Better way to integrating forces in btActionInterface?
Posted: Thu Mar 20, 2014 3:57 pm
In my hydrodynamic simulation, I am computing per face forces, multiplying them by the current time step, and applying them as impulses to my rigid body. I believe this is basically the euler method for integrating acceleration.
Given that the forces are not constant, it would seem that a different integration method could provide a more accurate simulation. I would assume that bullet uses a better integration method for forces already, but I've had issues every time I tried to apply forces directly to my model instead of using impulses inside of the btActionInterface and the the simulationTick callback. Is there a better way for me to be integrating hydrodynamic forces in my simulation with bullet? Should I be looking at some other bullet interface that plays more nicely with a solver?
Thank you for any direction you can provide.
Given that the forces are not constant, it would seem that a different integration method could provide a more accurate simulation. I would assume that bullet uses a better integration method for forces already, but I've had issues every time I tried to apply forces directly to my model instead of using impulses inside of the btActionInterface and the the simulationTick callback. Is there a better way for me to be integrating hydrodynamic forces in my simulation with bullet? Should I be looking at some other bullet interface that plays more nicely with a solver?
Thank you for any direction you can provide.