Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: New processorgenaration and chessprograms

Author: Hannu Wegner

Date: 15:05:41 02/08/99

Go up one level in this thread


On February 08, 1999 at 17:01:53, Dann Corbit wrote:

>On February 08, 1999 at 14:05:04, Alain Lyrette wrote:
>>
>>Well if we agree that if you double the speed of a computer ,you increase the
>>elo of program by roughly 70-75 points,then my guess is that if you go from a
>>k6-2 400 mhz to a 450 k6-3 you could expect a 20-25 elo jump.If you're talking
>>about the k7 it's too early to tell but my guess for a 500mhz k7 would be a
>>50-60 points increase.P.S:Sorry for my poor english..i keep trying to improve
>>it.
>I should be very surprised if the process is linear in MHz or even if there is
>any accurate predictor.  I really suspect that the only way to know is to
>benchmark.  On-chip cache, the way it pipelines, internal bus width, and a lot
>of additional factors will come into play.  In fact, I would be very surprised
>if two different programs get the same boost from the same hardware.


Exactly such a benchmark I was asking for:-). I know that diffeent programs will
get a different "boost" by a new AMD-processor. But I just was asking for some
info's. Anyway some info is better than no info!?

Best greetings,
Hannu Wegner



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.