Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Coparing two Identical Programs using Different Processors Speed !

Author: Uri Blass

Date: 23:38:23 01/28/01

Go up one level in this thread


On January 29, 2001 at 02:25:46, Severi Salminen wrote:

>>We can probably say that Crafty will do better against nimzo in 1 hour per game
>>on athlon1.2 and not in 2 hours per game on athlon600 so we can say that Crafty
>>likes more the faster hardware when we do the right comparison.
>
>But remember this: there are other factors than NPS. One is branching factor.
>Let's assume we habe programs A and B. A has average BF of 2 and B has 6. It is
>very likely that the NPS rate of A is much smaller than B's because doing more
>selectivity seems to "slow" things down in terms of NPS. But program A needs
>only 2x speed or time to reach an additional ply whereas B needs 6x the time or
>speed. For two additional plies the figures are 4x and 36x. Even if doubling
>speed results 3x increase in NPS for B and only 2x for A we can see that A earns
>more from speedup in terms of plies - in terms of strength, who knows. So again
>NPS doesn't tell you much.
>
>Severi

If you do the right comparison then it is clear that B can get better results
with faster hardware.

You only need to do the right comparison and use faster time control on the
faster hardware so A will calculate the same number of nps per move when B will
calculate more nps per move.

If we agree that for all programs more nps per move is better than it is clear
that B can get better result on faster hardware when you use the right time
control that is different for both hardwares.


Uri



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.