anti-roll bar model suggestions

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Recently on the TRB it has come to light that the simuv2 and simuv3
anti-roll bar models are slightly different.

Im simuv3, the anti-roll-bar uses a limit on the force applied through
the bar, implemented via the anti-roll-bar course (xMax). However, in
simuv2 the anti-roll-bar suspension course plays no role whatsoever. I
think this makes more sense, actually.

So, perhaps I should change simuv3's ARB to be like simuv2's. But while
we're at it, let's examine all the alternatives. So, what we have is:

simuv2: linear model of anti-roll bar
simuv3: linear model with force clamped at a maximum depending on the
course (xMax).

There are four other options:

1) 4th power model (corresponds to the tension of a stiff metal rod, so
is probably the most realistic)
2) piecewise 2nd power model: linear until xMax, then quadratic
3) piecewise linear model with 'play': no force until xMax, then linear,
with no upper limit,
This seems relatively realistic, since most such mechanical systems have
a bit of play.
4) make it an option! (complicates things for drivers)

If we change the models, I will go through all standard 7 cars to fix
the parameters if necessary, but I can't do it for all other cars.

My suggestion is to make simuv3 like simuv2 and add an option for other
suspension models.

tangent: a lot of options in torcs sim right now are implemented via ifs
or cases.  Would it be more efficient to implement them via function
pointers? (we are using straight C in any case for most of the sim)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFJYy41yBv6kELa61sRAltnAJ9u5oHkU7puvxmlt4YZ7osol3ixZACeMGVR
uH6L4kaw5fMiAHRCpOlpURM=
=3sip
-----END PGP SIGNATURE-----

------------------------------------------------------------------------------