LuxCoreRender Forums

Yes, it is the same with the cuda version and also the opencl only version.

Just updated Blendluxcore with the content of LuxCore commit 071e1cb (tentative fix for an OpenCL crash), but still get a crash when trying to render the Benchmark scene with GPU devices selected as OpenCL (Cuda works).

I have installed cuda sdk 10.2 for the new compilation of the new code. The old one I had installed was 8.0, IIRC. Could that be the source of the problem of the error messages?

epilectrolytics wrote: ↑Tue Apr 28, 2020 3:11 pm Just updated Blendluxcore with the content of LuxCore commit 071e1cb (tentative fix for an OpenCL crash), but still get a crash when trying to render the Benchmark scene with GPU devices selected as OpenCL (Cuda works).

Experience on my side...

1. OCL renders only final, but image intensity goes darker as rendering progresses/ viewport goes instantly black
2. OCL+CUDA together, show mixed result // Viewport is striped, tiled had this happening before selections were added
3. CUDA fine
4. CPU fine
BTW
UI wise, I think there should just simply be CUDA Device option added in dropdown menu, which would automatically deselect NVidia's OCL to prevent the confusion and extra settings

kintuX wrote: ↑Tue Apr 28, 2020 8:48 pm UI wise, I think there should just simply be CUDA Device option added in dropdown menu, which would automatically deselect NVidia's OCL to prevent the confusion and extra settings

The UI will be improved of course.

B.Y.O.B. wrote: ↑Tue Apr 28, 2020 8:54 pm

kintuX wrote: ↑Tue Apr 28, 2020 8:48 pm UI wise, I think there should just simply be CUDA Device option added in dropdown menu, which would automatically deselect NVidia's OCL to prevent the confusion and extra settings

The UI will be improved of course.

Indeed, I have faith in you... was just stating the obvious

I really find this engine becoming the first true hybrid in nature (all vendors) simply amazing.
But alongside (offtopic), I'm quite puzzled about how you'll define "Devices" while still offering a choice between OCL/CUDA for NVidia in a mixed system.

Respect

BTW
Should this topic be split as it already builds fine on Windows, and we've started to bug fix the version in development?

Hi, I test only small scenes but does anybody have the same performance boost like me?
It is about 1.8 times faster than OCL on my stone age GTX 760.
Thank you for implementing it @developers.

Cheers, mib

mib2berlin wrote: ↑Tue Apr 28, 2020 9:28 pm Hi, I test only small scenes but does anybody have the same performance boost like me?
It is about 1.8 times faster than OCL on my stone age GTX 760.

There seem to be a pattern: small GPUs gain a LOT from CUDA

mib2berlin wrote: ↑Tue Apr 28, 2020 9:28 pm Hi, I test only small scenes but does anybody have the same performance boost like me?
It is about 1.8 times faster than OCL on my stone age GTX 760.
Thank you for implementing it @developers.

Cheers, mib

I have a similar boost result on my old (gtx960 4gb)

Tell me, is cuda also now involved in the viewport rendering? It also feels like a faster viewport (no delays,lags, faster compiling)

zuljin3d wrote: ↑Tue Apr 28, 2020 11:56 pm

mib2berlin wrote: ↑Tue Apr 28, 2020 9:28 pm Hi, I test only small scenes but does anybody have the same performance boost like me?
It is about 1.8 times faster than OCL on my stone age GTX 760.

I have a similar boost result on my old (gtx960 4gb)

neo2068 wrote: ↑Mon Apr 27, 2020 6:21 pm I get errors with latest build and OpenCL and the food demo scene, too, e.g. PATHOCL:clFinish(CL_INVALID_COMMAND_QUEUE) or TILEPATHOCL: clEnqueueReadBuffer(CL_OUT_OF_RESOURCES). With cuda my GTX 780 is 1.5 times faster than the GTX 1080. Is it normal behaviour?

acasta69 wrote: ↑Tue Apr 21, 2020 5:36 pm Now, without native threads and samples/sec as reference, I have 12% faster performance with OpenCL [than with Cuda on GTX970]

It seems we have a wide variety of results, mostly improvements but not all of them.
Also a GTX780 faster than a 1080 is quite strange, isn't it? The 1080 has more Cuda cores and nearly double clock speed and memory bandwitdh than the 780.