Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Nimzo99 MMX - Hiarcs 6 P90 SSDF game 12/20 1-0 Now: 10 - 2

Author: James Robertson

Date: 21:59:56 05/29/99

Go up one level in this thread


On May 29, 1999 at 23:49:47, Melvin S. Schwartz wrote:

>What do you think Mr. Hyatt means when he says:
>
>"If program A on hardware B beats program D on Hardware E - does that say much
>about A compared to B? This belies the principles of science - you have to have
>a uniform platform for all participants to make any kind of judgement".

Yes.... if out of the blue you play program A on hardware B vs program D on
hardware E you have no clue how the program's relate to each other.

But.....
Say, hypothetically, that program AChess is rated 2000 on a P90.

Then suppose AChess on a P90 plays AChess on a P200, and loses by 60%. That
would put AChess on a P200 rated about 2100.

Now we have some concrete evidence; AChess on a P90 is 2000, AChess on a P200 is
2100.

Now suppose we play AChess on a P200 against BChess on a P90, and AChess wins by
60%. That puts BChess' rating at 2000.

We have discovered BChess on a P90's rating, and we have used uneven hardware.
We will reliably be able to predict a ~50-50 score vs AChess on a P90 vs BChess
on a P90.

Do you see how this works? Uneven hardware does not compare programs, it
compares program's on a certain hardware.

James


>
>Now, I understand that to mean exactly what I have said over and over again. And
>that is simply the fact you have to have chess programs running on the same type
>of hardware to make a proper judgement. Now, isn't that what Mr. Hyatt is
>saying? Or perhaps you feel this is the first time he is wrong and you are
>right?
>
>Mel





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.