Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: latest Arasan test suite

Author: Jon Dart

Date: 08:43:12 05/16/03

Go up one level in this thread


I think most are tactical, in the sense that the best move wins material or
produces some other clear advantage. But a program with material only eval
wouldn't likely do well. Some positions require some endgame knowledge to
evaluate accurately, for example. And tactical problems are often solved much
faster if the program understands something about king safety.

--Jon

On May 16, 2003 at 11:11:07, Uri Blass wrote:

>On May 16, 2003 at 10:06:32, Jon Dart wrote:
>
>>Following is a test suite of positions from recent Arasan games (there is an
>>another, older collection ("arasan2.epd"), available at
>>http://www.arasanchess.org/tests.zip).
>>
>>This one is a bit harder overall than the arasan2 test suite. Currently Arasan
>>itself gets around 32-35 right at 60 seconds per move on a fast machine
>>(XP2500+).
>
>Are all the position tactical position or there are some cases when the reason
>for the right move is positional?
>
>In other words do you expect programs with only material evaluation to solve all
>of them if they search deep enough?
>
>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.