Author: Uri Blass
Date: 05:52:59 02/25/02
Go up one level in this thread
On February 25, 2002 at 08:50:27, Odd Gunnar Malin wrote:
>On February 25, 2002 at 08:20:25, Benny Antonsson wrote:
>
>>Hi !
>>
>>I have a question about how to count the nodes when using alphaBeta and
>>quiscense searching. Here is what we do (pseudo code):
>>
>>function alphaBeta()
>>{
>> nodeCount++;
>>
>> value = quisce();
>>}
>>
>>function quisce()
>>{
>> nodeCount++;
>>}
>>
>>Wouldn't this result in counting the first quisce-node twice ??
>
>Yes.
>I count qnodes after I do the move.
>
>GenerateCapturemoves(...);
>while (movesreminding)
>{
> if (doLegalMove(..))
> {
> qnodes++;
> score=-qsearch(..);
> undomove(..);
> ...
> }
>}
>
>I count the normal nodes like you do.
>
>It was a discussion here some time ago that some was counting the move they
>generate regardless if this was pruned away or not. I think this would be
>inaccurate and only for better showup in the nps. Instead I have a movegenerator
>test for timing the move/undomove. This is also usefull for testing correctness
>of the movegenerator.
>
>Odd Gunnar
I have no moves that are pruned away.
I make only legal moves.
Uri
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.