Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: To Mel: you misquoted Dr. Hyatt.

Author: Dave Gomboc

Date: 12:23:00 06/02/99

Go up one level in this thread


On June 02, 1999 at 13:51:47, Melvin S. Schwartz wrote:

>Well, if I am wrong, than blame the people responsible for writing in both the
>manuals for Hiarcs 7 and Fritz 5.32 that the programs strength would be better
>by running them on a faster pc!
>
>Mel

Why would we blame them?  That their statement is correct does not make the
other statement incorrect.

program A running on a P90:       rating = 2400
program B running on a P90:       rating = 2410

program A running on a P200MMX:   rating = 2450
program B running on a P200MMX:   rating = 2440

program A running on a K2-450:    rating = 2480
program B running on a K2-450:    rating = 2480

Both programs play better as the hardware improves.  But which program is
better?  If you tested only on "equal hardware", you could come up with any of
three conclusions, depending on what hardware you happened to pick!  And your
answer might change every time a new CPU is tested.

There are other problems with "play a match on equal hardware" as well.
Situations such as "program A beats program B, program B beats program C,
program C beats program A" are common.  Which of these is the best program?
Tough to decide, isn't it?

So I hope that you can appreciate these arguments and agree that attempting to
come up with a player rating, where a player is comprised of hardware and
software, is a good solution to the question "how good is it?", and indeed, a
better solution than various alternatives such as playing short matches.

Dave



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.