Author: Tony Werten
Date: 00:53:36 01/05/02
Go up one level in this thread
On January 04, 2002 at 20:05:45, Dann Corbit wrote: >On January 04, 2002 at 17:40:54, Nino wrote: >[snip] >>Listen I dont have a programming language on my system. Is there any way you >>can send me just an exe file of this app?? I would be curious to see this demo. > >ftp://cap.connx.com/pub/les/Least/EGTB8.exe > >>BTW when you say "fuzzy" do you mena inaccurate?? And if so how can we count on >>that??? > >Not inaccurate. It is a limit. Just like in an alpha-beta search, you have a >window. In this case, it is like the Deep Blue search, in that the window is >one sided. So (for instance) if a position has a mate in 8 score, it could >possibly be a mate in 4 but it won't be any worse than a mate in 8. So, >absolutely for certain, you will still know that you have won the game. It could give some problems. If you receive a matescore you will research. In that research the pv might change. You run the risk of not finding the endposition in the table wich would leave you with: fail high, research, not resolved. Next ply: failhigh, research, not resolved etc. ( Been there) The idea will still be working since the pv that your opponent will have to follow to escape the recognition will have a worse score for him. Tony
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.