Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Misevaluation in endgame by Gambit Tiger 1.0

Author: Uri Blass

Date: 09:58:02 02/11/01

Go up one level in this thread


On February 11, 2001 at 12:45:07, Alex Patterson wrote:

>On February 11, 2001 at 12:24:21, Vincent Lejeune wrote:
>
>>in this position :
>>
>>[d]k7/p7/1p6/3p4/3K1B2/r7/8/4R3 w - - 0 1
>>
>>Gambit Tiger 1.0 gives +2.14 for white , I think it's a big overevaluation for
>>white !
>>
>>
>>
>>consequence : in this position
>>
>>[d]k7/p7/1p6/3p4/3K1p2/PP3r2/7B/4R3 b - - 0 1
>>
>>Tiger 1.0 wants to play Rf2??
>
>
>So? Rf2! is the best move, and that is backed up by crafty, TCB, and HIARCS.
>Anyway I dont see the problem with Rf2!.

Rf2 is not the best move of Crafty18.1
It prefers Rxb3 with evaluation of 0.00 at depth 15

After Rf2 Crafty can see a clear advantage for white (1.54 pawns at depth 13 and
1.68 at depth 14)
The problem is that tiger does not know that rook and a bishop with no pawns
usually cannot win.

It is a known problem of tiger and it is not news.

Uri






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.