I’ve started to experiment with ray tracing in ParaView and so far I have only been able to use OSPRay for CPU based rendering which is quite slow to be honest. I wanted to try the OptiX option for GPU rendering with my NVIDIA card but as soon as I select that option ParaView freezes and stops working. I have a NVIDIA GeForce GTX 1660 Ti (gaming GPU) and I am a windows user.
I assume that I might not have the right NVIDIA drivers or installed ParaView in the correct way. Actually I didn’t install it, I’ve just downloaded the .zip installer and executed the binary files because this downloadable version of ParaView has more python modules than the .msi. However I couldn’t find the right answer to my questions on the internet so I am writting this post hoping that I could get a better answer.
This mode will render any kind of surface geometry using ray tracing, to simulate advanced lighting effects like reflections or refractions in real-time.
If that does not work out-of-the-box, please check your log for errors. Updating the graphics driver is the first option to try.
Do you have another GPU in your system? In that case, check the Windows Graphics settings. You can assign the NVIDIA GPU as the default graphics processor to your ParaView installation. To verify, the NVIDIA GPU should be listed in the ParaView About box. This can help to resolve graphics conflicts between multiple GPUs.
BTW, to render volumes in real-time, there is another NVIDIA technology integrated into ParaView that is available as a plug-in: NVIDIA IndeX. It can visualize volumetric datasets, regular or irregular, and allows interactive manipulations including live colormap editing.
Thank you so much Steffen for your detailed reply. After some adjustments I could make it work but since my GPU doesn’t have accelerated ray tracing hardware the image quality is quite poor. For this reason I am trying to switch now to my linux laptop where I have indeed 2 GPUs: a NVIDIA Quadro RTX 4000 Mobile and an Intel UHD Graphics. And I can also see that the Intel GPU is the only recognized by ParaView so I can understant why it crashes when using Optix:
Client Information:
Version: 6.0.1
VTK Version: 9.5.2
Qt Version: 6.9.0
vtkIdType size: 64bits
Embedded Python: On
Python Library Path: /home/user/Documents/ParaView-6.0.1/lib/python3.12
Python Library Version: 3.12.7 (main, Sep 25 2025, 18:56:46) [GCC 10.3.1 20210422 (Red Hat 10.3.1-1)]
Python Numpy Support: On
Python Numpy Path: /home/user/Documents/ParaView-6.0.1/lib/python3.12/site-packages/numpy
Python Numpy Version: 1.26.4
Python Matplotlib Support: On
Python Matplotlib Path: /home/user/Documents/ParaView-6.0.1/lib/python3.12/site-packages/matplotlib
Python Matplotlib Version: 3.7.2
Python Testing: Off
MPI Enabled: On
ParaView Build ID: superbuild 8c0f929966f005396802f3b2957c734a6c68b3c1 (!1292)
Disable Registry: Off
Test Directory:
Data Directory:
SMP Backend: TBB
SMP Max Number of Threads: 12
OpenGL Vendor: Intel
OpenGL Version: 4.6 (Core Profile) Mesa 21.2.6
OpenGL Renderer: Mesa Intel(R) UHD Graphics (CML GT2)
Accelerated filters overrides available: No
Connection Information:
Remote Connection: No
How could I switch to my RTX GPU in linux? I should also add that I tried in the past to compile the paraview superbuild without much success but I mention it here because I see the reference in “ParaView Build ID“ but I am not using it. Should I do a hard reset of all my ParaView installations?
