Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: CM 6666 v CM-Queen+ latest score W8 L14 D9 1hr/game (NT)

Author: blass uri

Date: 06:53:50 04/18/00

Go up one level in this thread


On April 18, 2000 at 08:50:48, Jorge Pichard wrote:

>On April 17, 2000 at 19:13:41, Chessfun wrote:
>
>>On April 17, 2000 at 17:59:53, Jorge Pichard wrote:
>>
>>>On April 17, 2000 at 16:01:45, Chessfun wrote:
>>>
>>>>:
>>>
>>>I wonder if in any of your games CM Jorge or CM-Queen+  I would prefer to be
>>>called, has ever encountered itself in a position where CM6666 has two Bishops,
>>>a knight plus a pawn Versus CM-Queen+  with a Queen plus a pawn, in such an
>>>ending yesterday I pitted CM6666 Vs CM-Queen+ in a match of 50 games which I
>>>started saturday, in game in 60 minutes per side and my setting thought that it
>>>was winning due to the high value of the Queen, but indeed it was loosing the
>>>endgame, Therefore, I decided to change the value of the OQ to 9.5 Instead of
>>>9.9, but somehow since this position that not occur too often, the new value of
>>>the queen is affecting the final result of other games, therefore, I decided to
>>>change it back to the original seeting of OQ 9.9 and 9.7 respectively. It was
>>>logical to change the value of the Queen to 9.5, since the value of the Bishop
>>>was set up to 3.2 and the knight to 3.1, but somehow the new setting was not
>>>working for the averall percentage score of CM-Queen+.
>>>
>>>PS: That is probably the reason why I would prefer to call this controversial
>>>setting the CM-Queen+, due to its peculiar queen setting.
>>
>>
>>6666 won the next two so now W 10 L 14 D9.
>>It didn't seem logical to me to just lower the queen value,
>>based on losing a single ending.
>>While it may lose such an ending the consequences of changing
>>the value will effect all game play.
>>
>>After this 50 game match I will pit it against the default and
>>if it scores well there then against SS10.
>>
>>Thanks.
>
>What is the SS10, I Have not heard about it, please explain how do I get it?
>I would like to start testing against it today, in game in 60 minutes per side
>against my CM-Quen+ setting.
>
>Thanks.

I think that SS10 is simply chessmaster6000 with only one change:
selective depth=10 and not 6.

My impression is that the difference is in the depth that it starts using null
move.
chessmaster gives number like depth=2/8 and it probably means that the brute
force depth without null move prunning is 2 and with null move prunning is 8

Selective depth is probably an upper bound for the difference between the
numbers.
When selective depth=6 the difference between the numbers is always at most 6
and when selective depth=10 the difference between the numbers is at most 10.

The difference between the numbers is usually smaller in endgames because
chessmaster "believes" that it is a bad idea to trust null move prunning in
endgames.

Uri



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.