Deep Rybka-Deep Shredder, SSDF 2009

[jdart]

In this position:

[d] 1qrr4/3p1kp1/p1b1pp1p/1pN5/4PN2/2P1B1R1/PP2B1P1/2K2R2 w - -

Rybka played Bd4!, which looks bad because it allows .. e5, but in fact White is better.

The complete game:

[Event "SSDF"]
[Site "Tony Hedlund"]
[Date "2009.10.28"]
[Round "22"]
[White "Deep Rybka 3 Q6600"]
[Black "Deep Shredder 12 Q6600"]
[WhiteElo "3232"]
[ECO "B48"]
[Result "1-0"]

1. e4 {0s} c5 {book 0s} 2. Nf3 {0s (c3)} e6 {book 0s} 3. d4
{0s} cxd4 {book 0s} 4. Nxd4 {0s} Nc6 {book 0s} 5. Nc3 {0s}
Qc7 {book 0s} 6. Be3 {0s (g3)} a6 {book 0s} 7. Qd2 {0s
(Bd3)} Nf6 {book 0s} 8. O-O-O {0s} Bb4 {book 0s} 9. f3 {0s}
Ne5 {book 0s} 10. Nb3 {0s} b5 {book 0s} 11. Qe1 {0s} Rb8
{0.00/20 4:02m} 12. Qg3 {0s} O-O {book 0s} 13. Bd4 {0s} Bd6
{+0.58/22 7:20m} 14. f4 {0s} Nh5 {+0.56/22 3:45m} 15. fxe5
{0s} Nxg3 {+0.49/24 10:13m} 16. exd6 {0s} Qxd6 {+0.50/21
2:15m} 17. hxg3 {0s} Qxg3 {+0.42/20 1:00m} 18. Rh3 {0s}
Qf4+ {+0.49/22 10:29m} 19. Be3 {34:57m} Qe5 {+0.49/24 0s}
20. Bd3 {16:29m (Rf3)} Bb7 {+0.50/22 7:44m} 21. Nc5 {6:13m}
Rfd8 {+0.64/21 0s} 22. Rdh1 {1:49m (Ne2)} h6 {+0.62/22
5:40m} 23. Ne2 {0s} f6 {+0.46/21 4:46m} 24. Rg3 {58s} Kf7
{+0.46/18 15s} 25. Rf1 {11:33m (Rd1)} Qd6 {+0.67/20 6:33m}
26. Nf4 {4:28m} Bc6 {+0.69/20 8:52m} 27. Be2 {0s (c3)} Rbc8
{+0.75/20 5:20m} 28. c3 {3:08m} Qb8 {+0.60/19 47s} 29. Bd4
{3:35m (e5)} e5 {+1.81/19 9:17m} 30. Nxa6 {0s} Qa8
{+1.88/19 3:24m} 31. Bxe5 {2:18m} Bxe4 {+1.63/19 3:55m}
32. Bh5+ {4:38m} Kf8 {+1.61/20 0s} 33. Bd6+ {8:40m} Kg8
{+1.86/19 0s} 34. Nb4 {8:25m} g5 {+1.62/21 0s} 35. Bg6
{3:27m (Rd1)} Rc4 {+2.07/19 8:55m} 36. Bxe4 {2:40m} Rxe4
{+1.99/22 3:21m} 37. Nfd5 {10s} Rxb4 {+2.01/20 2:30m}
38. Nxf6+ {1:15m} Kg7 {+2.12/22 4:00m} 39. Bxb4 {2:09m}
Qxa2 {+2.33/22 0s} 40. Rgf3 {3:51m} d5 {+1.90/20 19s}
41. Be7 {19:40m (Nh5+)} Rc8 {+1.93/21 5:21m} 42. Bd6 {0s
(Nd7)} Qa1+ {+1.82/21 3:49m} 43. Kc2 {2:47m} Qa4+ {+1.82/20
0s} 44. Kb1 {12:03m} Ra8 {+2.19/23 0s} 45. g4 {2:07m} Kg6
{+2.11/21 1:38m} 46. Be5 {6:36m} Qa2+ {+2.16/22 0s} 47. Kc2
{4:34m} Rf8 {+2.91/21 2:19m} 48. Rh1 {2:16m} Qc4 {+2.99/21
5:09m} 49. Kd2 {0s} b4 {+3.27/21 11:08m} 50. Ke3 {0s} d4+
{+3.33/19 8:45m} 51. Bxd4 {39s (cxd4)} Qe6+ {+3.33/19
5:33m} 52. Kf2 {0s} Qb3 {+3.61/20 5:53m} 53. Re1 {0s
(cxb4)} Qxb2+ {+3.07/21 4:32m} 54. Kg3 {0s} Qa2 {+3.35/20
1:23m} 55. Re7 {2:44m (cxb4)} Rf7 {+2.92/19 2:10m} 56. Nd7
{1:01m (Rd7)} Rxf3+ {+2.21/18 1:26m} 57. Kxf3 {0s} Qa8+
{+2.27/18 50s} 58. Kg3 {2:27m (Ke3)} h5 {+4.73/21 1:53m}
59. Ne5+ {18s} Kf6 {+5.61/22 52s} 60. Nc6+ {2:34m} Kg6
{+6.82/24 0s} 61. Rg7+ {20:14m} Kh6 {+8.68/29 0s} 62. Ne7
{13:39m} Qb8+ {+8.84/25 0s} 63. Kg2 {9:13m} 1-0

