[ODE] Possible solution to unstable rotating bodies problem

Tyler Streeter tylerstreeter at gmail.com
Fri Apr 22 12:44:24 MST 2005


On 4/22/05, Aras Pranckevicius <nearaz at gmail.com> wrote:
> > 64 spinning objects
> > ------------------------
> > no spin fix: 555.1 fps
> > spin fix: 554.5 fps
> > 256 spinning objects
> > --------------------------
> > no spin fix: 206.1 fps
> > spin fix: 202.4 fps
> > 1024 spinning objects
> > ---------------------------
> > no spin fix: 23.9 fps
> > spin fix: 22.5 fps
> 
> I suggest turning renderer off to measure raw performance. Right now,
> I guess, you spend most of the time submitting draw calls and/or
> traversing the scene (or whatever else ogre3d does :)).

Ok, I ran some new tests with no graphics, just simulation.  This time
I used more spinning objects since the lack of graphics gave me a lot
more CPU time.  The results show about the same relative performance
as the tests with graphics on.

3600 spinning objects
------------------------
no spin fix: 83.37 fps
spin fix: 83.37 fps

6400 spinning objects
--------------------------
no spin fix: 42.72 fps
spin fix: 40.33 fps

10000 spinning objects
---------------------------
no spin fix: 24.47 fps
spin fix: 22.89 fps

Again, most people would rarely need this many freely-spinning objects at once.

Tyler



More information about the ODE mailing list