Re: [rtcweb] Prioritization

[Harald Alvestrand <[e-mail]>]

On 04/25/2014 07:06 PM, Matthew Kaufman (SKYPE) wrote:
> Harald Alvestrand:
>> Prioritization is a different matter.
>>
>> What I was trying to encapsulate was:
>>
>> - Higher priority means more packets get sent
>> - "More" has a numeric value, so behaviour is predictable (that's where the
>> x2 figure came from)
>> - Small packet flows get more packets than large-packet flows (counting
>> bytes not packets)
>> - Lower priority flows can't be starved completely
>>
>> This text captures that. I'd like to continue to capture that.
> The above is simply wrong, so the text shouldn't capture it at all.
>
> If I say that audio is more important than video, and I try to send 100kbps of audio and 3 Mbps of video over a 101kbps channel, how much audio should I drop at send time to let video through?
>
> Matthew Kaufman
Is either stream adaptive?

If I send high quality OPUS (100 Kbits/sec) and VP8 in HD (3 Mbits/sec), 
the suggested result would likely be OPUS scaled down to ~60 Kbits/sec 
and VP8 scaled down to ~30 Kbits/sec.

The latter is probably useful for recognizing that there was an 
intention to send a picture.
This MAY be a fine result, depending on the application's requirements.