Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Who fixed the original settings of CM8000 ?

Author: John Merlino

Date: 11:09:33 04/04/01

Go up one level in this thread


On April 04, 2001 at 14:08:07, Uri Blass wrote:

>On April 04, 2001 at 13:42:55, John Merlino wrote:
>
>>On April 04, 2001 at 05:52:17, Rainer Neuhäusler wrote:
>>
>>>Who fixed the standard settings of the CM8000? And ist it the strongest setting
>>>in the opinion of J. de Koning and J. Merlino ?
>>>
>>>Perhaps this could be simple questions to J. Merlino.
>>>
>>>Rainer
>>
>>The settings for the Chessmaster personality are believed to be the strongest
>>FOR ALL TIME CONTROLS AND ALL HARDWARE CONFIGURATIONS -- this last part is VERY
>>IMPORTANT to remember! CM8000's minimum spec machine is a PII-233 with only 48MB
>>of RAM. This is why the default personality only has a 1MB hash table. If
>>(knowledgable) users want to increase the strength for their hardware or
>>personal needs, then they can easily do so. The default personality is designed
>>for our default user, which is IN NO WAY representative of the typical member of
>>this board.
>>
>>The ONLY personality setting that MIGHT improve play UNDER CERTAIN GAME
>>SITUATIONS with modification (other than the increasing the hash table size,
>>obviously) would be the "Selective Search" value. But, a value of 6 is best for
>>ALL time controls. I promise you, if you play a blitz tournament on a PII-233
>>with one personality being the default and the other having a 32MB hash table
>>with SS=12, the default one will be stronger.
>>
>>People have confused the idea of strongest OVERALL with strongest FOR TOURNAMENT
>>TIME CONTROLS. These two are not NECESSARILY the same (although they might be --
>>nobody has proved anything either way).
>>
>>Not only that, the SS setting is now even in doubt, as some people have recently
>>posted personalities that they feel are the strongest that only have SS=8. So,
>>in other words, there's absolutely no proof that the default settings (apart
>>from the hash table size) are not the strongest for tournament time controls.
>>
>>jm
>Without testing chessmaster I am almost sure that the best setting is unknown.
>There are many parameters and finding the best numbers seems to me to be an
>impossible task.
>
>Uri

Which is why there are always threads like this.... <sigh>

jm



This page took 0.01 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.