RE: [gnugo-devel] address@hidden: computer-go: 13x13 C GoTournament 2002-03 (12)]

[Portela Fernand]

I wrote:

> I have often noticed (I'm not alone I believe) owl reading mistakes due to
> the too optimistic "getting deep, looks lively" judgement. I think the
> engine would problably benefit from raising this depth parameter.

Well, this is actually not that clear. A first test with --owl-distrust 11
showed more new problems than solved ones, what actually isn't very good for
the current status either (if the result is correct, then deepening the
search shouldn't change the result, since either the owl code should be able
to find sure life for the defender, or a WIN would still be returned in case
of nodes limit hit). I'm gonna do some more testing...


Arend wrote:

> I think it's right to increase node limits exponentially when you are
> increasing depth limits linearly. 

Definitely. Sorry, my sentence wasn't clear. I just meant that it wouldn't
pay off to increase so much the default owl nodes limit for level 10.


/nando