RE: [Bug-gnubg] Resign bug

[Ric]

Joern wrote:

That's an alternative solution, but a bit dangerous, since terminating the
game before end by an implicit resignation will be based on gnubg's
evaluator. People may argue that the evaluator is not perfect, and it's
perfectly imaginable that gnubg overlooks a winning 66-21-66-21-66-21-66-21
sequence if analysed at 0-ply (or perhaps even at 2-ply).

=================

Let me ask this: Is there a point where gnubg could recognize it has a 100%
probability of a win, no matter what is rolled from then on? Obviously the
game I submitted is such a case, or gnubg wouldn't have taken just one
chequer off. Is there a reliable event that indicates certainty within the
evaluation at that point that could trigger a popup or some other indicator?
I would think most such games would end during the bearoff, close to the
end, rather than eight rolls out as you suggest in your example, which may,
or may not, simplify the problem. 

In any event, my real point is about letting the user know what's going on
so that he doesn't think the program has gone nuts and is unreliable. Could
even be as simple as putting a note on the "About" menu or some such thing.
The problem with that is that if gnubg starts doing some weird things with
chequers, unrelated to what we're talking about, the user would just assume
that he's dealing with a won game rather than some new bug.

Anyway, the dead horse is beaten and buried. You folks do what you think is
best.

Ric