Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Speedup?

Author: Robert Hyatt

Date: 20:25:41 09/06/02

Go up one level in this thread


On September 06, 2002 at 22:06:25, Slater Wold wrote:

>On September 06, 2002 at 18:47:44, Robert Hyatt wrote:
>
>>>I don't think so.
>>>
>>>If you take the position of WAC41 you will see there is a 1.19x NPS speedup.
>>>However, there is a 16x speedup to ply 11!
>>>
>>
>>I'm not talking NPS at _all_ here.  I am talking about how long it takes to
>>complete depth X with 1 cpu and with 2 cpus.  Divide the 2cpu time into the 1
>>cpu time and that's the number...
>>
>>
>>>Then again, if you take the position of WAC76 you will see there is a 1.62x NPS
>>>speedup.  However, there is a 1.44x speedup to ply 11.
>
>I understand that.  It's just not reliable IMO.


I don't understand your reservation.  This is the most reliable form of testing
possible.  Because it is reproducible time after time since the N-cpu tests will
be forced to search to the same depth exactly as the 1-cpu test, which means
that the only variable is time...



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.