Acceleration by Cuda is not recognized in Blender

Accelerated Computing CUDA CUDA Programming and Performance
,
1

My system: Kde Neon, kernel: 5.11.0-41-generic GPU: GTX 1050ti

Fri Dec 10 02:50:54 2021       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 495.44       Driver Version: 495.44       CUDA Version: 11.5     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                               |                      |               MIG M. |
|===============================+======================+======================|
|   0  NVIDIA GeForce ...  Off  | 00000000:02:00.0  On |                  N/A |
| 40%   26C    P8    N/A /  75W |    368MiB /  4035MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+

Here I am having the same problem, I installed the driver, from the nvidia website, NVIDIA-Linux-x86_64-495.44.run, one thing I noticed when testing Debian is that with the 460.91.03 driver Blender recognizes the GPU(gtx1050ti) , since the 470 and 495 do not.

Here I am having the same problem, I installed the driver, from the nvidia website, NVIDIA-Linux-x86_64-495.44.run, one thing I noticed when testing Debian is that with the 460.91.03 driver Blender recognizes the GPU(gtx1050ti) , already the 470 and 495 no.

I’ve tried everything and still haven’t had success, it only works with the 460.91.03 driver…
“for hardware acceleration to work in Blender it is not necessary to install the cuda-toolkit, the video driver already comes with Cuda libraries for some purposes.”

2

You might want to see if you can get support from the blender community for this:

The decision to recognize or not recognize the GPU is a function of Blender. This is not a CUDA programming question.

3

I get confused because in a forum some Blender dev said not to me, but to another one who had the same problem, that it seemed like a problem in the driver… I don’t know what to do… I’m investigating nvidia-persistenced, which is giving a error in initialization of deamon, I don’t know if that would solve my problem, but I’m on that line for now!