Winboard Forum

Hello:

Why some engines like LG Revival ruins its games like this? This was an easy draw for LG, but instead of going for the draw, LG risks its position and finally loses the game. I would understand it, if the engine with the advantage try some risky move to avoid the draw, but the engine who is losing the game?

Look at the move 81. ... d4?? Why not just move the bishop back and forth?

[Event "Germany"]
[Site "CARLOS-V94YZ8V4"]
[Date "2005.10.08"]
[Round "1.3"]
[White "Anaconda 2.0.1"]
[Black "LG Revival"]
[Result "1-0"]
[ECO "B33"]
[PlyCount "203"]
[EventDate "2005.10.08"]
[TimeControl "240+2"]

1. e4 c5 2. Nf3 Nc6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 e5 6. Ndb5 d6 7. Bg5 a6 8. Bxf6 gxf6 9. Na3 b5 10. Nd5 f5 11. c4 Bg7 12. cxb5 Nd4 13. bxa6 h6 14. Nb5 O-O 15. b4 fxe4 16. Nbc7 Ra7 17. Rc1 e3 18. fxe3 Qh4+ 19. Kd2 Qf2+ 20. Be2 Bxa6 21. Nxa6 Rxa6 22. Nc3 Qxg2 23. Qf1 Qxf1 24. Rhxf1 Nxe2 25.Kxe2 e4 26. b5 Ra3 27. Nxe4 Re8 28. b6 f5 29. Rxf5 Rxa2+ 30. Kf3 Rb8 31. Rb1 Rb2 32. Rxb2 Bxb2 33. Rb5 Be5 34. Nf2 Kf8 35. b7 Bxh2 36. Ng4 Be5 37. Nxh6 Ke7 38. Ke4 Ke6 39. Nf5 Rxb7 40. Nd4+ Bxd4 41. Rxb7 Bc5 42. Kd3 d5 43. Rb8 Ke5 44. Rb5 Kd6 45. Ra5 Bb6 46. Ra6 Kc5 47. Kd2 Bc7 48. Re6 Bb6 49. Kc3 Bd8 50. Rg6 Bc7 51. Kd3 Ba5 52. Kc2 Bc7 53. Kc3 Bd8 54. Kb3 Ba5 55. Rf6 Bd8 56. Re6 Bc7 57. Kc2 Bg3 58. Rf6 Bc7 59. Rh6 Be5 60. Kd3 Bg7 61. Re6 Bb2 62. Rg6 Be5 63. Ra6 Bg7 64. Kc2 Be5 65. Rh6 Bg3 66. Kd3 Be5 67. Rg6 Bc7 68. Kc2 Bd6 69. Kd2 Be5 70. Re6 Bc7 71. Kd3 Ba5 72. Kc2 Bc7 73. Rf6 Be5 74. Ra6 Bh2 75. Kd3 Be5 76. Ke2 Bc7 77. Rh6 Be5 78. Rg6 Bc7 79. Rf6 Bd8 80. Rh6 Bc7 81. Kd1 d4 82. Rh5+ Kc4 83. Rh4 Be5 84. Ke2 Bg7 85. Kf3 Be5 86. Rg4 Bf6 87. e4 Be5 88. Rg8 Kc5 89. Rg6 Bh2 90. Ke2 Bd6 91. Rg5+ Kc6 92. Rd5 Bc5 93.
Kd3 Bb6 94. Kc4 d3 95. Rxd3 Bc7 96. Rd5 Bd6 97. e5 Be7 98. Kd4 Bb4 99. e6 Bd6 100. Ke4 Be7 101. Ke5 Ba3 102. Rd7
1-0

Any comments?

Regards,
Carlos

At the start of the game, it is possible that the position is drawn.
Will you feel horrified if one side loses?

In a similar way, a program may still attack from a drawn position.

It is not unlikely for a program to misevaluate some positions, and think it is ahead when it should strive for a draw.

