What's Next for IP-Based Systems?
Hot Button Discussion
What’s Next for IP-Based Systems?
by Michael Goldman
In the world of broadcast technology, if there is one key development that needs to keep pushing forward even in the face of a global pandemic and an economic downturn, it would be the industry’s ongoing transition to IP-based systems. At least that’s the view of Willem Vermost, design and engineering manager for Belgium’s public broadcast company, VRT, and a longtime European Broadcast Union (EBU) veteran. Vermost, also a veteran of SMPTE's TC 32NF drafting group on Studio Video over IP (SVIP), which is responsible for writing and now evolving SMPTE ST 2110 suite of standards for Professional Media over Managed IP Networks, says the transition to IP-based systems “is probably the biggest change in the last 30-plus years in terms of the history of professional video. By moving out every different type of cable and putting everything on IP, we can achieve our dream or goal, which is to have centralized data-centers...etc etc in order to make it flexible, scalable, and shareable. We can have centralized data centers full of off-shelf commodity servers with software only applications that we can deploy instantly at the moment we need them, including mixing capabilities, editing capabilities, or other processes—anything you can think of. The icing on the cake will be if we can move it all into the Cloud to have real scalability.”
Vermost emphasizes that he is talking about professional media technology, but he also points out that in today’s quarantine-based communication world, “people are making videos and going on the air from their living rooms, and they are often doing it with pretty cheap prosumer equipment. So there is [an irony] there—professional media is going through its biggest shift in decades, and yet, there is enough quality in prosumer devices [to broadcast]. The quality is not the same, but nevertheless, this brings an exclamation mark to the vendors implementing [SMPTE ST 2110] and the underpinning synchronization delivered by the IEEE 1588 standard (PTP) with the SMPTE ST 2059 profile. In my opinion, this means the longer professional media waits to come up with all the equipment it needs concerning the various layers of an IP-media [ecosystem]—the transport, discovery, and connection layers—the more prosumer equipment will take parts of the market. That may sound negative, but I mean it in a positive way—we should all collaborate and do it faster than ever before. We should focus on the layers on top that can and will make the [most] difference—managing and scheduling all those resources in a very granular way.”
He emphasizes that there is plenty of good news in this regard—thanks to SMPTE ST 2110, “the immediate transport layer is there; discovery and connection are there; time and synch are there, and so on.” However, there remains the issue of how the product testing process—an area Vermost is deeply involved with—has been altered by the current inability to hold in-person events. And, he adds, there are also a myriad of complicated business issues involved in transitioning to IT-based facilities—issues now made even more complex on the current landscape, thanks to the economic slowdown.
On the testing front, Vermost says the Joint Task Force on Networked Media (JT-NM) was able to move ahead with its Spring 2020 testing program for vendors looking to validate how their new products fit into an IP-based production structure despite the COVID-19 emergency, even while canceling their planned live event for this spring. However, the testing process was different this time, he points out, in that it added a self-tested element, meaning vendors performed their own tests, and then JT-NM published those results in its catalogues without conducting any sort of live validation session.
There was one exception to that approach, he adds. In the case of NMOS controllers specifically, a new protocol was devised whereby vendors used a VPN connection to link directly with a lab at CBC/Radio-Canada, where a JT-NM tested team observed and verified the results.
“That remote process went very smoothly—everyone did a marvelous job,” Vermost says. “But for the vendor self-testing process, every test was neatly laid out and described—what boxes you have to check off in order to attain a pass for a particular test. Everyone was made familiar with the test plans and were able to ask questions remotely. But they had to do the tests at home with tools we proposed or anything similar that can do the same things. So, the self-testing was helpful, because those [conducting them] were aware of everything demanded or required. However, for more physical types of testing on various aspects of ST 2110, for instance, we could only rely on results that people handed in to us. Testing those results remotely is not very easy to do in most cases.”
In any case, Vermost emphasizes that that the development of an IP-based media ecosystem requires far more than just SMPTE ST 2110, which represents, in essence, the media transport layer of a much larger construct. The other layers, as mentioned earlier, include time and synchronization with SMPTE ST 2059 as PTP Profile, configuration and monitoring, discovery and connection, and security. To build a meaningful IP-based media facility, all of these layers need to be properly addressed, he says, pointing to the EBU’s so-called Technology Pyramid graphic, which illustrates these layers.
“If you look at [that graphic], you see that you need much more than just SMPTE ST 2110 to make a full system work,” he says. “The subtitle of that graphic is pretty revealing—it calls [the pyramid an illustration of] ‘the minimum user requirements to build and manage an IP-based media facility.’ I often make the analogy with an old telephone system. You have seen those pictures where ladies [operators] are sitting at switchboards, patching cords. With SMPTE ST 2110, we are more or less in that situation. We have the signal flowing through a cord, but we still need to manually patch it. And that is why you need things like discovery and connection. If you plug a camera into a network, you need to know if it is a camera, and if it produces a picture, and if that picture is SD or HD or UHD, and if there is any audio flowing with it, among other things. And, of course, there is a vision mixer or a switcher that needs to get the video stream. Therefore, it is necessary to set up a connection between the camera and the video switcher.”
Thus, the ongoing testing process and various surveys of producers and content creators and distributors have, according to Vermost, made clear three specific attributes that the industry needs to make sure are built into IP-based broadcast equipment. Those attributes are flexibility, scalability, and shareability, Vermost says. “Flexibility is about [formats],” he explains. “You will need to be able to switch from any one to another—maybe SD to HD or HD to UHD, or whatever. These formats are coming at us faster and faster. So we need flexibility in order to change. I remember back in the SDI days, I worked on changing an SD facility to HD, and such projects are cumbersome. They can take one or two years to accomplish—you need to rip out all the cabling, all the devices, all the equipment, until only the physical rack remains. But with IP, it does not care about the format—you are able to go from one to another without ripping up your facility. And then, of course, IP brings you scalability, because if there is one network on Earth that is proven its ability to scale, that network is the Internet. So that is crucial.
