Applying joints to meshes vs. xforms?

Is there are a difference in where the joint is applied? I did a few experiments creating the same joints between meshes and afterwards replaced these with the ones between the xforms of the same meshes with same results.

Is there a standard methodology or certain differences to be aware of in creating joints between meshes <> xforms?

It only makes sense to talk about applying physics joints to rigid bodies - i.e. Bodies whose position is determined by physics simulation.

Those are specified by the Rigid Body API and that can be set either to a parent Xform or directly onto a mesh.

Technically there’s no fundamental difference between either approaches, but there’s a key advantage to setting to a parent, that is the ability to define your own set of colliders that are independent from the visual representation of the body. And this is an interesting take because High fidelity collision comes at a cost in physics simulation, so as a rule of thumb we try to simplify our colliders as much as we can get away with to boost our performance. For example, instead of using a high polygon mesh with intricate details about where masking tapes, small bumps and crevices are in a cardboard box, we just use a regular bounding 6-faces rectangular prism. The link above should give you more examples and details on this, as well as details about the joints API

1 Like

Perfect response, thank you 🙌

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.