Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: search extension

Author: Robert Hyatt

Date: 06:55:58 07/08/98

Go up one level in this thread


On July 08, 1998 at 06:20:18, Inmann Werner wrote:

>To make ma program faster I included Null-Move and a cutoff one ply
>before the horizon, if the position is "good".
>On the other hand, simple mates go beyond the horizon (also because of
>small quiet search).
>
>My idea:
>
>If I do a Null-Move it Cuts if greater Beta. OK
>if it is much lower than alpha I extend the search, because it could get
>to some interesting mate.
>
>The idea behind: In Null move the other side can do 2 moves in row. If this
>gets to a very high score (mate) it is interesting enough to extend.
>I made some tests, and the program really found the mates 1 to 2 plys faster.
>Also, the program does not get much slower.
>
>Is this a "good idea". Why not use Null-Move the other way round?
>
>Comments?
>
>Werner


I think the idea is one that has "been tried and found wanting".  It has
problems for a couple of reasons:

1.  many programs use PVS or Nega-scout algorithms, which means that 99.999%
of all nodes searched have alpha=beta-1, making it difficult to return a
value "very much lower than alpha" to trigger this.

2.  once you try it for a while, you find it has simply cost you a lot of
time.  I don't know of anyone using this method any longer, including
Chrilly...

But there is one useful thing you can do...  at ply N+1, you *can* return
an accurate score, namely "mate".  And this can be useful.  Bruce was the
first I know of that used this method, and it works well.  If you do nothing,
and you get mated, then there is obviously a "threat" in the position and it
is worth searching deeper.  I do this too, but *only* for mate, because I
can't detect when the score is "well below alpha" because it can't happen in
PVS.



This page took 0.1 seconds to execute

Last modified: Thu, 07 Jul 11 08:48:38 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.