Somehow it doesn’t seem quite appropriate to assess the networking technology year in terms of its ‘eventfulness’. Nonetheless, it is hard to resist describing 2014 as a highly eventful year for audio networking – possibly even the one in which we witnessed some form of tipping point away from traditional, point-to-point connectivity.
Of course, many long-serving technologies – notably MADI – remain in heavy usage. But with Audinate’s Dante continuing to record new licensees at a formidable rate (in excess of 180 at time of writing), ALC NetworX’s Ravenna starting to achieve market traction, and the AES67 networking standard providing further lustre to Layer 3-based networking, the impression of a proper breakthrough is no mirage.
What’s more, the upward trajectory looks set to continue in 2015. A forthcoming control standard may complement AES67, while the AVnu Alliance – flag-waver for the Audio/Video Bridging (AVB) movement – will benefit from a certified audio endpoint reference platform, a newly created Industrial market segment, and a brace of additional members (namely, Belden, General Electric and National Instruments).
But what about the situation at the ground level? Virtually everyone involved with networking talks about upping the educational effort to reach more potential end-users – and few would deny the desirability of that. However, the early adopters among the integrator community have already passed this point and are actively thinking about how to deploy specific networking technologies in their fixed install projects.
Finding a new way to approach the networking issue isn’t easy, but Installation decided that it might be insightful to address three specific scenarios and invite leading manufacturers to identify the initial considerations to be taken into account when plotting a networked audio system. These, then, are the ‘headline’ priorities that should be heeded; sadly, there isn’t scope here to go in-depth into multiple design options.
The first scenario revolves around a mid-size conference facility requiring a comprehensive audio networking solution that is fully integrated and easy to use. Picture one large (1,000-seat) capacity room and a smaller (500-seat) space, along with respective control rooms and back office areas. There should also be opportunity to expand the configuration relatively painlessly in the future.
A similar degree of flexibility is called for in our second scenario, which involves a multipurpose performance venue. Holding a capacity crowd of 1,500, the venue requires the ease of (re)configuration necessary to host everything from stand-up comedy sets to full band performances. The networking scheme should cover all console positions as well as control and rack room areas.
The final scenario centres upon what is apparently a significant growth area for networking vendors: a college of higher education. Coverage of long distances is a particular priority for this environment, with the audio networking design needing to accommodate an auditorium (approximate capacity 800-1,000), two radio studios and a couple of classrooms. With the site expected to undergo a development programme in the mid-term future, it should also be possible to achieve easy expansion of the system.
1) Conference facility
Andreas Hildebrand, senior product manager at ALC NetworX, highlights some of the key factors that would have to be taken into account when designing a networked system in a conference facility. “The first thing I would look at is whether my conference system is something that can live in a local network infrastructure,” he says. “If it is bound to the borders of a venue, you could potentially run a Layer 2 environment, which gives the option of using a technology like AVB. But if it needs to run across network boundaries, into several local area network segments, then you need to look at a Layer 3 solution. Performance-wise you wouldn’t experience much difference between a Layer 2 and a Layer 3-based solution.”
But, he continues, “current and future product availability would be another key issue here”. While Ravenna itself is a Layer 3-based technology successfully introduced in certain application areas (mostly professional broadcasting and high-end recording), Hildebrand admits that there isn’t much in the way of available, Ravenna-supporting product suitable for conference applications at this time. “But it is certainly a market we are aware of and looking at,” he confirms.
Maintaining low latency would be another preference for conference applications, says Audinate CEO Lee Ellison, who highlights the ability to put together a Dante-based conference facility system using a wide variety of vendors’ equipment. “In terms of the conference market, [there is Dante deployment] for products from Symetrix, BSS, Biamp, QSC, Shure, Audio-Technica and others. There is also a wide selection of I/O boxes and suchlike to make it easier for the installer to connect, use and change the system,” says Ellison.
The general Harman philosophy, explains Harman International senior manager for systems design Adam Holladay, is that “we don’t want to force the customer down a particular route. The emphasis, therefore, is on offering as many solutions as possible in order to meet the needs of a particular project – [not least] because in an installed sound system, certainly a larger one, we find that the IT network is often determined well in advance of the audio networking protocol.”
With that caveat in place, Holladay says that he would probably recommend a Dante-based deployment of Harman equipment for conference applications. “In a conference venue, the Ethernet or network infrastructure will probably have been determined by an IT division beforehand. This would basically rule out using AVB as the AVB solution we offer is only going to function on AVB-compatible switches, and at this time there are not many of those,” he says.
