After diverting video transmission to the ground OTT is now pushing it back up into the air with growing wireless transmission
The last year has witnessed unprecedented growth in OTT across the whole video chain from production through contribution to distribution, driven not just by the Covid-19 pandemic but also intersection of key technological and societal developments.
This virtuous circle has illuminated innovations across the field in CDNs and streaming, but also exposed deficiencies, or perhaps over optimistic assumptions of current capabilities. The latter applies particularly to 5G, which will be transformative for video but with much of the promise yet to be delivered as networks roll out and more advanced features such as network slicing become more widely available.
The pandemic has acted largely to accelerate trends already in train, whether in terms of consumption, production or distribution. On the consumption side, the lack of public cinema encouraged online viewing of content from major sites such as Netflix, Amazon Prime and Disney +, with growing amounts of binge viewing that can best be satisfied by online subscription or advertising supported VoD services. On the distribution side, the continuing improvement in streaming quality over mobile networks especially, even with 4G, has encouraged OTT viewing on the move. This will only increase as mobile services improve further and start to fill in for patchy fixed broadband networks in many developing countries, as well as remote areas in more affluent countries, delivering video services via Fixed Wireless Access (FWA). That is where the mobile network acts as a direct replacement for fixed broadband, capable of delivering IPTV as well as OTT, with WiFi still figuring for onward delivery to connected devices within a home.
Expanding Remote Production
There has also been a huge surge in remote production, and that will continue even as the pandemic eases, because it offers huge advantages in coverage, cost and flexibility. This is being enabled by 5G, which almost for the first time is offering good enough performance, capacity and reliability for outside broadcasting from the field. While dedicated fibre-based connections will still prevail for coverage of major events from locations such as premium sports grounds where the highest quality is essential, mobile based remote production promises to liberate a great long tail of lesser content from niche or lower tier sporting events for example. It also promises to enrich coverage at major events by enabling multiple camera angles from roving stations around the ground, as well as being able to cater for events such as golf tournaments or marathon running races that spread out over large areas and are notoriously hard to cover with consistently high quality. The ability to set up temporary remote production facilities at short notice is enabling broadcasters to cover breaking events in the field better than ever before.
Apart from the impact of mobile and 5G, the biggest single factor involved in advance of OTT on the distribution side has been the ongoing evolution of CDNs, combining connectivity with cached storage to optimize content delivery and minimize latency. CDNs are continuing to advance under pressure to meet demands for performance, availability and scalability as affordably as possible.
CDNs predate OTT but it is streaming video that is now dominating their traffic and driving growth. It is almost entirely video driving CDN growth, accounting for 72 percent of all internet traffic crossing CDNs by mid-2022, up from 56 percent at a much lower base in 2017, according to Cisco. This is reflected in sustained strong revenue growth at an average CAGR (Compound Annual Growth Rate) of 14.1% between now and 2025, according to Markets and Markets. That revenue will soar from $14.4bn in 2020 to $27.9bn in 2025 on that basis.
In technological terms, CDNs have been diversifying to support different specific applications, services and geographies, making choice harder for content owners or video service providers, which in turn has spawned an industry of CDN brokers. France’s Broadpeak is one vendor that has addressed that conundrum with its umbrellaCDN feature allowing customers to select the CDNs best suited to deliver each video request according to prespecified criteria. Each request can be determined by user location, content requested, time of day and other variables, on the basis of which delivery strategy can be adapted to optimize cost or deliver the best experience to a provider’s subscribers.
At the same time, CDNs have diversified into versions best suited to specific requirements. The likes of Netflix delivering exclusively on demand content are well served by dedicated proprietary infrastructures dovetailing with local broadband service providers. Some CDNs on the other hand are optimized for specific regions, such as Alibaba’s for China and CDNvideo for Russia.
The role of CDNs naturally varies between live and on demand content. For the later the emphasis is on judicious use of caching closer to subscribers to minimize consumption of wide area bandwidth through excessive unicast distribution of content. By storing cached copies of data around the world at strategic locations, scalability is increased and customer experience improved through smoother delivery and reduced start up lag down to five seconds or less.
Reducing Start-up Delay
Even for live streaming video, CDN caching is playing a growing role by allowing segments to be stored as they come, again reducing start-up delay. This is especially important for live sports where delays in transmission of key events such as goals ruins the experience if they have already been played out on someone else’s device nearby.
Proliferation of live video has though over 2021 underlined the harsh reality that CDNs can only go so far to reduce latency before running up against the laws of physics that determine ultimate end to end signal delay over any medium. Ultra-low latency can only be achieved locally and that is why the claim of 5G to bring delays to down 1 ms has to be set in context. That level of latency can only be achieved at short range and is really intended for deployment in engineering or manufacturing settings for wireless control of processes and UAVs (Unattended Automated Vehicles).
In those situations, the applications are local anyway, but latency can become a constraint in advanced video related applications such as VR (Virtual Reality), or more generally ER (Extended Reality). It means some gaming applications and services will have to be partitioned locally rather than available worldwide. As many gamers know, any end-to-end delay greater than about 7 ms is likely to cause motion sickness.
CDNs do have an important role to play minimizing their impact on latency through optimal design. This involves devolving computation to the edge where relevant and possible, dovetailing with 5G as it cuts delays over the local wireless loop. We have seen major CDN providers evolve their architectures and make sure that is reflected in branding, for example as an “intelligent Edge platform” in the case of Akamai.
