Hot Button Discussion
SDI in an IT World
By Michael Goldman
When he spoke with Newswatch in 2014 about the status of Serial Digital Interface (SDI) technology on an increasingly IT-centric landscape, John Hudson, Director of Strategic Technology and New Business Development for the Semtech Corporation and Chair of the SMPTE’s 32NF-40 working group on SDI interfaces and the 32NF-70 6G SDI drafting group, stated that he believed SDI would remain relevant for the foreseeable future as a crucial element of a burgeoning hybrid industry infrastructure.
At the time, Hudson believed it was “highly unlikely that there will ever be just Ethernet-based networks or just SDI-based networks for the transport of realtime base-band images [in the broadcast industry],” because different applications benefit from different capabilities and technologies.
Three years later, Hudson still feels that way, but additionally, he says the conversion to IP-based interfaces has shown if anything, that “the transition is going to take a little longer than people had originally [anticipated].”
“Also, there is increased recognition that when those systems roll out, they will be hybrid SDI/IP,” he adds. A difference [from a few years ago] is that we understand now that, for early adopters of Ultrahigh-definition [UHD], (i.e., South Korea and Japan, in particular, where they are currently building infrastructures for UHD, SDI is still a mainstream choice. That’s because IP standardization, workflows, and technologies have not fully matured to the point where they can be dependable.
“That represents a subtle shift in recognition that, while the transition to IP is moving ahead, people are a lot more realistic about the time frames.”
Much of the reason for this, Hudson suggests, has nothing to do with the growing potential and efficiency of Ethernet-based networks for many broadcast applications. Instead, it pertains to business considerations. The notion of revamping broadcast plant infrastructures from SDI into IP territory is not, necessarily, the most compelling priority for many broadcast entities right now, particularly when SDI continues to evolve at a highly efficient rate, and demands on broadcast networks are continually shifting.
“We saw 1-Gigabit Ethernet IP links widely deployed [rather quickly] into broadcast infrastructures for file transfers when [data] was quite heavily compressed or did not need realtime transfer,” he explains. “That was a cost-effective and widely used option. However, for realtime acquisition capture of non-compressed [base-band] images for live events, where it was necessary to get bits from the camera to the glass in the most direct, high-speed and cost-effective way, SDI was hard to beat, as it was specifically developed for that very purpose.
“As technologies developed, we went through 1.5-Gig SDI to 3-Gig SDI, but then, IP came along with 10-Gigabit Ethernet [GbE]. At that point, it became possible to deploy IP networks to handle the required realtime interface rates at a similar cost structure to SDI. In other words, a 10-GbE network could now be developed to transfer or manage live HDTV content in the same way with SDI, for a modest cost and complexity increase [10 GbE versus SDI].
However, when 10 GbE became available, the industry simultaneously moved to even higher image resolutions, higher bit depths, and even more bandwidth so that the cost-effectiveness of a 10-GbE network now faced that traditional problem of, “actually, what I really need is a minimum of 25-Gig Ethernet for UHD. I should probably be at 40-Gig or even 100-Gig, all of which comes with a significant cost and complexity increase. So, at the same time, SDI evolved and came up with its next-generation of interface rates.” Hudson states.
In other words, as IP arrives to meet new requirements and challenges, SDI continues to evolve simultaneously, and both continue to be relevant. Therefore, in a world where, outside of some Asian markets, there is not yet a mandate for broadcasters to transition from HDTV to UHDTV, Hudson suggests “it is difficult to see a value proposition” in accelerating the reconfiguration of broadcast plant infrastructures in the U.S. and elsewhere on any schedule that would render SDI obsolete anytime soon.
“There is a government mandate to move to 4K in Korea and Japan, but for many technical and business reasons, that has not happened in North America and much of the rest of the world,” he says. “It’s a confused environment, so I don’t know that transport technology advances are driving broadcaster business decisions about this format choice. If I’m a broadcaster and want to produce all my TV shows or send out my over-the-air broadcast in 4K, a business proposition is needed. Broadcasters are currently much more concerned with the ‘TV repack’, for example, resulting from the recent incentive spectrum auction, or the rollout of ATSC 3.0. After all, without getting 3.0 right, broadcasters may find it difficult to move to next-generation TV services. So the issue of ‘when do I start spending money on my internal infrastructure buildout’ probably comes after those other concerns for [major broadcasters].”