[Kirk]

In this position:

[d] 1qrr4/3p1kp1/p1b1pp1p/1pN5/4PN2/2P1B1R1/PP2B1P1/2K2R2 w - -

Rybka played Bd4!, which looks bad because it allows .. e5, but in fact White is better.

The complete game:

[Event "SSDF"]
[Site "Tony Hedlund"]
[Date "2009.10.28"]
[Round "22"]
[White "Deep Rybka 3 Q6600"]
[Black "Deep Shredder 12 Q6600"]
[WhiteElo "3232"]
[ECO "B48"]
[Result "1-0"]

1. e4 {0s} c5 {book 0s} 2. Nf3 {0s (c3)} e6 {book 0s} 3. d4
{0s} cxd4 {book 0s} 4. Nxd4 {0s} Nc6 {book 0s} 5. Nc3 {0s}
Qc7 {book 0s} 6. Be3 {0s (g3)} a6 {book 0s} 7. Qd2 {0s
(Bd3)} Nf6 {book 0s} 8. O-O-O {0s} Bb4 {book 0s} 9. f3 {0s}
Ne5 {book 0s} 10. Nb3 {0s} b5 {book 0s} 11. Qe1 {0s} Rb8
{0.00/20 4:02m} 12. Qg3 {0s} O-O {book 0s} 13. Bd4 {0s} Bd6
{+0.58/22 7:20m} 14. f4 {0s} Nh5 {+0.56/22 3:45m} 15. fxe5
{0s} Nxg3 {+0.49/24 10:13m} 16. exd6 {0s} Qxd6 {+0.50/21
2:15m} 17. hxg3 {0s} Qxg3 {+0.42/20 1:00m} 18. Rh3 {0s}
Qf4+ {+0.49/22 10:29m} 19. Be3 {34:57m} Qe5 {+0.49/24 0s}
20. Bd3 {16:29m (Rf3)} Bb7 {+0.50/22 7:44m} 21. Nc5 {6:13m}
Rfd8 {+0.64/21 0s} 22. Rdh1 {1:49m (Ne2)} h6 {+0.62/22
5:40m} 23. Ne2 {0s} f6 {+0.46/21 4:46m} 24. Rg3 {58s} Kf7
{+0.46/18 15s} 25. Rf1 {11:33m (Rd1)} Qd6 {+0.67/20 6:33m}
26. Nf4 {4:28m} Bc6 {+0.69/20 8:52m} 27. Be2 {0s (c3)} Rbc8
{+0.75/20 5:20m} 28. c3 {3:08m} Qb8 {+0.60/19 47s} 29. Bd4
{3:35m (e5)} e5 {+1.81/19 9:17m} 30. Nxa6 {0s} Qa8
{+1.88/19 3:24m} 31. Bxe5 {2:18m} Bxe4 {+1.63/19 3:55m}
32. Bh5+ {4:38m} Kf8 {+1.61/20 0s} 33. Bd6+ {8:40m} Kg8
{+1.86/19 0s} 34. Nb4 {8:25m} g5 {+1.62/21 0s} 35. Bg6
{3:27m (Rd1)} Rc4 {+2.07/19 8:55m} 36. Bxe4 {2:40m} Rxe4
{+1.99/22 3:21m} 37. Nfd5 {10s} Rxb4 {+2.01/20 2:30m}
38. Nxf6+ {1:15m} Kg7 {+2.12/22 4:00m} 39. Bxb4 {2:09m}
