Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Hiarcs 7.32 leads Fritz 5.32 on an inferior machine.

Author: Harald Faber

Date: 03:41:06 07/22/99

Go up one level in this thread


On July 22, 1999 at 05:43:38, hgkjhg wrote:

>Well, fritz 5.32 is the program that played at wmccc and at frankfurt open.
That is still WRONG as Uri pointed this out already.
In both events a parallel version on 4 processors AFAIK played.

I'm
>not saying it's the best in comp. vs. comp. games, shredder is.  But all these
>results from shep or from Tania Devora don't seem very reliable.  Chess Tiger
>and Hiarcs seem to be a lot better than other programs there, but they didn't do
>so well at wmccc, for example.  So may be the opening books make the difference.

The difference is that you conclude from few games. Like most people do.
If you would have played another tournament after the WCCC with exactly the same
programs without any changes, the result wouldn't have been the same again.
Of course we got an impression of the strength and the style. But nothing more.
If you want to find out if Hiarcs or Fritz is better, then take 2 PCs at about
P2-400 speed and play at tournament control

Fritz5-Nimzo2000
Fritz5-CM6000
Fritz5-MCP8
Fritz5-Rebel10
Fritz5-CSTal2
Fritz5-Shredder3
Fritz5-Junior5

at least 20 game match each which would give 140 games. (Even 20 game match
isn't very much...) Then add

Hiarcs-Nimzo2000
Hiarcs-CM6000
Hiarcs-MCP8
Hiarcs-Rebel10
Hiarcs-CSTal2
Hiarcs-Shredder3
Hiarcs-Junior5

and compare the results. Of course you may add Genius, WChess and Zarkov, the
more opponents the more games you get and the better is your comparison.



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.