Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Deep Fritz 8 Experimental hit 3.5 Millions Node Count !!!

Author: Vincent Diepeveen

Date: 08:39:41 10/14/02

Go up one level in this thread


On October 14, 2002 at 02:38:25, Gian-Carlo Pascutto wrote:

>On October 13, 2002 at 22:00:01, Robert Hyatt wrote:
>
>>Nobody in their right mind would do that unless forced to do so due to
>>some oddball constraint having nothing to do with chess.
>
>'Marketing'
>
>--
>GCP

2.2 MLN nodes a second *easily* on a dual K7 against me at world
champs. Now a 'peek' in some far endgame of 3.5 MLN a second.

No way that this is 8 processors getting used by fritz. More like
4 processors or 2 x 2.8 Ghz.

Fritz works great on those Xeons. It was originally optimized for the P3.

4 x 900 Mhz = 3.6ghz and that with parallel memory :)

this where 2 x 1.6Ghz K7 has no parallel memory.

On the other hand a well optimized program for 8 x 900Mhz Xeon
using fritz would peak easily at 6 million nodes a second.

Best regards,
Vincent




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.