IETF
rmcat@jabber.ietf.org
Thursday, November 17, 2016< ^ >
Magnus Westerlund has set the subject to: Coupled congestion for RTP Media
Room Configuration
Room Occupants

GMT+0
[06:05:21] Meetecho joins the room
[06:10:45] Jonathan Lennox joins the room
[06:15:14] Lorenzo Miniero joins the room
[06:15:35] Mahalingam Mani joins the room
[06:16:19] Xiaoqing Zhu joins the room
[06:16:40] Michael Ramalho joins the room
[06:16:56] csp joins the room
[06:17:17] Michael Ramalho leaves the room
[06:18:13] Anna Brunstrom joins the room
[06:18:26] Sergio Mena joins the room
[06:20:19] Simon Romano joins the room
[06:20:29] Michael Ramalho joins the room
[06:21:56] Randell Jesup joins the room
[06:23:03] Simon Pietro Romano joins the room
[06:23:24] Ashish Batajoo joins the room
[06:24:08] Bernard Aboba joins the room
[06:25:25] gorry fairhurst joins the room
[06:25:33] Ken Sakamoto joins the room
[06:26:23] jholland joins the room
[06:26:34] <jholland> jabber scribe=Jake Holland
[06:28:29] <jholland> Zaheduzzaman sarker speaking
[06:34:17] <jholland> mirja speaking
[06:34:48] <jholland> michael welzl
[06:35:50] <jholland> mirja: "but then you don't resolve any dependencies"
[06:36:14] <jholland> colin: the proposal is literally that we take these 2 paragraphs and move them to another document
[06:36:50] <jholland> (mirja is showing zaheduzzaman how it's generalized on his laptop)
[06:38:14] <jholland> (I think this is xiaoqin speaking remotely)
[06:39:16] gorry fairhurst leaves the room
[06:39:21] <jholland> welzl speaking
[06:41:04] <jholland> (zaheduzzaman asked the what is the concrete proposal question)
[06:41:10] gorry fairhurst joins the room
[06:43:03] <jholland> harald alvestrand speaking
[06:43:14] herver joins the room
[06:43:28] <jholland> mirja speaking
[06:44:14] <jholland> michael welzl
[06:44:45] <jholland> zahed
[06:46:19] <jholland> harald
[06:46:29] <jholland> michael
[06:50:21] Ashish Batajoo leaves the room
[06:57:37] m h joins the room
[06:58:30] varun singh joins the room
[07:09:41] m h leaves the room
[07:11:57] varun singh leaves the room
[07:13:30] Stefan Holmer joins the room
[07:15:15] Varun Singh joins the room
[07:15:32] Stefan Holmer leaves the room
[07:16:42] Simon Pietro Romano leaves the room
[07:16:47] Varun Singh leaves the room
[07:16:55] <Randell Jesup> I can see and hear her.  I saw a temporary "Network error, you may need to log in again" message
[07:17:11] Lorenzo Miniero leaves the room
[07:17:11] <Michael Ramalho> Same here ...
[07:17:17] <csp> I see the same: we can hear her well, though
[07:17:20] Simon Pietro Romano joins the room
[07:17:24] Lorenzo Miniero joins the room
[07:17:36] <csp> meetecho?
[07:17:51] <Meetecho> csp: we still see and hear Xiaoqing
[07:18:04] Stefan Holmer joins the room
[07:18:06] <csp> we list video in the room, but have audio
[07:18:07] <Michael Ramalho> +1
[07:18:09] <Meetecho> we're coming to reload the queue in the room
[07:18:12] Varun Singh joins the room
[07:18:12] <jholland> on the screen in the room we cannot see her.
[07:18:13] Ingemar Johansson joins the room
[07:18:17] <csp> thanks
[07:19:37] Bernard Aboba leaves the room
[07:20:03] <Varun Singh> meetecho resumed, so i can see the slides again.
[07:20:20] <Meetecho> Varun Singh: (y)
[07:20:28] <Ingemar Johansson> Does this mean that the warping is disabled when no packet loss detected ?
[07:21:21] <Ingemar Johansson> OK, thanks
[07:38:04] <Varun Singh> Is someone else hearing feedback from the floor mic?
[07:38:20] <Jonathan Lennox> Yes, it's ringing in the room.
[07:38:59] <Varun Singh> ok
[07:41:00] <jholland> meetecho, do you guys run the sound? it would be helpful to turn it down, i expect.
[07:41:34] <Meetecho> we don't but we can call the A/V team to look into that
[07:42:56] <Randell Jesup> 80 bytes, plus IP/UDP overhead of IIRC 28 bytes, and if encrypted (required in WebRTC) 10 bytes of SRTCP tag data
[07:43:01] <jholland> much better, thanks.
[07:43:11] <Meetecho> (y)
[07:43:30] <jholland> randell, is that a question for the mic?
[07:43:50] <Randell Jesup> I may say it directly in a bit
[07:43:53] <jholland> k
[07:46:00] <Varun Singh> This feedback does not just apply to per packet/frame feedback, however, can be send in longer feedback intervals.
[07:47:11] <Varun Singh> thanks
[07:56:38] <Varun Singh> There are two things I want to make sure it is covered:
[07:57:23] <Varun Singh> 1. burst losses are a reality, so 0xffff might occur in a burst because of that.
