Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: More CM 6000 settings tests.

Author: Dan Kiski

Date: 06:08:48 12/10/98

Go up one level in this thread


On December 10, 1998 at 07:55:31, Mark Young wrote:

>On December 10, 1998 at 07:51:35, Mark Young wrote:
>
>>On December 10, 1998 at 07:27:14, Micheal Cummings wrote:
>>
>>>
>>>Test 4 and CM6555 are pretty much the same arn't they apart from the queen value
>>>for Test 4 being 9.6 and the Cm6555 being 10.
>>
>>A\D is +4 not -4, Q is 9.6 not 10, R is 5 not 5.5.
>
>*Correction A\D is 0 not -4.*
>
>>
>>Small changes like this can be good or very bad. The only difference in CM6555
>>and test 6 is .1 change in the bishop and night settings over CM6555. Either
>>test 6 had bad luck or very small changes *can* push you over the edge. I just
>>go slow and try to fine tune things here and there and then see what happends.

I think that although the difference between CM6555 and test 6 is just .1 that
the main difference in the results is lack of games played, not that it pushes
over the edge. But simply that with the variations in opening available and such
a small sample any result can be discounted. This is why for me the nunn
positions are good because for sure both sides get an equal chance with the same
set-up position after move 10.

Dan Kiski



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.