Author: Vincent Diepeveen
Date: 06:31:49 06/22/99
Go up one level in this thread
On June 21, 1999 at 16:51:12, Michel Langeveld wrote: >Kind regards, > >Michel Langeveld An experiment of me some years ago. program only searching material played: 1.a3,d5 2.b3,e5 3.Ra2,Bf5 etc. Rating i'm not sure. it's awful bad... ...it searched 18 to 20 ply, but the trick in this deep search was it playing actually anti-positional, so it could search very deeply becasue of that. A big improvement is already a small table which says pieces must go to the centre, then it's already having a way higher playing level, however it's tough to search deep then. It's a major problem actually to search deep then, without much further work. My estimate: tactics only ==> 1100 piece square tables (without preprocessor) ==> 1800 better qsearch (better tactics overall) ==> 2100 also better preprocessor ==> 2400 excellent preprocessor ==> junior, genius Greetings, Vincent
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.