Computer Chess Club Archives




Subject: Re: Null move mate extension

Author: James Robertson

Date: 08:41:36 06/25/99

Go up one level in this thread

On June 25, 1999 at 07:48:31, Ulrich Tuerke wrote:

>On June 25, 1999 at 02:29:45, James Robertson wrote:
>>I implemented the threat exension in the null move (if the null move returns
>>that we were mated, then extend), and I almost _never_ get any extensions this
>>way. Maybe 3 or 4 times in an entire search, if that many. How many is normal?
>>Do I have a bug? Any help is appreciated.
>I get a lot of extensions this way, far more than I can afford to search. So I
>only use this very close to the horizon (depth <= 2).
>May be that you are doing something wrong ?

Yeah.... I think Bob put his finger on it. But, I strongly encourage you to
extend several plies on every threat..... It seems like my program gets withing
200 points of Comet, and then you release a new version. :(



This page took 0.03 seconds to execute

Last modified: Thu, 07 Jul 11 08:48:38 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.