Re: [AVT] draft-ietf-avt-rtp-svc-03.txt and RFC 3984

[Randell Jesup <rjesup at wgate.com>]

<Ye-Kui.Wang at nokia.com> writes:
>The backward compatibility issue regarding the single NAL unit mode has
>been well taken and addressed for non-layered-multicast use cases.

Excellent.

>However, for layered-multicast use cases, where an video/H264 RTP stream
>and one more more video/H264-SVC RTP streamss are conveyed in different
>RTP sessions, use of single NAL unit mode for the video/H264 RTP stream
>is very ineffcient because of the overhead caused by encapsulating each
>prefix NAL unit in its own packet. Prefix NAL unit is of small size, and
>it must be present for each coded slice NAL unit in the video/H264
>stream. 

That makes sense (given the prefix-NAL requirement).  Obviously in mode 1
they can be aggregated into a STAP.

>Moreover, are there really commercial multicast-capable receivers that
>only support the Single NAL Unit Mode? For which applications and
>network environments are those? The multicast-capable systems that we
>are aware of, namely DVB-H and MBMS, require support of non-interleaved
>mode - and even interleaved mode in case of MBMS.

Multicast-capable mode-0 only?  Not that I know of at the moment though
it's possible, though unlikely, that one or two are; as I mentioned
(indirectly) I was mostly talking to point-to-point use (videophones).

-- 
Randell Jesup, Worldgate (developers of the Ojo videophone), ex-Amiga OS team
rjesup at wgate.com
"The fetters imposed on liberty at home have ever been forged out of the weapons
provided for defence against real, pretended, or imaginary dangers from abroad."
		- James Madison, 4th US president (1751-1836)


_______________________________________________
Audio/Video Transport Working Group
avt at ietf.org
https://www1.ietf.org/mailman/listinfo/avt