Re: CRT Combat Model (really long)

[Lincoln Peters <peters2000 at mindspring dot com>]

On Sat, 2002-03-16 at 15:38, Jim Kingdon wrote:
> 
> Sorry, maybe I missed something, but what is the difference between
> this game and the standard game, which sets:
> 
> (add u* acp-per-turn  (. . . 0 1 1))     ;; base town city
> . . .
> (add places speed 0)
> . . .
> (table acp-to-attack
>   ;; Places can defend, but not attack.
>   (places u* 0)
>   )

That is essentially what I did that apparently caused the segfault.  All
places get 1 ACP per turn (there are no units that resemble bases), then
I set up the acp-to-attack table almost identical to the snippet you
sent, except there is a line before the "Places can defend" comment that
says "(u* u* 2)".  When I ran into the segfault problem, I took out the
"(places u* 0)" line, and the game ran fine again.

> 
> Was your comment specific to combat model 1?

No, this game uses combat model 0.

> 
> (I'm starting to wonder whether having multiple combat models is in
> fact not as good as a bunch of more specific tables).

I prefer model 0 because it gives me a lot more control over how the
units react to each other during combat (the game "future.g" uses that
kind of control really well).  However, I've never written a game that
was as big as "3rd-age.g" or "civ2.g", so I can't say from experience
that model 1 is better or worse in such games.