Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Hiarcs 10 in ECMGCP

Author: Tord Romstad

Date: 06:20:35 12/15/05

Go up one level in this thread


On December 15, 2005 at 07:21:18, stuart taylor wrote:

>On December 15, 2005 at 06:16:30, Tord Romstad wrote:
>
>>I just finished testing Hiarcs 10 in the ECMGCP test suite on an iMac G5
>>1.8 GHz (roughly equivalent to a Pentium IV 2.4 GHz) at 10 seconds per position.
>>The result was a remarkable 161/183!
>>
>>As a comparision, Glaurung 051213 scores 152, Shredder 9.1 137, and
>>Fruit 2.2 127.  The score of Hiarcs 10 is the highest I have ever seen.
>>
>>Tord
>
>Before we get too much into it, do you have any results of Rybca? CAN you?

I can't test Rybka, because a Mac version is not yet available.

>After that, if Hiarcs is still top, then we need to see how well, Hiarcs, knows
>how to apply its knowmedge in games, too.

Yes, the ECMGCP results by themselves mean nothing, of course.  They are
not even a good estimate of the tactical abilities of the programs.  My
own program scores better than most in ECMGCP, but in reality tactics is
probably its biggest weakness at the moment.  Fruit is tactically much
stronger than Glaurung, but scores far worse in ECMGCP.

The ECMGCP scores are a curiosity, nothing more.

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.