Author: Omid David Tabibi
Date: 00:36:47 12/12/03
Go up one level in this thread
On December 11, 2003 at 20:15:24, Aaron Gordon wrote:
>First of all, I'd like to say having EGTBs fixes this problem. However, take a
>look at what happens when Shredder 7.04 isn't using any EGTBs.
>
>Here is the game:
>
>[Event "ICS unrated blitz match"]
>[Site "freechess.org"]
>[Date "2003.12.10"]
>[Round "-"]
>[White "ShredderAMD"]
>[Black "FritzAMD"]
>[Result "1/2-1/2"]
>[WhiteElo "0"]
>[BlackElo "0"]
>[TimeControl "300+3"]
>
>1. e4 c5 2. Nf3 e6 3. Nc3 a6 4. d4 cxd4 5. Nxd4 Qc7 6. Be2 b5 7. O-O Bb7 8.
>Re1 b4 9. Na4 Bxe4 10. Bf3 Bb7 11. g3 Nc6 12. Bf4 Qa5 13. Nxc6 Bxc6 14.
>Bxc6 dxc6 15. Re5 Qd8 16. Qh5 g6 17. Qe2 Ne7 18. Rd1 Nd5 19. c4 bxc3 20.
>Nxc3 Be7 21. Bh6 Qd7 22. Qf3 Rg8 23. Ree1 g5 24. Qh5 Nf6 25. Qe2 Qc7 26.
>Kh1 Nd5 27. Qh5 Rg6 28. Ne4 g4 29. Bd2 O-O-O 30. Kg2 h6 31. Kg1 Kb7 32. b4
>Qb6 33. a4 Nf6 34. Qe5 Rd5 35. Qb2 Nxe4 36. Rxe4 Qd8 37. b5 cxb5 38. axb5
>axb5 39. Rb1 Qd7 40. Bf4 Rd1+ 41. Re1 Rxe1+ 42. Rxe1 Rg5 43. Bxg5 Bxg5 44.
>Qe2 h5 45. Qe5 Be7 46. Qxh5 Bb4 47. Rb1 f5 48. h4 Bc5 49. Re1 Qd4 50. Qh7+
>Kc8 51. Rf1 Qe5 52. Kh2 b4 53. Rc1 Qd4 54. Qg8+ Kd7 55. Qf7+ Be7 56. Kg1 b3
>57. Re1 Qd5 58. Qg7 Ke8 59. Rc1 Bc5 60. Qc3 Bf8 61. Rb1 Kf7 62. Qxb3 Qxb3
>63. Rxb3 Bc5 64. Rc3 Bb4 65. Rc7+ Kf6 66. Kg2 Ba5 67. Rc2 Bb4 68. f3 gxf3+
>69. Kxf3 Bf8 70. g4 fxg4+ 71. Kxg4 Bh6 72. Rf2+ Kg7 73. h5 Bc1 74. Kh4 Bh6
>75. Kg3 Bg5 76. Kf3 e5 77. Rc2 Kh6 78. Kg4 Be3 79. Rc6+ Kh7 80. Kf5 Bf4 81.
>Ra6 Bh2 82. Kg4 Bf4 83. Ra7+ Kg8 84. Kf5 Be3 85. Rb7 Kf8 86. Kxe5 Ke8 87.
>Kf5 Kd8 88. Rg7 Ke8 89. Kg6 Bd4 90. h6 Kd8 91. h7 Bxg7 92. Kxg7 Kc7 93.
>h8=Q Kb6 94. Qc8 Kb5 95. Qg4 Kc5 96. Kf6 Kb5 97. Ke6 Ka5 98. Qc4 Kb6 99.
