Megaconference

[Nicholas Thompson <>]

I want to add a side-note here:

Historically, the processing power required for both video and audio was high.  So from a programming perspective with those older software clients, the video took precedence where the audio portion was easily side-stepped by using a headphones.  I think the general idea was that, why would you blare a conversation out loud if you are in an open environment, so the processing power for echo-cancelation gave way to the video processing.  However in the last few years since the software client presence has grown, so has the processing power of machines on the market.  Which is why now, many of the software application do a much better job at this.  

Coupling this with the proprietary vs. non-proprietary:  Many of the issues that stemmed from the proprietary/non-propritary space had more to do with ease of user use and network functionality.  Standards (h323/SIP) can do firewall traversal, but from a vendor standpoint it is much easier to write all the custom code to handle this (making it proprietary) allowing for a much smoother user-experience without the worry of network ports, registration..etc.  Along with this custom code, also has to do with media transport which a vendor would do in a non-standard way, and as a result, could write more efficient code to allow for their software to do the EC well, versus the standards-based applications which need to follow protocol (no pun intended).  This all of course plays into Ben's statement of business model, resource…etc.

I also agree with Ben's statement about room systems.  Trying to use a software application to audio cancel multiple microphones and speakers is very difficult and resource intensive.  With hardware-based room systems, this is almost always eliminated by using the dedicated microphone-in's and audio-outs'.  The underlying DSP's and programming use the microphone-in's and audio-outs specially for audio cancelation.  Using Aux-In/Out's can result in non-echo cancelation depending on the how the system was designed.  It further becomes complicated when the system is using the local microphone to project into speakers into the local room as well as for the far-site.  With proper planning and design, the room system should have equal, if not better, EC than software clients since much of it is hardware based processing.

Nick~

Nicholas Thompson            60b

Operations Manager           2009 Millikin Road

Internet2 Commons     Columbus, OH 43210

Internet2

Work: 614-247-8066

Video/h.323 (GDS): 00111090720060

http://commons.internet2.edu

*----------------------*

From: Paul Bonnett <>
Date: Monday, December 3, 2012 9:27 AM
To: 'Rusty Presley' <>, Ben Fineman <>, Christian Helft <>
Cc: Simon Horne <>, "" <>
Subject: RE: [Megacon] Message from Platinum sponsor (instructions and notes)

We find all of the echo cancellation built into desktop VC systems can be easily confused so we always recommend people purchase a dedicated piece of kit like a Clear One, or Phoenix or one of these new excellent bargain devices:

http://www.amazon.com/Jabra-7410-109-Speakerphone-Microsoft-Communicator/dp/B0051V16L0/ref=sr_1_8?ie=UTF8&qid=1354544622&sr=8-8&keywords=jabra+echo+cancelling+microphone

 

They take a lot of processing off the PC as well so they can improve the quality of the whole experience.

 

Regards,

 

Paul Bonnett

Videoconferencing Technical Coordinator

Direct line: +44 1235 822 332

www.ja.net/janetvc

Janet, the UK’s research and education network.

 

From: [] On Behalf Of Rusty Presley
Sent: 03 December 2012 14:16
To: Ben Fineman; Christian Helft
Cc: Simon Horne;
Subject: RE: [Megacon] Message from Platinum sponsor (instructions and notes)

 

ScopiaDesktop (h.323 and sip software client) does an excellent job of echo cancellation, especially if the end user does not place their microphone right in front of their speakers!  A good bit of echo can be solved with a little end user education, especially from a computer based solution.

 

Excellent discussion on the topic!

 

Thanks – Rusty

 

==================

Rusty Presley

IT Specialist VI

Alabama Cooperative Extension System/Auburn University

67 Extension Hall, Auburn University, AL 36849

(334) 844-3504 Office

(334) 750-8863 Cell

 

 

 

From: [] On Behalf Of Ben Fineman
Sent: Monday, December 03, 2012 7:55 AM
To: Christian Helft
Cc: Simon Horne;
Subject: Re: [Megacon] Message from Platinum sponsor (instructions and notes)

 

Hi Christian,

 

I'm not sure how to interpret your question but I can take it one of two ways, both of which are interesting points:

 

1) Why do the proprietary (non-H323) softclients do echo cancellation well, while standards-based clients struggle?

 

I would say this: echo cancellation is hard, and developing good echo cancellation is expensive. There is no good open source solution available, and from what I gather it must be prohibitively expensive to license the good solutions that are available. This means that solutions with good echo cancellation are either very mature ones that have been around for a while, or solutions that come from big players (with big $$$). In both cases this often means proprietary solutions - business models often hinge on the proliferation of a walled garden approach (such as Skype), while solutions that "do the right thing" and embrace interoperability have historically lacked the same level of financial backing. And of course from a technical standpoint, if you only have to support connections with your own client and proprietary format, many things (including echo cancellation) get easier. 

 

2) Why do softclients have better internal echo cancellation than room systems?

 

I disagree with this premise, but maybe this isn't what you meant. It is the case the room deployments often benefit from external echo cancellation technology, while you never see soft clients plugging into a big 1RU Biamp. The simple reason is that soft clients generally have to deal with a single microphone and speaker pair. This is much easier to echo cancel than, say, a conference room with eight ceiling speakers and eight table mics. Room codecs work very well with one (or even two, sometimes three) mics. But once you start to scale up to many mics and speakers you really need a dedicated external echo canceler.  

 

Let me know if I missed the point completely. 

 

Regards,

Ben

 

/*-----------------------
Benjamin J. Fineman

Program Manager, Video Services
Internet2

(email/video/chat)

http://www.internet2.edu

734.352.4975 (desk)
734.417.0811 (mobile)

-----------------------*/

 

On Dec 1, 2012, at 1:33 AM, Christian Helft <> wrote:

Hi,

Le 30 nov. 2012 à 23:15, "Simon Horne" <> a écrit :

2.       Recommend Headphones or echo cancelling speakerphone. (for everyone’s sanity)

Can you comment on the fact that any reasonably recent videoconferencing "softphone" (you name it: VSee, zoom.us, Jabber video, Messages (for Mac), Vid, etc. (even Skype)) doesn't require external echo cancellation any more, while there seems to be a malediction with truly H323 compliant systems?

Regards,

Christian.
--
Christian Helft LAL - IN2P3 - CNRS
LAL Bat 200 BP 34 91898 Orsay CEDEX France
+33 1 64 46 84 12
Mobile +33 6 32 90 97 09
http://www.lal.in2p3.fr

 

Janet(UK) is a trading name of Jisc Collections and Janet Limited, a
not-for-profit company which is registered in England under No. 2881024
and whose Registered Office is at Lumen House, Library Avenue,
Harwell Oxford, Didcot, Oxfordshire. OX11 0SG. VAT No. 614944238