Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: hiarcs7.32 vs fritz 5.32 question

Author: Melvin S. Schwartz

Date: 07:03:06 06/13/99

Go up one level in this thread



On June 12, 1999 at 16:57:20, blass uri wrote:

>
>On June 12, 1999 at 16:31:34, john c cook wrote:
>
>>hi chess player i got hiarcs7.32  two day  a go and i play some game with
>>fritz 5.32 to see who is the king whit my two computer 256meg pii400
>>fritz5.32 beat all the game i gave both program 120meg for hash tables
>
>How many games did you play and what was the time control?
>
>If you played in fast time control then maybe Hiarcs7.32 was slowed down because
>you gave it too much.
>
>I think that you should not give Hiarcs7.32 120meg even at tournament time
>control because it cannot use them.
>
>>
>>after that i play on one computer a engine vs engine match whit 64meg of hash
>>and hiarcs 7.32 won all the game my question is should i play all hiarcs7.32
>>game whit only 64 meg hash i see fritz5.32 love 132 meg for hash
>
>
>I think that if you have 2 identical computers then the fair way to test is to
>do games between them because the programs can think during the opponent's time.
>
>You need to check if the spped of both computers is the same before you do a
>match between them(if one of them is slowed down by some reason then it can
>effect the result).

Hello Uri,

He did state above he had two computers and listed only the P400, so let's
assume for the moment that he has two identical computers. I have tested Hiarcs
with 131 megs and 64 megs and found it made the same moves. Hiarcs, I believe,
does not use HT as Fritz and therefore will only use a very small amount of even
64 megs. On the other hand, as I have stated to you before, Fritz is very
dependant on the amount of HT available. On a Pentium 400 Fritz needs 144 megs!
The testing run above with Fritz getting 120 megs is a lot more than what Tania
is giving Fritz. Without knowing the time control, although from testing done at
fast time controls Hiarcs clearly is better than Fritz, I would assume he tested
at a reasonable time control and hopefully at 40/2 but we'll have to wait for
his reply. I know his results are surprising but then the SSDF tested Hiarcs 7
against Fritz 5.32 at 40/2 and the results were 22-18 in favor of Fritz!

Mel
>If you use an autoplayer then it is important to check after the games if the
>moves can be reproduced manually.
>I know that Ed discovered in games of Rebel against Mchess that Mchess did poor
>result because of the autoplayer.
>
>Uri



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.