Author: Uri Blass
Date: 13:26:45 10/03/01
Go up one level in this thread
On October 03, 2001 at 16:17:55, Dann Corbit wrote: >On October 03, 2001 at 15:50:36, Sergei Smith wrote: > >>A couple of centuries ago they had never heard of >>"Najdorf" so we can assume that in the future new and >>better openings will be developed that >>would currently still be considered A00 irregulars and >>that new variations or completely new openings will be >>used that we currently do not even expect. >> >>My purpose was to build a shallow/broad opening book. >>I wanted to run a 6-ply or maybe an 8-ply brute >>force search and import all EPD into an experimental >>opening book for the ChessBase GUI. >>Here arises another problem since the GUI will build >>its book from moves but not from positions. >>To this end we need to look to "all possible moves" >>within 6|8 ply from the start position >>instead of "all possible positions". >>If you want to help with this project, be welcome. > >There is a truly monumental distance between 6 ply and 8 ply. >There is no purpose beyond analyzing the positions. If you can get to the same >board position 12,000 ways, why analyze it more than once? > >Here is the number of moves to get to just 6 ply: 119,060,324 > >88,319,013,353 is the total for 8 ply. That's billions with a 'B' ;-) > >I suggest that you use Bookup instead. It will import EPD and be happy with >that. > >There are almost 15,000 certain checkmates just within the first 5 plies. Why >will you want to try to analyze positions forward from that? In other words, >once you know some position is a mate in 3, there is no point to analyze the >next two plies from there. I can add that even if there is no checkmate there are position when it is obvious not to analyze. For example every position after 1.e4 Nf6 2.Qg4,1.e4 f5 2.Qg4,1.e4 Nf6 Qh5,1.e4 h5 2.Qg4,1.e4 g6 2.Qh5 Uri
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.