Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Researching after a fail-low

Author: Vincent Diepeveen

Date: 14:14:26 08/17/00

Go up one level in this thread


On August 17, 2000 at 14:43:08, J. Wesley Cleveland wrote:

>When you have a fail-low in a deep search where the value drops significantly,
>finding an alternate move can take a very long time. This is largely because the
>values in the hash table are largely useless, so in effect we are researching
>the entire tree. It seems to me one should use iterative deepening, and start
>from ply 1 again.

In DIEP a huge fail high takes usually more time as a huge fail low,
as from the root i use -infinite with PVS. If you don't start with -infinite
a fail low can seemingly take forever indeed as all kind of trees that
gave a cheap cutoff in the past are in fact bad moves and now are searched
till very deep in order to conclude it's not the best move!

In several programs they use something called i think internal deepening
or something.

Anyway to describe the trick:

if you have a position without a best move in hashtable, and
still a lot of plies to go, it might help if you search with the
current window to a reduced depth in order to ONLY get a best move
in the current position. You can obviously do this also in an 'iterative
deepening' form. The other advantages over having a best move is already
having searched a part of that tree.

The most horrible thing is if you have a position with a mate in 2
possible (which starts with a piece sacrafice, so a move the ordering
will not select soon), and it searches all moves real slow, then after
a huge while it finally finds it.

If you would have searched that position with a reduced depth first,
then it would already have found this mate in 2 way earlier.






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.