Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: ECM........#28.....RxN.....!? ... CM selectivity

Author: Johan de Koning

Date: 22:02:35 05/11/04

Go up one level in this thread


On May 11, 2004 at 04:18:09, Vincent Lejeune wrote:

>On May 11, 2004 at 01:17:32, John Merlino wrote:

>>Also, it looks like you have modified your SKR setting, and are using a
>>selective search value of at least 13. SKR is supposed to use a value of 12.
>
>Yes, sorry to not mention this, I always like big selective search, my SKR is
>modified with ss=16.
>
>Do you know if Johan plan to rise the possible value for the ss ?
>Or do you have results of a modified CM with ss>16 (24,32) ? is it a lot counter
>productive ?

I think you can hex-edit the appropriate byte in the personality file,
if you insist on going beyond SS=16. Or run The King with the "usual"
OPK fix and cm_parm settings. Compare the output of a sparse endgame,
to make sure it works.

But IMHO it makes little sense.
In the middle game TK will not search beyond 17 ply within a day.
In the endgame TK will suffer from zugzwangs, because it still(!)
relies on the full-width part of the search in endgames. That is
including the upcoming CMX.

... Johan



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.