IETF
rmcat@jabber.ietf.org
Friday, November 6, 2015< ^ >
Magnus Westerlund has set the subject to: Coupled congestion for RTP Media
Room Configuration
Room Occupants

GMT+0
[00:00:25] John Leslie joins the room
[00:02:36] Sergio Mena_2 joins the room
[00:03:34] Randell Jesup_3255 joins the room
[00:05:04] <Randell Jesup_3255> I don't seem to be able to join 'rmcat' on jabber.ietf.org from pidgin....
[00:05:30] <Sergio Mena> I'm typing this from my pidgin client
[00:05:38] <Sergio Mena> So it works for me
[00:07:31] Ray Atarashi joins the room
[00:08:07] Martin Stiemerling joins the room
[00:08:11] Ray Atarashi leaves the room
[00:08:19] <Meetecho> note to chairs: Randell in the virtual queue
[00:09:36] Ray Atarashi joins the room
[00:09:54] Ray Atarashi leaves the room
[00:11:32] <Martin Stiemerling> seen, noted to the chairs
[00:11:40] <Martin Stiemerling> sorry for not seeing this earlier
[00:12:10] David Hayes joins the room
[00:19:27] jesup joins the room
[00:28:41] <jesup> line is cut, but a few comments: even if RMCAT yields a single algorithm, having a standardized feedback message will help future work to improve on the algorithm, or to develop or deploy future algorithms
[00:29:31] <Martin Stiemerling> Randall, you will need to send this to the mailing list, as the lines are cut.
[00:31:22] <jesup> Also: it appears there are multiple implementers dealing with the bandwidth and frequency for RTCP traffic if the control is on the sender side.  Almost all algorithms *can* be moved to run entirely on one side or on the other.  Running it on the receiver side can dramatically reduce the feedback required (as Michael is saying)
[00:32:25] <jesup> We can build a generic arrival-time feedback message, or we can build a generic results-of-receiver-CC message
[00:33:08] <jesup> I'll post to the list
[00:43:52] <jesup> Sent email to the list about feedback packets
[00:45:20] <Sergio Mena_2> Yes. curremt traces in the source code amount to 10-12s only
[00:45:33] <Sergio Mena_2> they should be 1 minute long at least
[00:47:07] <jesup> hum adopt
[00:47:11] <Sergio Mena_2> (Y)
[00:47:32] <Sergio Mena_2> Stefan
[00:47:59] <Sergio Mena_2> If, when using the source code you see your approach has something we can integrate in syncodecs, don't hesistate to let us know
[01:04:30] <jesup> in the queue...
[01:04:51] Simon Romano_5828 joins the room
[01:05:47] <Meetecho> jesup: waiting for the chairs to give us the go
[01:09:31] Sergio Mena leaves the room
[01:09:56] Sergio Mena joins the room
[01:45:19] <jesup> This is the reason I was suggesting receiver-based CC -- especially as RTCP are often not directly included in congestion control; there's an implicit assumption that it's always "low"
[01:58:05] Simon Romano_5828 leaves the room
[01:58:33] Sergio Mena leaves the room
[01:58:48] Sergio Mena_2 leaves the room
[02:02:02] <Meetecho> Randell Jesup in virtual queue
[02:05:48] Simone Romano joins the room
[02:08:50] <Meetecho> to people in the room, are the chairs aware Randell's waiting in line?
[02:25:38] David Hayes leaves the room
[02:33:51] <jesup> stream cut out.... did it get dropped due to running over?
[02:33:59] <Meetecho> no
[02:34:19] <Meetecho> I can still hear it in our monitor
[02:34:22] <jesup> should I try rejoining?
[02:34:28] <Meetecho> yes please
[02:34:30] <Meetecho> sorry about that
[02:34:38] Martin Stiemerling leaves the room
[02:34:47] <jesup> "Too late"
[02:35:03] <Meetecho> d'oh
[02:35:16] <Meetecho> you can use the mp3 stream
[02:35:19] <jesup> maybe you tap off earlier
[02:35:20] <Meetecho> that's still on
[02:35:24] <Meetecho> no no
[02:35:29] <Meetecho> we only limit access to sessions
[02:35:33] <Meetecho> but streams remain on
[02:35:33] Randell Jesup_3255 leaves the room
[02:35:39] <Meetecho> session's over though
[02:36:28] <jesup> and they're done anyways.  Thanks in any case; they were about done.  Worked well; a few dropouts for a few seconds
[02:36:55] <Meetecho> sorry! you'll be able to catch up with the recordings :)
[02:37:00] <jesup> And remote queue/mic-lines seriously rocks :-) (WebRTC FTW!)
[02:37:23] <Meetecho> yeah! :D
[02:37:51] <Meetecho> you were its most passionate user, we should award you with a medal or something :)
[02:38:54] <jesup> Everyone should use it.  And being able to talk back and forth with the answerer/chairs/etc helps a ton
[02:39:26] <jesup> Plus you can make comments that you could never get typed in fast enough (and never get read out understandably)
[02:40:00] <Meetecho> yes, we enabled this in 4 rooms out of 8 here in yokohama
[02:40:08] <Meetecho> hopefully in the future it will be on for all of them
[02:40:19] <jesup> :-)
[02:41:17] <Meetecho> cheers, now it's time for the Meetecho team to see what Japan looks like!
[02:41:24] <Meetecho> see you in Buenos Aires!
[02:42:34] Meetecho leaves the room
[02:42:39] jesup leaves the room
[05:47:06] dtbphayes joins the room
[05:55:02] Simone Romano leaves the room
[06:00:04] dtbphayes leaves the room
[06:03:53] John Leslie leaves the room
[15:25:26] Sergio Mena joins the room
[15:25:37] Randell Jesup leaves the room
[15:25:37] Simon Romano leaves the room
[15:30:21] Sergio Mena leaves the room