Qxa2 {+2.33/22 0s} 40. Rgf3 {3:51m} d5 {+1.90/20 19s}
41. Be7 {19:40m (Nh5+)} Rc8 {+1.93/21 5:21m} 42. Bd6 {0s
(Nd7)} Qa1+ {+1.82/21 3:49m} 43. Kc2 {2:47m} Qa4+ {+1.82/20
0s} 44. Kb1 {12:03m} Ra8 {+2.19/23 0s} 45. g4 {2:07m} Kg6
{+2.11/21 1:38m} 46. Be5 {6:36m} Qa2+ {+2.16/22 0s} 47. Kc2
{4:34m} Rf8 {+2.91/21 2:19m} 48. Rh1 {2:16m} Qc4 {+2.99/21
5:09m} 49. Kd2 {0s} b4 {+3.27/21 11:08m} 50. Ke3 {0s} d4+
{+3.33/19 8:45m} 51. Bxd4 {39s (cxd4)} Qe6+ {+3.33/19
5:33m} 52. Kf2 {0s} Qb3 {+3.61/20 5:53m} 53. Re1 {0s
(cxb4)} Qxb2+ {+3.07/21 4:32m} 54. Kg3 {0s} Qa2 {+3.35/20
1:23m} 55. Re7 {2:44m (cxb4)} Rf7 {+2.92/19 2:10m} 56. Nd7
{1:01m (Rd7)} Rxf3+ {+2.21/18 1:26m} 57. Kxf3 {0s} Qa8+
{+2.27/18 50s} 58. Kg3 {2:27m (Ke3)} h5 {+4.73/21 1:53m}
59. Ne5+ {18s} Kf6 {+5.61/22 52s} 60. Nc6+ {2:34m} Kg6
{+6.82/24 0s} 61. Rg7+ {20:14m} Kh6 {+8.68/29 0s} 62. Ne7
{13:39m} Qb8+ {+8.84/25 0s} 63. Kg2 {9:13m} 1-0

Hiarcs shows that as equal

New game
1qrr4/3p1kp1/p1b1pp1p/1pN5/4PN2/2P1B1R1/PP2B1P1/2K2R2 w - - 0 1

Analysis by HIARCS 11.1 UCI:

1. = (0.05): 1.e5 f5 2.Nxa6 Qb7 3.Nc5 Qa8 4.Bh5+ Kg8 5.a3 g5 6.Ng6 Kf7 7.Rg1 Rg8 8.Nf4+ Ke7
2. = (0.05): 1.Nxa6 Qb7 2.e5 f5 3.Nc5 Qa8 4.Bh5+ Kg8
3. = (0.00): 1.Bd4 e5 2.Nxa6 Qa8 3.Bxe5 Qxa6 4.Bh5+ Kf8 5.Ng6+ Kf7 6.Nf4+
4. = (-0.04): 1.Bg4 b4 2.Ncxe6 Rg8 3.Nc5 bxc3 4.bxc3 Rgf8 5.Bd4
5. = (-0.21): 1.Bh5+ Kg8 2.Be2 Kh7 3.e5 f5 4.Nxa6 Qa8 5.Nb4 g5 6.Nh5 d6 7.Nf6+ Kg7 8.Bb6

[jdart]

Look at Shredder's scores - it apparently failed low after Bd4 and after 9 minutes has a score of +1.8 for White. I think it's the strongest move but not easy to see with a + score (unfortunately we don't have Rybka's scores here, for some reason).

