I did a bisect and found the first 'bad' commit was the one that added the log message and assert (so would require a more complicated bised that would involve putting this check into past…
If it helps, I examined some variables when the crash (assert abort) happens:
in function MTLContext::ensure_buffer_bindings:
- crash is due to call of this message print: "Shader 'eevee_lega…
I am currently working through attribute propagation as I write the new solver. Once I understand how to do it properly (hopefully fixing the bugs reported here and the other issue), I will…
I kind of like that idea, Bastien, and think I will try it out. For one thing, it makes it easier for me to experiment with some things without having to involve the platform maintainers (though I…
Here are some timings for a Release build on my Macbook Pro M3 Max (12 performance cores, 4 efficiency cores):
For a sphere-sphere difference where sphere1 has 200k faces (400k triangles) and…
I can see what I can do. Interestingly, Manifold internally keeps a tree of boolean operations, unevaluated, until one asks for the mesh output.
I did a very rough timing test yesterday…
Answering Jacques' earlier questions:
Overall, I think the Manifold library seems to be a good fit for procedural workflows, probably better than the other solvers we have already. Are there…
Niki,
Thanks for your response.
By "how large", I meant "how many polygons", but you answered that already.
Can I assume you are currently using the "Float" solver in the modifier? (I…
I'll answer Jacques' questions later, but Nika: would you like to explain your particular requirements on boolean here? Do you care about the output topology (do you care if they are all…