Author: Gian-Carlo Pascutto
Date: 15:07:05 01/05/01
Go up one level in this thread
On January 05, 2001 at 15:10:47, Severi Salminen wrote: >Hi! > >Today I programmed SEE and got also weird results (now it works). Are you using >PVS? The configuration I tested was: PVS+aspiration search nullmove (R=2) hash table ONLY for move ordering killer+history+LVA/MVV move ordering normal qsearch and *nothing* more. >My problem was to decide what to use as the best score if all the moves >were pruned because off futility pruning, also in qsearch. In qsearch I decided >to return alpha and nothing less. It ok to return the stand-pat score or alpha, depending on which is higher. If you increase alpha to the stand-pat score in the beggining of your qsearch it is ok to return alpha of course. >I don't know if it is right, but it worked. >And in normal search I use material_balance + FUTILITY_MARGIN to alter the >bestscore so far. I don't know whether I should include the material gain of a >move or exclude the MARGIN, but now things work. Maybe you should check out >those and experimant. I start from -inf and see if I can improve upon that, if not, I just return -inf and let the previous level fail high. Whatever score we find at a node with futility pruning is always a lower bound. It does not matter for my question though because all futility pruning was disabled. What do you mean by altering bestscore? Do you mean you increase alpha to the values you mention? -- GCP
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.