Elie Michel

Regarding the other issue I don't have precise ideas but for this one what do you think about the suggested fix of changing line 896 of autodiff.cpp? Did I misunderstand...

Hi Sebastien, Don't have CUDA set up here. I understand DiffArrays with simple floats received less attention, nevertheless they are explicitly implemented here https://github.com/mitsuba-renderer/enoki/blob/master/src/autodiff/autodiff.cpp#L1223 which means somebody though about supporting...

I understand, no worry, I worked it around by replacing the matrix vector product by a series of dot products for now.

Storing vertex attribute values on a per-vertex basis fragments the memory and does a lot of repeated allocations when adding a new attribute. It would be much more efficient to...

Agreed as well! *** Since we are on the way to breaking changes, it is also time to think about performances a bit. I really wanted to foster ease of...

Agreed, in its current state the API is not safe at all, it assumes that the user knows a bit what they're doing.

You can use BMesh at runtime for sure, but it includes no OBJ importer at the moment. This is an interesting suggestion though, you may build something based on this...

I noticed, indeed, but the "back from archive" feature really interests me. ;) edit: I started implementing it, and have a `archive_mask` argument slightly spreading in the code. Hope it...

@timurhai Which one do you think is the best: - Using a single instance of `JobContainer` holding `Job`s loaded from different archives. Requires the `JobContainer` to be able to hold...

@timurhai Your last comments corresponds to my first bullet point ;) @sebastianelsner Making IDs unique is hard, because the archive lives independently from the rest of the program, which will...