Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Testposition: Fritz6 see not mate in 10, Tiger 14 directly!

Author: Mike S.

Date: 13:00:55 07/13/01

Go up one level in this thread


On July 13, 2001 at 00:38:43, Robert Hyatt wrote:

>(...)

>This is a very sensitive part of all chess engines.  I believe mine does worse
>tactically than most others.  Because I rely a lot on hardware to help with
>tactics, Crafty generally does poorly when compared to other programs on equal
>hardware...
>
>Every program will have a few positions it does well on, and a few it does
>poorly on.  Fritz has more of the former than the latter.  Crafty is the
>opposite, in general.

My impression is, that Crafty is more than competitive compared to Fritz and
other top engines, in terms of tactical speed. Maybe Fritz solves some more
within blitz time ranges, but Crafty looks like a solid "finder" when given
longer time of some minutes. My test contains mostly positions with easy to
medium difficulty (by today's standards), and I've tested on single cpu always.

I don't know if I would come to another conclusion, if I would use a lot of
extremely difficult positions only... I doubt that this is better to predict the
strength in practical play, compared to tests with a "bandwith" (but those
cannot predict it either anyway :o).

Regards,
M.Scheidl

Btw. I think programs are still not obliged to announce any # in 10 under
standard time conditions (I'd expect any # in 5, and most # in 6). Maybe we can
add one move for 40/2h on fast computers. Not talking of database endgame
positions of course, or of positions near of that.



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.