 Re1e7+ Ka7b6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Re8e1+ Kb1a2 Rc6e6 Bg2h3 Re6d6 Bh3c8 Nf6d5 Bc8b7 Re1e6 f3f4 Rd6d8 Bb7xd5 Rd8xd5 b2b4 a6a5 Ka2b2 c5xb4 a3xb4 a5xb4
7/33+ 00:00 7,534 0 -0.17 Na5c6+
7/33 00:00 8,466 0 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Rd6xc6 Re1e7+ Ka7b6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Re8e1+ Kb1a2 Rc6e6 Bg2h3 Re6d6 Bh3c8 Nf6d5 Bc8b7 Re1e6 f3f4 Rd6d8 Bb7xd5 Rd8xd5 b2b4 a6a5 Ka2b2 c5xb4 a3xb4 a5xb4
8/33 00:00 12,012 0 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Rd6xc6 Re1e7+ Ka7b6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Re8e1+ Kb1a2 Rc6e6 Bg2h3 Re6d6 Bh3c8 Nf6d5 Bc8b7 Re1e6 f3f4 Rd6d8 Bb7xd5 Rd8xd5 b2b4 a6a5 Ka2b2 c5xb4 a3xb4 a5xb4
9/33 00:00 20,069 557,000 -0.21 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 c6c7 Kb6xc7 Re1e7+ Kc7b6 Re7xf7 Nf6d5 f3f4 h7h6 Rd1e1 Nd5e3
10/33 00:00 28,182 640,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Re8e1+ Kb1a2 Rc6e6 Bg2h3 Re6d6 Bh3c8 Nf6d5 Bc8b7 Re1e6 f3f4 Rd6d8 Bb7xd5 Rd8xd5 b2b4 a6a5 Ka2b2 c5xb4 a3xb4 a5xb4
10/33 00:00 47,790 955,000 -0.23 Rd1xd4 Ka7b6 b2b4 Qd6xf4 Rd4xf4 Nf6xd5 Rf4xf7 c5xb4 a3xb4 Nd5xb4 Na5b3 Ba8d5 Rf7f6+ Nb4c6 f3f4 Rh8f8 Rf6xf8 Rd8xf8 Bh3d7 Rf8d8 Bd7xc6 Bd5xc6 Re1e7 a6a5 Re7xh7 a5a4 Nb3c1 Rd8d2 Nc1d3 Bc6e4 Kb1c1 Rd2e2
11/33 00:00 75,562 956,000 -0.28 Rd1xd4 Ka7b6 b2b4 Qd6xf4 Rd4xf4 Nf6xd5 Rf4xf7 c5xb4 a3xb4 Nd5xb4 Na5b3 Ba8d5 Rf7f6+ Nb4c6 f3f4 a6a5 Nb3d4 Rh8f8 Rf6xf8 Rd8xf8 Nd4xc6 Bd5xc6 Re1e7 h7h5 Re7e6 Rf8g8 Kb1b2 a5a4 Re6d6 Kb6c5
11/33 00:00 80,150 1,001,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 Rf7f4 g6g5 Rf4f5 d4d3+ Rd2xd3 Re1e2+ Kc2d1
12/33 00:00 104,712 1,203,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 Rf7f4 g6g5 Rf4f5 d4d3+ Rd2xd3 Re1e2+ Kc2d1
13/33 00:00 163,981 1,012,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 Rf7f4 g6g5 Rf4f5 d4d3+ Rd2xd3 Re1e2+ Kc2d1
14/33 00:00 272,890 1,421,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 Rf7f4 g6g5 Rf4f5 d4d3+ Rd2xd3 Re1e2+ Kc2d1
15/33 00:00 458,011 1,982,000 -0.21 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Rd6xc6 Re1e7+ Ka7b6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 f3f4 Re1e3 c3xd4 Rd8xd4 Rd2xd4 c5xd4
16/33 00:00 804,437 2,812,000 -0.21 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 f3f4 Re1e3 c3xd4 Rd8xd4 Rd2xd4 c5xd4
17/33 00:00 1,452,674 3,853,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 Rf7f4 g6g5 Rf4f5 d4d3+ Rd2xd3 Re1e2+ Kc2d1
18/35 00:00 2,746,309 5,011,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 c2c3 Re8d8 Bg2h3 Re6e1+ Kb1c2 Nf6e8 Rf7f4 g6g5 Rf4f5 d4d3+ Rd2xd3 Re1e2+ Kc2d1
