Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Bruce Moreland on collecting the principle variation. Is this correct?

Author: José Carlos

Date: 09:32:30 04/29/02

Go up one level in this thread


On April 29, 2002 at 11:47:01, Russell Reagan wrote:

>On April 29, 2002 at 09:50:57, Daniel Clausen wrote:
>
>>On April 29, 2002 at 02:59:56, José Carlos wrote:
>>
>>[snip]
>>
>>>
>>>  If you want to do PVS or Aspiration Search or any other windowed search at the
>>>root, you must handle the root moves differently, but if you do a plain
>>>AlphaBeta with initial -inf,+inf window, the above code must work with no
>>>problem, and give a PV always.
>>>
>>>  José C.
>>
>>Note that if you call your search() with depth=0 (and therefore only make a
>>qsearch) you also may return w/o a move but only a score.
>>
>>Sargon
>
>How do most people handle this? Do they include the qsearch moves in their PV?
>
>Russell

  I don't, but others may do it. Particularly I don't find qsearch moves
interesting for the PV, and with today's search depths (~12 plies + extensions)
I don't think those are needed at all for the user. Note that you won't see them
in winboard due to the window width.

  José C.



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.