UK commercial FTA broadcaster ITV is a strong advocate for 5G based remote production.
Expanding Immersive VR
Despite the caveat over latency, the VR field, and immersive streaming generally, started to expand rapidly from quite a low base during 2021. This was largely driven by Facebook’s launch of the Oculus VR console at prices from $299, almost an order of magnitude cheaper than existing systems costing upwards of $2000. The impact of this dramatic fall in price has initially been confined largely to the consumer sector, but there are signs of it opening up industrial applications as well, in conjunction with deployment of 5G in some cases. The latter will also boost Ultra HD and 4K video transmission, which requires a minimum of 25 Mbps and preferably above 40 Mbps at present.
On this front, advances in video codecs will also make an impact with potential to halve those bit rates for a given video quality. Unfortunately, 2021 did little to resolve the confused state of the video codec field as the MPEG side remains beset by a complex royalty picture.
Among broadcasters, H.265 (HEVC) remains the most popular codec but was debuted in 2013 and is increasingly considered inadequate in the face of Ultra HD, while suffering from ongoing patent issues with three separate pools of vendors owning IP related to the codec. VVC (H.266), has been developed as the sequel to H.265 and can reduce the bandwidth to transmit given video by around 38% compared to HEVC but with 10 times greater encoding complexity and still royalty issues.
Meanwhile, AV1 was developed by major technology players, including Amazon, Apple, ARM, Facebook, Google, Intel, Microsoft, Mozilla, Netflix, Nvidia and Samsung, under the Alliance for Open Media to sidestep the patent issues. With this weight behind it, and being almost 30% more efficient than HEVC according to tests conducted by Moscow State University, AV1 has gained a lot of traction in the streaming world, with Netflix starting to transmit titles over that codec.
At the same time though relief has come for users of legacy codecs via the Low Complexity Enhancement Video Coding (LCEVC), an ISO/IEC video coding standard developed by MPEG. This adds two layers that can be combined with an existing codec such as H.265 or its predecessor H.264 to boost efficiency. There is a base layer decodable by a hardware decoder, and an enhancement for software processing with sustainable power consumption. The enhancement layer improves compression capability and encoding/decoding complexity for existing codecs, both for live streaming and linear broadcasting. As a result, it has appeal for broadcasters with legacy codecs, such as H.265 or even H.264. On the other hand, it will gain little or no traction in the streaming world or for new deployments because it is not a stand-alone codec and is no use on its own.
Among other significant developments whose impact was felt strongly during 2021 for the first time was fragmentation of the global streaming content field, followed by consolidation around five or six major poles. Netflix has started to feel this impact as was widely predicted, as a result of being stripped of content by other major producers that have now launched their own dedicated OTT services. These include Disney with Disney+, AT&T/Warner with HBO Max and Comcast with Peacock Premium, while Apple has also gained ground with Apple TV+. Disney+ of these has made the strongest impact, recording 118 million subscribers by December 2021 just two years after its launch in November 2019, a far higher rate of growth than Netflix ever enjoyed at a comparable age, admittedly at a time when streaming was in its infancy.
This streaming fragmentation has in turn led to growing subscription fatigue among consumers and demand for some way of aggregating content again, as it used to be under Netflix to a large extent, or indeed under many pay TV packages of the past. A global survey from Apester, an interactive platform for audience engagement, found that 60.1% were fed up with having so many streaming services to choose from and only 16.5% approved of there being so many competing services.
This in turn is driving growth in pay per view and advertising as ways of monetizing content outside subscription paywalls, and that trend is expected to gather steam during 2022.
Back on the production front, a major trend during 2021 has been towards use of 5G to extend the scope of coverage as well as reach more remote locations cost effectively. In the UK, commercial broadcaster ITV has been promoting use of private 5G networks owned by broadcasters alone, and also dedicated slices of public 5G networks with guarantee capacity and performance.
The greatest and most immediate impact of 5G on this front has been in sports production, having been used for the first time at the Summer Olympics in Japan for both the opening and closing ceremonies, as well as coverage of several of the events. But this was over the Non-Stand-Alone version of 5G which is built on a 4G core, with limitations over flexibility and capacity. The second phase of 5G called Stand Alone will be more revolutionary, built on new core networks that allow dedicated slices to be allocated temporarily to broadcasters during events such as Olympic Games. Such Stand Alone 5G should be available both for the Winter Games in Beijing during February 2022, and at the end of the year for the FIFA World Cup in Qatar.
You might also like...
Our sports media COO featured in this article continues to reflect on how the D2C business opportunity drives their decisions about where content is made available, how content is created and produced for different audiences, and how the “D2C…
This is a story about the COO of a media business, that shines a light on the thinking underway at the leading edge of the media industry, where the balance shift from Linear Broadcasting to D2C Streaming is firmly…
What we’ve seen as ATSC 3.0 deploys and develops is just the tip of the NextGen TV iceberg.
Broadcasters are experimenting with many new TV business models to monetize new NextGen TV technologies.
Synamedia, the London-headquartered video services and technology company spun out of Cisco in 2018, has acquired UK content discovery firm Utelly to beef up its offering for aggregated search, navigation, and recommendation.