Simon, so you can pass me a .vdb filename in a new ".file" property instead of the ".data" property in density grid texture, right ?
Can you post a test .vdb exported by Blender for testing ?
I think we can release beta1 with the new volume rendering code and beta2 with .vdb support (the only time consuming part of adding OpenVDB support is adding OpenVDB library to Linux and Windows binary dependency repositories).
Support LuxCoreRender project with salts and bounties
Dade wrote: ↑Sun Mar 25, 2018 10:24 pm
Simon, so you can pass me a .vdb filename in a new ".file" property instead of the ".data" property in density grid texture, right ?
Yes.
Dade wrote: ↑Sun Mar 25, 2018 10:24 pm
Can you post a test .vdb exported by Blender for testing ?
I'll prepare a few variations in the evening.
Dade wrote: ↑Sun Mar 25, 2018 10:24 pm
I think we can release beta1 with the new volume rendering code and beta2 with .vdb support (the only time consuming part of adding OpenVDB support is adding OpenVDB library to Linux and Windows binary dependency repositories).
I think we can release beta1 with the new volume rendering code and beta2 with .vdb support (the only time consuming part of adding OpenVDB support is adding OpenVDB library to Linux and Windows binary dependency repositories).
On linux and osx you do not even need the libs explicite, only headers for compiling.
The module pulls its symbols from the host app just like it does with python.
Avoid this lib delivery overkill.
Dade wrote: ↑Sun Mar 25, 2018 10:24 pm
Can you post a test .vdb exported by Blender for testing ?
Here is the same simulation baked with 3 different compression methods.
I also added one version with half float format (blosc compression) - the others support this, too.
Download button in the upper right corner. https://drive.google.com/open?id=1eWqt7 ... RKKgkTNjcJ
Dade wrote: ↑Sun Mar 25, 2018 8:22 pm
I'm working on the new volume rendering code. this is a cube with a volume defined by an FBM procedural texture:
That‘s great news. What are the improvements compared to the old code?
It is a lot faster and it works
I must admit I have still aliasing problem with sharp edges but there aren't usually a lot of sharp edges in most common heterogeneous volumes (i.e. smoke, etc.).
Just compile the latest sources and try to use an heterogeneous volume with density grid (btw, I added the OpenCL support for density grid texture too).
Support LuxCoreRender project with salts and bounties
Dade wrote: ↑Mon Mar 26, 2018 10:04 pm
It is a lot faster and it works
I must admit I have still aliasing problem with sharp edges but there aren't usually a lot of sharp edges in most common heterogeneous volumes (i.e. smoke, etc.).
Just compile the latest sources and try to use an heterogeneous volume with density grid (btw, I added the OpenCL support for density grid texture too).
OpenCL support is really great. Thanks for that. I tried to implement that some time ago but only got frustrated as debugging the code was a nightmare. Is there an automatic step size estimation for heterogeneous volumes? Aliasing was bad with the old code, too, and multiple scattering volumes wasn't as bright as rendered with other renderers. I have to try it.
i7 5820K, 32 GB RAM, NVIDIA Geforce RTX 2080 SUPER + GTX 1080, Windows 10 64bit, Blender 2.83.5
Support LuxCoreRender project with salts and bounties