Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Singular extensions

Author: Robert Hyatt

Date: 13:20:43 01/08/98

Go up one level in this thread


On January 08, 1998 at 12:34:26, Don Dailey wrote:

>On January 08, 1998 at 11:48:22, Bas Hamstra wrote:
>
>>On January 07, 1998 at 22:51:59, Robert Hyatt wrote:
>>
>>Bob, a while ago we had a short conversation about singular extensions.
>>
>>You said something like if a move is way better than the rest research
>>that move till Depth+1.
>>
>>But what is the basic idea of singular extensions? Is it something like
>>q-search but then not for captures but forcing moves? How could it lead
>>to searching at greater depth *average* as was said of Deep Blue?
>>
>>Could you tell something more about this, about the general idea, not
>>the implementation? (I'll be back for that later hopefully).
>>
>>
>>Regards,
>>Bas Hamstra.
>
>
>Bas,
>
>Singular extensions is the concept of granting an extension if there
>is only a SINGLE clearly best move.  "clearly best" is defined by
>some positional margin for instance 1/2 pawn.  Determining singularity
>is determined by a reduced depth search with zero window to prove or
>disprove singularity.  There  are some implementation complications
>you should be aware of too but this is the basic high level idea.
>
>- Don


Note that the reduced depth trick is only used on "fail-high singular
moves"
and not PV-singular.  IE you search a node and the first move would fail
high.  Are you *sure* it is way better than the others?  If not, you
have
to somehow test this.  And you can't do it with normal depth, or you
turn
the search into pure minimax.  So for the fail-high nodes, you simply
test
the singularity with a reduced depth and accept the errors it
produces...



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.