Yeah it's an AI project, I'm going to investigate social AI and develop
a system to make background characters in games more believable. I'm
sure you've noticed that in most games they either sit about repeating
some simple behaviour loop, or patrol a path with little variation.
Those that have been more ambitious tend to have their characters act
like f**king idiots (eg. Fable), so there is definately room for
improvement :biggrin: There are things that demonstrate believable socal AI,
but not often running in the background characters of a real time game.
Gotta look into some basic psychology papers and presentations to get a
grip on key behavioural traits, determine what is feasible to implement
and what implementation technique is most suitable, and then develop a
demonstration. Should be fun :smile:
We haven't had any formal classes on AI through our course (yeah...it's
a bit of a joke :sad: ), so its a new area to me but one I find more
interesting than the technical side of things...
...at which I suck.
Don't think I knew you were going for a masters Crono, do you have an honours or degree already?
Funny you should mention the Tic-Tac-Toe thing, as we did something
somewhat related near the start of the course. We had to write a
program that worked out how many legal board layouts there were in
Tic-Tac-Toe if crosses start. A lot of people got different answers,
but mine were 19683 positions in total, 5869 of them legal, and 820 of
them entirely unique. Looking back at that program it's like I didn't
know globals could be local :biggrin: I appear to have used variable names as
if they were pointers, when I quite clearly wasn't using any. Madness.
[img]http://card.mygamercard.net/sig/Default/reno84.png[/img]
Designer @
Haiku Interactive |
ReNo-vation.net