Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: DEEP BLUES AVERAGE PLY?

Author: Vincent Diepeveen

Date: 11:28:35 08/21/02

Go up one level in this thread


On August 21, 2002 at 11:21:38, Robert Hyatt wrote:

>On August 21, 2002 at 10:45:51, Gian-Carlo Pascutto wrote:
>
>>On August 21, 2002 at 10:35:13, Vincent Diepeveen wrote:
>>
>>>2 * sqrt(40)^18 = 524288000000000 nodes needed to search it *minimum*.
>>
>>That is pure alphabeta. Simple PVS will already reduce this tremendously.
>>
>>I'm quite sure we already did all the math five times, actually, but
>>you keep 'forgetting' the details which don't coincide with your
>>point of view.
>>
>>--
>>GCP
>
>
>You got it... :)

They used normal alfabeta in the software part as you can read in IEEE99.

Probably you forgot that too?

In 1999 you tried to prove, that using normal alfabeta didn't matter
much as it was all done in the hardware and they only did a few ply
in software anyway so who cares for it.

Best regards,
Vincent



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.