[07:58:14] <Varun Singh> 2. A frame on average (as per our observations) in the >500 kbps tend to be 20-50 packets
[07:58:44] <gorry fairhurst> Gorry->MIchael R: I may have missed this, but I wanted to check if this method could potentially hide the presence of a CE-Mark (specifically receiving a CE-marked packet and not reporting seems like something we may wish to worry about).
[08:00:40] <gorry fairhurst> P.S. please could you ask at the Mic - I don't have audio here. :-)
[08:01:36] <jholland> i am in line, yes. after varun…
[08:02:03] Stefan Holmer leaves the room
[08:03:33] <Xiaoqing Zhu> i have the same question as Randell:  the traces I just presented has ~5KB per frame at 1Mbps
[08:03:53] <gorry fairhurst> Thanks for realying - I can follow up after that to check what is happening at end of meeting.
[08:04:20] <Randell Jesup> hear hear to the shout-out
[08:08:20] <Varun Singh> +1 to Mo
[08:08:22] <Randell Jesup> agreed, it does need to handle bursts
[08:08:53] <Randell Jesup> but that wasn't what he said... probably was what he meant to say though :-)
[08:10:22] <Varun Singh> Early morning here. not on to my first coffee yet.Did mean that >500 kbps there are spikes of 20-50 packets.
[08:10:33] <Varun Singh> I am happy to share logs from callstats.io
[08:11:26] <Varun Singh> The ATO is relative to report time.
[08:23:33] <jholland> off-mic from jonathan: why not?
[08:24:59] <Ingemar Johansson> I had the same understanding of trr-int , .i.e one can use it to control the transmission of compound RTCP
[08:25:26] <jholland> you want that mirrored? the timing was off, maybe due to delay
[08:25:46] <jholland> and for context: that's the same understanding as jonathan?
[08:26:18] <Ingemar Johansson> Yes, I avoid going to the mike
[08:26:25] <Ingemar Johansson> please
[08:26:29] Bernard Aboba joins the room
[08:28:34] m h joins the room
[08:30:13] Sandra Murphy joins the room
[08:32:38] <Varun Singh> We have to note that the applications might want to send other feedback messages, NACKs, FIRs or similar
[08:32:56] <Varun Singh> no mic needed. just leaving it as a note here
[08:33:00] <jholland> k
[08:38:19] <Varun Singh> off-mic: RTT measurements incase someone needs them. ATO gives inter-arrival times.
[08:39:52] <gorry fairhurst> +1 that there are CC tradeoffs with reduced reporting interval - however codec ability to change is important ;-) (don't need to say at Mic!)
[08:41:22] <Bernard Aboba> mic:  the other things to consider here is that loss and delay is not the only thing needed for audio:  Opus has FEC, so info on burst loss is critical for adaptation.
[08:44:45] <Randell Jesup> thanks colin
[08:50:43] Ingemar Johansson leaves the room
[08:51:08] m h leaves the room
[08:51:09] <Jonathan Lennox> Meetecho: video has frozen
[08:51:23] <Jonathan Lennox> Now gone blue-frame
[08:51:29] <Jonathan Lennox> Now back, n/m
[08:51:31] <Meetecho> Jonathan Lennox: just unfrozen
[08:52:06] <gorry fairhurst> <Video K here - may be a local problem>
[08:53:07] <Sergio Mena> Thanks
[08:53:10] <Sergio Mena> Colin
[08:53:59] Mahalingam Mani leaves the room
[08:54:02] Jonathan Lennox leaves the room
[08:54:06] Michael Ramalho leaves the room
[08:54:27] Lorenzo Miniero leaves the room
[08:54:37] Xiaoqing Zhu leaves the room
[08:54:46] Sergio Mena leaves the room
[08:54:46] Bernard Aboba leaves the room
[08:54:46] Sandra Murphy leaves the room
[08:54:46] Varun Singh leaves the room
[08:54:46] Randell Jesup leaves the room
[08:54:46] Simon Pietro Romano leaves the room
[08:54:46] Ken Sakamoto leaves the room
[08:54:46] Anna Brunstrom leaves the room
[08:54:47] gorry fairhurst leaves the room
[08:55:05] Meetecho leaves the room
[08:56:51] csp leaves the room
[08:58:14] Simon Romano leaves the room
[08:58:28] herver leaves the room
[08:58:50] jholland leaves the room
[09:22:58] csp joins the room
[09:26:44] csp leaves the room
[10:01:05] jholland joins the room
[10:18:29] jholland leaves the room: Disconnected: closed
[14:08:13] jholland joins the room
[14:09:26] jholland leaves the room
[14:12:01] jholland joins the room
[15:22:28] csp joins the room
[15:33:26] Simon Romano joins the room
[15:54:45] Simon Romano leaves the room
[22:09:18] jholland leaves the room
[23:14:26] Simon Romano joins the room
[23:32:21] Simon Romano leaves the room: Disconnected: Replaced by new connection
[23:32:21] Simon Romano joins the room
[23:40:47] Simon Romano leaves the room
[23:46:35] Simon Romano joins the room
[23:49:07] Simon Romano leaves the room
[23:53:18] Simon Romano joins the room
[23:59:55] Simon Romano leaves the room
Powered by ejabberd - robust, scalable and extensible XMPP server Powered by Erlang Valid XHTML 1.0 Transitional Valid CSS!