With the global movement towards IP, Rajat Nigam, Group CTO of Network 18 Media, tells Vinita Bhatia that the non–availability of standards needs to be addressed urgently
The adoption of IP, owing to its fast-changing face, is similar to any other new technology standard or format that has kept India’s broadcasting, production, and media and entertainment industry on tenterhooks. This adoption resonates with the global movement towards IP and convergence into the Internet of Things (IoT) scenario. Besides the ease of managing IP systems, the increasing demand for higher bit rates and the lowering cost of IT-based elements are driving adoption. Interestingly, IP has gained momentum only over the past couple of years despite IP signal flow being around for more than a decade.
Globally, the manufacturing, BFSI, and retail sectors have adopted IP in a big way. Can the broadcasting industry too embrace IP absolutely, including on real-time video production? Yes, but when that happens remains to be seen.
Across industries, today’s work life is all about being connected and accessing devices remotely, all of which is being driven by IP protocol. Increasingly, this technology is driving operations in the engineering and manufacturing industries. The broadcast industry is no exception and has been adapting workflows around IP. The signal processing happens from the very first step of baseband signal generation, however, and that is what makes the campaign for full IP adoption look tricky for the video content industry.
IP-based real-time production has been around for a few years now but with solutions that offer a logistical point of view. For instance, multi-camera productions using an IsoCam output, transported in the IP mode to remotely located PCRs, to facilitate live and complex productions has been done for a while.
Similarly, controlling camera operation besides output has been possible for at least a decade. Sports productions have now been widely adopting IP-based production techniques, which entail a common PCR; while the venue of the matches shifts, the crew remains consistent. Therefore, IP-based productions have been adopted at different levels and at different organisations without needing the extra-push from a video-over-IP campaign.
The Right Signals
IP has been the mode of signal processing in the broadcast technology chain over different segments over more than a decade. While IP-based production has been a reality for a few years now, IP signals have been the mode of signal transportation for contribution as well as transmission for decades. The advantages of IP are efficiency in managing, monitoring, and measuring the signals but the real benefit lies in the signal-carrying capacity of the cable with the IP signal. This is where the significance of video over IP becomes pronounced as video resolution is changing generations. Various algorithms and standards are being developed to make the signal transport more efficient on bandwidth consumption and resilience.
A comparison is usually made between a standard definition interface (SDI) routing switcher and an IP switch. Even though the two devices primarily have the same objective—input and output management—they have two different used cases. The SDI router is for single-port signal flow with consistent time delays and performance while the IP switch is meant for managing the input and output ports for multiple streams. The two have a defined purpose and design specifications with proven used cases and applications. With IoT set to become a big part of our lifestyle in the time to come and IP switches with throughput of 25Gbps to 40Gbps and 50Gbps on the anvil, there is a case for getting video transported over the IP network.
Migrating to Video over IP
As the adoption of IP while processing video over the production value chain has been done already, I feel a call for migration of video standards from SDI mode to IP completely is a bit premature. Fundamentally, the business rationale–with the apparent hurry behind this call–itself lacks confidence, particularly for traditional TV operations that are currently operating on standard definition (SD) or even in high definition (HD). There is no sure advantage for these operations on migrating to IP.
News channels and sports productions, which operate live productions, have a downside of adopting IP standards over traditional SDI as the baseband signal flow imparts relatively more flexibility in routing and bypassing of video sources for a high level of availability.
Another factor that is being advocated in favour of video over IP campaign is physical facility optimisation. Though the IP-converted signal has a lower attenuation index on the distance, baseband audio-video streams were travelling efficiently over the required length in a facility even during the analogue days by use of HF equalisers. The tech fraternity observed the advent of the embedded signal in SDI mode as a boon that helped to reduce the size of the cable bunch in a facility and eased the architecture.
The most significant hurdle for IP adoption in a cost-conscious country such as India is the availability of skilled IT manpower. The question is whether organisations that have been in business for long have the conviction to replace the existing tech people who have been managing the facility and delivering the business needs for years. Getting additional IT professionals to manage the complete production and transmission chain would enhance manpower costs without any possibility of revenue upside.
IP and SDI can work together in harmony by adopting a hybrid workflow. In fact, hybrid workflows suit most optimally in broadcasting operations based on the benefits of video over IP and video as SDI. The campaign to completely migrate the facility to be IP or IP enabled that too at a cost with a compromised chain design is beyond tech conviction. A hybrid system chain is, therefore, the only possibility at this stage. Controlling costs is important for media businesses and technology budgets are always challenged. A mature facility would not have the conviction to upgrade the facility to IP and, therefore, the hybrid environment is likely to be around for at least the next five years. Hence, future-built chassis for station backbone equipment such as routers, which can allow the mix and match of IP and SDI input/output boards, could make sense but has a cost implication.
All in all, it may not be out of place to state that the call for ‘SDI must die’ is untimely.
There is a fundamental difference between how media flows in SDI and IP. In the case of SDI, the information data is sequential and unidirectional; in IP, media is wrapped with ancillary information over core data and sent as packets while being bidirectional. The call for across-the-board adoption of IP is premature as the protocols and standards are far from getting aligned. Various industry associations are working on reaching compatibility and defining the standardisations but are understood to be not yet ratified by any authorised body, including the Society of Motion Picture & Television Engineers (SMPTE). Audio, though a part of TR-03 protocol, remains a concern for production design and workflow. Acceptance and application of the common timing signal as PTP is also to be concluded even when the concerns on synchronisation errors (lip sync/video switching jerk/monitoring synch) hovers in the minds of technologists. The TR-03 protocol, in principle, is an encouraging effort by the VSF (Video Services Forum) but the conviction in the next steps of IP workflow adoption is a bit low. A lot of work has happened in the recent past towards facilitating the switching time sync for IP sources using the same cable to provide clock reference (PTP) and source to sync standards used to achieve seamless operational obviating multicast signal-switching errors. Multicast source switching issues have thus been addressed as a solution but not as standards in a reasonably big production facility.
Indian broadcast channels have another level of dependence in converting into an IP-based facility as, currently, the industry-adopted audience measurement system—Broadcast Audience Research Council (BARC)—is not yet IP enabled.
Technology solutions entailing video over IP have an immediate advantage on resources and medium bandwidth and, hence, come with an inherent upside. Digital and IT companies have been speaking loudly about the IP era but most of them have restricted their scope of presence to networking and consumption platforms and devices, with the broadcast manufacturers acting at interface points thereby questioning the need for a native IP facility. The IP ecosystem allows more efficiency and capability with the basic building block of cables and switches being bidirectional and catering to multiple streams simultaneously. While that is a huge advantage, by adapting the same and stacking more and more in a single cable and single switch, we are looking at putting all the eggs in the same basket.
The adoption of video over IP for the entire value chain is challenged at this point in time by primarily non–availability of standards and a lack of skilled IT manpower.
So, in my view, an IP ecosystem for video content production and distribution comes with system and process efficiency but is a distant proposition for the industry to migrate to fully.