I think we discussed this a while ago: cubeless winning chances in match play are taken from the 2s dB, janowski formula converts them into cubefull then plies are applied. Hence cubeless winning chances from the 2s dB in match play replace 0ply NN evals.
See this: http://lists.gnu.org/archive/html/bug-gnubg/2009-06/msg00230.html 2009/8/28 Michael Petch <[email protected]>: > > There has been a significant amount of discussion on BGO about this position > of Neil¹s. > > http://www.bgonline.org/forums/webbbs_config.pl?noframes;read=47213 > Position ID: gwEAQCoAAAAAAA Match ID: cAmgACAAGAAA > > > Gregg raised a question: > > ³I'm confused. Shouldn't GNU get the CPW for the position from the database, > and stick with that? It seems to get some rather wildly different chances of > winning on the different plys, when it should know the exact number. > > Why isn't doing a lookup in the bearoff database?² > > I responded with: > > " Gregg, I may be mistaken, but it was my understanding that the Bearoff > Databases had cubeful results for money play, but not for match score. > > If you had this in a money session, I am sure that it would go right to the > DB (if it doesn't that would be an error). > > My gut feeling says that it doesn't go right to the DB (and won't until its > a dead cube), and match score is what is making this difficult for the bot. > > Someone please correct me if I am wrong. I'd be willing to admit that my > understanding here is not correct." > > ---- > > The question I have, when do values get taken from the Bearoff database when > analyzing a match play cube decision (not money play) when one is using > Cubeful checker evaluations on a cube analysis. > > > > > _______________________________________________ > Bug-gnubg mailing list > [email protected] > http://lists.gnu.org/mailman/listinfo/bug-gnubg > _______________________________________________ Bug-gnubg mailing list [email protected] http://lists.gnu.org/mailman/listinfo/bug-gnubg
