Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: I think that DJ result in the worst case is better than 96DB.

Author: Chris Carson

Date: 11:16:34 07/14/00

Go up one level in this thread


On July 14, 2000 at 13:55:23, blass uri wrote:

>On July 14, 2000 at 09:29:13, Chris Carson wrote:
>
>>.
>
>Explanation:
>
>Deep Junior is commercial and the players got a version that is almost the same
>and can produce almost all the moves in the games.
>
>Kasparov could not buy something that produce most of the moves of Deep blue,so
>the result of Deep Junior is better even if Deep Junior's performance is
>slightly lower than the performance of Deep Blue(96).
>
>I also believe that Deep Junior is better than Deeper Blue if you assume the
>same number of nodes per second(It means that Deep Junior is 50-100 times
>faster).
>
>I believe that the quality of the knowledge in the evaluation function of Deep
>Junior is better than the quality of the knowledge of Deeper blue's evaluation
>function.
>
>It is possible that Deeper Blue had more knowledge but the quality of knowledge
>is important and not the quantity.
>
>Uri

I agree with you.  IBM set out to beat Kasparov and di so in the
second match.  However, Deep Junior has played a variety of Super
GM's and done at least as well with only a fraction of the
hardware advantage.  Create ASIC's from DJ source and you would
would have a very scary monster.  :)

Best Regards,
Chris Carson



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.