Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: A proposed WAC replacement for testing

Author: Dann Corbit

Date: 17:11:09 09/18/01

Go up one level in this thread


On September 18, 2001 at 17:15:34, Gian-Carlo Pascutto wrote:

>On September 18, 2001 at 17:09:08, Jon Dart wrote:
>
>>There is already a corrected version of ECM (ECM98) that
>>has been analyzed in detail. It still contains some easy
>>problems but many "cooked" positions have been eliminated
>>and solutions corrected.
>>
>>I'd recommend starting with this rather than the uncorrected
>>ECM if you are looking for a smaller test suite.
>
>I started from ECMF5.EPD, which already is a corrected version.
>
>But even then, there were some positions with questionable
>solutions, which I've thrown out. There might still be left,
>but there certainly shouldn't be much...

I am guessing that at least 10% will have other solutions and at least 5% will
have wrong solutions.

This is a very interesting and fertile ground for investigation.  If only we all
had Slate's monster machines!
;-)



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.