Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: How to use a [cough] EPD test suite to estimate ELO

Author: Paulo Soares

Date: 23:18:09 02/11/99

Go up one level in this thread


On February 11, 1999 at 15:41:25, Dann Corbit wrote:

>Andreas Schwartmann asked an interesting question in r.g.c.c.:
>"I wonder if anyone can enlighten me on how to use various test suites, like
>LCT, LCT II and Covax. There are ceratin formulas on how to calculate the
>playing strength according to these test suites, right?"
>
>Now, ignoring the fact that they are full of bugs and the measures are probably
>bogus, how *does* one arrive at an ELO from a test suite evaluation?
>
>What is the actual mathematical basis for the calculations?


The history shows that many scientific truths had been
concluded through the experimentation.  I compare the tests with that,
We are still in the phase of the experimentation and far of reaching a
definitive reply, a mathematical formula capable to find the relative
force between programs, if this formula exists, inside of the
mathematics for us known.
	On the other hand, exactly knowing that the tests are far to
evaluate, with small margin of error, the relative force between the
programs, I believe that the first thing that the majority of us makes
when buying a new program, is to test it in a suite of any tests.
I, for example, like the LCTII suite.  I think that these tests also
help to sell programs, and that is important for that they like computer
chess.
	Summarizing, that more tests become, of preference with more
knowledge, so that if it diminishes the margin of error, and so that
let us can in them amuse more.
	Can the proposal of Bill McGaugh next to the reality?  I think
yes, and I go to test it!!!

Best regards,

Paulo Soares, from Brazil.



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.