Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: A New Self-Play Experiment in Computer Chess: one interesting notice

Author: blass uri

Date: 05:58:13 05/25/00

Go up one level in this thread


On May 25, 2000 at 08:37:12, Mogens Larsen wrote:

>On May 25, 2000 at 08:19:31, Colin Frayn wrote:
>
>>On May 25, 2000 at 05:03:33, blass uri wrote:
>>>No
>>>In the ssdf games the result is about 40% for white 30% draws and 30% for black.
>>
>>There's too many external factors here to make any sort of meaningful deduction.
>> Remove all opening books and there might be a more interesting result...
>>
>>Cheers,
>>Col
>
>Be careful about suggesting things like that Colin. I've been told many times in
>this forum that playing without an opening book is a pointless exercise in
>futility and that incorporating opening knowledge in a program isn't necessary.
>The same goes for endgame knowledge of course. If only there was a way to avoid
>or "solve" the middlegame everyone would be happy and we would have an automated
>database program. Who knows, Chessbase 8.0 could eventually become the strongest
>chess program.
>
>Sarcasm aside, I think it's a very good idea and an interesting experiment,
>which derserves more attention than it's getting. Especially since it's been a
>personal agenda of mine for some time :o).
>
>Best wishes...
>Mogens

I am not against the general idea of playing without the opening books.

I am simply against the idea of doing it without the agreement of the
programmers.

I think that you should at least tell the programmers one year before you do
this test in order to give them time to prepare a better version for this test.

It is not fair to test a program in conditions that the programmer did not think
about.

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.