Author: Sune Fischer
Date: 08:52:53 09/03/02
Go up one level in this thread
On September 03, 2002 at 11:25:04, Robert Hyatt wrote: >On September 03, 2002 at 10:42:31, Eduard Nemeth wrote: > >>[D]2b3k1/1p1p1p1p/1P1P1P1P/8/8/8/6P1/K4B2 w - - 0 1 >> >>when will be black mates (in how many moves) ? > >No idea. Crafty says Bd3 is winning big-time. +23 is the score: > > > 24-> 7:18 22.80 1. Bd3 Kh8 2. g4 Kg8 3. g5 Kh8 4. g6 > hxg6 5. Bxg6 Kg8 6. Bxf7+ Kh7 7. Bd5 > Kxh6 8. Be4 Kg5 9. f7 Kf4 10. f8=Q+ > Ke5 11. Bd5 Kd4 12. Qxc8 Ke5 13. Qxd7 > Kd4 14. Qxb7 Kd3 > > >It has stuck with that move since very early searches, with the evaluation going >higher each successive iteration... I like your pv better than mine: 24 1004 78907 527798889 1.Be2 Kf8 2.g3 Kg8 3.g4 Kf8 4.Kb2 Kg8 5.Bd3 Kh8 6.Ka1 Kg8 7.g5 Kh8 8.g6 hxg6 9.Bxg6 Kg8 10.Bxf7+ Kxf7 11.h7 Kxf6 12.h8=Q+ Kg6 13.Qxc8 I don't understand why white is wasting so many moves (g2-g3 and later g3-g4, Be2 and later Bd3, Kb2 and then back with Ka1), that's four moves wasted for nothing. Do I have a bug or are you doing something special to 'optimize' the pv? Any idea what could cause this behavior? Another perhaps related question; When frenzee is about to lose material, eg. a knight it will often sacrifice a pawn first, that way the loss of the knight is pushed out of search reach. That doesn't improve the position obviously, is there any way clever way to avoid it, like extending the search along these lines? thanks, -S.
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.