Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Crafty and EGTB Question

Author: Mike Byrne

Date: 18:04:07 06/16/04

Go up one level in this thread


On June 16, 2004 at 20:52:37, Ed Trice wrote:

>Hi Mike,
>
>I am familiar with the exit and egtb line items in the .rc file. However, the
>position I am tested cannot be played properly unless the R + P vs. R database
>is PROBED IN RAM during the search.
>
>Even strong programs can search for a sufficient length of time and play the
>incorrect move, unknowingly giving away the draw from a won position.
>
>The idea is not to just play perfectly once in a theoretical win, but play
>perfectly in a postion where a pair of pawns must come off of the board, but the
>correct exchange is not obvious to a program.
>
>I have a 7 piece positon, R + 2P vs. R + P, and a deep search with the 5-piece
>database of R + P vs. R should solve it from the parent position if the EGTBs
>are being probed properly.

please post position in FEN format , I will let you know what I get ...



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.