Author: P. Massie
Date: 11:59:00 09/04/03
Go up one level in this thread
It's also interesting that the results vary depending on the program. I tested on the Centrino 1.5 ghz, as mentioned before. I also ran the same set of tests on a PIV 3.0 ghz machine I had for a bit. The PIV was slightly faster on Fritz, but actually a little slower for Hiarcs 8. I don't remember the exact numbers, and they were in the same general area, but it's clear that actual peformance varies somewhat depending on the specific characteristics of the individual programs. I think it's safe to say a Centrino 1.5 is generally comparable to a PIV running at 2.8 (or maybe 2.9, if there were such a chip) ghz for chess programs, but that's a general statement that may be slightly incorrect based on the specific performance of a particular program. Paul On September 04, 2003 at 14:52:20, Anthony Cozzie wrote: >On September 04, 2003 at 14:40:39, Terje Vagle wrote: > >>On September 04, 2003 at 14:07:22, Roy Eassa wrote: >> >>>On September 04, 2003 at 05:41:10, Enrique Irazoqui wrote: >>> >>>>On September 04, 2003 at 02:00:34, P. Massie wrote: >>>> >>>>>I re-ran the benchmarks on a Centrino 1.5 ghz machine. The software was Fritz >>>>>8, version 8.0.0.23, with 64 MB of hash. I got a result of 1135 for the >>>>>Fritzmark and 1025 kn/s. >>>>> >>>>>Paul >>>> >>>>With Fritz 8.0.0.23 and 64MB hash, on a P4-3.06 I get 1238 Fritzmark and 1120 >>>>KNS. It seems that the Centrino 1.7 is slightly faster than the P4-3.06 for >>>>chess. >>>> >>>>Enrique >>> >>> >>>What makes the Centrino so much faster at a relatively low GHz, even than the >>>Athlon or P3? >> >>It seems to be the 1 MB On-Die Level 2 (L2) Cache. What else could it be? >> >>> >>>Can one buy a desktop system using a Centrino instead of a P4? > >don't forget branches: the predictor is larger and the misprediction penalty is >smaller.
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.