So I have no problems with programs losing in drawn positions. In fact, I rather prefer that to passively taking the draw. Give me crazed pirates over the wall, despite impossible odds, over the handshake any time.

It is clear that in the relevant position it is good to take the draw if there is a draw but I am not sure if there is a draw

I think that you should ask nalimov what is the losing error

Unfortunately I cannot find the tablebases at

http://www.k4it.de/index.php?topic=egtb&lang=en

Anaconda 2.0.1 - LG Revival, Germany CARLOS-V94YZ8V4 2005
8/2b5/7R/2kp4/8/4P3/8/3K4 b - - 0 1

Analysis by Fruit 2.2:

81...Be5
+- (1.50) Depth: 1/5 00:00:00
81...Be5 82.Kd2
+- (1.47) Depth: 2/5 00:00:00
81...Be5 82.Kd2 Kc4 83.Rc6+ Kb5
+- (1.78) Depth: 3/5 00:00:00
81...Bd6 82.Kd2 Be5
+- (1.70) Depth: 3/5 00:00:00
81...Bd6 82.Rh8 Be5 83.Rc8+ Kb5
+- (1.59) Depth: 4/8 00:00:00
81...Be5 82.Rh5 Kd6 83.Ke2
+- (1.48) Depth: 4/9 00:00:00
81...Be5 82.Rh5 Kd6 83.Ke2 Ke6 84.Rh6+ Bf6
+- (1.68) Depth: 5/9 00:00:00
81...d4 82.Rh5+ Kc4 83.Rh4 Be5 84.Kd2
+- (1.58) Depth: 5/11 00:00:00
81...d4 82.Rh5+ Kc4 83.Rh4 Be5 84.Kd2 Kd5
+- (1.72) Depth: 6/13 00:00:00 8kN
81...Be5 82.Rh5 Kd6 83.Ke2 Bf6 84.Rh7
+- (1.64) Depth: 6/13 00:00:00 9kN
81...Be5 82.Rh5 Kd6 83.Ke2 d4 84.e4 Ke6 85.Rh6+ Bf6
+- (1.67) Depth: 7/13 00:00:00 14kN
81...d4 82.Rh5+ Kc4 83.Rh4 Be5 84.Kd2 Kc5 85.e4
? (1.28) Depth: 7/13 00:00:00 23kN
81...d4 82.Rh5+ Kc4 83.Rh4 Be5 84.Kc2 Kd5 85.e4+ Kc4 86.Rh3
? (1.39) Depth: 8/14 00:00:01 31kN
81...d4 82.Rh5+ Kc4 83.Rh4 Be5 84.Kd2 Kc5 85.e4 Kc4 86.Rh3 Bf4+ 87.Ke2
? (1.40) Depth: 9/16 00:00:01 46kN
81...d4 82.Rh5+ Kc4 83.Rh4 Be5 84.Kd2 Kc5 85.e4 Kc4 86.Rh3 Bf4+ 87.Ke2