Suggesting a possible workflow, Holladay says that Dante could be used to network between Soundcraft consoles and BSS Soundweb signal processing. “For ease of use, I would then suggest our BLU Link protocol to daisy-chain between BSS and Crown amplifiers; in essence to turn the rack room into a large matrix taking the audio off the network. This means that you can use the network for system-wide distribution, but then for processing audio from the processor box to the amplifier box, there is no need for audio from the network because they are right next to each other in the rack,” says Holladay.
2) Performance venue
Again pinpointing some of the main requirements for a venue of this kind, Hildebrand says: “It would be good if the selected networking solution could offer some interoperability schemes. For example, this would mean it is possible to extract some of the individual streams for an OB van in the event that a performance is to be broadcast.”
This need for interoperability would probably lead the consultant and venue operator in the direction of a Layer 3-based solution. “While you could use some sort of bridging or gateway technology from the mixing desk to produce an output that is suitable for the OB set-up, the more natural approach would be to use a Layer 3 approach right away.”
The specific advantages of implementing Ravenna in this case, suggests Hildebrand, would be “a very high flexibility in setting up the streaming formats and adapting to the latency requirements”.
UK-based audio interface specialist Focusrite is a long-term licensee of Dante technology. Invited to consider the roadmap for an installation of this kind, Focusrite product manager Will Hoult says that as a manufacturer of high-quality mic preamps “we would be looking to add the number of boxes required to satisfy the channel count, then depending on what audio workstation is used we would be able to connect to it. So for example, if it’s a Pro Tools HD system we can bridge directly in to it with a 32-channel RedNet 5 interface.”
Any such venue will inevitably include a mixing console as part of the network, “and we provide bridging interfaces that allow people to use pretty much any console, whether it has a network connection or not. It’s worth noting that one of the drivers behind developing the AES and MADI bridges that we now offer is to be able to connect equipment that is not endowed with a network port, to a Dante network.”
While network design is inevitably impacted by the maximum Ethernet cable length of 100m, Hoult points to the availability of fibre modules that allow the user to cover much greater distances. “For example, you can get a 40km fibre module capable of achieving a single mode fibre connection up to 40km long, which allows you to [bring the network] to a variety of different areas,” he says.
3) College of higher education
In a large, potentially cross-campus deployment as might frequently be found in an HE facility installation, a Layer 3-based solution may again be preferable. “You would probably go Layer 3 as you would need to route audio, video and data across network boundaries,” says Hildebrand. “You rarely have a facility like that sitting on a single, big local network segment, so you would need routing capabilities, and that means you need Layer 3.”
He continues: “If wide area connections are also part of the setup, the networking solution needs to be capable to offer high flexibility in the choice of operating parameters on individual routes to different destinations in order to deliver satisfying performance with the lowest possible latency, matching the individual jitter characteristics of the various WAN routes.”
Holladay confirms that, once again, “the chances in a college of higher education are that the IT specification is not going to be under the audio designer’s control. Since it is highly unlikely that the IT department would have chosen a [Layer 2-based] AVB-compatible switch as there are relatively few of them, that means a [Layer 3 design] would be the preferred option.”
In a college of higher education, remarks Hoult, the ability to deliver audio quickly and efficiently where it is required is an obvious priority. Once again, he suggests, a Dante-based deployment can come into its own in this environment. “Often you would be looking to move the audio equipment around the facility on something like a 12U rack, and in that regard the ease of use of Dante makes that a real possibility,” he says. “It’s based on the flexible location of devices and their identity, so it remembers which device audio was being received from previously. It might now be in a completely different location, but audio would still be received properly, and that makes a mobile rack-based approach – something that would be ideal for an HE college – a realistic possibility.”
Time of transition
Anecodtal evidence aside, it is quite difficult to ascertain precisely how widely the newer technologies are being used in real-world applications. But the experience of interface, conversion and routing technology products developer DirectOut does underline the current transitional state.
The company is currently completing work on its first Ravenna-based product – “we are finalising that now and expect to be able to announce more details shortly” – but DirectOut CTO Stephan Flock confirms that MADI conversion technology remains the bedrock of its current offer.
“It’s a slightly odd situation to be talking about the benefits of MADI at the same time we are also pursuing the road of audio over IP,” admits Flock. “But with MADI, you have defined point-to-point connectivity and very low latency. There is also the fact that it is a standard with a weight of history behind it, and it is very open with regards to selecting equipment and putting together a system design. There is a sense of reassurance that you are going to have a compatible way of working, and that can still be a challenge with networked solutions.” And that, in a nutshell, is why MADI will doubtless remain an integral part of the landscape for many years to come.
But as the above responses indicate, Dante, in particular, is now making dramatic inroads into all manner of install applications. Dependable, Layer 3-based networking is bringing unheralded-of flexibility to the built environment – so expect to see it applied widely to many more than the three scenarios outlined in this feature.
By David Davies, SVG Europe managing editor and freelance pro-AV writer