Author: Jon Dart
Date: 12:01:29 11/13/05
This game had an interesting ending:
[Event "?"]
[Site "chessclub.com"]
[Date "2005.11.12"]
[Round "?"]
[White "Arasan 9.0"]
[Black "thebaron"]
[Result "0-1"]
[ECO "B42"]
[WhiteElo "2599"]
[BlackElo "2506"]
[TimeControl "1200+10"]
1. e4 c5 2. Nf3 e6 3. d4 cxd4 4. Nxd4 a6 5. Bd3 Nf6 6. O-O Nc6
7. Nxc6 dxc6 8. Bf4 Bd6 9. e5 Nd5 10. Qg4 Nxf4 11. Qxg7 Rf8 12. exd6
Nxd3 13. cxd3 Qxd6 14. Nc3 f5 15. Qxh7 Bd7 16. Qh5+ Kd8 17. Rfe1 Kc7
18. Qe2 c5 19. Rad1 Rh8 20. Qe5 Rh4 21. Qxd6+ Kxd6 22. g3 Rg8 23. Kg2
b5 24. f3 Bc6 25. a3 e5 26. Rd2 Rf4 27. Re3 b4 28. axb4 cxb4 29. Nd1
a5 30. Rde2 Rd4 31. Nf2 Re8 32. Nh3 a4 33. Ng5 Bd5 34. Kf2 Ra8
35. Re1 a3 36. Rxe5 a2 37. R5e2 Rxd3 38. Ra1 b3 39. h3 Rc8 40. Kf1
Rc2 41. Ke1 Rdc3 42. Rd1 0-1 {ArasanX resigns}
In this position, White is already in trouble, but 36. Rxe5 looks
like a mistake. Chess Tiger avoids it immediately, and gives a
+2 or so plus score for Black after forcing Rxe5. But several other
programs, including mine, want to play it.
[D] r7/8/3k4/3bppN1/1p1r4/p2PRPP1/1P3K1P/4R3 w - -
At move 38, we have this position for Black:
[D] r7/8/3k4/3b1pN1/1p6/3r1PP1/pP2RK1P/R7 b - -
Here b3 is indeed the best move, but Arasan doesn't find it
readily. This is my lastest build in multi-variation mode,
1 hour per move:
result: Kd7 3600 seconds. score: +1.38 2038387785 nodes. ** error
Kd7 Kg1 Bb3 Kf2 Rd8 Ree1 Rd2+ Ke3 Rxh2 Kf4 Rf8 Re5 Rd2 Rb5 Rd4+ Ke5 Rc4 Ne6
result(2): Rc8 3600 seconds. score: +2.22 2000296738 nodes.
** error
Rc8 b3 Bxb3 g4 Rg8 h4 fxg4 fxg4 Rf8+ Ke1 Rd4 Ne4+ Kc6 Nf2 Bc4 Raxa2 Bxa2 Rxa2 Rh
8 Rd2 Rxd2
result(3): b3 3600 seconds. score: +3.00 2030428316 nodes.
++ correct
b3 h4 Rc8 Ke1 Bxf3 Nxf3 Rxf3 Rd2+ Kc5 Rc1+ Kb6 Rd6+ Ka7 Rd7+ Kb8 Rcd1 Rxg3 Kf1 R
h3 Kg1 Rxh4
So it actually had to be forced to eliminate Kd7 and Rc8 before
selecting b3. I am still looking into the reason. Chess Tiger and
Pharaon play b3 very quickly here, but Crafty 19.17 prefers Rg8 with
an even score, at least for several minutes.
--Jon
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.