Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Killer Move Heuristic Questions

Author: Robert Hyatt

Date: 11:43:31 06/03/99

Go up one level in this thread


On June 03, 1999 at 12:59:02, William Bryant wrote:

>Every time I try to improve my program, I seem to break it in innumerable ways.
>
>The killer move heuristic should be rather easy to implement.
>
>At every ply, any move that generates a cutoff (I interpret as a beta cutoff, a
>fail high), this move is added to the current killer table.
>
>Is it an alpha cutoff (score > alpha) more appropriate?
>

score > alpha isn't a cutoff, unless it is also >= beta.  There are two
reasons for killer updates.  (1) current move fails high (>= beta);  (2)
current move gets backed up as a best move (ie score > alpha, < beta, so it
is a potential PV move).


>Also, does this move automatically displace previous killer moves, or do you use
>some qualifier such as the search score, keeping the moves with the greates
>score?
>

I store 2 killers.  when one of the two causes a cutoff, it is moved to the
front to try the next time.  New moves always replace the second killer.




>Finally, in using the killer moves for move ordering the ranking I am using is:
>1. Hash Table Move
>2. PV Move (in not the same as #1)
>3. Captures with a net positive score (captured piece > capturing piece)
>   Promotions
>4. Killer moves and Captures with an equal score (exchange)
>5. Loosing Captures
>6. All other moves--sorted by the history heuristic.


get rid of 5.  Let those drop into 6...


>
>Any help will be appreciated, I am missing something simple because this is
>slowing down, not speeding up my search.
>
>
>William
>wbryant@ix.netcom.com



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.