Author: Gian-Carlo Pascutto
Date: 07:49:45 09/20/01
Go up one level in this thread
On September 20, 2001 at 10:34:55, Uri Blass wrote: >I doubt if Diep has a better evaluation function that Goliath. >searching less nodes per second does not mean better evaluation function. Vincent will *love* to hear this ;) >It does not happen but it also does not happen that a crappy positional program >can play great chess. > >If you use only material evaluation you have no chance against the top programs >even if you use the best hardware that is available against 386. With only material, I'd doubt it. But piece-squares and very basic passed pawns at 200Mnps should hold up vs. Rebel or whatever you prefer on an 386sx16. >>Sometimes the search will take ages to see something which is >>'eval' in a positional program. But the connection is certainly >>there. >> >>My program can solve several 'positional' positions from the Louget >>II test set with a material-only eval, by just looking deep enough. > >My guess is that it is not going to solve most of the 3 positions that I posted. >I guess that Be2 is the only position that may be an exception but it is going >to need a very long time to see it by only material evaluation. 'Not going to solve' is a very broad statement. I say it will. I won't say if it will finish before the universe collapses. Tactics and positional evaluation are, in essence, the same for a chessprogram and improving one will improve the other. -- GCP
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.