>Qb4+ Kc6 100. Qb1 Kc5 101. Qe4 Kb6 102. Qb4+ Kc6 103. Ke7 Kd5 104. Kf6 Kc6
>105. Ke6
>{Game drawn by mutual agreement} 1/2-1/2
>
>As you can see, this is *NO* draw. What I'd like you guys to see is this
>particular position (Remember, when analyzing w/ S7.04 disable EGTBs):
>
>[D]8/2k3KP/8/8/8/8/8/8 w - - 0 93
>
>Play h8=Q and let shredder think on that a bit... :)
>
>
>Analysis by Shredder 7.04:
>
>93...Kd6
> +- (12.30) Depth: 1/1 00:00:00
>93...Kd6
> +- (12.30) Depth: 1/1 00:00:00
>93...Kd6 94.Qc8
> +- (12.84) Depth: 2/3 00:00:00
>93...Kd6 94.Qe8
> +- (12.98) Depth: 2/3 00:00:00
>93...Kd6 94.Qf8+ Kd5 95.Qe7
> +- (12.90) Depth: 3/5 00:00:00
>93...Kd6 94.Qh5 Kc6 95.Qa5
> +- (13.11) Depth: 4/11 00:00:00
>93...Kc6 94.Qh5
> +- (13.10) Depth: 4/11 00:00:00
>93...Kc6 94.Qh5 Kb6 95.Qd5
> +- (13.10) Depth: 4/11 00:00:00
>93...Kc6 94.Qc8+ Kd5 95.Kf6 Kd4 96.Qe6
> +- (13.34) Depth: 5/13 00:00:00
>93...Kd6 94.Qd8+ Kc5 95.Qe7+ Kc4 96.Qe6+ Kb4 97.Qd6+ Ka4 98.Qa6+ Kb3 99.Qb7+ Ka3
> +- (13.32) Depth: 5/13 00:00:00
>93...Kd6 94.Qh2+ Kc5 95.Kf6 Kc4 96.Qb2
> +- (13.12) Depth: 5/13 00:00:00
>93...Kd6 94.Kf7 Kc6 95.Qc3+ Kd5 96.Kf6 Kd6
> +- (13.38) Depth: 6/15 00:00:00
>93...Kd6 94.Kf7 Kd5 95.Qc3 Ke4 96.Ke6
> +- (13.44) Depth: 6/15 00:00:00
>93...Kd6 94.Kf7 Kd5 95.Qc3 Ke4 96.Kf6 Kd5 97.Qe3
> +- (13.48) Depth: 7/17 00:00:00 53kN
>93...Kd6 94.Kf7 Kd5 95.Qc3 Kd6 96.Kf6 Kd5 97.Qd3+ Kc6 98.Ke5
> +- (13.51) Depth: 8/19 00:00:00 89kN
>93...Kd6 94.Kf7 Kd5 95.Qc3 Ke4 96.Ke6 Kf4 97.Qe5+ Kg4 98.Qf6
> +- (13.67) Depth: 9/21 00:00:00 256kN
>93...Kd6 94.Kf7 Kd5 95.Qc3 Ke4 96.Ke6 Kf4 97.Qd3 Kg4 98.Qe4+ Kh5 99.Qf3+ Kh6
>100.Qf4+ Kh5
> +- (13.75) Depth: 10/23 00:00:00 380kN
>93...Kd6 94.Kf7 Kd5 95.Qc3 Ke4 96.Ke6 Kf4 97.Qh3 Ke4 98.Qg4+ Kd3 99.Kd5 Kd2
> +- (13.78) Depth: 11/25 00:00:00 584kN
>93...Kd6 94.Kf6 Kd5 95.Kf5 Kc4 96.Ke4 Kb3 97.Qd4 Ka2 98.Qd3 Kb2 99.Kd5 Ka2
>100.Qc2+ Ka3 101.Qc3+ Ka2 102.Qb4 Ka1 103.Qa3+ Kb1 104.Qc3 Ka2 105.Qb4 Ka1
>106.Qa3+ Kb1 107.Qc3 Ka2
> +- (13.87) Depth: 12/27 00:00:03 3186kN
>93...Kd6 94.Kf6 Kd5 95.Qa8+ Kd4 96.Qa4+ Ke3 97.Ke5 Kf2 98.Qa3 Kf1 99.Qf3+ Kg1
> +- (14.11) Depth: 13/27 00:00:03 3643kN
>93...Kd6 94.Kf6 Kd5 95.Qa8+ Kd4 96.Qa4+ Ke3 97.Ke5 Kd3 98.Qe4+ Kc3 99.Qe3+ Kb4
>100.Qd4+ Kb3 101.Qd3+ Ka4 102.Qc4+ Ka5 103.Kd5 Kb6
> +- (14.12) Depth: 14/29 00:00:04 4080kN
>93...Kd6 94.Kf6 Kd5 95.Qa8+ Kc4 96.Qe4+ Kb3 97.Qd3+ Ka4 98.Qc4+ Ka3 99.Qc3+ Ka4
> +- (14.37) Depth: 15/29 00:00:04 4283kN
>93...Kd6 94.Kf6 Kd5 95.Qa8+ Kc5 96.Qa4 Kd6 97.Qd1+ Kc5 98.Ke6 Kc4 99.Qa4+ Kd3
