Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Tortoise or Hare?

Author: Chris Carson

Date: 10:34:41 06/22/99

Go up one level in this thread


On June 22, 1999 at 10:58:42, paul bedrey wrote:

>Hi Chris, thanks for your comments. I was thinking of the just completed Micro
>Championship. Some of the Hares were running on souped up hardware yet the
>tortoises more than held there own. It would be interesting to see what would
>have happened if it was reversed and the Tortoises had the superior hardware.
>Someone had said in a previous post that speed is only good if your pointed in
>the right direction.

I see two things discussed here.  S/W speed (evaluation) in Nodes Per
Second (NPS) and HW speed (faster HW).

Given the same HW (like SSDF) you can compare (not perfect, but best
I know of) SW strength.  A fast Searcher will get rating Xs and a
Slow Searcher will get Ys ratings.  In my opinion, High NPS or Low
NPS is irrelevant (although mine is a fast searcher, high NPS), the
bottom line is how strong the program is.  Just my opinion.

If you put the Xs or the Ys program on a faster machine and play against
the other program on a slower machine, then either X or Y will see a
ratings increase due to the faster hardware.  I am talking single
processor now.  If a program is multi-processing, then adding more
processors may add significant strength.  I am not an expert, this is
just my opinion.

What has been your observation?

Best Regards,
Chris Carson



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.