[ODE] Racing game settings (update)

[Jon Watte]

> > Calculate a front vector for the wheel as:
> >    fDir1 = normalize( cross( wheelAxis, groundNormal ) )

> Sorry you misunderstand - I mean if I scale the slip parameter by the 
> velocity.

Yeah, the velocity that makes sense (to me) is the linear 
velocity of the wheel object itself, at its center.

> As slip/mu vary I can make it better but I've never 
> found a particularly good mix for driveability, and certainly not 
> managed to get it to do any nice handbrake turn sorta stuff!

> > Real cars don't drive straight. Especially not at the forces 
> > and speeds that you'll cram into a typical arcade game. If you 
> > want smoother, more game-like physics, I recommend the box 
> > with four ray casts.

> Yes I realise this. Its a very small error but it is noticeable.

I'm starting to think that the problem is mostly in my 
expectations with my model. I want the model to be physically 
modeled, but behave like an arcade racer. If I keep courses, 
speeds, forces, and everything to "normal" parameters, the car 
drives reasonably well.

Also, the specific controls you add to it do a lot. I'm only 
using arrow keys, and I had to work a bit on how to map controls 
over time to actual steering/throttle values. Basically, I use a 
very quick (100 ms) ramp-up, and a similarly quick ramp-down for 
steering, but much slower ramp-down for throttle (unless you 
down-arrow).


Anyway, the main source file and the executable for my toy buggy 
is available at http://www.speakeasy.net/~hplus/carworld-040730.zip 
if you want to take a look, although I doubt it's any better than 
yours. After you run it, there will be a file called "car.twk" 
which you can tweak parameters in, for example the default car is 
not very high-powered.

Oh, and there's a reason the art is boxes with noise ;-)

Cheers,

			/ h+