Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Hiarcs 10 - Glaurung 051213: 76-24!

Author: Tord Romstad

Date: 02:42:05 12/15/05

Go up one level in this thread


On December 14, 2005 at 22:13:46, Jaime Benito de Valle Ruiz wrote:

>Hi again,
>
>To be honest, I didn't expect less from Hiarcs.

I am also not really surprised.  I had very high expectations for Hiarcs 10,
and so far it doesn't disappoint.

>A program that manages to stay at the top ten with such ridiculous depths (
>I'm talking about V. 9) must have an amazing evaluation/extension scheme

Yes.  Hiarcs 10 has a slightly lower nodes/second count, but searches
similar depths as 9.6.  It looks like the tactically strongest program
I have ever seen.  In the match against Glaurung, it often found some
winning combination two or three full moves before my program realized
that it was lost.

>(I've always had a blind spot for Hiarcs, I have to say).

Me too, as you know.

>I haven't tested Rybka, but I know that programs like Fruit are
>just extremely stable in their eval, and their search provide them with even
>extra stability, so they can be deadly at almost any time controls, but programs
>like Hiarcs use a more aesthetically elegant approach that I find more
>appealing; probably even more now with the version 10.
>Again, we have this thing about different programs' approach working differently
>against different opponents, so maybe Hiarcs' approach is just the super
>anti-Glaurung one, who knows?.

It's possible, but I don't think so.  We'll have to wait for results against
other engines to be sure.  Perhaps I will find the time to run matches against
Fruit and Shredder myself.

>In any case, I am sure that Hiarcs 10 games will
>be worth the waiting.
>In the meantime, and for as long as your engine is 400 ELO above mine, I still
>acknowledge with admiration the job you've done with your engine to remain among
>such bests.

400 Elo?  I don't think so.  At the UEL list, the current difference between
Ayito 0.2.994 and Glaurung Mainz is exactly 200 Elo points.  Considering your
rapid rate of improvement over the last year, neither Glaurung nor Hiarcs can
be said to have a comfortable lead over Ayito.  :-)

Tord



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.