NVIDIA Optical Flow SDK returns unsupported device for GTX 1650 Ti

Gaming and Visualization Technologies General Topics and Other SDKs
1

I am trying to use NVIDIA Optical Flow SDK 2.0.23 on my Dell XPS. It has the following output for `nvidia-smi.

+-----------------------------------------------------------------------------+
| NVIDIA-SMI 450.80.02    Driver Version: 450.80.02    CUDA Version: 11.0     |
|-------------------------------+----------------------+----------------------+
| 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  GeForce GTX 165...  Off  | 00000000:01:00.0 Off |                  N/A |
| N/A   49C    P8     3W /  N/A |    994MiB /  3914MiB |      9%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+

I build the NvOFBasicSamples and try to execute the AppOFCuda sample using the command - ./AppOFCuda --input=<folder>/*.png --output=<folder> and get the following output

GPU in use: GeForce GTX 1650 Ti
NvOFCudaAPI : m_ofAPI->nvCreateOpticalFlowCuda(m_cuContext, &m_hOF)returned error 2 at Optical_Flow_SDK_2.0.23/Common/NvOFBase/NvOFCuda.cpp;35

To my understanding, the error code 2 matches to NV_OF_ERR_UNSUPPORTED_DEVICE error code enum. However, the GTX 1650 Ti is a Turing architecture GPU supported by the SDK.

Am I missing something here?

2

@dusty_nv Any leads would be appreciated

3

GTX 1650 Ti → TU117

NVIDIA Optical Flow SDK Dependencies:
NVIDIA GeForce, Quadro and Tesla products with Turing (except TU117) and Ampere generation GPUs

1 Like