Consequently, Hudson emphasizes, even where North American broadcasters are currently investing in infrastructure revamping, “their main eye is on their current HDTV services, rather than necessarily building up to support the transition to UHD. After all, what is the catalyst for a broadcaster's decision to change its infrastructure? Traditionally, it has always been a [government mandate] to switch to a next-generation television service or a major quality requirement.”
Thus, while IP-related in-plant infrastructure upgrades are certainly technically possible right now, he emphasizes that for the foreseeable future, broadcast entities will need “extremely deep pockets” to even entertain the thought. And with SDI innovation still underway, more than sufficient for maintaining current HDTV services in many cases, broadcasters continue to have a vibrant option that they can sew together with incremental IP-based changes in the long run.
Hudson points to numerous SDI advances targeting next-generation television services just in the few years since he last spoke with Newswatch. He points out that standardizing interfaces for UHD now takes place in the SDI world. “In SMPTE, we have the 6-Gig and 12-Gig SDI standards [ST 2081 and ST 2082], and we also have two 10-Gig standards for UHD [ST 2036-3 and ST 2036-4]. The ITU has its version of those standards as part of its ITU-R Recommendation BT.2077-2, which extend to 24-Gig SDI. So there will be next-generation standards and speed grades of SDI.”
The next priorities from both standards and technology perspective for SDI are focused on keeping pace with ongoing improvements to the next-generation broadcast image—higher dynamic range, wider color gamut, and higher frame rates, in particular.
“These [enhancements] apply to both HD and UHD, so in both cases, the focus of both SMPTE and ITU has been on adding HDR, higher frame rates and wider color gamut capabilities to those existing SDI interfaces,” Hudson says.
“One area of interface technology transition that could offer significant benefit is the transition to optical interfaces,” he adds. “But, even there, you can’t predict there won’t be a new generation of coaxial cable that could handle longer reaches of 24-Gig. For instance, we are only just starting to see multiple coaxial cable vendors introduce optimized coaxial cable for 12-Gig. There are new generations of cable from multiple suppliers, including Belden and Canare, for example, where you can get over 100 meters at 12-Gig SDI. Indeed, even today, at 24-Gig, you can easily achieve 30-40 meters on existing coaxial cables and 75 ohm BNC connectors. And, there is always the option of optical SDI, which is abundant, even at lower data rates. That could be a good solution for higher SDI data rates going forward, and it would also be compatible with 10, 25, and 100 GbE networks, as well.
”Hudson fully expects hybrid infrastructures to be the industry’s direction for the foreseeable future, and for SDI’s traditional strengths to continue to be enhanced and increasingly useful.
SMPTE ST 2110 for IP-Based Networks Moves Ahead
TV Technology recently explained that SMPTE has approved the first standards within the SMPTE ST 2110 suite—standards for Professional Media over Managed IP Networks. SMPTE ST 2110, built on the Video Services Forum (VSF) Technical Recommendation for Transport of Uncompressed Elementary Stream Media over IP (TR-03), is designed to aid the push toward a single, interoperable, IP-based protocol for professional media work. SMPTE President Matthew Goldman noted in the article that the process of effectuating ST 2110 should “enable a broad range of media technology suppliers to move forward with manufacturing and meet the industry’s high demand for interoperable equipment, based on the new suite of standards.”
Fastest Transatlantic Internet Cable Ever
According to a recent report in TechNewsWorld, Microsoft has completed a massive undersea, transatlantic Internet cable project that is expected to provide interoperable connections across great distances at extremely high speeds—the highest-capacity subsea transatlantic cable ever. The so-called Marea undersea cable stretches 4,400 miles between Virginia Beach, Virginia, and Bilbao, Spain, some 17,000 ft below the surface of the ocean. It was a joint project between Microsoft, Facebook, and telecommunications infrastructure provider Telxius. The project finished, which finished ahead of schedule, was initially designed for completion in early 2018. The cable will reportedly allow 160 Tbits/sec connectivity and can carry far more data than existing cables in other parts of the world, according to the article.
Court Rules on Drones
A recent report in Videomaker says a Federal district court has made a significant ruling on the use of drone technology in the U.S. The case addressed the issue of whether cities and states can make regulations for drone use that are different than those imposed on the Federal level by the Federal Aviation Administration (FAA). The case, which could impact the use of drones for filmmaking purposes, among other applications, revolved around the city of Newton, Massachusetts, enacting a rule different from FAA regulations regarding how low drones can fly over a private or city-owned property. The court ruled the Federal law took precedence, which could impact other local and state laws regarding drone use going forward.