? (1.40) Depth: 10/17 00:00:01 70kN
81...d4 82.Rh5+ Kc4 83.Rh4 Be5 84.Kd2 Kc5 85.e4 Kc4 86.Rh3 Bd6 87.Rf3
+- (1.41) Depth: 11/18 00:00:01 117kN
81...d4 82.Rh5+ Kc4 83.Rh4 Be5 84.Kd2 Kc5 85.e4 Kc4 86.Rh3 Bf4+ 87.Ke2 Kc5 88.Rb3
+- (1.41) Depth: 12/20 00:00:01 185kN
81...d4 82.Rh5+ Kc4 83.Rh4 Be5 84.Kd2 Bf6 85.Rf4 Be5 86.Rg4 Kd5 87.e4+ Kc4 88.Rh4 d3
? (1.07) Depth: 13/20 00:00:01 363kN
81...d4 82.Rh5+ Kc4 83.Rh4 Be5 84.Ke2 Kc3 85.Rh5 d3+ 86.Kd1 Bg7 87.Rg5 Bf6 88.Rf5 Be7 89.Rf2
? (0.94) Depth: 14/22 00:00:02 580kN
81...d4 82.Rh5+ Kc4 83.Rh4 Be5 84.Ke2 Kc3 85.Rh5 d3+ 86.Kd1 Bg7 87.Rf5 Kc4 88.Rf4+ Kc3 89.Rf2 Be5 90.e4
? (0.97) Depth: 15/24 00:00:02 852kN
81...d4 82.Rh5+ Kc4 83.Rh4 Be5 84.Ke2 Kc3 85.Rh5 d3+ 86.Kd1 Bg7 87.Rc5+ Kb4 88.Rc6 Bc3 89.e4 Be5 90.Kd2 Kb5
? (1.38) Depth: 16/26 00:00:02 1346kN
81...d4 82.Rh5+ Kc4 83.Rh4 Be5 84.Ke2 Kc3 85.Re4 d3+ 86.Kd1 Bd6 87.Re8 Bb4 88.Rc8+ Kb3 89.e4 Ka4 90.Rd8 d2 91.e5 Kb5
+- (1.48) Depth: 17/27 00:00:02 2061kN
81...d4 82.Rh5+ Kc4 83.Rh4 Be5 84.Ke2 Kd5 85.e4+ Kc4 86.Rh3 Bd6 87.Rf3 Be5 88.Rd3 Bf4 89.Ra3 Bd6 90.Rh3 Be5 91.Ra3
+- (1.47) Depth: 18/29 00:00:03 3243kN
81...d4 82.Rh5+ Kc4 83.Rh4 Be5 84.Ke2 Kd5 85.e4+ Kc4 86.Rh3 Bd6 87.Kd2 Bf4+ 88.Kc2 Be3 89.Kd1 Bf4 90.Rf3 Bd6 91.Ke2 Be5 92.Kd2
+- (1.47) Depth: 19/30 00:00:04 4456kN
81...d4 82.Rh5+ Kc4 83.Rh4 Be5 84.Ke2 Kd5 85.e4+ Kc4 86.Rh3 Bd6 87.Kd2 Bf4+ 88.Kc2 Be3 89.Kd1 Bf4 90.Rf3 Bc7 91.Kd2 Ba5+ 92.Kc2 Bc7 93.Ra3
+- (1.48) Depth: 20/31 00:00:04 6190kN
81...d4 82.Rh5+ Kc4 83.Rh4 Be5 84.Ke2 Kd5 85.e4+ Kc4 86.Rh3 Bd6 87.Kd2 Bf4+ 88.Kc2 Be3 89.Kd1 Bf4 90.Rf3 Bc7 91.Kd2 Bd6 92.Ke2 Be5 93.Kd2
+- (1.47) Depth: 21/31 00:00:06 9179kN
