Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Triple a-pawn and wrong bishop

Author: Robert Hyatt

Date: 06:37:55 03/29/00

Go up one level in this thread


On March 28, 2000 at 23:49:23, Kai Skibbe wrote:

>>Look at the last PV.  Bishop is getting taken on c7.  Leaving a k+rookpawn vs
>>king.
>
>Hi Bob,
>
>why don“t you have code for recognizing king and rookpawns against king ? When I
>implement the wrong bishop in Gromit I realized that you have to implement it,
>because otherwise you have no chance to check the other code with the wrong
>bishop on board.
>
>Kai


It actually does, after I looked at it.  The problem was that I had some
qualifiers that could be subtly defeated.  I require that the lone king
by at least on the same rank as the rook pawn, and within 1 file.  In this
case, 1 file wasn't enough as after the Bc7 check, KxB leaves the king two
files away.  Which still means the pawn can't possibly promote unless it is
at a7.  I fixed this one line and the problem went away although there is still
an issue of an occasional fail-high but no PV when searching this to very deep
levels...



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.