Replies: 1 comment 4 replies
-
I'm not sure what you mean by delay, do you mean stream latency, or do you mean startup time? Overall, comparing a Javascript WebRTC implementation with mulitple hops to a native WebRTC stream is of course going to provide some additional stream latency overall, however, the code is typically with 1 second of native WebRTC in my testing, and stream startup time is generally 2-3 seconds total, vs about 1-2 seconds for native WebRTC. If you are talking about startup time, well, we'd need more data on exactly what your are comparing. If you are comparing running a script, which includes starting Node, initialization Javascript and all the dependencies, authenticating to the API, create the stream control channel, starting the WebRTC stream, firing up ffmeg, etc. to an app that already has all of this done. |
Beta Was this translation helpful? Give feedback.
-
If I compare the output from the Ring Android app and custom code written with this library, there is at least a 5-10 second delay in the video received via
camera.streamVideo()
. Is this expected?I tried the stream-example and also tried a custom UDP version with ffplay:
ffplay udp://127.0.0.1:5455
Edit: I am referring to the delay in the stream and not the time taken to run the script or to initially fetch and render the stream.
In other words, once the stream is up and running, the delay between the content shown in the Ring app and the content picked up by the example code is significant in my case - in the 5-10s range.
Beta Was this translation helpful? Give feedback.
All reactions