Author: Koundinya Veluri
Date: 12:53:12 10/27/01
[D]7n/7P/8/8/Q7/2qN4/7K/1k6 w - - 0 1
Hi,
I noticed that from this position, Crafty (along with several other programs,
mine included), doesn't see the very simple win 1. Qb4+! by eval. By that I
mean, given enough time, it will find that move once it searches far enough to
see the promotion, but it doesn't recognize that trading queens here is a good
idea. Just to see if it would win in a blitz match without TB's, I set up a
blitz 40/5 game starting from this position, Crafty vs Yace (Yace sees that move
immediately by eval). The game ended in a draw!! I don't know if it missed any
other chances to win, but the first move definitely should have been Qb4+ I
think. It's the simplest win.
As the game progressed, Crafty's eval actually increased while Yace's decreased.
Perhaps it's not a big deal, since this is very easy when using endgame
tablebases, but IMHO it should be easily solvable without TB's.
[Event "Computer chess game"]
[Site "KOUNT"]
[Date "2001.10.27"]
[Round "-"]
[White "Crafty-18.12"]
[Black "Yace 0.99.56"]
[Result "1/2-1/2"]
[TimeControl "40/300"]
[FEN "7n/7P/8/8/Q7/2qN4/7K/1k6 w - - 0 1"]
[SetUp "1"]
{--------------
. . . . . . . n
. . . . . . . P
. . . . . . . .
. . . . . . . .
Q . . . . . . .
. . q N . . . .
. . . . . . . K
. k . . . . . .
white to play
--------------}
1. Qe4 Kc2 2. Nf4+ Kb3 3. Kg2 Qg7+ 4. Kf3 Qf8 5. Qd3+ Kb2 6. Qd4+ Ka2 7.
Kg4 Qc8+ 8. Kg3 Qb8 9. Qd5+ Ka3 10. Kg4 Qa7 11. Qd6+ Kb3 12. Qe6+ Ka3 13.
Ng6 Qxh7 14. Qd6+ Kb3 15. Qb8+ Kc2 16. Nxh8 Qe4+ 17. Kg3 Qe3+ 18. Kg4 Qe4+
19. Kh5 Qf3+ 20. Kh6 Qe3+ 21. Kg6 Qd3+ 22. Kg5 Qe3+ 23. Kh4 Qe4+ 24. Kg5
Qe3+ 25. Kf5 Qf3+ 26. Kg6 Qd3+ 27. Kg7 Qc3+ 28. Kg8 Qb3+ 29. Qxb3+ Kxb3
{Draw} 1/2-1/2
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.