LuxCoreRender Forums

SIRIO DEPERO wrote: ↑

Thu Apr 02, 2020 5:49 pm

I noticed the problem is with the windows.

Sometime, Windows Update breaks NVIDIA OpenCL driver installation. It is something we have seen to happen in the past. Can you try to re-install NVIDIA drivers ?

juangea wrote: ↑

Thu Apr 02, 2020 9:32 am

I also think it’s worth it.

But don’t forget about clamping for light groups, they are not so useful without clamping (just mentioning it, not trying to get a thread-hijack )

Do you mean a separate clamping parameter for each light group ? It may be very easy to add.

Sharlybg wrote: ↑

Wed Apr 01, 2020 5:49 pm

Btw simplify node make displacement export time horrible.
It is really slow.

It is a complex and iterative operation but, indeed, it is related to the initial subdivision too (i.e. it depends on initial triangles count and on the final requested triangles count).

Dade wrote: ↑

Tue Mar 31, 2020 10:46 am

This is likely to be a duplicate of: https://github.com/LuxCoreRender/LuxCore/issues/309

You are using a negative scale and that is the source of the problem, as work around, you can apply all transformation to the object and it should fix the problem.

I should have fixed this problem.

Sharlybg wrote: ↑

Tue Mar 31, 2020 12:50 pm

for those of us with 32 to 128 giga of RAM it is limitating.

It is also GPU ram, not only CPU.

Sharlybg wrote: ↑

Tue Mar 31, 2020 12:00 pm

Is there any technical reason that can prevent Photon GI to work with Light groups ?

It would require a separate PhotonGI cache for each light group, so the answer is: memory usage.

This is likely to be a duplicate of: https://github.com/LuxCoreRender/LuxCore/issues/309

You are using a negative scale and that is the source of the problem, as work around, you can apply all transformation to the object and it should fix the problem.

The error sounds like a corrupted file, try to erase and re-install BlendLuxCore.

So "soon" we might see if luxcore on cuda is going to be faster than on opencl? If NVIDIA has not crippled its OpenCL drivers, the performance should be about the same ... IF ... We will see, anyway the motivation beyond this work is all in having access to NVIDIA-only goodies like RTX, Optix, etc....

Introduction The support for CUDA image pipeline (instead of OpenCL) is now available on sources after the merge of "cuda_imagepipeline" branch. Having CUDA support, instead of OpenCL, for the image pipeline is not particularly useful but it was a perfect test case to check if it was possible to ad...