[ODE] FDir1 and Slip for car wheels

[Arturo Colorado Garín]

>> It looks like it always takes mu and ignores mu2.

Ok, problem corrected! I was setting flag dContactFDir1 but not dContactMu2!
That's why it was ignoring mu2 and was always
using mu. Now it works as I expected. Unbelievable, I've read several times
that part of the
documentation but never noticed that flag until the last time.


>And, if you want car rotation in the air, then chances are you aren't
>really after really physical cars; you want "arcade" or "stunt" cars,
>at which point entirely different mechanisms could be used.

Correct, I'm after a stunt car. But stunt cars are real, too!  ;-)


>1) Add aerodynamic load. A constant times your forward velocity squared
>should push the main body of the car in the car's down direction (not
>the gravity direction!). This will help with stability at high speeds.

Yes, I will add it soon or later, thanks.

Another thing I've found that might help in car stability: assuming you use
a box for mass distribution (dMassSetBox() and dMassSetBoxTotal()
functions), make its width longer than the actual car width. It has the
effect of "making harder" for the car to rotate along its longitudinal axis
(dMassSetBoxXXX() functions modify the moment of inertia I, which opposes to
rotation movements pretty much like mass "opposes" to translation
movements). But beware, this makes also harder for the car to recover when
starting to flip over. This said, I must confess I don't use this trick
anymore (my car is quite stable now) but I think it is worth trying.

Thanks for your help!

--Arturo Colorado