Author: blass uri
Date: 15:55:37 10/14/98
Go up one level in this thread
On October 14, 1998 at 18:20:23, Robert Hyatt wrote: >On October 14, 1998 at 13:26:50, blass uri wrote: > >> >>On October 14, 1998 at 07:30:53, odell hall wrote: >> >>> >>> If anyone is interested I.M. Jonathan Schroer expressed a desire to play a >>>match at 40/2hrs against any program. He told me he would be glad to take on any >>>computer. I asked him if he thought that he could defeat rebel, his answer was >>>that he doesn't know!. >> >>If rebel is a deterministic program like crafty or fritz5 it is very easy to >>beat it in a match. >> >>You can play 1.f3 to get rebel out of theory and after it to play good moves you >>learned before the match. >>I am not interested in a match between humans and commercial programs if the >>commercial programs are deterministic. >> >>Uri > > >Crafty is not deterministic either. I remember that you posted before some monthes that crafty needs a big book because it is deterministic and if it plays without a book it is going to lose after players learn it. Did you change it? > First, my opening book has several >*thousand* games with f3 in it. I thought that nobody plays it. I know that even in the powerbook of fritz5 there is no 1.f3 Uri > But even without that, it does some "learning" >along the way and will vary from game to game, without a book of any kind... > >Can't survive on a server without that..
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.