BEPUphysics

This forum is a little old and stinky. I'm making it readonly in favor of github discussions!

v1: https://github.com/bepu/bepuphysics1/discussions
v2: https://github.com/bepu/bepuphysics2/discussions

The phpBB forum will remain visible for archival purposes for the foreseeable future.

To lock rotation, you can set the inverse inertia tensor to all zeroes: inertia.InverseInertiaTensor = default; You can technically lock individual rows in the tensor by setting them to zero too. Be wary, this can result in unsolvable constraints sometimes. A single body constraint that used the rel...

A few things: 1) The index passed into IntegrateVelocity is not a handle, but rather an index into the contiguous active set. The integrator only touches bodies that are active, so it doesn't do the handle indirection by default. If you want the handle, you'll need to look it up through the Simulati...

Not at the moment, but querying the broad phase for bodies with overlapping bounding boxes is a good start. Something like: var broadPhaseEnumerator = new BroadPhaseOverlapEnumerator { Pool = BufferPool, References = new QuickList<CollidableReference>(16, BufferPool) }; Simulation.BroadPhase.GetOver...

I don't see a stack overflow locally; are you using latest master or the latest nuget prerelease package (2.3.0-beta10)? If I remember correctly, there was a semirecent fix to crashes with degenerate mesh input. The underlying problem is that the mesh is degenerate. All vertices have the same x and ...

Using a CompoundBody (which is v1's named entity type for an entity with a CompoundShape) would work: https://github.com/bepu/bepuphysics1/bl ... iesDemo.cs

Other than the standard recommendations I already mentioned (play around in the demos and look at the documentation to get a feel for how things are set up), I don't have much to add, especially with respect to unity integration. If you have a specific non-unity question, I might be able to help more.

Setting a position is equivalent to teleportation- there is no velocity component, so things will only move out of the way due to penetration recovery. If the ground teleported up a little bit (and the affected bodies are awake), they would scoot up a bit in response to get out of penetration. If th...

The fixed point fork of bepuphysics1 is supposed to be cross platform deterministic, but I didn't make that fork and haven't used it so I can't offer much commentary there. I also don't use unity, so I'm of limited help- but at the core, there's not much to it. There's stuff in a simulation, you can...

Sometimes I forget how v1 works :P It's a numerical issue caused by the small scales involved. By default, there are thresholds configured to work well with sizes ranging from 0.5 to 10. Values significantly outside that will be more likely to encounter issues. For a simulation primarily concerned w...

When you say stepped 5 times, do you mean reduce the solvers iteration limit to 5? No, just calling Space.Update() more frequently to compensate for the lower amount of time simulated per timestep. If you're using internal timestepping (Space.Update(dt)), it'll happen automatically. The solver iter...

If I had to guess, it's related to the size of the ball, the timestep, and the default slight softness of collisions. If it hits hard enough, the constraint may not fully eliminate all penetrating velocity in one timestep. Once the object is in contact, CCD can't help, so if enough velocity is left ...

It would probably be worth showing what's actually possible, yes Will be a while before I get to it, though- authoring real fancy content is pretty effortful!

Yup! It's pretty neat. Incidentally, bepuphysics2 has advised using substepping and lower velocity iteration counts for the same underlying reasons (since well before the paper, if I might toot my own horn :P). It's an extremely strong stabilizer and enables higher frequency responses. You could, in...

There is no exact mapping of coefficient of restitution/bounciness to v2's parameters. -Damping ratio of 0 is undamped, which corresponds to a full bounce. In practice, the timestep durations necessary for real time simulation will introduce extra damping when the true version of the simulation can'...