Author: Tony Werten
Date: 13:08:50 03/03/06
Go up one level in this thread
On March 03, 2006 at 16:01:12, Robert Hyatt wrote: >On March 03, 2006 at 13:05:18, Tord Romstad wrote: > >>On March 03, 2006 at 12:20:43, Tony Werten wrote: >> >>>On March 03, 2006 at 09:46:12, Tord Romstad wrote: >>> >>>>What's the problem with a shared history table? >>> >>>Problem is a big word, it probably isn't that bad. >>> >>>Suppose thread 1 want to add 1 to the counter: >>>1) load memory into register >>>2) add 1 to register >>>3) move register to memory >>> >>>If the thread is interrupted between 1 and 3 and thread 2 adds to the same >>>entry, you have lost 1 add. >> >>Yes, this can happen, of course, but it doesn't bother me at all. >>Who cares if the history counters are only approximately correct? >>The deviations will be really tiny, and I would be very suprised if >>they have any measurable impact on move ordering at all. I really >>don't understand why anybody would want to copy the entire history >>table to each thread at all split points, instead of just sharing it. >> >>Tord > >Please ignore all this crap. No idea what I was thinking about. The history >counters (fh and non-fh) are shared, not local. When I added this I decided to >make the normal history counters global as well. > >When I wrote my comment above I apparently was living in the past and had >forgotten the change, since the killer moves are still in thread-local memory... > >Yes, there are potential interleaved-update problems, but missing a count here >and there doesn't seem critical enough to justify locks... I agree, Tord is right. Our thoughts are probably just a reflex reaction based on the bad experience from nonlocking hashtable sharing. BTW on my TODO TESTING list is using the percentage of failed highs (already doing) but then corrected with a local failedhigh percentage based on ply. ie reduce when the move has a bad fh-percentage, but not if the fh-percentage is good on this specific ply. Tony
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.