Author: Dan Andersson
Date: 15:10:04 02/06/01
Go up one level in this thread
I have only tried it out pn/pn^2 a little as I cannot find a useful application for it in classical chess. I still use ab-search in crazyhouse but also pn instead of q-search. Its use in crazyhouse is predicated on the following observation. One cannot search all that deep in crazyhouse as the amount of possible moves are high (It may be that the number of transpositions make up for that, but I doubt it) hence I wanted something that would signal if a leaf position was a forced win or loss (Obviously the number of nodes to be searched need to be limited, in a dynamic way) else I return the heuristic value of the leaf node. Used in conjunction with iterative deepening and ETC the tree will be smaller due to absoulute value cutoffs. I must mention that there might be better win-loose deciders that pn but I decided for it as I felt lazy. Regards Dan Andersson
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.