Author: Andrei Fortuna
Date: 04:41:14 07/10/03
Go up one level in this thread
On July 10, 2003 at 07:02:15, Jens Kahlenberg wrote: >On July 10, 2003 at 06:27:07, Andrei Fortuna wrote: > >>Perhaps I was didn't express my goal as I should have. >> >>While my ultimate goal is to teach my chess engine to play good chess, I cannot >>achieve this without understanding myself as a human player the modern chess >>play. Therefor the need for books to study. >> > > >I guess that Watson's "Secrets ..." won't help you in teaching your engine, but >it will help you in understanding current GM practice a lot. Watson and Dorfman >have nearly totally different viewpoints, so Watson's "review" might be a bit >coloured by emotions (e.g. "... with his Mostly Inapplicable Pedagogical >Oberservations"). Probably you are right. >Watson is IMHO not implementable, but Dorfman might be. I prefer not to think in this way. I don't want to take a chess book and expect to find a coherent rule system which I can implement in my engine. I don't think this is possible either. I agree that I would have a hard time extracting knowledge from Watson's "Secrets ... " that I could use in my chess program. Perhaps this is not the way ! I want to reach a chess knowledge level when I see my program playing a move, detect if it's bad and understand the reason why it's bad. Then I can tune the eval function accordingly. >Here's a link to a more >uncomitted review: http://www.chesscafe.com/text/review313.pdf Thanks, I guess there might be some adepts in Dorfman's methods. Still I think it is very controversial and moreover it cannot be ordered through chessco or thechesshouse (the ones I'll use for my current order) so I'll have to pass investigating it. Thank you for signalling it to me though ! Cheers, Andrei
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.