Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: New paradigm ???

Author: Robert Hyatt

Date: 17:48:18 11/10/00

Go up one level in this thread


On November 10, 2000 at 19:20:09, Peter Skinner wrote:

>>That is the kind of thing I would worry about, if Crafty was doing it.  Crafty
>>is presently not particularly fast.  I am pretty sure that I could make it at
>>least 2x faster, if I was ready to cast a lot of things into stone, and _really_
>>muck up the readability/understandability of the source.
>
>I actually find the Crafty 17.14 version very slow, and results are proving that
>possibly something might be wrong. In a match vs RoboElvis(C) on FICS using
>Crafty 17.14 on a dual PII 300, vs my Gambit Tiger on a PIII 500, I scored
>81W-15L-30D, this is rare, as usually Roboelvis destroys me.
>
>When he uses Crafty 17.13, the win% is lower, and way more draws.
>
>>And that incurs a penalty I don't mind, because I can pick up that factor of two
>>with the right hardware...
>
>Crafty used to be a brute. What is the explanation as to why it has fallen
>behind so far? Not to say really far, but it seems to be having more problems
>with commercial programs than before.


17.14 is not appreciably slower than 17.13...  The benchmark positions take
longer, but that is very common.  The eval changes simply change the shape of
the tree it searches...  Some positions are faster, some are slower.  The
endgame positions in the benchmark suite take more nodes to search, with the
new endgame stuff.

As far as win/lose/draw differences, there should be little difference other
than 17.14 should recognize new classes of won/lost endings that 17.13 would
not understand...



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.