Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Null-Move: Difference between R = 2 and R = 3 in action

Author: Uri Blass

Date: 10:50:49 07/15/02

Go up one level in this thread


On July 15, 2002 at 11:33:22, Omid David wrote:

>>Junior does not use null move pruning(I know that it did not use it few years
>>ago and based on Amir's post he worked mainly on evaluation after Junior5).
>>
>>I wonder if amir tried null move pruning in Junior.
>>
>>Uri
>
>I talked to Shay recently, they still aren't using null-move pruning, and won't
>in the near future. They're quite happy with their half-ply algorithm.
>
>Omid.

I think that their algorithm does not contradict null move pruning.
Every program can search if there is threat to reduced depth and decide if to
prune lines based on the result(the definition of depth is not important).

The only question is if null move can help their program.
The question is if they tried and found that it is not productive.

Even if it was not productive some years ago it may be productive today when the
hardware is faster.

Considering important moves as half plies can be eqvivalent to extensions(an
extension does not have to be by a full ply and can be also by half ply).
The problem is only to define the important moves that should be extended
because otherwise the program may waste too much time about extending
unimportant lines.

Uri



This page took 0.08 seconds to execute

Last modified: Thu, 07 Jul 11 08:48:38 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.