Author: margolies,marc
Date: 18:36:16 02/08/04
Go up one level in this thread
On February 08, 2004 at 16:44:19, Aaron Gordon wrote:
>On February 08, 2004 at 16:35:43, Ingo Bauer wrote:
>
>>On February 08, 2004 at 16:18:49, Aaron Gordon wrote:
>>
>>>>That only shows how bad a P4 3.0 GHz is.
>>>>
>>>>A Athlon @ 2.4 GHz (Real 2.4 not a 2400+) has a Shreddermark of 2784 (440kN/s)
>>>>at 64 MB Hash. With 409 MB Has (To compare with the 2 GHz Opteron) it is 1237
>>>>(191 kN/s).
>>>>
>>>>I would be higly interested in a Shreddermark on that opteron with 64 MB Hash.
>>>>
>>>>Bye Ingo
>>>
>>>Here is my over year old Athlon XP 2100+ (1.73GHz) running 2505MHz,
>>>200fsb(400DDR), Epox 8RDA motherboard.
>>>
>>>ShredderMark8:
>>>64mb hash : 3712 @ 503kn/s
>>>409mb hash: 2227 @ 309kn/s
>>
>>If that is true there is either something wrong with this "Shreddermark" or with
>>your or my computer!
>>
Ingo is right-- shreddermark is broken...here are fritzmarks for 2ghz amd
opteron
using WMfritz8 (8.0.26) 64 meghash --> fm 1339 ; 1201 kns
128 meg hash -> fm 1339 ; 1194 kns
256 meg --> fm 1297 ; 1167 kns
512 meg 1238 1112
-marc
>>Your Comp is 4% (or 100MHz - in fact its less I am running 14*173=2422 - FSB for
>>Chess doesnt matter much) faster than mine and is performing 33% better in
>>Shreddermarks but only 14% in kN/s (at 64 MB)
>>
>>In all the other test I did (e.G Solutiontime for a position) my two Athlons
>>(2400 + 1850MHz /142 FSB) behave completly linear (less than 1% difference). If
>>I have a value for the 1850 I can calculate the value for the 2400 (even if one
>>has only 256 L2 Cache and the other 512)
>>
>>I tend to a broken Shreddermark!!!
>>
>>Ingo
>
>The test only takes 0.3 seconds and the timer is probably not that accurate as
>well. It COULD be an accurate test if it ran a good minute (or more) and timed
>down to the millisecond.
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.