WebRTC One To Many Broadcast (SFU)
Why Stack Overflow recommends it
Stack Overflow users link it as a starting point for one-to-many or few-to-many WebRTC streaming and as the basis for Node.js wrtc server implementations.
Recommended by 1 people across 1 different months, 2021 to 2021.
What it is
A 12-minute tutorial by Coding With Chaim showing how to build a one-to-many WebRTC broadcast using a Node.js server acting as a simple SFU. The presenter's tracks are added to each viewer's peer connection. Final code is on GitHub.
What people said
Recommends Kimberlee Johnson · 2021-06 · Stack Overflow · on “How to Broadcast (From one to multiple peers) using Webrtc and Firebase”
I haven't created a WebRTC + firebase one-to-many streaming application myself yet, but here are a few resources I'd look at if I was: Coding with Chaim [this video] Doug TV site walkthrough [LINK] awesome-webrtc repository [LINK] has a lot of gems Since you asked about other npm packages, there's also daily-js [LINK] , but I'm biased. Full disclosure: I work at Daily. You could use the daily-js call object to handle some of the WebRTC complexity. We built a React + daily-js webinar, which could be similar to the one-to-many use case you're building (here's the first blog post [LINK] and sample code [LINK] ).
Cites Squiba · 2021-08 · Stack Overflow · on “Making group video call application wrtc vs mediasoup”
I am making a group video call nodejs application. I come to know about two npm packages wrtc - from this video [this video] mediasoup the mediasoup is very large 167 mb ( [LINK] ) compared to wrtc 82kb ( [LINK] ). I want to know why there is such a large difference which is better suitable for the purpose of developing a group meeting webapp.
Cites Matej Ukmar · 2021-12 · Stack Overflow · on “WebRTC few to many broadcast server, how to resolve high CPU usage?”
I created a "few to many" WebRTC server using NodeJS package wrtc , that broadcasts a presenter (or a few presenters) to hundreds of viewers (ideally). My implementation was inspired by this video, [this video] , where the server simply adds audio and video tracks of the presenter RTCPeerConnection to all of the viewer's RTCPeerConnection instances. The implementation actually works fine, except I'm experiencing high CPU consumption. Even just single presenter and viewer take up more than 30% CPU, with 2 presenters CPU consumption increases to about 90%. I assume the underlying C++ WebRTC library must be decoding and re-encoding video streams (once the track gets passed from the presenter's RTCPe