Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Comments on SSDF by Mr.Diepeveen

Author: Uri Blass

Date: 01:28:25 03/06/04

Go up one level in this thread


On March 06, 2004 at 04:17:14, Sandro Necchi wrote:

>On March 06, 2004 at 03:25:50, Thorsten Czub wrote:
>
>>On March 06, 2004 at 03:10:22, Sandro Necchi wrote:
>>
>>>They do their best to give reliable data without favoring any program and
>>>checking the games before recording the score.
>>>This is how they always did.
>>
>>this is what we are discussing Sandro, if they give their best. this is part of
>>the critics for years now. because people have different kind of standards
>>concerning this "they do their best".
>
>I agree on this, but they cannot find something that everybody will agree on.
>I was stating that they test the commercial programs with the strongest
>settings. To do that they ask the programmer to advise the strongest settings.
>
>I said that I did not always agreed with them, so I am not saying that I think
>how they do it is the best, but I know they have been trying to find a way to
>get the most reliable data.
>
>Of course suggestions are welcome. When I gave them some, most of the time they
>were accepted and everytime they lesson and talked about them.
>
>Since they are independent, we cannot force them to accept our suggestions, but
>if you give them good suggestions I am sure they will consider them.
>
>>
>>IMO it would be within a normal range of "doing the best" to make sure that a
>>company is not getting special advantages due to a special autoplayer system.
>>also IMO it is within the normal range of "doing the best" to make sure
>>double-games are not counted again and again, therefore you need to collect the
>>data and make them public so that all testers can see when they play double
>>games. We have seen this list beeing presented like a holy grail for many years.
>>
>>and in the years of the dedicated chess computers the double-game topic was IMO
>>very relevant due to the fact that the books were small and the variety of
>>openings and games was very limited.
>
>Yes, but on this matter I do not agree with you. If a program has no learning
>feature and will play a lost game more than once it has to be tested as it is
>and the double game included in the list.
>Why?
>Because the owner will be facing the same situation in a match against it and
>since this effects the streght of the program it must be included to get more
>realistic data.

There is a second possibility
Not to include programs that behave like that in the list.

Suppose that program A has no learning and can lose the same game again and
again.

Program B played against A 40 games when program C played against A only 15
games
Is it fair that B will get better rating than C?

Uri



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.