[Freeciv-Dev] Re: patch 3: ai passive diplomacy + alliances fixes (PR#1277)

[Raahul Kumar <raahul_da_man@xxxxxxxxx>]

--- "Per I. Mathisen" <Per.Inge.Mathisen@xxxxxxxxxxx> wrote:
> On Tue, 26 Feb 2002, Raahul Kumar wrote:
> > I 'm curious about that. I thought SDI was city specific. Exactly how does
> > SDI in city 1 stop an attack on city 2?
> 
> Apparently SDI has a range effect. If the nuke hits one of your units or
> cities within this range, then the nuke is cancelled. I merely extended
> this to cover your allies' units and cities too.
> 

Great. I never knew about the range effect. I thought it was only city
specific.
Could be good for SAM aircraft defence as well.

> > > Major questions:
> > >  -> Should allied units in your cities help enforce martial law?
> >
> > No, I wouldn't like that.
> 
> Actually, I thought that would be a nice feature. My city is in unrest,
> and you have three mech inf standing outside... "Can you please squash
> that rebellion for me, Raahul?" :)
>

I could ;). OK, you've won me over.
 
> > >  -> Should it be possible to teleport or airlift to allied cities?
> >
> > Makes sense.
> 
> Yes, I think so too.
> 
> > >  -> Should it be possible for allied caravans to help build wonders?
> > >
> >
> > Of course. Otherwise people will send in units to the allied city, change
> > ownership and disband to help build a wonder.
> 
> Good point.
> 
> > > This part of the code works by magic, as far as i am concerned.
> > > (unit->ai.charge is both city and unit id??) I do not believe it needs
> > > change to work with peace/alliance.
> >
> > Yes it does. I demand to know how unit->ai.charge can be both city and unit
> id.
> 
> That puzzled me to no ends as well. It cannot possibly be correct.

Can you look it up? Maybe this is a bug. 

> > > Correct. We don't want to stay to defend allied cities. Not our problem!
> ...
> > > Changed. AI now looks to allied players' units nearby as
> > > possible reinforcements during attacks.
> ...
> > So the units that are not going to stay to defend allied cities are going
> to be
> > looked on as possible reinforcements. Interesting approach.
> 
> Hehe. I think the point of this code is: we want to know whether there
> will be other units to carry on with further attacks, if the unit
> currently being considered attacks, and, say, manages to almost kill a
> defending unit. Otherwise the attack was terribly futile. So, if there are
> allied units around, they can finish the job too. But it then becomes
> _their_ city. Which they will, of course, have to defend as their own.
> 

OK. We won't want to make a habit of this. I can see human players exploiting
AI's by stationing a 2-3 move unit, and after the AI has softened up the
defenders, sneaking in and taking the city at no risk to themselves.

> Yours,
> Per
> 
> "What we anticipate seldom occurs: but what we least expect generally
> happens." -- Benjamin Disraeli
> 
> 


__________________________________________________
Do You Yahoo!?
Yahoo! Greetings - Send FREE e-cards for every occasion!
http://greetings.yahoo.com