Computer Chess Club Archives


Search

Terms

Messages

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

Author: Aaron Tay

Date: 06:21:05 01/26/03

Go up one level in this thread


On January 26, 2003 at 07:19:01, Christopher A. Morgan wrote:

>
>Hmmmm, Johan has his own tablebases (transposition tables) that
>work only in his GUI. Seems fair to me. You take TheKing out of
>its own GUI so it can't use its tablebases, then blame CM 9000/Johan
>because it can't use someone else's tablebases in someone else's
>GUI, then play TheKing against other engines that use someone else's tablebases,
>knowing you are handicaping TheKing, and publish the
>results.

A few points

1)Even in Chessmaster GUI, CM 9000 does not probe tablebases in search. It can
only use them when the positions covered by the tablebases are reached.

2)In  Chessbase GUI, whenever a position covered by the tablebase is reached,
the GUI takes over. This is regardless of the fact that the engine has tablebase
access

3) Given 1)+2) , CM9K doesnt lose anything when used in Chessbase GUI.

4) Tablebases probably result in a very small (if at all) increase in strenght.
A minority are of the opinion that in certain positions, probing the tablebases
can even weaken play.



>What do your test results tell us?  Try some tournaments
>without tablebases so we can really compare engine strength.
>Until then I don't know what your tournament results mean.








>
>
>On January 26, 2003 at 06:46:47, Kurt Utzinger wrote:
>
>>On January 26, 2003 at 06:30:24, Christopher A. Morgan wrote:
>>
>>>
>>>TheKing outside of its own GUI cannot use tablebases.  If the King
>>>plays outside its own GUI then to make the engines comparable, no
>>>tablebases.
>>>
>>>Kurt, do you use tablebases in games against the TheKing when it
>>>is used outside its own GUI?  If so you give TheKing a handicap.
>>
>>Yes, we use tablebases when playing with TheKing outside its own GUI. And it's
>>true: out of 30 games, there are about 2-3 games that would bring other (better)
>>results for TheKing and it is indeed somewhat a handicap for TheKing. On the
>>other hand I think that Fritz & Co. are not to blame for having implemented this
>>feature, it's a pity that TheKing cannot use tablebase in its thinking process.
>>I made the experience that people accept using the same opening book (remis.ctg)
>>but there is less acceptance for not using tablebases. In principle I have the
>>same opion like Sedat but we do not want to hear from people all over the world
>>that our games are not much worse because of missing tablebases.
>>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.