Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Me Vs Some Programs

Author: Ralf Elvsén

Date: 03:37:25 09/18/99

Go up one level in this thread


On September 17, 1999 at 20:32:23, Peter Kappler wrote:

>
>Ralf,
>
>I believe your results, but you really need to play more games before drawing
>any serious conclusions.  I *promise* you that both of those programs are well
>above 2000 strength at 40/2 time controls.
>
>It's possible that your style of play is very effective against computers, and
>that this allows you to perform better than your estimated 1700 rating would
>suggest.
>
>It's also possible that you just got lucky in the first few games, and that you
>will lose the next 20 in a row.  :-)
>
>Check back with us after another dozen games or so...
>
>--Peter

Thanks for not flaming me, Peter :)

Yes, I need to play more games, but it is very time consuming.
And believe me, I *know* these programs can play well above
2000. If I try to moderate my statement to an extreme we
might both can agree on, it would go something like this:

If e.g. a 2300 player plays a number of games and tries to
win (and then he have to take risks), the programs will outplay
him tactically often enough to gain
their "rightful" performance rating = Rx , much above 2000.

When they play someone like me, their poor long term planning
fails sometimes and they get a performance rating Ry. Ry might
be much higher than what I indicated (1700 + 265), but...

Ry < Rx , right?

Note: for those of you who doesn't quite understand the
rating system: this does NOT imply that I would play
"better" than the 2300-player.

Ralf



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.