Author: Ryan B.
Date: 11:44:16 02/01/06
Go up one level in this thread
On February 01, 2006 at 05:50:19, Uri Blass wrote: >On February 01, 2006 at 04:42:56, Ryan B. wrote: > >>On February 01, 2006 at 02:17:40, Uri Blass wrote: >> >>>On February 01, 2006 at 01:50:23, Ryan B. wrote: >>> >>>>On January 31, 2006 at 23:07:08, walt irvin wrote: >>>> >>>>>in a rybka vs hydra match ,,,with rybka using a athlon 3800 12 game match 40 >>>>>moves in 2 hours what is the predicted score ??????????? >>>>> >>>>>mine is hydra 9/3 >>>>> >>>>>walter >>>> >>>>I think your prediction is very good. What I think many people fail to realize >>>>about Hydra is that it is a dedicated chess machine so it does not suffer the >>>>same overheads that PC chess engines have. Also Software optimized for just one >>>>exact platform can be very fast compared to traditional PC programs. >>>> >>>>Ryan >>> >>>I do not agree. >>> >>>By that logic Crafty with 8 processors can also beat Rybka with one processor >>>but it is probably not the case. >>> >>>The fact that hardware that is dedicated for one task has a potential advantage >>>does not prove that hydra is better. >>> >>>The facts are that Rybka is leading the CEGT list even if you consider only 32 >>>bits and not one of the better hardware participants. >>> >>>This is from the CEGT rating list. >>> >>>2 Rybka 1.01 Beta 13-13b 32-bit 2886 45 45 189 74.1 % 2704 28.6 % >>>6 Deep Shredder 9.12 x64 2CPU 2822 35 34 283 64.7 % 2717 31.1 % >>> >>>14 Shredder 9 2754 8 8 4776 63.3 % 2660 31.8 % >>>34 Crafty Cito 1.4.2 64-bit 2CPU 2695 59 58 106 63.7 % 2598 27.4 % >>> >>>As you can see even the combination of 64 bits and 2 cpu does not help when you >>>have inferior program. >>> >>>If 64 bits+2 cpu gives does the program 3 times faster then you have: >>> >>>Crafty*3<Shredder >>>Shredder*3<Rybka and in other words >>> >>>Crafty*9<Rybka. >>> >>>Uri >> >>I have no idea how anything I said had anything to do with crafty. Because we >>do have a Hydra sitting around to test out in our homes we go by what e believe. >> Based on what I have read here including posts by Chrilly and results of games >>Hydra has played in the past I believe Hydra to be much stronger than Rybka or >>any other known chess entity at this time. I hope to see Zappa take that crown >>in the future though. >> >>Ryan > >I wonder what are the results that you talk about? > >I know that in long time control hydra could not even get into the quarter final > >http://www.chessbase.com/newsdetail.asp?newsid=2446 > >I know that it lost in the first 2 correspondence games against GM nickel and >based on the report it has only equality in the third game. > >Note that commercial chess engines that are clearly weaker than rybka could beat >GM nickel 3.5-2.5 > >for the games see >http://www.chessfriend.com/ > > >Uri Chrilly said Hydra got 75-80% vs Rybka and I have no reason not to trust him on such issues. I may feel like he can misslead people about what is inside Rybka or what to look for but he does not lie from what I can tell. I trust Chrilly, maybe you do not. Ryan
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.