It seems that FFmpeg is inserting an auto_scaler before the encoder which converts from yuv420p10le (my input format) to p010le (the 10-bit format accepted by NVENC). But a reduction in throughput on this scale makes me think that I'm doing something wrong.
Is there anything I'm missing in my command for 10-bit HEVC encoding? I understand that with 10-bit, because of the increased bit-depth, the encoding will take longer. ffmpeg -h encoderh264nvenc Encoder h264nvenc NVIDIA NVENC H.264 encoder: General capabilities: delay hardware Threading capabilities: none Supported hardware devices: cuda cuda Supported pixel formats: yuv420p nv12 p010le yuv444p p016le yuv444p16le bgr0 rgb0 cuda h264nvenc AVOptions: -presetIn the example, I'm using CBR as rate-control mode but I also tested and obtained similar results (in terms of encoding throughput) with VBR and constant QP modes. But for 8-bit sequences I'm getting a much higher throughput of around 450-500 fps with similar preset and rate control modes. For 1920x1080 10-bit sequences, I get around 50-60 fps with HEVC encoder. The results are similar for all the 10-bit sequences I encoded. I also tried with different presets and different sequences. A logfile from the ffmpeg output from the above command is here. However, I only get around 15 fps encoding speed with this command. My goal is to achieve as low latency as possible therefore I set the preset to low-latency high-performance. ffmpeg -hide_banner -benchmark -loglevel debug -y -f rawvideo -s:v 3840x2160 -r 50 -pix_fmt yuv420p10le -i ParkRunning3_3840x2160_50fps_10bit_420.yuv -c:v hevc_nvenc -preset hp -rc cbr -profile:v main10 -b:v 10M output.mp4
For an example sequence and configuration, I use the following command.
I'm encoding some 4K 10-bit YUV test sequences using NVENC HEVC encoder.
0 kommentar(er)
