Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Swiss Tournament (25min+5s)

Author: Kurt Utzinger

Date: 05:18:13 01/26/03

Go up one level in this thread


On January 26, 2003 at 08:01:29, Christopher A. Morgan wrote:

>
>Kurt,
>
>I certainly agree with you that Johan has in effect handicapped
>himself in playing against other engines ouside his GUI.

That's not the point. As to tablebases it makes no difference whether TheKing
runs under CM9- or Fritz-GUI. The engine has no access to tablebases during
thinking process and that's indeed sometimes a handicap.<

>And that>to measure strength of an engine it is best to test it as it was
>meant to be played, with tablebases. But we're stuck with the fact
>we can't with TheKing. Unless you make them comparable then I believe
>your tournament results should have an asterick which says TheKing
>did not use tablebases, as it would in its own GUI, while other
>engines did use tablebases.
>
>Then, I think some very useful information would be obtained if you ran
>the same tournaments without tablebases. That at least tests engine
>strength alone, assuming opening book has no effect.

You are right here but considering the great number of games we are playing,
there is simply not enough time for that. And therefore, it's ok that Sedat is
testing TheKing the other way round.
>
>On January 26, 2003 at 07:38:34, Kurt Utzinger wrote:
>
>>Hi Christopher
>>TheKing [not using tablebases in it's thinking process] has the same handicap
>>when playing under its own GUI. I can't change things. When playing under
>>Fritz-GUI, TheKing has access to the tablebases as soon as a position stored in
>>the tablebases is reached over the board. I see no reason to give all other
>>engines a handicap they do not have in normal play just because of TheKing. If
>>Johan de Koning would ever participate at WCCC he would have to fight vs
>>programs using tablebases during thinking process. For these reasons, I think
>>the real playing strength between chess programs on the market is better
>>comparable the way we are testing. The same thing cannot be said about opening
>>books. We all know that the books delivered with ChessBase and other engines are
>>in no way identical with the well prepared books used at WCCC and therefore it
>>makes much more sense to use the same book for both engines in
>>engine/engine-matches.
>>Kurt
>>
>>Homepage "Kurt & Rolf Chess": http://www.beepworld.de/members39/utzinger



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.