[Fluxus] commits + oops

[Artem Baguinski]

On Sun, Nov 30, 2008 at 4:05 PM, Artem Baguinski <artm at v2.nl> wrote:

> my cylindrical physics still doesn't look right, so I'll continue
> investigating this.

I found what goes wrong. Fluxus cylinder starts at y=0 and goes to
y=length, ODE's cylinder is from -length/2 to length/2. So, the visual
representation and physical/collision geometry aren't aligned
correctly. spheres and cubes are centered in fluxus, so there is no
problem. fluxus convention is intuitive for building trees from
cylinders, but inconsistent. But for compatibility with existing
scripts I'll try to fix it on the physics side, not geometry (i assume
cylindrical collisions are little used, since this problem still
exists).

-- 
cheers,
artm

http://lab.v2.nl/