“Shareability is another important topic, but maybe we're not quite there with it yet. The problem is, if you move to IP, you have to move everything, not just audio and video, but all the other things you see in a studio. For example, that would include things like black burst becoming PTP [Precision Time Protocol] in an IP facility; and RS-422 [a serial data specification] would need to become something like [an Ethernet cable].” This would enable us to share resources over different studios and increase the usage. Instead of driving around and plugging the equipment in other productions this will require scheduling access management to those resources. This would also better unlock the possibility of distributed production, which we've had some experience with during the COVID-19 pandemic.
Moving forward, Vermost thinks new broadcast plants will need to fit into what he calls “a whole roadmap, a new business perspective” from the broadcaster’s point of view. And for now, with the global economy in turmoil, the business proposition of figuring out when and how to commit to reinventing or building from scratch broadcast infrastructures on an IP foundation will, in some ways, be a more vexing problem to resolve than developing and installing the actual technology itself.
Nevertheless, at press-time, Vermost estimated that throughout Europe, “about 36 broadcasters are in the process of making plans or building new facilities.” While it isn’t yet clear how many of those projects have been delayed or scaled back due to the global pandemic, he says that the general existence of such projects and strategies illustrates the maturity of the technology to move forward as the foundational core of future broadcast facilities across the globe.
“Many of those facilities were originally built in the 1960s, and most are old and [broadcasters] were thinking about refurbishing or starting from scratch anyway,” he says. “But a good example [of the decision-making process] is the new CBC/Radio-Canada broadcast center [in Montreal, a 419,000 sq. ft. facility almost entirely running on IP-based technology]. About five years ago, they had a choice to make a new facility—more SDI or should they go with IP? François Vaillant [executive director of engineering solutions at CBC/Radio-Canada] had to make a tough decision about that. If he went with [SDI], they would have a legacy system the moment they went live, or they could take a giant leap and have a flexible, future-proofed IP system. Even five years ago, they made the decision they had to go forward with IP.”
The new CBC/Radio-Canada facility began construction in 2017 and, at press-time, was slated to open some time this year, although the pandemic’s consequences have left it unclear as to the exact timing of the opening. In terms of other recent examples, Vermost points out that European broadcaster BCE Luxembourg has a brand-new 36,000-sq. ft. IP-based facility called RTL City “now rocking and rolling,” and the BBC recently added a new 155,000 sq. ft. facility called BBC Wales Headquarters in Cardiff, Wales.
As the pandemic eventually recedes, Vermost fully expects to see more IP-based broadcast facility deployments moving ahead, but he says even those will need to retain what he calls “gateways” to the longstanding SDI world because SDI continues to play a role in the industry and “isn’t about to disappear any time soon.”
In particular, live-event broadcasting continues to lag behind other forms of programming in terms of functioning well in an IP-based environment, he emphasizes, and therefore, Vermost suggests facilities of the future will need to eventually figure out a way “to merge nonlinear and linear flows. Right now, they are strictly two different things if you look at most existing facilities. We should be able to merge those two things together, but the file-based and live broadcasting are not yet coming together.”
COVID Cancels IBC
As widely reported recently, hopes that the COVID-19 related wave of cancelations of major industry events would ease up by the fall in time for IBC's 2020 scheduled September date to hold were dashed this month when the IBC announced that this year’s show would not be held or rescheduled, but rather canceled. In a Q&A explanation on the IBC site, IBC CEO Michael Crimp explained that, although the event was not scheduled until September, there was simply no way for the organization to “guarantee that we would be able to deliver a safe and valuable event” due to all the uncertainty surrounding the COVID situation. Crimp went on to explain why the event was not canceled earlier, how exhibitors have been reacting to the news, and he discussed the organization’s plans for some form of a virtual event and hopes for holding IBC 2021 as scheduled.
Oscars Could Get Postponed
Meanwhile, on the heels of the IBC cancelation, as well as announcements about important industry shows like Cinegear and Siggraph also being canceled, there are rumblings that such disruptions could linger into next awards’ season. A recent Variety report, for example, indicated that the Academy of Motion Picture Arts and Sciences is currently discussing scenarios for possibly postponing the mother of all awards shows—the Academy Awards. The event, currently set for February 28, 2021, has a good chance of being pushed back, the report suggests. Even before that report, the effective closure of the entire movie exhibition industry during the pandemic had already forced the Academy to implement rule changes for eligibility of certain films, among other things.
IATSE Seeks Expert Guidance on Disease Control
In terms of figuring out when and how to open up the entertainment industry again when the time is right, the International Alliance of Theatrical State Employees (IATSE)—one of the most important labor guilds for show business craftspeople—recently announced it would be hiring a team of three epidemiologists to consult with the union on setting up procedures for permitting industry employees to return to work safely when the time comes. IATSE International president Matthew D. Loeb announced that while the Guild wants to get its unemployed workers back on the job as soon as possible, “we need to do it right” since creative jobs such as hair, makeup, and wardrobe, among others, pose unique challenges in terms of avoiding contact and disease transmission. Therefore, the hiring of a team of experts was considered a priority. The three doctors who will consult with IATSE are David H. Wegman, M.D., M.P.H; Letitia Davis ScD, EdM; and Gregory R. Wagner, M.D.