Author: Dieter Buerssner
Date: 11:52:39 09/22/03
Go up one level in this thread
On September 21, 2003 at 09:15:06, Mikael Bäckman wrote:
>int qSearch(int wtm, int alpha, int beta)
>{
> score = evaluate(wtm);
bestscore = score;
> if (score > alpha) {
> if (score >= beta) return score;
> alpha = score;
> }
>
> pLast = genAllCapMoves(wtm, pFirst);
> nLegal = 0;
>
> while (pMove = nextMove(pFirst, pLast)) {
> if (makeMove(wtm, pMove)) {
> nLegal++;
> score = -qSearch(wtm^1, -beta, -alpha);
> unmakeMove(pMove);
if (score > bestscore)
bestscore = score;
> if (score > alpha) {
> if (score >= beta) return score;
> alpha = score;
> }
> }
> }
return bestscore;
/* instead of
> if (!nLegal) return score;
> return alpha;
No need to check nLegal, unless you want to do something
specific in case of check */
>}
I think, I found the problem. You will typically return alpha here in cases,
where all subsearches and the eval are smaller alpha. This basically makes the
whole search fail hard.
Regards,
Dieter
This page took 0.01 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.