Mikael Hermansson

Awesome. I'm working on it as we speak. > Beware that it costs up to 25% extra memory per mesh, so you may want to put this behind another project...

Looks to be working well enough (#955) but I've only done some very basic testing so far. The whole `GetLeafShape` dance and dealing with `SubShapeID` remainders does feel a little...

Fair enough! That all sounds reasonable to me.

To anyone who might still be subscribing to this issue, support for `face_index` can now be found in the newly released [0.14.0-stable](https://github.com/godot-jolt/godot-jolt/releases/tag/v0.14.0-stable) release. Note that the release on Godot Asset...

To start with, I don't have a ton of experience with parallel programming, outside of your average mutex/atomic. I would certainly much rather rely on Jolt's existing locking. If it...

Not that I'm aware of. You have [this](https://docs.godotengine.org/en/stable/contributing/development/core_and_modules/godot_architecture_diagram.html) for Godot and for Jolt I guess you have the "Simulation Step in Detail" section in [this](https://jrouwe.github.io/JoltPhysics/) document. Neither of them will...

I'm not very familiar with how DOTS works, so I can't speak to any specific discrepancies, but from what I understand `PhysicsServer3DWrapMT` aims to mostly fulfill the safety part of...

> May I ask what the status of this is? I haven't worked on this at all yet, and I don't currently have any plans to. It's looking likely that...

> I'm assuming Jolt supports this under a similar assumption (correct me if not though). Yes, Jolt supports this just fine. > But with GodotJolt, from the last discussion we...

As mentioned by @cridenour already, thread-safety is a thing in the newly merged Godot engine module, due to being able to make use of Godot's `PhysicsServer3DWrapMT`. With `PhysicsServer3DWrapMT` effectively just...