Author: Dieter Buerssner
Date: 01:48:42 09/29/00
Go up one level in this thread
On September 29, 2000 at 03:08:25, Bernhard Bauer wrote:
>On September 28, 2000 at 10:18:53, Dieter Buerssner wrote:
>
>>On September 28, 2000 at 04:26:54, Bernhard Bauer wrote:
>>
>>[much snipped]
>>
>>>My modification has also to do with null move, but I do another change.
>>>Instead of the original statement in search.c
>>>if (do_null && !tree->in_check[ply] && pieces
>>> && (pieces>5 || depth<421)) {
>>>
>>>I use the following statement
>>>if (do_null && !tree->in_check[ply] && pieces>0 && depth>60 &&
>>> && (pieces>9 || depth>301) ) {
>>
>>This is interesting. The original Crafty code uses null move at the leaves of
>>the tree, when close to the endgame. Your code uses null move at at the front of
>>the tree. Has this shown to be better in test suites or in games?
>>
>>Also, you don't allow null moves at the last ply. Does this reach higher depth
>>with the same node count?
Bernhard, I really read your source snippet different, then your description.
>For many pieces (pieces>5) the original Crafty code will always use null move.
Yes, this is like I read the source snippet of original Crafty.
>I don't use null move for the first ply hoping to get a better move ordering.
But wouldn't this mean to use "&& ply > 1" instead of and "depth > 60"
>If only a few pieces remain I don't do null move for the first 5 plies.
Again, wouldn't this mean to use "&& ply > 5" instead of "&& depht > 301"
For your description I would use:
if (do_null && !tree->in_check[ply] && pieces>0
&& ply>1 /* No null move at first ply */
&& (pieces>9 || ply>5) /* Little material: not for the first 5 plies */
) {
Where is my misunderstanding?
Nevertheless, I think depth instead of ply seems superior, for example because
of hash table consistency. But then your ">" seems the wrong way around.
Remember that depth is the remaining depth, or have you changed the meaning of
depth? I would read "depth > 60" only one ply is remaining (which could make
sense, because there is not much you can cut off in the last ply).
-- Dieter
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.