Author: José Carlos
Date: 01:41:41 08/20/01
Go up one level in this thread
On August 20, 2001 at 02:37:55, Artem Pyatakov wrote: >I have a (somewhat) strange question about null move and WAC.001. > >Here is what happens when I run my program: >Depth=4 (Correct PV here) +MATE2 >Depth=5 Fail Low !!! (Aspiration window +MATE2-50 +MATE2+50) >Depth=5 +MATE2 > >I investigated and discovered why this strange fail low happens: it happens >because when I start search search on depth 5, I search the move in the PV from >depth 4 (the correct mating move) first. After making the move, I do a null move >search (R=2) - and this is where it returns beta. > >So, I have two questions: >1) Is this behavior inherent to null-move or is there a way around it? >2) How would I avoid this kind of fail-low behavior? (Perhaps, I should write >different code for null move when a search returns -MATE/+MATE?) > >Thank you > >Artem I might be wrong, cause I'm very new to null-move, but I've found that my program had a very strange behaviour when first implemented null-move. I was doing it everywhere. Then I tried not doing it in the PV, and the strange things went out. The tree got smaller and the search deeper. I don't know if this is the correct way of doing it (never do a null-move in the PV), cause I like to try myself before looking other's sources to see how they do it. But looks reasonable. In the PV, we don't expect to fail high, so why trying a null move? José C.
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.