19/38 00:00 4,908,460 6,022,000 -0.22 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Rd6xc6 Re7xf7 Rh8e8 Rd1d2 h7h5 Bh3g2 Rc6e6 f3f4 Nf6g4 Rf7b7+ Kb6a5 b2b4+ c5xb4 a3xb4+ Ka5xb4 Rd2xd4+ Kb4a5 h2h3 Ng4e3 Bg2f3 Ne3c4
20/40 00:01 10,344,418 7,134,000 -0.18 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Kb6xc6 Re7xf7 Rh8e8 Kb1c1 h7h5 f3f4 Kc6b6 Bh3g2 Re8e2 Rd1d2 Re2e1+ Rd2d1 Re1xd1+ Kc1xd1 Nf6g4 Kd1d2 Ng4e3 Rf7b7+ Kb6a5 Bg2h3 Ne3c4+ Kd2c1 Rd6c6 Bh3d7 Rc6f6 Bd7h3 Nc4e3 Kc1d2 Rf6d6 b2b3 Rd6d8
21/42 00:02 20,698,104 7,976,000 -0.19 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Kb6xc6 Re7xf7 Rh8e8 Kb1c1 h7h5 f3f4 Kc6b6 Bh3g2 Re8e2 Rd1d2 Re2e1+ Rd2d1 Re1xd1+ Kc1xd1 Nf6g4 Kd1d2 Ng4e3 Rf7b7+ Kb6a5 Bg2h3 Ne3c4+ Kd2c1 Rd6c6 Bh3d7 Rc6f6 Bd7h3 Nc4e3 Kc1d2 Rf6d6 b2b3 Rd6d8
22/45 00:05 42,379,863 8,225,000 -0.21 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Kb6xc6 Re7xf7 Rh8e8 Kb1c1 h7h5 f3f4 Kc6b6 Bh3g2 Re8e2 Rd1d2 Re2e1+ Rd2d1 Re1xd1+ Kc1xd1 Nf6g4 Kd1d2 Ng4e3 Rf7b7+ Kb6a5 Bg2h3 Ne3c4+ Kd2c1 Rd6c6 Bh3d7 Rc6f6 Bd7h3 Nc4e3 Kc1d2 Rf6d6 b2b3 Rd6d8
23/54 00:16 132,706,888 8,215,000 -0.18 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Kb6xc6 Re7xf7 Rh8e8 Kb1c1 h7h5 f3f4 Kc6b6 Bh3g2 Re8e2 Rd1d2 Re2e1+ Rd2d1 Re1xd1+ Kc1xd1 Nf6g4 Kd1d2 Ng4e3 Rf7b7+ Kb6a5 Bg2h3 Ne3c4+ Kd2c1 Rd6c6 Bh3d7 Rc6f6 Bd7h3 Nc4e3 Kc1d2 Ne3c4+ Kd2c1 Nc4e3
24/54- 00:29 256,124,787 8,548,000 -0.25 Na5c6+ Ba8xc6
24/61 01:11 629,868,144 8,862,000 -0.23 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Kb6xc6 f3f4 Rh8f8 Bh3g2+ Nf6d5 c2c3 d4xc3 b2xc3 c5c4 Kb1b2 Kc6c5 Bg2xd5 Rd6xd5 Rd1xd5+ Kc5xd5 a3a4 Kd5c6 Kb2a3 Kc6b6 Ka3b4 a6a5+ Kb4a3 b5xa4 Re7d7 Kb6c6 Rd7e7 Kc6b5 Re7e5+ Kb5b6 Ka3xa4 Rf8d8 Re5b5+
25/61- 01:18 700,574,044 8,885,000 -0.29 Na5c6+ Ba8xc6
25/68 02:47 1,534,780,264 9,180,000 -0.37 Na5c6+ Ba8xc6 Qf4xd6 Rd8xd6 d5xc6 Ka7b6 Re1e7 Kb6xc6 f3f4 Kc6b6 Re7xf7 Rh8e8 Kb1c1 c5c4 f4f5 Re8e2 f5xg6 h7xg6 Rd1d2 d4d3 c2xd3 c4xd3 Kc1d1 Re2e3 b2b4 Nf6e4 Bh3g2 Ne4c3+ Kd1c1 a6a5 b4xa5+ Kb6xa5 Bg2f3 Re3e1+ Kc1b2 Nc3a4+ Kb2a2 Na4c5 Ka2b2 Rd6e6 Rf7a7+ Ka5b6
25/68 03:31 1,932,497,852 9,148,000 -0.29 Rd1xd4 Ka7b6 b2b4 Qd6xf4 Rd4xf4 Nf6xd5 Rf4xf7 c5xb4 a3xb4 Nd5xb4 Na5b3 Ba8d5 Rf7f6+ Nb4c6 f3f4 Rh8f8 Rf6xf8 Rd8xf8 Bh3d7 Nc6b4 Re1e7 Rf8f7 Re7xf7 Bd5xf7 Kb1b2 Nb4d5 Nb3d4 Nd5e3 Bd7e6 Bf7e8 Kb2c3 Ne3f1 Nd4f3 Kb6c5 Kc3b2 Nf1e3
26/68 04:57 2,771,423,701 9,314,000 -0.30 Rd1xd4 Ka7b6 b2b4 Qd6xf4 Rd4xf4 Nf6xd5 Rf4xf7 c5xb4 a3xb4 Nd5xb4 Na5b3 Ba8d5 Rf7f6+ Nb4c6 f3f4 Rh8f8 Rf6xf8 Rd8xf8 Bh3d7 Nc6b4 Re1e5 Rf8d8 Re5e7 Bd5g8 Re7g7 Nb4d5 f4f5 Nd5f6 Bd7e6 Bg8xe6 f5xe6 Rd8e8 e6e7 Nf6d5 Rg7xh7 Nd5xe7 Nb3d4 Ne7d5 Kb1b2 Nd5e3 Rh7f7 Ne3c4+

