The <translate> and <rotate> under <shape> in physics should be local to the bound node so it can be reused. I don't think that's the case in these files (see darpa-arm.dae and neuronics-katana.dae for what I've looked at.)
I think it's not good to use the full geometry for because it takes a long time to prepare for simulation and might be slow to simulate. Some of these have very detailed visuals that would not make sense for for collision detection.
EDITED: The translate/rotate stuff is for offsetting basic shape geometries. When a mesh is used it's typically not required to be offsetted unless a collision model is being constructed from a kit of ready-made shapes.
The
<translate>and<rotate>under<shape>in physics should be local to the bound node so it can be reused. I don't think that's the case in these files (see darpa-arm.dae and neuronics-katana.dae for what I've looked at.)I think it's not good to use the full geometry for because it takes a long time to prepare for simulation and might be slow to simulate. Some of these have very detailed visuals that would not make sense for for collision detection.
EDITED: The translate/rotate stuff is for offsetting basic shape geometries. When a mesh is used it's typically not required to be offsetted unless a collision model is being constructed from a kit of ready-made shapes.