RE: [Bug-gnubg] bug in 3 ply equities

[Misja Alma]

gnubg plays the 'contained side' of this position very poorly.
For instance, in positions like YttuAQDb7BoACA it does not come up to the
edge of the prime.
Not doubling in the initial position does not cost very
much in the rollout, because gnubg will take always later on, even with one
checker against a six prime: 69wVAAhi23YQAA
As a result of all those errors on the contained side, a 2 ply rollout at
level world class (108 rollouts) comes up with the result that this position
is a pass and almost too good. But this really is a clear take.

Misja

-----Oorspronkelijk bericht-----
Van: address@hidden
[mailto:address@hidden Joseph Heled
Verzonden: Wednesday, December 03, 2003 11:49 PM
Aan: Misja Alma
CC: Hugh Sconyers; Oystein Johansen; gnubg (E-mail)
Onderwerp: Re: [Bug-gnubg] bug in 3 ply equities




Misja Alma wrote:

> Joseph mentioned the possibility of having a separate net for these
> positions, a sub-class of the crashed. But the position I noticed,
> ObsdAAhi2zYQEA, is not crashed yet. It is just about containing one
checker
> with an outer prime.
>
>

I find it hard to believe GNUbg plays this position poorly. Sure, It
makes mistakes, but so poor as to require another net?

Please elaborate. My (0ply 1296) rollout shows 75.58+11 24.42+6.8

I looked at 10 continuations played at 0 ply, analyzed at 2ply. All
seems normal.

-Joseph



_______________________________________________
Bug-gnubg mailing list
address@hidden
http://mail.gnu.org/mailman/listinfo/bug-gnubg