Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Help with Aspiration Windows and Researches

Author: Peter Kappler

Date: 10:38:21 04/25/99

Go up one level in this thread


On April 25, 1999 at 05:30:29, Steve Maughan wrote:

>William
>
>The problem that you are encountering is that in the research (Beta-1,
>Infinity), the new parameters are allowing some variations to 'kick in' that
>were previously pruned.  In addition these new variations are pertinent to the
>PV, the result being a fail low.
>
>I would simply search (Alpha, Beta) then on fail high (Alpha=Beta-1, Beta =
>Infinity) and then on fail low (Alpha=-Infinity, Beta = +Infinity).  This should
>cope with all cases.
>
>Regards
>
>Steve Maughan


I still don't use aspiration search in my program.

Do either of you have an estimate of how much it speeds up your search?

--Peter



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.