Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: What makes Kasparov Vs DJ more attractive than Kramnik match !

Author: Peter McKenzie

Date: 18:19:30 09/03/02

Go up one level in this thread


On September 03, 2002 at 17:17:08, Mig Greengard wrote:

>The point is not when Kasparov gets anything, it's that the Junior team can
>bring *anything they want* to the table. There are no constraints on which
>version they use, so Kasparov's "fresh copy" could be as different from what
>they use in Jerusalem as a version from a year ago. It's not as if the Junior
>team had to start with the copy they gave Kasparov!
>
>Assuming that what they had in Maastricht was their best, that's still not a lot
>of time to make meaningful changes and test them thoroughly, which I pointed out
>in my article. But it's not at all the same as having a completely
>version-frozen situation as in the Deep Fritz - Kramnik match. If Kramnik can
>get the openings he wants just one or two times there is no reason why he can't
>play entire winning games from memory. This is improbable, but possible. And
>just doing it once turns the match into a bit of a joke.

This is based on the assumption that that Fritz is strongly deterministic, which
could easily be incorrect.  For example if Fritz slightly randomised its
thinking time for the first few moves out of book, it could easily be enough to
send the game on a completely different track.

And there are other ways to make a program less deterministic, like adding small
random values to piece square tables before a game.  I don't know if Franz used
any of these methods, but it strikes me as useful insurance.

cheers,
Peter

>
>My full article is here, by the way:
>
>http://www.chessbase.com/columns/column.asp?pid=149
>
>Saludos, Mig
>http://www.chessbase.com
>
>
>On September 02, 2002 at 18:27:38, Joachim Rang wrote:
>
>>Kasparov also receive a fresh copy of Deep Junior a few days ago. So the
>>circumstances will be the same (it's still impossible to change much in  a few
>>weeks).
>>
>>I suppose that both matches will be very close, but I think the humas will
>>survive.
>>
>>Does anybody knows something about the hardware which will bi used?



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.