If we consider encoders and packetizers as transformative elements, a number of metrics only make sense in one end of the element; for instance, there is no obvious volume in a packet that has audio in it; the volume property (if it exists!) belongs to the stream at the input to the encoder. The obvious exception is MediaStreamTracks, which have 2 ends of the same form, but with different metrics, because it is a transformative element (scaling, volume sampling, muting). I'd like to look at only one end for each object, so that we push the input volume up to being a property of the source object (and thus, by definition, becomes the same value for all tracks from the same source). -- GitHub Notification of comment by alvestrand Please view or discuss this issue at https://github.com/w3c/webrtc-stats/issues/353#issuecomment-400668407 using your GitHub accountReceived on Wednesday, 27 June 2018 13:19:44 UTC
This archive was generated by hypermail 2.4.0 : Saturday, 6 May 2023 21:19:44 UTC