Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: cometb40 v gromit382 7-1-0

Author: Günther Simon

Date: 09:58:38 04/21/02

Go up one level in this thread


On April 20, 2002 at 20:59:30, pavel wrote:

>It is a fact and stated well by the programmer of Gromit3.8.2 that Gromit was
>not optimised to play good as ponder=off.
>In other words Gromite3.8.2 (or maybe every versions) doesnt play as good with
>ponder=off as it does with ponder=on.
>
>just wanted to note it.
>
>pavs

I dont believe the problems had to do with pondering at all!
There was something definitely broken in this match and it
was not the pondering issue because i.e. I could prove that
the log showed a horribble garbage move at search depth 4 ply
in 19 sec! when it only needs 0.220 sec under WB to understand
that this move is unplayable and moreover reach ply 8 easily
on a PIII650 though running a lot of apps at the same time.

Take a look there:
http://www.f22.parsimony.net/forum41668/messages/10907.htm

Regards,
günther



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.