Re: [gnugo-devel] engine/influence.c (and DFA)

[Arend Bayer]

On 12 Sep 2002, Dave Denholm wrote:

> > I haven't done 2), because I think if anything we should fix the fact
> > that the order makes a difference. (I think I can also explain why this
> > does make a difference, in case anybody is interested -- I believe it is
> > a rare problem and not much worth worrying about.)
> >
>
>
> Hmm - without thinking too hard about it, I had decided that it
> was an inherent problem with the algorithm, in that it only
> looks at each outlying point once, and so it depends on the
> path taken to reach that point on whether the influence is
> spread from it. ie if we reach a point from which influence
> has already spread, the new contribution is not distributed
> further.
>
> But I'm probably mistaken...
No, I absolutely agree with you. But unless I am mistaken, this
can happen only in rare configurations. (There need be some spots in the
surrounding with permeability == 0.0, in such a configuration that we
drop some paths, and among the remaing paths there are some "fast paths"
and "slow paths"...

Arend