Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: endgame position, score +8 "easy" win, and spoiled to draw by engines

Author: Dieter Buerssner

Date: 11:08:28 09/16/02

Go up one level in this thread


On September 16, 2002 at 03:15:24, Peter McKenzie wrote:

>Maybe I will use one of these for PET version 2, is any one more suitable than
>the other do you think?

I fear, none of them are really good test positions for a test suite. An engine
should proof, that it can win this position vs. any defense - something not so
easy. In the first positions, the engine may do the correct move, and still have
NxB in the line later (all just transpositions). Perhaps the last position,
formulated as an am position will trick many engines to move NxB.

8/4PK2/6p1/5pB1/5P1P/5n2/4k3/3q4 b - - am Nxg5;

I am not sure, if other moves as Qd7 win, too.

Also, if my analysis is correct, and you assume, that an engine will not go
through an repetition:

5K2/3qP3/5Bp1/5p2/5P1P/4kn2/8/8 b - - bm Kxf4;

Cheers,
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.