Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Tests for PC programs.

Author: Dann Corbit

Date: 15:37:36 09/09/99

Go up one level in this thread


On September 09, 1999 at 17:42:56, Bradley S., Short wrote:
>Are there any fairly accurate rating tests for PC programs?  I have the
>One-Hour CCR test from the Winter '94 review but I'm uncertain if its even in
>the ballpark.
>Chessmaster 5500 on a Celeron 366 scored 2632 while Chessmaster 6000 managed
>only 2608.   The real suprise was Fritz 5.0 when it brought up the rear with
>2592.
>I thought 6000 was stronger than 5500.  If it is why did it not do as well on
>the test?  If I had the hardware available I'd play the two against each other
>in a match just to see.  Has anybody ever played 5500 and 6000 against one
>another on equal hardware?  If so I'd be interested in the results.
That tactical suite gives a rough measure of performance.  However, there is no
certain connection to playing strength.  Play a few thousand games between the
programs if you really want to determine which one is strongest.  By the time
you have completed (of course) new iterations of the products will be out and
the tests will have no value.
;-)




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.