The Underscore Archives

On 1 Dec 2006, at 17:36, Stuart Gallemore wrote:
> I think you may want some game physics rather than real physics by the
> sound of it (i.e. faked ones that look right but are quick to
> calculate by iterative changes through a time frame and cartesian
> coordinates).
>
> In my experience of applying physics to widgets, games flash etc is to
> understand the concepts, then screw the actual physics and make
> something that just feels right and runs fast. Usually give your self
> a lot of parameters and tweak them until it all looks lovely.

Yeah... that's pretty much the way I'm going.  I mocked up a particle- 
based system (with Verlet Integration), but Javascript in Firefox is  
just too damn slow for the number of particles, rods and sticks I'd  
need for a *decent* simulation of what I want. So, I was hoping to  
get enough of a real solution for the polygon collision just for the  
first frame after impact to initialise sensible motion and then scale  
back to the less accurate model for subsequent frames.

Then I realised how little I know about the real physics of the  
thing, and thought it'd be worth swotting up on it first :)

If I don't get much further with the cheap way, I think I'll be  
looking into that book you suggested.

Cheers,

-- 
Tom Gidden
http://gidden.net/tom/