Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Pretty amazing FritzMark score?

Author: Slater Wold

Date: 21:31:49 03/08/02

Go up one level in this thread


On March 08, 2002 at 21:33:48, Dann Corbit wrote:

>On March 08, 2002 at 21:31:26, Wayne Lowrance wrote:
>
>>On March 08, 2002 at 21:08:09, Slater Wold wrote:
>>
>>>On March 08, 2002 at 20:58:21, enrico carrisco wrote:
>>>
>>>>On March 08, 2002 at 20:56:20, Slater Wold wrote:
>>>>
>>>>>On March 08, 2002 at 20:30:01, Kris Jordan wrote:
>>>>>
>>>>>>wow, not at all. It's an athlon xp 2000+ overclocked to 1755mhz!!! This
>>>>>>fritzmark is with fritz 7 though.;)
>>>>>
>>>>>You never said that.
>>>>>
>>>>>My AMD 1.2Ghz gets *about* those numbers on Fritz 6.
>>>>>
>>>>
>>>>You better check again..  I'd say closer to the 900s with a 1.2GHz T-Bird.
>>>>
>>>>-elc.
>>>
>>>Just ran it.
>>>
>>>Fritz 6 Fritzmark AMD 1.2Ghz 64MB hash:
>>>
>>>781(9.0s)
>>>1123kN/s
>>>
>>>I was ONLY looking at nps.  That number is worthless.
>>>
>>>Don't have F7 on that computer.
>>
>>Which is worthless ? nips or fritz mark ?
>
>Yep.
>;-)
>
>Actually either can be useful as long as you know the version number of Fritz
>and the exact hardware, OS, etc.
>
>But (as they say) the proof of the pudding is in the eating.
>
>NPS is a pretty meaningless figure.  FritzMark is not really comparable even to
>other FritzMark figures.

When I ask someone how the new "hottest" hardware stacks up, I usually expect it
to be in NPS, not fritzmark.

Everyone benches their software off NPS.  Crafty, Sunsetter, hell even TSCP.
But Chessbase thought a formula would make it better somehow.  Not sure how.

NPS does = ELO.  In some small, fractorial way.  ;)



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.