Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: I'm wrong about 10-0 vs 60-40

Author: Bruce Moreland

Date: 09:20:13 02/05/01

Go up one level in this thread


On February 04, 2001 at 22:25:52, Ralf Elvsén wrote:

>On February 04, 2001 at 20:04:42, Bruce Moreland wrote:
>
><snipped>
>>
>>Often, we start out with two programs and we want to be able to make comparative
>>statements about the two programs:
>>
>>1) A is stronger than B.
>
>This is what I find so hard to understand in many cases. To have the
>information "X is stronger than Y" is to me uninteresting unless
>I know how much stronger it is, i.e. I want an estimate of the
>average score. Some people seem to run matches and then declare
>which program is strongest, but doesn't it really matter to them
>if the average score is 50.000001% or 99.99999% ? And if you
>want this number it has to be given with an estimate of the
>uncertainty... Back to square one.
>
>Contrary to what you are saying (which I snipped and now
>am too lazy to restore) the information "X is stronger than Y"
>only makes sense to me if e.g. you as a programmer were running
>two close versions of your program against each other.

A lot of people want to be able to say this, for whatever reason.

bruce



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.