Author: Robert Hyatt
Date: 09:51:09 01/30/03
Go up one level in this thread
On January 30, 2003 at 09:02:06, Joel wrote: >Hey All, > >Just wondering whether anyone here has had any success with some kind of history >heuristic ordering approach as compared to say MVV/LVA? > >Also, are there any good sites which talk about implementing a SEE? I have been >playing around on paper, but I suspect I am using more data structures than I >really need. Should I expect to be losing some degree of accuracy when >implementing a SEE as compared to a simple MVV/LVA approach? > > >Regards, > >Joel Using SEE you should see _more_ accuracy, not less. SEE is inherently more accurate because it can detect that QxP wins a pawn while QxR loses a queen for a rook. MVV/LVA will like QXR better (most valuable victim first). You can look at the Crafty source, "swap.c" to see how it might be implemented... It is not a lot of code.
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.