[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [AVT] Request for feedback on draft-valin-celt-rtp-profile-01.txt
Hello,
o Jean-Marc Valin [05/11/09 22:02]:
Hi,
We have just updated the CELT RTP profile to take into account the
suggestions we got:
http://www.ietf.org/internet-drafts/draft-valin-celt-rtp-profile-01.txt
Among the changes, we have replaced b:AS= with a bitrate= parameter and
only one frame size is now allowed for each dynamic payload. This should
solve the problem with early media at the expense of using possibly many
payload types. Anything we forgot to address or any new issues?
I realize this has been discussed before and may still be an unsolved
question, but I still find the description of multiple frames per RTP
packet in the 01 draft confusing, especially the discrepancy between the
diagram in "3. RTP payload format for CELT" and "3.3. Multiple CELT
frames in a RTP packet". The first one looks like there would be no
length bytes, just the plain payload to follow the RTP header. Because
the length bytes then only appear in the section specific to multiple
frames in a packet, one might think that they are transmitted only in
this case. Only later, the Low-Overhead mode is defined, which when used
may omit the length bytes.
My suggestion is to remove the first diagram (anyway it is clear that
the payload follows the RTP header), or add the length bytes to it to
make clear that this is the usual mode of packetization.
In 5.2. Low-Overhead Mode, the text says:
"One the a=fmtp: parameter low-
overhead: is defined and contains a single frame size, followed by a
'/', followed by a comma-separated list of the number of bytes per
frame for each stream defined in the channel mapping."
but the example gives:
a=fmtp:97 low-overhead=256/1/86,86,43,30;mapping=2,2,1,1/
L,R,LR,RR,C,MLFE/ITU-RBS.775-1
What is the extra /1 supposed to mean in 256/1/86,86,43,30 ?
Also, a simpler example (e.g. for mono and stereo streams) could be
illustrative.
Thanks
Stefan Sayer