Author: Dieter Buerssner
Date: 09:51:16 04/13/02
Go up one level in this thread
On April 12, 2002 at 18:14:58, Gian-Carlo Pascutto wrote: >It's my undestanding that for MTD to work well you have to >do the adjustment too. > >So why aren't all MTD programs weak? Why should they? This as a direct followup to my post seems to indicate, that you assume, that I believe that (almost) allways trusting previous search results (if it was a fail high, it must be true) is a bad idea. I don't think so, it is a compromise in search, with pros and contras. And actually, it will make the search more stable, not less stable (but may need now and then more depth to see the same thing). Very typical of many things used to improve plain alpha-beta search. Regards, Dieter
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.