[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [AVT] SVC ad-hoc meeting , February 19 remind +1-678-7955235
Ye-Kui,
You were missing an important point in your email sent on 02/07/2008:
CL-DON does not support fragmentation units of the non-interleaved mode. We
can have a long discussion whether fragmentation is useful or not, but if
using FEC or retransmission for pre-encoded content you want to benefit from
having application layer fragmentation instead of IP fragmentation. Further
you want to allow this without the burden of supporting the interleaved mode
for the (RFC3984 backward compatible) base layer session.
And there is still the following point sent by myself as reply on Colin?s
email from 02/10/2008:
Actually 8.1.1. describes the usage of session multiplexing for a layered
video codec with RFC3550. Besides the requirement of using the SDP
dependency signaling instead of standard SDP RFC4566, 8.1.1. does only rely
on mechanisms defined in RFC3550. Thus, I still believe that this mode
should be always allowed, whatever the exact gain using any other method may
be.
And another point we haven?t discussed:
Isn?t CL-DON relaxing the differentiation between non-interleaved mode and
interleaved mode packetization modes as defined in RFC3984? I mean CL-DON
would also allow for interleaving within one session with the definitions
present in the current draft. Thus, I think we should connect the CL-DON to
at least using interleaved mode in one of the other sessions.
Thomas Schierl
-------------------
Fraunhofer HHI
From: avt-bounces at ietf.org [mailto:avt-bounces at ietf.org] On Behalf Of
Ye-Kui.Wang at nokia.com
Sent: Tuesday, February 19, 2008 11:59 AM
To: avt-bounces at ietf.org; avt at ietf.org
Subject: Re: [AVT] SVC ad-hoc meeting , February 19 remind +1-678-7955235
It would be great if we can make some progress in the biggest issue, i.e.
cross-layer decoding order recovery, which has made the progress of the
draft very slow, because other parts, in particular the signaling part,
which need major work, are heavily dependent on this.
I therefore would like the issue to be discussed with the following context:
1) my long email sent on Feb. 7th summarizing the pros and cons of the two
approaches existing in the draft, as well as the discussion initiated by
Mike's proposal sent out on Feb. 8th, 2) Mike's earlier proposal and the
recent discussions between Jonathan, Colin and myself for generic approaches
for cross-layer decoding order recovery.
BR, YK
________________________________________
From: avt-bounces at ietf.org [mailto:avt-bounces at ietf.org]
Sent: Tuesday, February 19, 2008 12:18 PM
To: avt at ietf.org
Subject: Re: [AVT] SVC ad-hoc meeting , February 19 remind +1-678-7955235
I?ve had a quick scan of the e-mails on the list since I sent some comments
on the draft on 8th February. Of those comments, the following five perhaps
need some discussion.
1. Introduction and Scope ? should some of the text in the Scope be moved to
the introduction?
6. Timestamps across sessions. These should have different offsets, but
should they be sampled from the same clock?
11. Various bits in PACSI, such as the A bit. Can these bits be defined in a
meaningful way without specifying constraints on the contents of other
aggregation packets than the one in which the actual PACSI is present?
15. Repeated SEI messages in PASCI and the other NAL units in an aggregation
packet. These should be discouraged, but there appears to be no restriction
in H.264/AVC/SVC on repeated SEI messages.
25. In which stream should repeated SEI messages be present (and other
non-VCL NAL units). This is also open issue 9 in the draft.
Also, Alex raised the issue of whether the S and E bits and TL0PICIDX should
use the same flag rather than the X and Y flags; and Jonathan and Colin
mentioned the possibility of defining a generic RTP extension for
cross-session decoding.
Best regards
Mike
Mike Nilsson
Multimedia Analysis and Coding
BT Group Chief Technology Office
___________________________
Sirius House (B54-MH), Room 92
Adastral Park, Martlesham Heath, Ipswich, IP5 3RE, UK
Tel: +44 1473 645413
Mobile: +44 7917 025433
Fax: +44 1908 862365
Email: mike.nilsson at bt.com
________________________________________
From: avt-bounces at ietf.org [mailto:avt-bounces at ietf.org] On Behalf Of Even,
Roni
Sent: 19 February 2008 08:27
To: avt at ietf.org
Subject: [AVT] SVC ad-hoc meeting , February 19 remind +1-678-7955235
Hi,
The next conference call: February 19, at 7:00 AM PST, 17:00 Israel/Finland,
16:00 Germany, 15:00 UK
The conference bridge number is +1-678-7955235
Regards
Roni.
----
Visit us at
GSMA Mobile World Congress / Barcelona, Spain / 11 - 14 February 2008 / Broadcast Mobile Convergence Forum / Booth 2B82
http://www.mobileworldcongress.com/homepage.htm
OFC 2008 / San Diego, California, USA / 26 - 28 February 2008 / Booth 411/415
http://www.ofcnfoec.org
_______________________________________________
Audio/Video Transport Working Group
avt at ietf.org
http://www.ietf.org/mailman/listinfo/avt