Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Rybka Benchmarks

Author: Aaron Gordon

Date: 13:09:31 12/07/05

Go up one level in this thread


On December 07, 2005 at 15:24:09, Keith Hyams wrote:

>On December 07, 2005 at 15:19:22, Aaron Gordon wrote:
>
>>On December 07, 2005 at 13:42:49, Keith Hyams wrote:
>>
>>>2.7 GHz Celeron   56180 nps     Surprisingly slow??
>>
>>It may be normal. I'll see if I can get a test run on a friends Celeron 2.4GHz
>>(should scale linearly with your results). I think Rybka may depend on L2 cache
>>size/speed. Traditionally Pentium 4's have been rather slow for chess and the
>>cache issue on the Celeron may compound the issue. Thanks for your result
>>though, I'll try to look into this further.
>
>Thank you Aaron. If you find anything and matters have moved on here, perhaps
>you would be kind enough to email me.
>                               Keith

Unfortunately it does look like the Celeron (P4 core) has problems in Rybka. I
just had a friend test his Celeron 2.4GHz and it got 31408 (his Athlon XP 2.2
Barton got 107k). That is the older Celeron, not the Celeron-D. Do you know if
yours is the original Celeron or the newer Celeron-D (-D should have more cache,
Prescott core, etc)? If you're unsure you could try downloading CPU-Z from here:
http://www.cpuid.com/download/cpu-z-131.zip



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.