- From: Harald Alvestrand via GitHub <sysbot+gh@w3.org>
- Date: Tue, 20 Feb 2018 16:39:04 +0000
- To: public-webrtc-logs@w3.org
I don't know if the value 2.5 is magical, arrived at by experience, or just a guess. Do we have statistics that show that this occurs often when what we are looking for happens (slide change, scene change?), and rarely in other cases? Concur with @henbos worry about stats at the beginning of the call - I would tend to define "average" as a decaying average over the frames encountered with modulus 50 - that is, average(p) = average(p-1) * 0.98 + size(p)*0.02. but 50 is another random number. What did you use when experimenting? -- GitHub Notification of comment by alvestrand Please view or discuss this issue at https://github.com/w3c/webrtc-stats/issues/330#issuecomment-367037813 using your GitHub account
Received on Tuesday, 20 February 2018 16:39:39 UTC