I discovered that rendering, not decoding, is a bottleneck for Jetson NANO displays.
We have been working to improve rendering performance in a variety of ways.
I used the NVIDIA SDK or tried OpenGL myself.
However, as a result, it is not possible to rendering more than a maximum of 60 fps per process.
Our goal is to rendering up to 60fps per thread (per camera) independently.
I want to rendering up to 8 channels @ 30fps at the same time.
Can you give me any advice for rendering up to 60fps or more per process?