Author: Will Singleton
Date: 17:22:20 01/22/03
Go up one level in this thread
On January 22, 2003 at 20:19:45, Will Singleton wrote:
>On January 22, 2003 at 15:54:11, Jon Dart wrote:
>
>>Here is a recent test game against Tao 5.4:
>>
>>[Event "Computer chess game"]
>>[Site "SHIVA-DT"]
>>[Date "2003.01.15"]
>>[Round "11"]
>>[White "yin54"]
>>[Black "Arasan version 6.3"]
>>[Result "1-0"]
>>[TimeControl "40/2400"]
>>
>>1. e4 c5 2. c3 d5 3. exd5 Qxd5 4. d4 Nf6 5. Nf3 Bg4 6. Be2 e6 7. Na3 Be7 8.
>>Nb5 Na6 9. c4 Qe4 10. Nc3 Qf5 11. Qa4+ Kf8 12. Be3 Bxf3 13. Bxf3 cxd4 14.
>>Bxd4 Nc5 15. Bxc5 Bxc5 16. O-O Rb8 17. b4 Bd4 18. Rac1 a6 19. b5 Be5 20.
>>Rfd1 Qf4 21. Qb4+ Kg8 22. bxa6 Qxh2+ 23. Kf1 Bxc3 24. Qb6 Ba5 25. Qxa5 bxa6
>>26. Rd8+ Ne8 27. g3 Rxd8 28. Qxd8 g6 29. Qxe8+ Kg7 30. Qa4 Qh3+ 31. Kg1 Qf5
>>32. Qa3 h5 33. c5 h4 34. gxh4 Rxh4 35. c6 Qg5+ 36. Kf1 Qb5+ 37. Be2 Qg5 38.
>>Qb2+ Kh7 39. Bf3 Rb4 40. Qc3 Rf4
>>{Black resigns} 1-0
>>
>>Arasan played 22 .. Qxh2+ and 23 .. Bxc3, both with a score of about -0.25 for
>>Black. At move 25, it failed low and could see it was lost. IMO Black shouldn't
>>take the pawn at move 22, and also shouldn't follow up with 23 .. Bxc3, although
>>it seems Black doesn't have any good alternatives. In these positions, it
>>doesn't seem at all easy for programs to see disaster coming.
>>
>>This is the move 22 position:
>>
>>[D] 1r4kr/1p3ppp/P3pn2/4b3/1QP2q2/2N2B2/P4PPP/2RR2K1 b - -
>>
>>and this is the move 23 one:
>>
>>[D] 1r4kr/1p3ppp/P3pn2/4b3/1QP5/2N2B2/P4PPq/2RR1K2 b - -
>>
>>Crafty was still selecting Qxh2+ after 50 minutes, although it was failing low
>>at the point (depth 14). The score was about +1.1 for White at that point.
>>Crafty also doesn't ever seem to switch off Bxc3, although it got a -2.38 score
>>for Black on ply 14 after 25 minutes. (Arasan also likes Bxc3 for a very long
>>time).
>>
>>How long does your favorite engine take to be really unhappy playing these
>>moves?
>>
>>--Jon
>
>I don't know if these are true am positions, since there aren't alternatives. I
>tried a few programs on the first one, and Tao was the only one that was unhappy
>immediately.
>
>Tao 5.4, amd 1.6ghz:
>
>7 -155 681 269876 f4h2 g1f1 e5c3 b4c3 b7a6 c3a5 h7h6 a5a6
>7. -155 871 374189 f4h2 g1f1 e5c3 b4c3 b7a6 c3a5 h7h6 a5a6
>8 -123 2263 1097896 f4h2 g1f1 e5c3 b4b6 c3d4 b6d4 b7a6 d4b6 f6d5
>8. -123 2584 1275282 f4h2 g1f1 e5c3 b4b6 c3d4 b6d4 b7a6 d4b6 f6d5
>9 -186 6149 3165105 f4h2 g1f1 e5c3 b4e7 c3b4 d1d8 b8d8 e7d8 b4f8 a6b7
>h2h1 f1e2 h1c1 b7b8 c1c4 e2e1
>9. -186 8753 4508404 f4h2 g1f1 e5c3 b4e7 c3b4 d1d8 b8d8 e7d8 b4f8 a6b7
>h2h1 f1e2 h1c1 b7b8 c1c4 e2e1
>10 -186 15573 8018156 f4h2 g1f1 e5c3 b4e7 c3b4 d1d8 b8d8 e7d8 b4f8
>a6b7 h2h1 f1e2 h1c1 b7b8 c1c4 e2e1
>10. -186 17606 9167210 f4h2 g1f1 e5c3 b4e7 c3b4 d1d8 b8d8 e7d8 b4f8
>a6b7 h2h1 f1e2 h1c1 b7b8 c1c4 e2e1
>11 -198 34910 18406458 f4h2 g1f1 e5c3 b4e7 c3b4 d1d8 b8d8 e7d8 b4f8
>a6b7 h2h1 f1e2 h1c1 b7b8 c1c4 e2e1 c4c1 d8d1
>
>
>Amateur does OK here, at least at getting a bad score (18 secs).
>
>Amateur 2.4, amd 1.6ghz:
>
>6 15 54 154304 Qxh2 Kf1 Bxc3 Qxc3 bxa6 Qa3 a5
>7 10 161 476653 Qxh2 Kf1 Bxc3 Rxc3 bxa6 Rd6 Rc8 Rxa6
>8 -29 433 1316573 Qxh2 Kf1 Bxc3 Rxc3 bxa6 Rd6 Rc8 Rd3 Re8 Rxa6
>9 -118 1863 5779425 Qxh2 Kf1 Bxc3 Qe7 Nd5 cxd5 Bf6 Qxb7 Bc3 Qe7
>10 -154 4654 14439576 Qxh2 Kf1 Bxc3 Qe7 Bb4 Rd8 Rxd8 Qxd8 Bf8 axb7 Qh1 Ke2 Qxc1
>Qb8 Qxc4 Ke1 Qc1 Ke2
>
>
>Surprisingly, Tao's pv at ply 9-11 is identical to Amateur's at ply 10.
>
>Will
I just noticed, Tao's time and nodes are all screwed up. I mentioned that to
Bas last weekend, perhaps he can take a look.
Will
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.