Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: My $.02 for what its worth

Author: Mike S.

Date: 11:29:50 04/06/02

Go up one level in this thread


On April 06, 2002 at 01:34:49, Les Fernandez wrote:

>(...) If the player
>has tremendous tactical knowledge far superior then all the others then the
>relevance of positional understanding is not as important.  Having said that who
>cares what makes a particular programs rating high lets just apply the same
>rules that have applied to all chess players over the years.

Very true. Ratings should be compared on the same basis, IOW. the practical
performance from lots of games should be compared. Wins, draws, losses. I think
it's more than obvious what programs can achieve nowadays.

In a similar discussion (on a german usenet group), I was pointed to the Tim
Krabbé website which has a lot of interesting anticomputer examples, intended to
show me what computer still can't and still don't understand etc. - But the
problem is, some people seem not to see that these are just exceptions, which
cannot stop programs from achieving GM performances.

Or some people just don't want to see that, for whatever reason. If matches end
Program vs. GM, 99.0 - 1.0 they will still show trojan sacs or the like, and say
comps are not GM strength.

They don't apply the same rules. It is unfair against the programmers IMO.

Maybe it is recommendable for such discussions, to say "Computers are already
*much better* than human GMs - you can't prove the opposite when you don't apply
the same rules. So I say comps are better when there's no fair evidence against
it, but a lot of matches and tournaments which prove it"

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.