Author: Artem Pyatakov
Date: 14:20:13 08/03/01
Go up one level in this thread
On August 03, 2001 at 17:05:37, Dan Andersson wrote: >I have noticed a certain gain by using ETC. The time to ply has improved by some >percent. But more importantly the ETC version is better on tactical test sets >and real positions. I.e. the ETC never makes a worse move (in 'very few' cases >it does, but that is mainly because the program played the right move for the >wrong reason) and in some cases provides a better answer. Thanks for the reply. Would you mind expanding on how do you implement ETC? More specifically: 1) Do you apply it to all moves? 2) How do you actually use it inside the search? Do you actually use it to exit the search early or do you just use it to "guide" the search to sucessors that will cause a cutoff? Artem Pyatakov
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.