Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Test position (opposite colour bishop)..very hard for chess engines

Author: Dieter Buerssner

Date: 08:03:33 09/09/05

Go up one level in this thread


On September 08, 2005 at 11:15:45, masros tukiran wrote:

>[D]5k2/5p1K/3p1p2/1B2pP2/P3P2P/8/8/6b1 w - - 0 1
>
>Seem no chess engine can find 1.Kh6!.Almost all top engines(S9,J9,H9,F8) say
>1.a5?? is winning.

The right idea was found (Kh6 delayed after a5 a6 as pointed out by Albert) -
but it really took a long time:


5k2/5p1K/3p1p2/1B2pP2/P3P2P/8/8/6b1 w - -

Engine: Yace via ssh (700 MB)
von Dieter Buerssner

33/57  294:29 +0.98    1.a5 Lf2 2.h5 Le1 3.a6 Lf2 4.h6 La7
                       5.Kh8 Le3 6.Ld3 (1.036.545.689) 58
                       TB:191.095

33/59  446:55 +0.98    1.a5 Lf2 2.h5 Le1 3.a6 Lf2 4.h6 La7
                       5.Kh8 Le3 6.Ld3 Lxh6 7.a7 Ke7 (3.920.047.597) 146
                       TB:345.793

34/59  566:53 +1.38++  1.a5 Lf2 2.a6 Le1 3.a7 d5 (190.678.928) 5
                       TB:468.134

34/59  651:13 +2.38++  1.a5 Lf2 2.a6 Ld4 3.Kh6 d5 4.exd5 Lf2 (2.406.623.183) 61
                       TB:545.394

34/59  667:03 +2.38    1.a5 Lf2 2.a6 (3.575.303.602) 89
                       TB:584.916

My engine threw away a half point against Brutus (now Hydra) in a very similar
position - it moved h5 and prisoned the own king. At the time, I thought about a
solution, but came to the conclusion, that it would be too complicated/not
efficient enough. It probably would decrease the strength.

Not only, that most engines will not understand, that the K is prisoned, the may
even give a bonus (mine does :-(). The passed pawn is given a bonus because it
is supported by the K. Also, in verly late endgame, I give another bonus, when
the K is in front of the passed pawn.

Regards,
Dieter



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.