Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Test suites - can they reliably predict ELO?

Author: Dave Gomboc

Date: 01:07:59 12/12/99

Go up one level in this thread


On December 11, 1999 at 21:14:14, Christophe Theron wrote:

>On December 11, 1999 at 20:18:45, John Warfield wrote:
>
>>On December 11, 1999 at 19:46:50, Bertil Eklund wrote:
>>
>>>On December 11, 1999 at 17:52:56, Tom King wrote:
>>>
>>>>Which of the well known test suites predicts the strength of chess programs most
>>>>accurately?
>>>>
>>>>I ask this, because I recently made some *slight* mods. to the evaluation
>>>>function in my program, Francesca. I ran the LCT-2 suite, and the results
>>>>indicated that it was a wash - the modification gave me about 5 ELO points,
>>>>apparently.
>>>>
>>>>I then ran a series of fast games against another amateur program. I realize
>>>>it's important to play a large number of games, to reduce the margin of error,
>>>>so I ran two matches of 65 games. The result was this:
>>>>
>>>>MATCH 1
>>>>"Normal" Francesca scored 37% against the amateur program.
>>>>
>>>>MATCH 2
>>>>"Modified" Francesca scored 45% against the amateur program.
>>>>
>>>>Quite a difference! It implies that the modification is worth over 50 ELO. I
>>>>guess I need to play more games, against a variety of programs to verify whether
>>>>this improvement is real, or imaginary.
>>>>
>>>>Anyhow, beware of reading too much into ELO predictions of test suites..
>>>>
>>>>Cheers All,
>>>>Tom
>>>
>>>Hi!
>>>
>>>Mr Irazoquis secret test-suite is very impressing! I think it´s about 111
>>>positions. He can predict a new programs strength better than any other test I
>>>have seen so far. If his predictions remains as good as his previous results, I
>>>hope we can stop publishing our list and just play for fun.
>>>
>>>Bertil SSDF
>>
>>  Why is this Test secret??
>
>
>Because this guarantees that nobody will tune his program to do well in this
>test.
>
>I don't know how he does it exactly, but Enrique was able to predict the ratings
>of the new SSDF list a few days before it was published, with a typical margin
>of error of +/- 5 elo points...
>
>
>    Christophe

Maybe he buys an SSDF tester a few beers?... ;-)

Dave



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.