Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: CEGT 40/40 ratinglists updated

Author: Heinz van Kempen

Date: 08:22:57 12/21/05

Go up one level in this thread


On December 21, 2005 at 10:24:00, Vincent Lejeune wrote:

>
>1  Rybka 1.0 Beta 64-bit SP   2924   38   37    376   78.6 %   2698   15.7 %
>…
>4  Rybka 1.0 Beta 32-bit      2837   15   15   1582   69.3 %   2695   28.3 %
>
>87 points between 64bit SP and 32bit VP !!!
>
>The full strengh of Rybka has been underestimate with the default (32 bits VP),
>I think !
>
>
>On December 20, 2005 at 10:12:54, Heinz van Kempen wrote:
>
>>
>>>My Nunn2 testing, as published here at CCC, suggests that the SP (slightly
>>>positional) setting might yield slightly better results. My sample is not large,
>>>160 games for each setting, and doesn't include the VT setting at all.
>>>
>>>                                        Albert
>>>
>>
>>Hi Albert,
>>
>>Wilhelm Hudetz played a lot of Blitz games with the settings and those also
>>indicate that SP might be the strongest. So far not tested in CEGT 40/40.
>>
>>http://www.husvankempen.de/nunn/eloblitz.html
>>
>>Best Regards
>>Heinz

Hi Vincent,

yes, we will see how the 64-bit version will do in the 40/40 match against
Hiarcs 10. Hiarcs took the lead here, but very few games played.

Hiarcs 10 vs. Rybka 1.0 Beta-64bit (3:2)
Hiarcs 10 vs. Deep Shredder 2CPU 512MB (2,5:2,5)
Hiarcs 10 vs. Fruit 2.2.1 (3,5:1,5)

Seems that I am the first CEGT tester who got the Hiarcs DVD. So the others will
also start matches in the next days.

Best Regards
Heinz



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.