81...d4 82.Rh5+ Kc4 83.Rh4 Be5 84.Ke2 Kd5 85.e4+ Kc4 86.Rh3 Bd6 87.Kd2 Bf4+ 88.Kc2 Be3 89.Kd1 Bf4 90.Rf3 Be5 91.Ke2 Bc7 92.Kd2 Ba5+ 93.Kc2 Bc7 94.Rh3
+- (1.48) Depth: 22/35 00:00:12 13427kN
81...d4 82.Rh5+ Kc4 83.Rh4 Be5 84.Ke2 Kd5 85.e4+ Kc4 86.Rh3 Bd6 87.Kd2 Bf4+ 88.Kc2 Be3 89.Kd1 Bf4 90.Rf3 Be5 91.Kd2 Bc7 92.Kc2 Bd6 93.Rd3 Kc5 94.Rb3
+- (1.48) Depth: 23/36 00:00:15 19366kN
81...d4 82.Rh5+ Kc4 83.Rh4 Be5 84.Ke2 Kd5 85.e4+ Kc4 86.Rh3 Bd6 87.Kd2 Bf4+ 88.Kc2 Be3 89.Kd1 Bf4 90.Rf3 Bh2 91.Kd2 Bd6 92.Rd3 Bf4+ 93.Kc2 Bd6 94.Rf3 Be5 95.Kd2
+- (1.47) Depth: 24/39 00:00:24 32965kN
81...d4 82.Rh5+ Kc4 83.Rh4 Be5 84.Ke2 Kd5 85.e4+ Kc4 86.Rh3 Bd6 87.Kd2 Bf4+ 88.Kc2 Be3 89.Kd1 Bf4 90.Rf3 Bd6 91.Kc2 Bb4 92.Rh3 Bd6 93.Rb3 Be5 94.Rd3 Bc7 95.Ra3
+- (1.48) Depth: 25/41 00:00:34 48567kN
81...d4 82.Rh5+ Kc4 83.Rh4 Be5 84.Ke2 Kd5 85.e4+ Kc4 86.Rh3 Bd6 87.Kd2 Bf4+ 88.Kc2 Be3 89.Kd1 Bf4 90.Rf3 Bd6 91.Kc2 Bc5 92.Kd2 Bb4+ 93.Ke2 Bd6 94.Rh3 Bc7 95.Ra3 Be5 96.Rf3
+- (1.47) Depth: 26/42 00:00:50 75544kN
81...d4 82.Rh5+ Kc4 83.Rh4 Be5 84.Ke2 Kd5 85.e4+ Kc4 86.Rh3 Bd6 87.Kd2 Bf4+ 88.Kc2 Be3 89.Kd1 Bf4 90.Rf3 Bd6 91.Kc2 Bc5 92.Kd2 Bd6 93.Rd3 Kc5 94.Ke2 Kc4 95.Rf3 Be5 96.Kd2
+- (1.47) Depth: 27/46 00:01:14 114630kN
81...d4 82.Rh5+ Kc4 83.Rh4 Bb6 84.Kd2 Bc5 85.Kc2 Kd5 86.e4+ Kc4 87.Rh3 Be7 88.Rf3 Bd6 89.Kd2 Bb4+ 90.Ke2 Bc3 91.Rh3 Bb2 92.Kd2 Bc3+ 93.Kc2 Ba5 94.Rf3 Bc7 95.Kd2 Bd6 96.Ke2 Be5 97.Ra3
+- (1.47) Depth: 28/52 00:02:15 211383kN
81...d4 82.Rh5+ Kc4 83.Rh4 Bb6 84.Kd2 Bc5 85.Kc2 Kd5 86.e4+ Kc4 87.Rh3 Be7 88.Rf3 Bc5 89.Kd2 Bd6 90.Rh3 Bf4+ 91.Kc2 Be3 92.Kd1 Bf4 93.Rf3 Bd6 94.Kd2 Bb4+ 95.Ke2 Bd6 96.Rd3 Be7 97.Rd2
+- (1.47) Depth: 29/52 00:03:21 314034kN

