Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Thanks! (and some additional discussion)

Author: Dan Andersson

Date: 10:33:11 08/17/01

Go up one level in this thread


I have tried another hashing scheme. I used one hash table with one entry per
lookup key. I replaced the hash table entry if the minimum of distance in ply to
root node or distance to leaf node of the new entry was less than the current
entry. The rationale of this is to save the, relatively few, early nodes that
can cause huge cutoffs and free the rest of the hash table for use by the
transpositions close to the end of the search. It works ok, but I tried to
better it by using some kind of aging mechanism. One based on aging the entry
the other by allowing a replacement if the distance was one ply greater in
distance thus facilitating a slow creep up or down. Wonder if you have any other
suggestions?

MvH 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.