Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: SSDF. Gandalfs LEARNING is destoyed when aborting!!

Author: Mike S.

Date: 14:44:05 10/18/00

Go up one level in this thread


On October 18, 2000 at 16:53:05, Enrique Irazoqui wrote:

>On October 18, 2000 at 14:49:09, Bertil Eklund wrote:
>(...)
>>You can't be serious?! Do you mean we should accept opening books like Mchess6
>>that' always (90%)of the games plays 1.e4-e5 2.Bc4.

I know that the variety of the M-Chess book is, or was, very low... I've even
once written an article comparing books, where I mentioned that. But I think,
nowadays it's the usual method to set up for an automatic engine match for let's
say 10 games - And without (book) learning, you may i.e. see afterwards that you
have 4 unique games and 6 doubles. Another method would be to *start* as many
games as necessary, to get 10 different one's (and just stop those which turn
out to be a double). I'm aware that this is much more time consuming.

>>I remember an old tournament of Enrique (stand-alone computers).He played 20
>>game matches with the different programs, except with one program where it was
>>impossible to play more than 4 rounds because of the limited opening-book.
>
>It was quite frustrating. Wasn't it Conchess the guilty machine?

It seems the variety of the opponents must have been quite limited too... I
mean, assuming that it played 1.e4 always, there is e5,e6,Nf6,c6, or even d6.
And when the machine had black, there is 1.e4,d4,c4,Nf3. Only 4 rounds?

Regards,
M.Scheidl



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.