[Freeciv-Dev] Re: Unit-vs-stack sim (was: Cache win_chance)

[Raimar Falke <hawk@xxxxxxxxxxxxxxxxxxxxxxx>]

On Mon, Feb 25, 2002 at 12:56:12PM +0000, Gregory Berkolaiko wrote:
>  --- Raimar Falke <hawk@xxxxxxxxxxxxxxxxxxxxxxx> wrote: 
> > On Sat, Feb 23, 2002 at 07:00:27PM +0000, Gregory Berkolaiko wrote:
> > >  --- Raimar Falke <hawk@xxxxxxxxxxxxxxxxxxxxxxx> wrote: 
> > > > On Sat, Feb 23, 2002 at 06:29:10PM +0000, Gregory Berkolaiko wrote:
> > > > > >
> > > > > > How well does it handle unit stacks? Example: pdef1 is a good
> > > > > > defender, pdef2 is an expensive unit, pdef3 is a cheap one. All in a
> > > > > > fortress. The attacker will likely destroy pdef1, will also destory
> > > > > > pdef2 and but will be destroyed by pdef3.
> > > > > 
> > > > > What is your question here?
> > > > 
> > > > Will the AI correctly attack even if it knows that the attacker will
> > > > be destoryed but has killed pdef2 before?
> > > 
> > > What is "correctly"?
> > 
> > > If the choice of defenders is pdef1, pdef2, pdef3, it should attack twice
> > > only.  Why would it be destroyed?
> > 
> > AFAI read the code: if you set occupychance to 100% and the target
> > tile has a fortress or is a city the attacker will continue its
> > attack.
> 
> AI doesn't know about occupychance because whoever introduced this rule did
> not bother/dare to update AI.
> 
> I don't know about occupychance either.

I was wrong. occupychance is for entering a city _without_ any enemy
units. So unit-vs-stack is always a unit-vs-unit combat. This makes
implementing this really more complex. Because you can't estimate an
_exact probabilities_ for a combat you have to introduce abstract
attack and defense powers. And these number are neither fixed nor easy
do calculate. Doh! I wanted to avoid this.

        Raimar

-- 
 email: rf13@xxxxxxxxxxxxxxxxx
  Two OS engineers facing a petri net chart:
        "dead lock in four moves!"