Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Modern engine test results with the BS-2830 test suite.

Author: Dann Corbit

Date: 16:20:28 06/19/02

Go up one level in this thread


On June 18, 2002 at 02:21:20, Slater Wold wrote:

>On June 18, 2002 at 02:08:47, Klaus Friedel wrote:
>
>>On June 17, 2002 at 19:57:46, Aaron Gordon wrote:
>>
>>>Just got bored and ran some tests of a few popular engines. I need to clean out
>>>the waterblock & radiator for my computer so until then I'll be running 1.76
>>>instead of 1.86ghz. Maybe when I get bored again I'll retest at 1.86GHz(or when
>>>I get a Sledgehammer). Anyway here they are...
>>>
>>>CPU: AthlonXP 1.76GHz unless otherwise specified.
>>>BS-2830 test. 15 minutes per position.
>>>
>>>Engine & hash             ELO              Comments
>>>
>>>Nimzo 8 320mb             2801.75          Thunderbird 1.5GHz
>>>Hiarcs 8.0 320mb          2778.79
>>>Fritz 7.0.0.8 320mb       2765.65
>>>Shredder 6.02 320mb       2759.11          Thunderbird 1.5GHz
>>>Chesstiger 14 192mb       2747.26
>>>Junior 7 128mb            2741.51          Celeron 1GHz
>>>Gambit Tiger 2 192mb      2736.82
>>>Crafty 18.15 CB 288mb     2652.01
>>Hagrid 0.7.41 200mb        2597             Athlon XP 1700+
>>
>>
>>I don't realy trust in the such testsets. Hagrid is far weaker than 2600 ELO...
>>
>>Bye,
>>Klaus Friedel
>
>Did ya think for 1 second that Nimzo is 2800+ Elo?

ELO is (of course) an arbitrary score.  Is Nimzo 2800+ FIDE strength against
people?  I have no idea.  I certainly think it is possible.  It also might be
considerably weaker.  Given the lack of data, I don't think we can say.

It certainly did well on a human calibrated test.  But I think extrapolation to
ELO is pretty wild.



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.