(, 18.10.2005)

Crafty analysis with all 3-4-5 man tablebase files:

Analysis from E:\drawn.epd
10/17/2005 4:51:10 PM Level: 20000 Seconds
Analyzing engine: Crafty-1921i

1)
Avoid move:
Best move (Crafty-1921i): d5-d4
Not found in: 5:33:20
15 01:28 1.093.878 12.336 +1.15 1. ... d4 2. e4 d3 3. Rh7 Be5 4. Rd7 Kc4 5. Rd5 Bg3 6. Rg5 Bc7 7. Rf5 Kd4 8. Rf7 Be5
15 01:33 1.259.451 13.409 +1.15 1. ... d4 2. e4 d3 3. Rh7 Be5 4. Rd7 Kc4 5. Rd5 Bg3 6. Rg5 Bc7 7. Rf5 Kd4 8. Rf7 Be5
16 01:45 2.193.625 20.808 +1.17 1. ... d4 2. Rh5+ Kc4 3. e4 d3 4. Rd5 Bg3 5. Kd2 Bf4+ 6. Ke1 Bg3+ 7. Kf1 Bf4 8. Kg2 d2 9. Kf3 Bh6 10. Rd7 Bg5
16 01:54 2.571.521 22.474 +1.17 1. ... d4 2. Rh5+ Kc4 3. e4 d3 4. Rd5 Bg3 5. Kd2 Bf4+ 6. Ke1 Bg3+ 7. Kf1 Bf4 8. Kg2 d2 9. Kf3 Bh6 10. Rd7 Bg5
17 02:23 5.202.092 36.145 +1.28 1. ... d4 2. Rh7 Be5 3. e4 d3 4. Rd7 Kc4 5. Ke1 Kc3 6. Kf2 Bd4+ 7. Kf3 Kc4 8. Rd5 <HT>
17 02:36 6.102.937 38.954 +1.28 1. ... d4 2. Rh7 Be5 3. e4 d3 4. Rd7 Kc4 5. Ke1 Kc3 6. Kf2 Bd4+ 7. Kf3 Kc4 8. Rd5 <HT>
18 02:52 7.954.332 46.066 +1.28 1. ... d4 2. Rh5+ Kc4 3. e4 d3 4. Rd5 Bg3 5. Kd2 Bf4+ 6. Ke1 Bg3+ 7. Kf1 Bh4 8. Kg2 Bf6 9. Kf2 Bd4+ 10. Kf3 d2 11. Ke2 Kc3 12. Kd1
18 03:08 8.976.280 47.513 +1.28 1. ... d4 2. Rh5+ Kc4 3. e4 d3 4. Rd5 Bg3 5. Kd2 Bf4+ 6. Ke1 Bg3+ 7. Kf1 Bh4 8. Kg2 Bf6 9. Kf2 Bd4+ 10. Kf3 d2 11. Ke2 Kc3 12. Kd1
19 03:54 13.683.298 58.370 +1.23 1. ... d4 2. Rh5+ Kc4 3. e4 d3 4. Rd5 Bg3 5. Kd2 Bf4+ 6. Ke1 Bg3+ 7. Kf1 Bh4 8. Kg2 Bf6 9. Kf3 Bd4 10. Rd6 Kc3 <HT>
19 04:17 15.659.970 60.893 +1.23 1. ... d4 2. Rh5+ Kc4 3. e4 d3 4. Rd5 Bg3 5. Kd2 Bf4+ 6. Ke1 Bg3+ 7. Kf1 Bh4 8. Kg2 Bf6 9. Kf3 Bd4 10. Rd6 Kc3 11. Rd7 <HT>
20 04:45 21.249.465 74.449 +1.32 1. ... d4 2. Rh5+ Kc4 3. e4 d3 4. Rd5 Bg3 5. Kd2 Bf4+ 6. Ke1 Bg3+ 7. Kf1 Bh4 8. Kg2 Bf6 9. Kf2 Bd4+ 10. Kf3 Bc3 11. Ke3 d2 12. Rd7 Kc5 13. Rxd2 Bxd2+ 14. Kxd2
20 05:59 28.875.758 80.433 +1.32 1. ... d4 2. Rh5+ Kc4 3. e4 d3 4. Rd5 Bg3 5. Kd2 Bf4+ 6. Ke1 Bg3+ 7. Kf1 Bh4 8. Kg2 Bf6 9. Kf2 Bd4+ 10. Kf3 Bc3 11. Ke3 d2 12. Rd7 Kc5 13. Rxd2 Bxd2+ 14. Kxd2
21 07:24 40.832.030 91.964 +1.25 1. ... d4 2. Rh5+ Kc4 3. e4 d3 4. Rd5 Bg3 5. Kd2 Bf4+ 6. Ke1 Bg3+ 7. Kf1 Bh4 8. Kg2 Bf6 9. Kf2 Bd4+ 10. Kf3 Bg1 11. Kg4 Be3 12. Rd7 d2 13. e5 <HT>
21 08:43 49.560.832 94.762 +1.25 1. ... d4 2. Rh5+ Kc4 3. e4 d3 4. Rd5 Bg3 5. Kd2 Bf4+ 6. Ke1 Bg3+ 7. Kf1 Bh4 8. Kg2 Bf6 9. Kf2 Bd4+ 10. Kf3 Bg1 11. Kg4 Be3 12. Rd7 d2 13. e5 <HT>
