Python in game development?

[Moritz Voss]

"Will Ware" <wware at world.std.com> schrieb im Newsbeitrag
news:FxzACB.KEB at world.std.com...
> > What exactly *are* these advantages? (Very curious...)
>
> simultaneously, and rapidly switch between them. This could be useful
> in a game if you want to model simultaneous behavior of a large number
> of agents, such as soldiers in an army, ants in an anthill, airplanes

I'd still queue that up! Because I'd have to maintain so much
synchronization (collisions, etc, one object influencing another, accessing
samme resources (be that a tank or something in memory), I'd just quickly
runt through them all in a straightforward way. They have ALL to be done
when the next frame will be drawn, so thread management will only produce
more overhead....right?



> http://world.std.com/~wware/uthread.html
<reading>


That brings up another question - how safe/restrictive is Python?

I want to allow the user to implement own micro AI scripts. These will run
on the serve ronly, though, and I want to prevent
a) the server from crashing due to malicious or buggy python scripts
b) the user from cheating (e.g. having that AI download huge tactical info
over the net, or maybe a mapfile or something that lets it eval its
situation better.)
c) the unit from screwing things up, as intearction with othe runits is
possible, but not ALL units. Other units can communicate with other ones,
respectively.
d) allow each AI script to only have a certian short period ooff processing
time - making long scripts idle the unit longer while they work, while
shiort scripts have far more throughput per runtime unit.


I don't understadn the concept of python thisfar, and I am reconsidering my
own virtual machine concept - yet I think that maybe too much work to
implement in reasonable quality... :-(



--
--
Moritz "Thygrrr" Voss
Client-Server & OpenGL Coder
spirit.link.studios - http://o2.ods.org