[webrtc-encoded-transform] Frame packetization (#233)

AfonsoVilalonga has just created a new issue for https://github.com/w3c/webrtc-encoded-transform:

== Frame packetization ==
I have a question that is probably not entirely suitable for this repo, but I wasn't sure where else to ask it.

Basically, it seems that the maximum size of a VP8 frame that the browser (Google Chrome) can send before packetization is 1174 bytes, but I am not entirely sure why. 

I tested this by reading the RTP packets in a Pion application that are being received from a web browser application (Google Chrome). The size of the frame is manipulated using the WebRTC Encoded Transforms API, and for sizes greater than 1174 bytes, the frame is divided into two packets.

However, this doesn't make much sense to me, since the size of the RTP header being sent is 12 bytes (as seen in Wireshark), and 4 additional bytes are added to each packet for the VP8 payload descriptor. This means that the maximum size should be 1200 - (12 + 4) = 1184 bytes. Is there any insight I might be missing, or am I testing something incorrectly?

Thank you in advance for the help! 

Please view or discuss this issue at https://github.com/w3c/webrtc-encoded-transform/issues/233 using your GitHub account


-- 
Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config

Received on Friday, 25 October 2024 13:23:48 UTC