22 10:40 77.049.077 120.389 +1.25 1. ... d4 2. Rh5+ Kc4 3. e4 d3 4. Rd5 Bg3 5. Kd2 Bf4+ 6. Ke1 Bg3+ 7. Kf1 Bh4 8. Kg2 Bf6 9. Kf2 Bd4+ 10. Kf3 Bg1 11. Re5 Bc5 12. Rf5 d2 13. Ke2 Be3 14. Kd1
22 14:24 122.922.790 142.271 +1.25 1. ... d4 2. Rh5+ Kc4 3. e4 d3 4. Rd5 Bg3 5. Kd2 Bf4+ 6. Ke1 Bg3+ 7. Kf1 Bh4 8. Kg2 Bf6 9. Kf2 Bd4+ 10. Kf3 Bg1 11. Re5 Bc5 12. Rf5 d2 13. Ke2 Be3 14. Kd1
23 26:32 501.996.075 315.324 +1.39 1. ... d4 2. Rh7 Be5 3. e4 d3 4. Rd7 Kc4 5. Ke1 Bc3+ 6. Kf2 Bd4+ 7. Kf3 Bf6 8. Ke3 Bg5+ 9. Kf2 Bf6 10. Kf3 Bd4 11. Rc7+ Kb3 <HT>
23 41:14 1.324.822.218 535.498 +1.39 1. ... d4 2. Rh7 Be5 3. e4 d3 4. Rd7 Kc4 5. Ke1 Bc3+ 6. Kf2 Bd4+ 7. Kf3 Bf6 8. Ke3 Bg5+ 9. Kf2 Bf6 10. Kf3 Bd4 11. Rc7+ Kb3 <HT>
24 56:51 2.458.272,212 720 +1.39 1. ... d4 2. Rh7 Be5 3. e4 d3 4. Rd7 Kc4 5. Ke1 Bc3+ 6. Kf2 Bd4+ 7. Kf3 Kc3 8. Rd6 Kc4 9. Rd5 Bc5 10. Rd7 Bd4 <HT>
24 1:27:28 4.644.367,367 884 +1.39 1. ... d4 2. Rh7 Be5 3. e4 d3 4. Rd7 Kc4 5. Ke1 Bc3+ 6. Kf2 Bd4+ 7. Kf3 Kc3 8. Rd6 Kc4 9. Rd5 Bc5 10. Rd7 Bd4 <HT>
10/17/2005 6:59:35 PM, Time for this analysis: 02:08:20, Rated time: 5:33:20

0 of 1 matching moves
10/17/2005 6:59:37 PM, Total time: 2:08:26 AM
Rated time: 5:33:20 = 20000 Seconds

analyzing this game I find that black could draw after 81...d4 by 83...Kd3

White could play 83.e4 instead of 83.Rh4 but I am not sure if white could win by that move

I do not have the KRP vs KBP tablebases so I do not know exactly what mistakes the sides did.

It is clear that white had no plan in the game but I am not sure that there was no good plan against bishop moves and we need tablebases to be sure.

Uri

Hi guys:

Very interesting analysis. Thanks both for your time. It would be interesting if someone with the 6-men EGTBs could tell us if the position is a draw.

Best regards,
Carlos

I didn't have any problem getting a draw against the computer by following your ideas, so I suspect you are right.

My guess for the reason d4 is preferred by the computer programs is that they are following a heuristic of trying to get all the pawns off the board if they are behind, and it just doesn't work here.

If it is not generally a good idea to get the pawns off the board when you are behind, and I really wouldn't know, then I'm wrong.

Toga II likes d4 at until it changes to Be5 at depth 32, and the evaluation changes to a drawish -0.28.