Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: centrino performance?

Author: Aaron Gordon

Date: 14:47:48 06/20/03

Go up one level in this thread


On June 20, 2003 at 15:53:13, Tom Kerrigan wrote:

>On June 20, 2003 at 14:20:14, Aaron Gordon wrote:
>
>>On June 20, 2003 at 12:48:21, Tom Kerrigan wrote:
>>
>>>On June 19, 2003 at 17:32:00, Aaron Gordon wrote:
>>>
>>>>On June 19, 2003 at 08:46:45, martin fierz wrote:
>>>>
>>>>>On June 19, 2003 at 05:10:25, Aaron Gordon wrote:
>>>>>
>>>>>>On June 19, 2003 at 03:55:49, martin fierz wrote:
>>>>>>
>>>>>>>i was just reading this computer magazine which says that the new pentium M
>>>>>>>(also sold as "centrino" together with other components) is very fast compared
>>>>>>>to older notebooks. does anybody have data on this (e.g. crafty benchmarks)?
>>>>>>>
>>>>>>>cheers
>>>>>>>  martin
>>>>>>
>>>>>>From what I've seen so far a Centrino cpu is a Pentium 3 with 1mb of L2 cache.
>>>>>>The extra L2 makes no difference for chess, though.
>>>>>
>>>>>if this is true, how come pichard's 1.3GHz centrino runs ruffian faster than my
>>>>>2.4GHz P4?
>>>>>
>>>>>cheers
>>>>>  martin
>>>>
>>>>Pentium 4's are much slower MHz for MHz than an Athlon or Pentium 3. This is why
>>>>an old Celeron 1GHz (with a P3 style core) is the same speed as a Pentium 4
>>>>1.5GHz. An old Athlon XP 1900+ 1.6GHz is about as fast as a Pentium 4 2.8GHz in
>>>>Crafty... and my Athlon XP overclocked to 2507MHz is about the speed of a
>>>>Pentium 4 @ 4GHz in Crafty if a P4 such as that existed.
>>>
>>>No, 1.5ghz vs. 1ghz is significantly different from 2.4ghz vs. 1.3ghz. The
>>>answer to Martin's question is that the Pentium M is quite different from the
>>>Pentium 3.
>>>
>>>I don't know about Crafty, but I don't think the performance difference between
>>>the AXP and P4 is that big on average. I know that for my program, my AXP 2000+
>>>runs slightly slower than the 2.2GHz Northwood I have at work.
>>>
>>>-Tom
>>
>>The Crafty & P4 speeds hold true with every other chess program I've seen, but
>>yours. Not sure what you're doing.. perhaps it's the simplicity of your program
>>that helps.
>>
>>In the tests that I've done for chess, I see...
>>
>>Athlon 2.5Ghz == P4-4.0GHz (w/o HT)
>>Athlon 2.2GHz > P4-3.25ghz
>>Athlon XP 1.86GHz == P4-3GHz
>>Tbird 1.68GHz > P4-2.53GHz
>>Pentium 3 1.12GHz > P4-1.6GHz
>>Celeron 1.0GHz > P4-1.5GHz.
>
>What other programs have you tested? Do you keep track of benchmark scores? It'd
>be interesting to see all of your data in, e.g., an Excel spreadsheet or
>something.
>
>-Tom

Most all of the results I've done have been posted on here, but I know those can
be a pain to find sometime. I'll try to get the results I have in Excel when I
get the chance. Right now I've been doing a bit dual Athlon testing. Looking
good so far, getting excellent speedups in everything besides Crafty (which
Hyatt has already pointed out it does the same on his box, and is working on a
fix).



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.