[Opensim-dev] BulletSim testing

[Bri Hasp]

I have 2 test setups that switch to ODE quickly for comparison.
(we do a Havok comparison with copied vehicles) 

a Win 8/.Net ivy-bridge/16gb D2. (opensim-61b816e)
a Suse12.2 /mono sandy-bridge/16gb on OSGrid (opensim-61b816e)

Initially checked basic region asset and Avatar actions. Was only marginally
different from ODE. 

Ran a 1 prim way-point vehicle with identical rectilinear route and
parameters.
speed was slower and it ran the last side with pauses. Bullet (108 sec) ODE
(58 sec) Havok (57 sec).

Tried two sailboats and concluded the up-wind tacking switchover lag is too
large to do any meaningful tests.  
Friction did feel lower running downwind. Perhaps scripting can improve the
boat's vector changes.
(tacking is pointing a boat into the wind and running a zig-zag course)

Bri Hasp



--
View this message in context: http://opensim-dev.2196679.n2.nabble.com/BulletSim-testing-tp7578304p7578317.html
Sent from the opensim-dev mailing list archive at Nabble.com.