Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: ETC vs. incremental move generator

Author: Robert Hyatt

Date: 12:33:33 01/15/06

Go up one level in this thread


On January 15, 2006 at 11:49:55, Tommi Rimpiläinen wrote:

>
>Could some one explain to me, how to use ET-cuttoffs with incremental move
>generator? Is this simply impossible? I have some ideas, but I don't know if
>they actually work. One could without doubt skip the first move in ETC, because
>it is searched soon enough in any case.
>
>Tommi Rimpiläinen


The ideas are diametrically opposed.  Incremental move generation attempts to
generate the moves one at a time, as they are searched, to avoid the extra work
done in generating moves that get skipped due to cutoffs.

ETC wants to search the move that leads to a quick hash hit and cutoff, but to
do so, you have to try all the moves to see which one will hit the hash table
properly...

So it sounds to me like something that could be combined, but then the
incremental generation would be useless until you get to the q-search (assuming
you don't do ETC there).



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.