Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: EG-database contribution to Kasparov-v-ROW?

Author: Robert Hyatt

Date: 06:47:20 09/14/99

Go up one level in this thread


On September 14, 1999 at 08:48:18, guy haworth wrote:

>Peter Marko at http://bbs.msnbc.com/bbs/kasparov-team/posts/cb/66952.asp is
>asking for volunteers to work on EG-database generation to assist the World Team
>v G Kasparov.

I don't want to be rude, but somebody doesn't have _any_ idea about what this
is all about.  You don't just generate a 6 man tablebase with 2 pawns in a day
or two.  This will be a huge file, will require a 64 bit file addressing
capability, and a _bunch_ of memory.  And a couple of weeks per file I would
suspect...

KQPKQP is one heck of a database.  Tell those guys that this is not some
simple thing they can run overnight and be done with it.


>
>The game is heading for KQPKQPP (and maybe KQPKQP and KQQKQQ) which is of course
>a lot harder to work on than KQPKQ.  Most of the World Team are now assuming
>that the game will arrive at:
>
>    7Q/1p6/3p2K1/6P1/8/8/8/1k1q4, wtm ... aka 'Position D'
>
>Is there any interest at CCC?  I think the task is ambitious but maybe some
>sub-problem EG-databases which could assist the World Team if not completely
>remove the need for human inspiration.
>
>If it possible to work on KQQKQQ, or on KQ(g5-7)PKQP(b7-2 or d6-2) - even maybe
>assuming P=Q for any conversions, this would be a good contribution.
>
>Is anyone following the game?  Any views on the event, game and current
>position?



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.