>100.Ke5 Kc3 101.Qe4 Kb3 102.Qd3+ Kb2 103.Kd4 Ka2
> +- (#29) Depth: 15/29 00:00:04 4420kN
>93...Kd6 94.Qh5 Kc6 95.Kf6 Kd6 96.Qe5+ Kd7 97.Qc5 Kd8 98.Ke6 Ke8 99.Qc7 Kf8
>100.Kd6 Kg8 101.Qc4+ Kf8 102.Qc8+ Kg7 103.Qc3+ Kh6 104.Qe3+ Kg6
> +- (#2) Depth: 15/29 00:00:10 10488kN
>93...Kd6 94.Qh5 Kc6 95.Kf7 Kd6 96.Qf5 Kc7 97.Ke6
> -+ (-#2) Depth: 16/29 00:00:13 13343kN
>93...Kd6 94.Qh5 Kc6 95.Kf6 Kc7 96.Ke6 Kb6 97.Qd5 Kc7 98.Qb5 Kd8 99.Kd6 Kc8
>100.Qb4 Kd8 101.Qa4 Kc8 102.Qd7+ Kb8 103.Ke6 Ka8 104.Qb5 Ka7 105.Qb3 Ka6 106.Qb4
>Ka7 107.Kd5 Ka8
> -+ (-#13) Depth: 17/31 00:00:16 16668kN
>93...Kd6 94.Qh5 Kc6 95.Kf6 Kc7 96.Qc5+ Kb8 97.Ke6 Kb7 98.Kd6 Kb8 99.Qb4+ Ka7
> -+ (-#25) Depth: 18/29 00:00:17 17507kN
>93...Kd6 94.Qh5 Kc6 95.Kf6 Kc7 96.Qc5+ Kb8 97.Ke6 Kb7 98.Kd6 Ka6 99.Qb4 Ka7
>100.Qa5+ Kb8 101.Qd5
> -+ (-#32) Depth: 18/32 00:00:18 19145kN
>93...Kd6 94.Qh5 Kc6 95.Kf6 Kc7 96.Qc5+ Kb8 97.Ke6 Kb7 98.Kd5 Ka8 99.Qb4 Ka7
>100.Qb5 Ka8 101.Kd6 Ka7 102.Qa4+ Kb8 103.Qb3+ Kc8 104.Qb4 Kd8 105.Qa4 Kc8
>106.Qd7+ Kb8 107.Ke6 Ka8
> -+ (-#45) Depth: 19/33 00:00:20 21040kN
>93...Kd6 94.Qh5 Kc6 95.Kf6 Kb6 96.Ke6 Ka6 97.Kd6 Kb6 98.Qe5 Ka7 99.Qa5+ Kb8
>100.Qc5 Kb7 101.Qb5+ Kc8 102.Qb4 Kd8 103.Qc3 Ke8 104.Qg7 Kd8 105.Qe7+ Kc8
>106.Qe6+ Kb7 107.Qe5 Kb6
> -+ (-#53) Depth: 19/35 00:00:25 25638kN
>93...Kd6 94.Qh5 Kc6 95.Kf7 Kb7 96.Qb5+ Kc8 97.Qb4 Kc7 98.Qc5+ Kd7 99.Kf6 Ke8
>100.Ke6 Kd8 101.Qe7+ Kc8 102.Kf6 Kb8 103.Qd7 Ka8 104.Qb5 Ka7 105.Ke6 Ka8 106.Kf7
>Ka7 107.Ke6 Ka8
> -+ (-#66) Depth: 20/34 00:00:26 27210kN
>93...Kd6 94.Qh5 Kc6 95.Kg6 Kc7 96.Qc5+ Kb7 97.Qb5+ Kc8 98.Kf6 Kc7 99.Qc5+ Kd8
>100.Qe7+ Kc8 101.Ke6 Kb8 102.Qd7 Ka8 103.Qb5 Ka7 104.Kd5 Ka8 105.Kd6 Ka7 106.Ke6
>Ka8 107.Kd6 Ka7
> -+ (-#67) Depth: 20/37 00:00:41 42271kN
>93...Kd6 94.Qh5 Kc6 95.Kg6 Kb6 96.Qd5 Kc7 97.Qc5+ Kb7 98.Kf6 Kb8 99.Ke6 Kb7
>100.Kd6 Ka6 101.Qd5 Ka7 102.Qb5 Ka8 103.Ke6 Ka7 104.Kf7 Ka8 105.Ke6 Ka7 106.Kf7
>Ka8 107.Ke6 Ka7
> -+ (-#79) Depth: 21/31 00:00:46 47048kN
>
>
>Hit space bar and play through the moves for a while.. you'll see white will get
>into a position to play a mate in 1 but does NOT play it! If you clear the hash
>and then search it'll find the mate in 1. I'm thinking it is a problem with
>hashtable collisions or something of the sort.
I had similar problems in Falcon. I fixed the problem by storing the checkmate
score as a bound only. Not an efficient solution of course, but haven't
encountered any such bugs ever since.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.