Author: Dan Newman
Date: 15:43:31 03/11/00
Go up one level in this thread
On March 10, 2000 at 20:41:55, Jorge Pichard wrote: >On March 10, 2000 at 20:14:25, Aloisio Ponti Lopes wrote: > >>I'd like to see a benchmark using chess programs with these processors: >> >>1) Dual Celerons >>2) Athlons >>3) Pentium III (or II) >> >>I like Rebel's benchmark page, but there's no dual celerons there, as Rebel >>still don't use SMP (and I don't know if it will use it in the near future). >> >>Are there any Crafty page on benchmarks? (or Junior?) >> >>A. Ponti > >I do not believe that Rebel's Benchmark is too accurate, if you take a close >look at the performance given by different people, for instance it is impossible >to believe that the Celeron 550 Mhz is equal to the perfarmance of the ATHLON >550 Mhz, NOR the performance of the K6-3 600 Mhz BY Kryotech to be Higher than >the performance of the ATHLON 600 Mhz. I truely believe that in both cases the >same person that provided the information on the celeron lied. > >If you have doubt check the different permance benchmarks of the Celeron which >in some instance it is even rated higher than a full Intel processor with the >same Mhz capacity. Compare it yourself at this website >http://www.rebel.nl/bech.htm My program (Shrike) is about 6% faster on the Celeron vs PIII (and my Celeron only has a 66 MHz bus, the PIII 100 MHz...). I think performance can vary widely depending on a lot of factors: 1) program size, 2) algorithms and data structures used, 3) cache sizes, 4) cache speed, 5) bus speed, 6) processor type, 7) compiler used, 8) target processor, etc. So I'm not surprised at all to find the Celeron equal to the Athlon or better than the PIII... -Dan.
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.