[AdminX]

Rybka played Bd4!, which looks bad because it allows .. e5, but in fact White is better.

True white is better, but If allowed to reach depth 20, it will revert back to 1.e5 +- where white is winning! Sometimes simple is better than Complex ...

[d]1qrr4/3p1kp1/p1b1pp1p/1pN5/4PN2/2P1B1R1/PP2B1P1/2K2R2 w - - 0 1

Analysis by Rybka 3:

1.Bg1 Rf8 2.Rg4 b4 3.Nh5
+/= (0.57) Depth: 6 00:00:00 18kN
+/= (0.42) Depth: 8 00:00:00 90kN

1.e5 f5[] 2.Nxa6 Qb7
+/= (0.49) Depth: 8 00:00:02 288kN

1.Bg4 b4 2.Ncxe6 Bxe4 3.Nxd8+ Rxd8 4.Bd4
+/= (0.60) Depth: 8 00:00:02 315kN

1.e5 f5[] 2.Nxa6 Qb7
+/= (0.49) Depth: 9 00:00:02 369kN
+/= (0.49) Depth: 10 00:00:04 694kN

1.Nh5 Rg8
+/= (0.56) Depth: 10 00:00:06 1042kN

1.e5 f5[] 2.Nxa6 Qb7
+/= (0.49) Depth: 11 00:00:13 2101kN

1.Bd4 e5[] 2.Nxa6 Qa8 3.Bxe5[] Bxe4 4.Bh5+ Kf8 5.Bd6+[] Kg8 6.Nb4[] Qa7 7.Be7 Kh8 8.Bxd8 Rxd8 9.Be2 f5 10.Bd3 Re8 11.Bxe4 fxe4 12.Nfd5 Qb8 13.Re3 Qd6 14.Kc2 Kg8 15.g3 Qe6
+/- (0.76 !) Depth: 11 00:00:17 2706kN
+/- (0.99) Depth: 19 00:39:35 459mN

1.e5 Qxe5 2.Bd4 Qg5 3.Rxg5 hxg5 4.Bh5+ Kg8 5.Nfd3 a5 6.Rf2 Bd5 7.Nb3 a4 8.Nd2 Bxa2 9.Nb4 Bb3 10.Nxb3 axb3 11.Be2 Rb8 12.Na6 Rb7 13.Nc5 Rbb8
+/- (1.22 !) Depth: 19 00:55:58 594mN
+- (2.28) Depth: 21 03:29:26 1934mN

[AdminX]

Here is another variation from a different system.

Analysis by Rybka 3:

1.Bg1 Rf8 2.Nh5
+/= (0.55) Depth: 6 00:00:00 30kN

1.Kb1 b4 2.Nh5 Rg8 3.cxb4 Bb5
+/= (0.59) Depth: 6 00:00:00 51kN
+/= (0.44) Depth: 7 00:00:00 80kN

1.Bg1 Qd6 2.Rg4 Ba8
+/= (0.54) Depth: 7 00:00:00 99kN

1.a3 a5
+/= (0.60) Depth: 7 00:00:00 159kN
+/= (0.41) Depth: 9 00:00:01 360kN

1.Bg1 Qd6 2.Rg4 Rg8 3.Kb1 a5
+/= (0.52) Depth: 9 00:00:02 484kN

1.e5 f5 2.Nxa6
+/= (0.34) Depth: 10 00:00:12 2990kN

1.Bd4 e5 2.Nxa6 Qa8 3.Bxe5[] Bxe4 4.Bh5+ Kf8[] 5.Bd6+[] Kg8 6.Nb4[] Kh8 7.Re1 Bh7 8.Rge3 Qa7 9.Be7 Rg8 10.Bf7 Bf5 11.a3 Qd4 12.Bxg8 Kxg8 13.Nfd5 Kh7 14.Bxf6 gxf6 15.Ne7 Qc5 16.Nxc8
+/- (0.86 !) Depth: 10 00:00:13 3040kN
+/- (1.12) Depth: 19 00:23:08 420mN

