Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Endgame trap; some progs see it - How is it found? e.g. Crafty?

Author: Robert Hyatt

Date: 21:26:34 05/28/99

Go up one level in this thread


On May 28, 1999 at 17:32:54, Paul Richards wrote:

>On May 27, 1999 at 20:38:48, Robert Hyatt wrote:
>
>>2.  if not, all I get for any K move, from ply 1 on is 0.00...  ie Kxh6 is a
>>draw and it doesn't vary.
>
>Actually I note the same thing, but only up until about 15 ply (which is
>only a few seconds for this position).  The eval for Kxh6 is 0.00, but
>the position is not actually a draw after Kxh6, because after white goes
>Kf6 and black's king is cut off from the queening square and will
>eventually be forced away from the rook file, and so the pawn will promote.
>Crafty realizes this around 15 ply and gives it a +12, until it then sees
>Kg8!! and a definite draw.  I'm curious as to why it gives Kxh6 a
>draw at first.  In any case all of this happens very quickly and
>Crafty handles this position well, unlike poor Fritz. :)


the only problem I see in the code (which has been there for a long while) is
that it is remotely possible that the eval can conclude that the king can reach
the queening square when it might not quite be able to do so, assuming the
opposing bishop/king are positioned correctly...

I'll check...



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.