[Freeciv-Dev] Re: (PR#6977) Goto into unknown area

["Gregory Berkolaiko" <Gregory.Berkolaiko@xxxxxxxxxxxxx>]

<URL: http://rt.freeciv.org/Ticket/Display.html?id=6977 >

On Thu, 20 May 2004, Jason Short wrote:

>
> <URL: http://rt.freeciv.org/Ticket/Display.html?id=6977 >
>
> Gregory Berkolaiko wrote:
> > <URL: http://rt.freeciv.org/Ticket/Display.html?id=6977 >
> >
> >>[i-freeciv-lists@xxxxxxxxxxxxx - Mon Jan 12 17:57:44 2004]:
> >>
> >>No unknown positions aren't safe.
> >>
> >>I have also fixed the spelling Greg pointed out.
> >>
> >>Missing is the increase of the EC of unknown positions. Suggestions
> >>for the amount welcome. It is possible that it works even without the
> >
> > Should pretend that they are hills, return
> > 2 * SINGLE_MOVE * PF_TURN_FACTOR
> >
> > If you don't like magic 2, go for SINGLE_MOVE * PF_TURN_FACTOR.  It will
> > be like grassland, so the unknown will be investigated but not too much.
> >
> > Without any callback the user goto will be attracted to unknown way too
> > much.
>
> Elsewhere 2*SINGLE_MOVE+1 is used for "dangerous" positions.  We
> abstracted this into DANGER_MOVE (in gotohand.c IIRC).

Ah.  Good memory.  Multiply it by PF_TURN_FACTOR though, in an EC
callback.

> Also, triremes should consider unknown terrain to be "dangerous".  Now
> that we have the TER_UNSAFE terrain flag, maybe everyone should.

Mmmh.  Yes and no.  No because then all PFs will go through danger-PF and
the CPU might(*) explode.  Having a EC-callback returning a big value on
TER_UNSAFE is another matter.

(*) And might not -- it'd be interesting to actually see the effect.