Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Standard hardware for rating program strengths.

Author: Micheal Cummings

Date: 03:46:09 11/02/98

Go up one level in this thread


I think all software should be tested on the same hardware. It is only way to
give a true indication.
I know some programmers write their engines to perform better with certain
hardware, but not all people have the same hardware that the programmer uses to
perfect his engine.
So I say tough luck to those programs if used on different hardware perform
worse.

AlI programs should be tested on the same hardware, and I do get annoyed when I
see a program on a lesser processer with less memory and hash comming right
behind one that is using 25% more power. I sit their and think what if that
other program had the same hardware maybe it would be sitting in front of the
other in rankings and not below.

Maybe I am wrong and their are some things I do not know, some of those programs
might have limited hash , but the programs that have limited abilities compared
to other programs should at least be used on the same hardware then you know how
strong the program is.

So in final, it should be only a battle of the software and not a battle of the
hardware as well. otherwise rankings are worthless in my view.



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.