Re: [AVT] Synchronization question abt RFC 2250

["Greg Herlein" <gherlein at herlein.com>]

It's a good question though.  I've seen several different
implementations of RTP encapsulation of MPEG2 transport streams.  One
in particular tries to interpolate the relative PTS of the data in the
RTP packet (this was a commercial implementation).  I'm interested in
what is considered to be 'standard' in this respect.

As a separate question I'd like to ask the old-timers what the
original thinking around this was.  It 'breaks' the concept of layered
separation - the idea that a lower level of protocol is relatively
insulated from the details of the payload.  This need to stamp
PTS-derived RTP timestamps is one argument I get from some vendors
*against* RTP (scaling costs) - they argue that TS-over-UDP is vastly
simpler.  Without the PTS-derived RTP timestamp RTP is not all that
much more complicated than using raw UDP.

<note:  I already grok the advantages of RTP so please no flames - I
was stating the arguments of others>

Greg

On Sun, Aug 17, 2008 at 7:21 PM, Ross Finlayson <finlayson at live555.com> wrote:
> According to RFC 3550, RTP timestamp is defined as sample time of each
> unit. It also reflacts the presentation time.
>
> But for some of payload type (e.g. mpeg1/mpeg2 vedio, RFC2250 ),  RTP
> timestamp is not the presentation timestamp any more.
>
> RFC 2250, Page 4:
>
> The RTP timestamp will not be passed to the MPEG
>    decoder.  This use of the timestamp is somewhat different than
>    normally is the case in RTP, in that it is not considered to be the
>    media display or presentation timestamp. The primary purposes of the
>    RTP timestamp will be to estimate and reduce any network-induced
>    jitter and to synchronize relative time drift between the transmitter
>
>    and receiver.
>
> That paragraph refers only to the encapsulation of MPEG System and Transport
> Streams - *not* Elementary Streams - over RTP.  For MPEG-1 or 2 Audio or
> Video Elementary Streams - described later in RFC 2250 - the RTP timestamp
> *does* correspond to presentation time, as usual.
>
> --
>
> Ross Finlayson
> Live Networks, Inc.
> http://www.live555.com/
> _______________________________________________
> Audio/Video Transport Working Group
> avt at ietf.org
> https://www.ietf.org/mailman/listinfo/avt
>
>
_______________________________________________
Audio/Video Transport Working Group
avt at ietf.org
https://www.ietf.org/mailman/listinfo/avt