Author: Dann Corbit
Date: 13:01:16 01/09/06
Go up one level in this thread
On January 09, 2006 at 15:48:33, Gabor Szots wrote: >On January 09, 2006 at 15:41:27, Uri Blass wrote: > >>On January 09, 2006 at 15:33:33, Dann Corbit wrote: >> >>>On January 09, 2006 at 15:29:58, Uri Blass wrote: >>> >>>>On January 09, 2006 at 15:14:29, Gabor Szots wrote: >>>> >>>>>On January 09, 2006 at 14:45:01, Austin A. Biberink wrote: >>>>> >>>>>When Rybka was first announced the author said "its strength is 90-150 Elo above >>>>>Shredder". The rating lists say otherwise, but you never demanded to correct >>>>>that lie. >>>> >>>> >>>>The rating list do not say different things >>>> >>>>http://www.husvankempen.de/nunn/cegtrating4040all.html >>>> >>>>2 Rybka 1.0 Beta 64-bit 2857 >>>>13 Shredder 9 2756 >>>> >>>>You can see 101 elo difference. >>>>90<101<150 >>> >>>I guess that .001% of the PCs have 64 bit Windows. So that measurement is >>>probably not too useful for them. >> >>I agree but the claim that rybka is 90-150 elo stronger was not a lie because it >>was not claimed specifically about the 32 bit version and people who downloaded >>rybka or bought it could see that there is 64 bit version even if they could not >>use it. >> >>Uri > >And Shredder has SMP version but I believe that in the comparison the 1-CPU >version was used. The SMP version would put the difference under 90, that's >sure. If you allow small test sets, here are the current CEGT measurements: no Program Elo + - Games Score Av.Op. Draws 1 Deep Shredder 9.12 64-bit 2CP 2881 74 73 60 65.8 % 2767 35.0 % 2 Rybka 1.0 Beta 64-bit 2857 21 21 742 68.5 % 2722 32.7 % 3 Rybka 1.0 Beta 32-bit 2826 10 10 3519 68.8 % 2688 31.0 % 4 Deep Shredder 9 2CPU 512MB 2805 17 17 1198 61.9 % 2720 30.6 % 5 Fruit 2.2 2788 11 11 2992 67.5 % 2661 32.8 %
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.