[Freeciv-Dev] Re: [PATCH] [1.2] move some common AI equations to kill_desire() (PR#1279)

[Petr Baudis <pasky@xxxxxxxxxxx>]

Dear diary, on Mon, Feb 25, 2002 at 04:05:36PM CET, I got a letter,
where Raahul Kumar <raahul_da_man@xxxxxxxxx> told me, that...
> 
> --- Petr Baudis <pasky@xxxxxxxxxxx> wrote:
>  Dear diary, on Mon, Feb 25, 2002 at 03:35:07PM CET, I got a letter,
>  where Raahul Kumar <raahul_da_man@xxxxxxxxx> told me, that...
>  ..snip..
>  
>  OBTW snipping the parts of mail you don't reply to would be nice feature to
>  have commited into you as well ;)).
> 
> Your will is my command. 

Oh well this is too much! :) The '>' chars are ok, I just didn't like that it
wrapped the test on next line w/o inserting the '>', thus messing it up. It was
better than this, however ;). Now let's see if I can dig out your reply from
this :).

>  +Compute how much we want to kill certain victim we've chosen.
>  ..snip..
>  I can't tell from this comment what happens when a fortress is the
>  target tile.
>  
>  That depends on the code where kill_desire() is used, kill_desire()
>  itself doesn't care. Take findvictim() or find_something_to_kill() as a
>  reference context code. It looks some defense value somewhere is computed    
>  
>   
>  properly for that case and it somehow reflects in the parameters of 
>  kill_desire()
>  -
>  at least I hope so :).
>  
>  I was hoping someone decides to audit the AI combat code and see if it
>  actually makes the right decisions  in accordance with the following
>  factors
>  
> Attacking 100 units stacked with a mech inf over 1 unit stacked with a
> musket.  I'm pretty sure the AI code does not weigh large stacks as being
> sufficiently good for attacking at all costs.
>  
> Are 100 units stacked with a mech inf frequent case? Stacks themselves are
> pretty rarely more than 3 units at one tile, IMHO.
> 
> Late game, I frequently stack 15 units with a mech inf in front of an AI city,
> before attacking. 

Oh well.. :) I don't, you do. You're thus the right person to do the work :).

>  As for the someone who audits the code, I think you're familiar with him.
>  You've probably seen him in the mirror before, and his last name is Baudis.
>  
>  Oh, I'm so sorry to disappoint you.. but as I said above, I don't think it's
>  worth it, and cleanups are my priority now. Also, I'm going to cleanup
>  assess_danger() after this patch and then my freeciv period will fall down
>  again and I'm going to have a pause for about two weeks or a month (or I'll
>  do
>  another patch before or I won't finish assess_danger(), depending how I'll
>  feel, what will get my interest and how much time other things will take).
>  
> 
> Ok. It was worth a shot. My innate laziness. Now I'll have to try and sucker
> Greg or Raimar into it. They were already discussing combat results.  

Hehe :)

-- 

                                Petr "Pasky" Baudis

* elinks maintainer                * IPv6 guy (XS26 co-coordinator)
* IRCnet operator                  * FreeCiv AI hacker
.
No one can feel as helpless as the owner of a sick goldfish.
.
Public PGP key && geekcode && homepage: http://pasky.ji.cz/~pasky/