Complete.Org: Mailing Lists: Archives: freeciv-ai: April 2003: [freeciv-ai] Re: [RFC] Ferry code

# [freeciv-ai] Re: [RFC] Ferry code

[Top] [All Lists]

 To: Gregory Berkolaiko Cc: Freeciv AI development Subject: [freeciv-ai] Re: [RFC] Ferry code From: astuckey@xxxxxxxx Date: Tue, 08 Apr 2003 00:14:25 -0500 (CDT)

```Quoting Gregory Berkolaiko <Gregory.Berkolaiko@xxxxxxxxxxxx>:

> On Sat, 5 Apr 2003, Jason Dorje Short wrote:
> > Also consider how a unit (explorer) might get *back* from an
> uncolonized
> > island, or move from one uncolonized island to another.
>
> In the present state of the code, there would be no explorer on
> an uncolonized island.

Civil War?  Or losing cities due to a war?  They seem common enough causes.

> > I think the only real solution to the ferry problem is a global one.
> > You can't consider units on a case-by-case basis.  But if you look at
> > all your units and where they need to go, you can assign boat B1 to
> > ferry units U1, U2, and U3 from point A on continent 1 to point B on
> > continent 2, while boat B2 ferries unit U4 from C on continent 1 to D on
> > continent 2.  Then you just tell B1, U1, U2, and U3 to all meet up at A,
> > while B2 and U4 meet up at C.
>
> This is sci-fi, as far as I am concerned.  If you have any practical
> ideas, please explain them in detail.  Mind you, very often you just
> need
> an estimate of the time to get from A on cont 1 to Z on cont 2.

I think I agree with JDS.  Imagine a map with all paths plotted.  The most
heavily travelled lanes are the ones where ferryboats should work, and knowing
how many paths traverse a given tile is "easy".  The minimal amount of
ferryboats to service the population can be calculated, and will vary from turn
to turn.  Units can keep asserting their want, but if it takes 40 turns to get
enough free time to rescue our stranded explorer, so be it.  Sadly, I think that
this calculation is exponential.

```