Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: About history pruning...

Author: Robert Hyatt

Date: 06:58:05 11/01/05

Go up one level in this thread


On October 31, 2005 at 14:37:22, Vincent Diepeveen wrote:

>On October 31, 2005 at 12:53:33, Robert Hyatt wrote:
>
>>On October 30, 2005 at 04:21:05, Mridul Muralidharan wrote:
>>
>>>On October 28, 2005 at 22:32:05, Robert Hyatt wrote:
>>><snip>>
>>>>Now it is "dubious forward pruning won't work".  Of course, Shredder, Fruit,
>>>>Zappa, Chess Tiger, and others all apparently have it working quite well...
>>>
>>>Are you sure that Zappa is using forward pruning ?
>>>I did not see Anthony Cozzie mention this anywhere (for WCCC'05) ...
>>>Ofcourse , I am not talking about null move here.
>>>
>>>- Mridul
>>
>>
>>He had mentioned it somewhere along the way, yes.  I think that was the addition
>>to produce zappa 2.0...  Anthony can comment, but the search depths he has
>>reported during the WCCC certainly support that as well...
>>
>>Any program that is going beyond 14 or 15 plies in middlegame positions is doing
>>forward pruning beyond pure null-move...
>
>Bob, perhaps start reading back old postings.
>
>Your memory is in a far state of deterioration.
>
>Zappa got 14-15 ply first moves out of book at 4 processors with 3.3 - 4 million
>nps.
>
>So if you would say he's getting that with singular extensions turned on, then
>that makes more sense than saying he's getting that with forward pruning other
>than nullmove.

Depends on the time control.  But did you read what I _wrote_?  And I quote:

"any program going _beyond_ 14 or 15 plies in middlegame positions is doing
forward pruning beyond pure null-move..."

Do you see that "_beyond_" in the quote?

Didn't think so...




>
>Fruit single cpu at 1 mln nps is getting typically 16-17 ply in those positions.
>
>First move out of book Zappa was obviously outsearched many plies by about
>everyone.

who cares about first move out of book.  Care to bet whether zappa is using any
forward pruning or not???  Care to bet whether Anthony and I have compared
search times for the same positions, using my 8-way WCCC box I had for the WCCC?

I'd rather have a deteriorated memory than a non-existant reasoning
capability...



>
>Vincent



This page took 0.01 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.