Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: strongest program

Author: Thorsten Czub

Date: 04:13:02 11/20/00

Go up one level in this thread


On November 20, 2000 at 01:09:16, Uri Blass wrote:
>I disagree.

>If this was the case he could only publish the names of the programs without
>publishing the hardware.

at the time he published the hardware, at the beginning,
he was not sure if junior or df would win.


>He also could not be sure that faster hardware will help to win and there are a
>lot of examples when programs with slower hardware won tournaments.

true.

>You said in the past that the fact that Fritz can use more processors is not
>important because it can only help it to see the draw scores faster.

right. give gambit-tiger ONE 800 cpu and it will kill df on 2 or 4 of these
cpus.
but giving gambit-tiger 500 when others have 1000 and 1100 machines
is really unique method to test the strength.

even on this cpu it made a 3rd rank.


>Did you change your mind about it?

no.

>Do you think that Deep fritz or Deep Junior is the best for people who have
>machine with more than one processor?

i prefer using shredder5.

>I do not know it.
>The programmer does not know it and he said that both programs are at the same
>level.

>I understood from Jeroen that Gambittiger is better but not with the commercial
>book that was build for tiger13 and not for gambittiger.

right.
better make your own book.

>It is impossible to do fair competition
>Equal hardware is also never fair.

equal hardware in not fair. aha.

now i know why we have problems to discuss things ...

>If you use equal hardware of one processor Deep Fritz is going to
>suffer(remember that part of the programming of Fritz was to make it able to use
>more than one processor and without this effort it is possible that they could
>create a better engine).

why not give df 2x500 Mhz ! :-))

>If you use equal hardware with more than one processor programs like tiger that
>can use only one processor are going to suffer  because people with machines
>with only one processor are more interested in the results when all program use
>one processor.
>
>I am satisfied with the fact that we know about the hardware of the players.

yes. i am satisfied too.


>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.