> On Aug 24, 2016, at 6:59 PM, Jonathan Lennox <[email protected]> wrote:
> 
> 
>> On Aug 24, 2016, at 6:35 PM, Timothy B. Terriberry <[email protected]> wrote:
>> 
>> Jonathan Lennox wrote:
>>> One thing I would like to see in the document (probably in the error 
>>> resilience section, 3.2.5) is an explicit requirement to support 
>>> independently-decodable subframes that can be MTU-targeted by an encoder.  
>>> This is very important for error resilience for RTP-packetized streams.  
>>> (See RFC 2736, to which an informative reference probably wouldn’t be out 
>>> of order.)
>> 
>> In general I support this idea, but I'm a little wary of writing this 
>> requirement in a way that specifies a solution (independently-decodable 
>> subframes) instead of the problem it solves. Any ideas for actual text we 
>> could use?
> 
> I’m not sure how best to phrase it, maybe something like:
> 
> The codec should support mechanisms such that individual packets sent in 
> common network protocols can be decoded even when other packets are lost.
> 
> But it’s not clear to me that a) it’s clear what this text means if you 
> haven’t read this thread, or b) whether the text seems to imply that you have 
> to be able to decode successfully even if you’re missing a reference frame, 
> which clearly won’t be the case.
> 
> Any wordsmithing suggestions?

I am informed that the term I’m looking for (meaning when you can’t decode a 
packet if you’re missing a previous one) is “bitstream dependency”.

So perhaps" “The codec should support mechanisms that allow individual packets 
to be sent in common network protocols without bitstream dependency between the 
packets.”  Thoughts?
_______________________________________________
video-codec mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/video-codec

Reply via email to