Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Verified Forward Pruning

Author: Premraj Natarajan

Date: 16:04:38 10/24/05

Go up one level in this thread


On October 24, 2005 at 17:47:22, Gerd Isenberg wrote:

>On October 24, 2005 at 15:41:28, Robert Hyatt wrote:
>
>>On October 24, 2005 at 13:38:02, Premraj Natarajan wrote:
>>
>>>I just came across a article that tells about verified Forward pruning instead
>>>of normal forward pruning. I am not sure if today's chess engines have
>>>implemented i hope developers would look into this and implement if it really
>>>helps.
>>>
>>>http://www.cs.biu.ac.il/~davoudo/pubs/vrfd_null.html
>>>
>>>bye
>>>Premraj
>>
>>
>>This is old.  Several tried it.  I don't know if anyone uses it, as it really is
>>an efficiency issue that slows null-move search down due to the verification
>>search overhead...
>
>I have a kind of conditional, reduced confirmation search for nullmove fail
>highs since years. Similar seems Vincent's double nullmove idea.
>Yes, it is overhead - how many elo points do you gain from an let say 5%
>speedup. How many points do you gain to detect a decisive zugzwang here and
>there inside the search interior node or root, but your opponent not?
>
>No idea ;-)
>
>CTT4-11 Crafty-IsiChess
>[D] 8/8/1p1r1k2/p1pPN1p1/P3KnP1/1P6/8/3R4 b - - ; bm Nxd5

Yup in decisive zugzwang the outcome of game is different so its important.

Raj



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.