It is hard to get models out of Blender into Godot (correctly and without loosing half the work).
Gamedev relevant open-source projects should come together to create a new 3D interchange format. More in the GitHub discussion.
I’ve submitted this proposal to other open source projects as well:
From the proposal:
I don’t want to bash on USD, but its use is mostly in the VFX world, which vastly differs from game-dev.
Also, last month, I tried to get a model from Blender 4.2 to Unity 2022 in USD. It omitted the main armature. I quadruple checked all settings^^
That Godot plugin you linked uses Blender to import USD files to convert them to glTF 2.0, which then will actually get imported in Godot.
Collada has been deprecated in Blender.
What you said is patently wrong.
As someone who works with USD daily digging through the bowels of the code and working with the format at a root level, it is 100% something that gaming should (and is) adopting. The Godot team would be frankly stupid not to put full native USD support as a high priority. The ability to store rigs, transform animation, and easily encapsulated asset variations alone makes the format godly. It is also very clean in how it manages asset linking and scenegraphs which makes traversal a breeze to encode. Also, it makes an DCC agnostic format which allows game developers to engage with freelance artists and not care what they use. Houdini, Blender, Maya, C4D, and any future DCC are going to be able to export to it. Hell, Houdini (my preferred DCC) has an entire context dedicated exclusively to working with, rendering, and exporting USD format.
I would have thought that applied to ANY universal 3D format. For everything you might want to work, to work, the feature-set that must be implemented is truly mind-boggling.
I can’t imagine a format only supporting a subset of what USD does, ever becoming universal.