Computer Chess Club Archives




Subject: Re: Crafty CCT6 notes

Author: Robert Hyatt

Date: 08:00:08 02/02/04

Go up one level in this thread

On February 02, 2004 at 10:48:10, Tord Romstad wrote:

>On February 02, 2004 at 10:30:16, Robert Hyatt wrote:
>>There are some "impressions" I get when playing other programs.  IE with
>>my dual xeon, when playing "deep shredder" I always have the feeling that
>>Shredder knows everything my program knows, but searches deeper.  I win
>>some games, but it always feels like I am pushing a bus uphill to make any
>>The new hiarcs seems to feel a lot like shredder,
>(Another big snip)
>>The search never "crumbled" as sometimes happens on my xeon when my
>>opponent has better hardware and a better search too (shredder/hiarcs for
>This raises the obvious question:  What do Shredder and Hiarcs do to
>achieve the bus pushing effect without having a massive hardware advantage?
>How are they able to outsearch strong and (NPS-wise) much faster programs
>like Crafty?  They do forward pruning, of course, but I wish I had a better
>idea of precisely what tricks they use.  One of the many reasons I look
>forward to the Palm OS version of Hiarcs is that I will finally have the
>chance to play with the engine myself and see if I can figure out some of
>its secrets.

I hope you manage.  :)  As to what they do, I don't know.  But I can certainly
say with some authority that it works pretty well.  :)  But it is more than
search alone.  IE Shredder seems to have a significant "feel" for a position as
well.  IE in several Crafty games yesterday, someone would fire up shredder to
see what it thought.  One such case was where Crafty had +1.0, and Fritz was
saying "0.0".  Shredder weighed in with +1.0.  And this was all about a
king-side pawn majority in the Crafty-Hiarcs game.  Hiarcs fits the same mold,
IMHO.  IE it was at about -1.0 at that same point.  So doing a good evaluation
_and_ having a good search, is impressive.

Perhaps one day someone will figure out their "tricks".  Or someone else will
develop them again, and avoid the "go commercial" tendency and really produce
some improvements in computer chess overall.

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.