Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: SSDF Rating list 03-07-06(For Amir Ban)

Author: Uri Blass

Date: 09:15:08 07/07/03

Go up one level in this thread


On July 07, 2003 at 10:38:13, Louis Fagliano wrote:

>On July 06, 2003 at 04:46:28, Thoralf Karlsson wrote:
>
>>Five months ago Shredder 7.0 appeared on the first
>>place of the SSDF Rating list after a rating increase
>>of more than 40 points. Now Stefan Meyer-Kahlen and
>>Sandro Necchi can be proud of another 40 points, which
>>places them 48 points ahead of the nearest competitor!
>>
>>After 465 games Shredder 7.04 UCI Athlon 1200 MHz, played
>>under the Classic Shredder-interface, has a rating of 2810!
>>Congratulations!
>>
>>We have started to play with Deep Junior 8.0, on a single
>>Athlon 1200 MHz, and will present the result on next
>>rating list, which will appear in August or September.
>>
>>Thoralf Karlsson
>
>Why not Junior 8 on a single Athlon 1200 MHz?  Deep Junior 8 on a single
>processor runs more slowly than Junior 8.  If you are going to use Deep Junior 8
>at all, use it on multi processors or don't use it at all and use Junior 8.

Are you sure about it?

see http://www.talkchess.com/forums/1/message.html?304955

What is the reason for using more than one thread for a single processor
machine?

Is it a bug?

In that case it is hard to imagine that a programmer who develops a special
version for one cpu is going to use 2 cpu's by accident so the only possible
reason that I can think about is that chessbase simply put the wrong file in
Junior8 and instead of using a file of Junior8 used a file of Deep Junior8 but
in this case it is easy to fix it.

The fact that chessbase probably did not fix it(otherwise mike byrne could use
Junior8 and not Deep Junior8 in his tournament) suggest that it is not the
reason so what is the reason?


A possible reason that I can imagine is that chessbase know that people use
duals and may want to test Junior8 against other programs in a dual.
In other words the reason is cheating the customers to think that Junior is
better than it's real level.

Fortunately posters like Mike byrne noticed the problem but Mike did not notice
the problem in the first game and there may be other people who did not notice
the problem and I wonder how many people ordered Junior8 based on reading
results like mike byrne's results or hearing results of their friends without
knowing that Junior8 is using more than one cpu.

Maybe Amir Ban can explain better what happened.

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.