Latency Measurements for Full Video Encode Pipeline

Autonomous Machines Jetson & Embedded Systems Jetson TX1
1

Hello. I was curious if anyone has standardized latency measurements for each individual part of the encode pipeline.

For example TI on the 8168 has the following breakdown for their product:[url]http://processors.wiki.ti.com/index.php/Latency_Measurement_on_Capture_Encode_Decode_Display_Demo#Summary[/url]
The only latency measurements I have seen anyone do were performed by RidgeRun and shown here:
Jetson TX1 GStreamer Pipelines | Jetson TX1 ARM Load

What I am curious of is what are the expected latencies for:
mipi capture to v4l2src
4:2:2 to 4:2:0 down conversion
H.264 encode
Mpeg2tsmux
udp stream

also for the decode end:
capture of udp stream via udpsrc
mpeg2 demux
h264 decode
render to display

2

Hi x1tester62,
We don’t have the implementation. Wish other users can share experience.

Here also is a related post:
[url]https://devtalk.nvidia.com/default/topic/989881/jetson-tx1/profile-gstreamer-plugins-/post/5066525/#5066525[/url]