[Freeciv-Dev] Re: [IAMBACK] [PATCH] [1.3] cleanup of proccess_*_want() (PR#1295)

[Raahul Kumar <raahul_da_man@xxxxxxxxx>]

--- Gregory Berkolaiko <Gregory.Berkolaiko@xxxxxxxxxxxx> wrote:
<snip>

> > What does it do? Petr comment is a reasonable explanation. Don't be so
> vague. How do you want it changed?
> 
> I don't really understand that.  Trace the tree up from this function.  
> It's really weird.  
> 
> No, leave the comment as it is.

You complain about the comment, but have no suggestions for improving it.
Grr!

> 
> > > +        /* Citywalls give a defensive bonus of 300%. So for units that
> lack the ability to ignore city walls, the lack of a city wall makes
> them 3  * times as dangerous. Yet in this check we multiply by 9.
> > > WHY????????
> > > +         * (Pulls out hair and screams). -- Raahul */
> > > +        /* FIXME: Use acity->ai.wallvalue? --pasky */
> > > +        vuln *= 9;
> > > +      }
> > > 
> > > Raahul, please stop ruining your hairstyle!
> > > City walls give bonus of 200% (that's multiplying by 3)
> > > vulnerability is quadratic, so we perform mental calculation
> > > 3*3 = 9 and fill in the result
> >  
> > Back the truck up there G. I'm not quite sure I follow. Yes, vulnerability
> is quadratic. From there to multiplying city wall value by 3 is beyond me.
> Explain it to me without the missing steps.
> 
> I was just trying to save your hair... :((
> 

I'm no longer going prematurely bald.  Thanks to GB. Yeah Yeah ;-). I will not
need Advance Hair Studios.

> Ok, vulnerability is Defense_Power * HP * FirePower squared
> Walls increase Defense_Power three-fold.
> Therefore vulnerability would increase nine-fold.
> 
> This is taking into account that although there no city-walls in the enemy 
> city yet, the enemy will probably build them before we get there.
> The existing city walls are taken into account by 
> unit_vulnerability_virtual2 what_a_lovely_name_for_a_function.

Finally it makes sense.

> > > All of my complaints are about your comments which is non-essential.
> > > I haven't checked line-by-line correspondence but if you run a few 
> > > autogames, I think the patch can be safely committed and human resources 
> > > directed to pastures new (like the dreaded and bugged f_s_t_k).
> > 
> > Not the f_s_t_k. Anything but that. What have I done to deserve this?
> 
> Don't be afraid.  We will cover you back ;)
> 

OK. I'm going in. Expect that patch soon. I'm not noticing any behaviour change
after my patch, and I *SHOULD* be seeing some difference. I've built a
lighthouse, magellan, and given that over to the AI. I don't see any brilliant
performances. I'm using unit_move_turns instead of the insanity that is the
current f_s_t_k implementation.

I'll send that patch in anyway. So what if the AI is no smarter? It should be.
No doubt there is some brain damage elsewhere that stops my patch from doing
good things.

__________________________________________________
Do You Yahoo!?
Yahoo! Tax Center - online filing with TurboTax
http://taxes.yahoo.com/