Christopher Crouzet

Hi @Luke-Skycrawler, thanks for reporting this issue and apologies for any inconvenience caused! ~The intention is for vectors to be row-major, so the kernel constructor is incorrect.~ I'll discuss how...

Hi @Luke-Skycrawler! Apologies for the year-late reply but we finally did address this issue in commit [62daadd](https://github.com/NVIDIA/warp/commit/62daadda34f794bcb8c8ece4fe40c95224005b26) by introducing new `wp.matrix_from_cols()` and `wp.matrix_from_rows()` functions, that can be called from both...

Hi @biansy000, are you running `import omni` from within an Omniverse application? The `exts` folder is dedicated to extensions for the Omniverse platform, so it won't run in a standalone...

Sincere apologies @lmtss for not replying sooner. I could repro it on my end and this does look like a bug indeed.

Apologies once again @lmtss, I skimmed too quickly over your description and didn't notice the part about `mul` being commented out. What you observed is to be expected. For Warp...

Hi @cadop, I personally can't run `self.points = self.particle_grid(1024, 1024, 256, (0.0, 0.5, 0.0), self.point_radius, 0.1)` because my GPU is running out memory, so I'm assuming the same might be...

Hi @jinz2014, this is by design. See the related entry in the FAQ: [Why aren’t assignments to Warp arrays supported outside of kernels?](https://nvidia.github.io/warp/faq.html#why-aren-t-assignments-to-warp-arrays-supported-outside-of-kernels).

Note that you can call `arr.numpy()` to return the corresponding NumPy array on the CPU, and from there you'll be able to index/assign normally to it. If the `wp.array` data...

That's right but the API exposed by `wp.array` needs to work for both GPU and CPU arrays. So if something isn't optimal in one case or the other, we don't...

You can already initialize Warp arrays on the CPU by passing them a block of data at once, e.g.: `wp.array((1.0, 2.0, 3.0), dtype=wp.float32, device="cuda")`. You can also initialize a Warp...