I'm not sure what your point is, It is clear Houdini does not think the sac is winning. That was my point. Great game, but not the best ever.

You stated the sac does not work, but infact it does work as it's playable, and perhaps the best try for a win. The fact that Rxd4 is playable was my only point. I never said it was winning, but was impressed to see Houdini play it. But I agree, nice game, but hardly the best.

The game I posted isn't certainly the best from a perfect play point of view.

I remember an interview with Kasparov about that, and he said that he went into deep thinking about the sacrifice (about 15-20 ply deep!!) and he couldn't predict if there was going to work on the end because it was so complex.

The greatness of that move is the K played even then he knew it may could not work and a refutation could be found, be he knew it was going to give him great chances of success because the refutation is very hard to find: ie not accepting the sacrifice is not an easy move to play. Seing the winning line and the refutation one over the board in this position for a human in tournament conditions and having the guts to play that against a 2700+ player is what makes this game a great one. Not the best?. No game is.

As has been said, engines refute the sac, but we're talking about "Greatest Human Chess Game Ever Played". Even then, I don't see how taking the sac is a bad move per se, as refuting it is not a good move either (although a better move than accepting it).

I am not sure engines refute the sac. It seems the sac does not win outright, but as I said before it may be the best attempt for the win. Houndini 1.5 likes it! Also, I watched this game live and it was one of my favourites. I had a postal GM/OTB master sitting beside me and his jaw hit the floor after that move. It's power was obvious as Topalov failed to find the correct defense.

[George Tsavdaris]

That is a good one, but it was so long ago - it does not remind me of the game I was thinking about. I remember the queen sacrifice as being almost positional and hard to believe - in this beautiful game it's not hard to believe there might be immediate tactics.

I don't really remember such a Fischer's game but i do remember one of Tal's.
He gave his Queen for N+B for positional purposes only. No immediate attack or gain of material.
I remember after days of analysis that the sacrifice was sound but no program of the time was able to play it. These were the Deep Fritz 8/Shredder 8 days i remember, and i was able to find that black gets better after countless of variations.

Don't know if analysis with todays programs would change the soundness of the move or if they can find it(if it of course proves correct with them also).

Tal played 11...Nxd5 here giving his Queen.
Bobotsov-Tal was the game.
[d]r1b2rk1/3nppbp/p2p1np1/qppN4/2PPP3/4BP2/PP1QN1PP/1K1R1B1R b - - 0 11

[Don]

That is a good one, but it was so long ago - it does not remind me of the game I was thinking about. I remember the queen sacrifice as being almost positional and hard to believe - in this beautiful game it's not hard to believe there might be immediate tactics.

I don't really remember such a Fischer's game but i do remember one of Tal's.
He gave his Queen for N+B for positional purposes only. No immediate attack or gain of material.
I remember after days of analysis that the sacrifice was sound but no program of the time was able to play it. These were the Deep Fritz 8/Shredder 8 days i remember, and i was able to find that black gets better after countless of variations.

Don't know if analysis with todays programs would change the soundness of the move or if they can find it(if it of course proves correct with them also).

Tal played 11...Nxd5 here giving his Queen.
Bobotsov-Tal was the game.
[d]r1b2rk1/3nppbp/p2p1np1/qppN4/2PPP3/4BP2/PP1QN1PP/1K1R1B1R b - - 0 11

Komodo briefly considers this move but seems to prefer Qd8. Komodo seems to view black as being down about a pawn. I will let it run a while. If it does not change it's mind I will force Nxd5 and see what the score is after a deep search.

[George Tsavdaris]

That is a good one, but it was so long ago - it does not remind me of the game I was thinking about. I remember the queen sacrifice as being almost positional and hard to believe - in this beautiful game it's not hard to believe there might be immediate tactics.

I don't really remember such a Fischer's game but i do remember one of Tal's.
He gave his Queen for N+B for positional purposes only. No immediate attack or gain of material.
I remember after days of analysis that the sacrifice was sound but no program of the time was able to play it. These were the Deep Fritz 8/Shredder 8 days i remember, and i was able to find that black gets better after countless of variations.

Don't know if analysis with todays programs would change the soundness of the move or if they can find it(if it of course proves correct with them also).

Tal played 11...Nxd5 here giving his Queen.
Bobotsov-Tal was the game.
[d]r1b2rk1/3nppbp/p2p1np1/qppN4/2PPP3/4BP2/PP1QN1PP/1K1R1B1R b - - 0 11

Komodo briefly considers this move but seems to prefer Qd8. Komodo seems to view black as being down about a pawn. I will let it run a while. If it does not change it's mind I will force Nxd5 and see what the score is after a deep search.

If you have to force a move it's better to let it analyze the forced continuation(after 11...Nxd5) 11...Nxd5 12.Qxa5 Nxe3 in order to save some time.

The attack black creates after some quiet moves like Bb7 and Rfc8 and the possible Pawn march is very strong i think for white and it's a lot far away from engine's horizon. Tough position anyway.