Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: some null-move questions

Author: Harald Lüßen

Date: 05:48:48 09/25/04

Go up one level in this thread


On September 25, 2004 at 07:19:21, Andrew Platt wrote:

>On September 25, 2004 at 06:35:57, martin fierz wrote:
>
>>4) i never thought much about verified 0-move pruning. but a post i read last
>>week made me think: i always thought i could do the following:
>>
>>if(pieces > 0)
>>  if nullmove-fails-high
>>    return value;
>>if(pieces == 0)  // do a verification search
>>  if nullmove-fails-high
>>     verify result with a normal search with reduced depth.
>>       if verification-search-fails-high
>>          return value;
>
>The standard verification search doesn't have your condition above. It just says
>if it fails high, verify.
>
>>now there was this post saying that this won't work because you have to disable
>>the nullmove in your entire verification search. i don't understand why that
>>would be? can somebody explain?
>
>I don't know if this was from the posts I was making about problems I was having
>but that's not exactly the solution you want. The key point is that even when
>verifying a null move search you are still doing null moves lower down the tree
>and they could be causing problems too. However, you won't verify those becuase
>you are in verify mode. So now you are doing an R=3 null move and could miss
>some things.

What mode? Are the nullmove algorithms in deeper plies dependent
on a decision of a shallow ply? I thought they are independent
with the exception of avoiding double nullmoves. A possible
deep ply nullmove is done when there are pieces and verified
when it fails high.
And when a search with a small depths with or without nullmove
backs up its score I could trust it whether I am in a normal
search or in a verification search. That is the trick with
recursion. Where am I wrong? Is it explained in the original
verification search document?

Harald



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.