Author: Henrik Dinesen
Date: 14:52:12 06/24/05
Go up one level in this thread
On June 24, 2005 at 17:00:29, Steve B wrote: >On June 24, 2005 at 16:56:30, Claude Le Page wrote: > >>I totally agree that 12 moves are sufficient >>Why 12? and not 10 ?Because of 2 openings : Open Ruy Lopez and KID main line >>Else ,programs are quite able , either to find theretical moves or to find >>better ones >>I remember i was making tries on a "minimal" opening book: >>There was a match between Junior7 and ChessTiger15 ( 2 games same line , colours >>reversed ) anopening book gave only : >>1 Nf3 Nf6 2 c4 >>In Ju7/CT15 I got a Neo-Grunfeld , and in CT15/Ju7 ,more amazingly , a >>Blumenfeld gambit >>But for Ruy Lopez and KID it doesn't work if your book doesnt give 10 moves , >>and in2 cases preferably 12 > >there is a real puzzlement to me then >it seems as if many knowledgeable people here agree on a limited book and yet >none of the leading programs use this concept except for Hydra > >i am simply dont know enough about PC progs to understand why this is > >Steve Intentions, I think. When a commercial engine is released, the enclosed is exspected - by the customer - to both suit the engines style (of course) and present "common" chess openings. But it's unclear to the developers how the program will used by the user, so they make a kind of mainstreambook, but deep enough for the engine to play other engines (and not loose because of bad choices too early in a blitzgame). Since most GUI's allow changes in the way the books are handled, this makes some sense if the user know how to adjust the booksettings. Donningers approach makes sense against a GM, but I doubt it works just as well against strong engines with good and deeper books. So the question and the answer is, what is the purpose - the intention - with book? A specific event or a release (and a release for what purpose). Henrik
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.