1.e5 f5 2.Bh5+[] Kg8 3.Nfd3[] Kf8 4.Bg6 Re8 5.Bxf5[] exf5 6.Rxf5+[] Kg8[] 7.Bxh6[] Re7[] 8.Rfg5[] Rf8 9.Bxg7[] Rf5 10.R5g4 Rxg7 11.Rxg7+ Kh8 12.Nxd7 Bxd7 13.Rxd7 Rxe5 14.Nxe5 Qxe5 15.Rf3 Kg8 16.Rdd3
+/- (1.27 !) Depth: 19 00:29:42 506mN
+- (2.24) Depth: 21 01:31:24 1426mN

[Eelco de Groot]

Go ahead laugh at silly Ancalagon If anyone has a clue what kind of bug I might be looking at please say. I have narrowed it down to somewhere between build 292 and build 382 and probably in search. Maybe the bug is older but just not as frequent in older versions. But I hope it is fairly recent. I do find it strange though how search can just seem to miss an obvious recapture here but otherwise it does not crash the engine. Staring at my sources but there are quite a few changes between these versions. Maybe there are several things at play, I know there where some bugs before the last changes that I had not really looked at, with a similar result so it is not just the latest change that does this.

1qrr4/3p1kp1/p1b1pp1p/1pN5/4PN2/2P1B1R1/PP2B1P1/2K2R2 w - -

Engine: Ancalagon 1.3 Weak Squares 180 Board Control middlegame 50 endgame 50
Build 382 (Athlon 2009 MHz, 256 MB) by Romstad, Costalba, Kiiski, de Groot

2.00 0:00 -2.50 29.Nxa6 (41.349) 110

2.00 0:17 -0.92 29.Bh5+ Kg8 30.Bf3 b4 31.c4 Bb5
32.Be2 Qe5 33.Nh5 Rxc5 34.Rxg7+ Kf8
35.Bxc5+ Qxc5 36.Rxf6+ Ke8 (7.484.020) 416

2.12 0:28 -0.91 29.Bd4 (11.622.787) 412

2.22 0:30 -0.91 29.Nh5 (12.650.742) 412

2.36 0:56 +M4 29.Rxg7+ Kf8 30.Nfxe6+ dxe6 31.Nxe6+ Ke8
32.Bh5+ (23.497.452) 419

3.01 0:57 +M4 29.Rxg7+ Kf8 30.Nfxe6+ dxe6 31.Nxe6+ Ke8
32.Bh5+ (23.880.315) 417

4.01 0:57 +M4 29.Rxg7+ Kf8 30.Nfxe6+ dxe6 31.Nxe6+ Ke8
32.Bh5+ (24.139.630) 416

5.01 1:01 +M4 29.Rxg7+ Kf8 30.Nfxe6+ dxe6 31.Nxe6+ Ke8
32.Bh5+ (25.266.382) 412

6.01 1:13 +M4 29.Rxg7+ Kf8 30.Nfxe6+ dxe6 31.Nxe6+ Ke8
32.Bh5+ (29.521.373) 399

7.01 1:54 +M4 29.Rxg7+ Kf8 30.Nfxe6+ dxe6 31.Nxe6+ Ke8
32.Bh5+ (42.689.461) 374

8.01 48:13 +140.75 29.Rxg7+ Kxg7 (1.199.349.622) 414

9.01 101:36 +108.51 29.Rxg7+ Kxg7 (2.550.653.618) 418

best move: Rg3xg7 time: 268:53.563 min n/s: 425.125 nodes: 6.858.480.254