So CUDA and cudnn are installed correctly, including visual studio, but when rendering I only see activity on my CPU (screenshot below).

I assume this is because the app only needs the VRAM, as my card is running at full speed, 75°C, but I'm just confused that utilisation is around 0 - 2%.

I'm rendering a 25fps video on 4x interpolation, resized to 480 as anything over doesn't work.

Specs if they help:

CPU: i5-9600kf clocked at 4.9GHz (hex core chip)

GPU: RTX 2070 8GB VRAM

RAM: 16GB DDR4 2133MHz

Storage: Intel 660p 1TB M.2 SSD

Thanks,

Jake

Similar behaviour here. But if you click on GPU to view the details you can see some activity in the memory copy section. My guess is that the algorithm itself uses a lot of memory accesses but not so much computations. That's generally not ideal for GPU acceleration. However, the application seems to be some kind of straightforward build from the python code so the GPU code most likely comes from python libraries. So there definitely is a room for optimization but it would probably require rewriting the app from scratch in CUDA C/C++.

Same situation
cpu maintained 50 ~ 60%
gpu only 0 ~ 5%

windows 10 1809

CPU: i7-9700

GPU:GTX 1070 8g

 RAM:16G -2666

CUDA 10.2

Just to clarify, this is what I'm now seeing as a result of karuchie's suggestion.