Video content encoding is largely skewed towards AVC or H.264 at present. The need for streaming high-quality video content on networks with limited bandwidth is pushing content providers to adopt new video codecs with high compression ratios.
Encoding video content with these codecs allow content providers to achieve a smaller file size without loss of video quality. However, complex value chain of video content delivery and ongoing royalty tussles amongst different camps has left no clear indication as to what the future might hold.
While media firms and broadcasters seem to favor the traditional choice for now i.e. HEVC, they’re still contemplating their next move. They have three codec options to achieve higher video compression ratios – HEVC (H.265), VP9 and AV1. A timeline of codec development and their launch dates is below:
Among the three, VP9 and AV1 are royalty-free codecs. However, HEVC (H.265), the successor of the existing standard – AVC (H.264), is a royalty-based video codec. This means that video encoders pay a hefty licensing fee to patent owners through the licensing firms.
HEVC Licensing Model and its Tumbling Future
HEVC was introduced back in 2013. It was a royalty-based codec, much like its predecessor AVC, and it promised a 50% improvement over AVC compression. Interestingly, Apple has largely been the flag bearer of the HEVC camps and it is also a key patent owner in one of the licensing pools. It provided native support in Apple devices as well as the Safari browser. Thanks to Apple, HEVC has seen higher adoption in terms of hardware support in devices. This meant that Apple devices were optimized for HEVC encoding/decoding at the chipset level.
HEVC royalty is managed by three licensing pools – MPEG LA, HEVC Advance and Velos Media. Such patent pools enable companies to acquire patent rights necessary for the HEVC standard from multiple patent holders in a single transaction as an alternative to negotiating separate licenses. Their role within the industry, often forgotten, is key to adoption of any new technology. In case of video codecs, they’ve played a key role in cementing the foundation of the existing industry standard AVC, against the previous challengers such as – On2’s TrueMotion VP3, Xiph’s Theora, Microsoft’s VC-1, and many others. Below are the key licensors of the existing HEVC licensing pools:
There was already too much confusion regarding royalty pay-outs among licensors. On top of that, licensing firms came under immense pressure when key licensees joined hands to support a royalty-free codec. This was a major setback to the video codec licensing business. Infact, the chairman of MPEG LA, Leonardo Chiariglione, himself shared his doubts in a personal blog about the crumbling licensing model and the challenges ahead for the video codec industry.
“At long last everybody realizes that the old MPEG business model is now broke, all the investments (collectively hundreds of millions USD) made by the industry for the new video codec will go up in smoke and AOM’s royalty free model will spread to other business segments as well.”
Leonardo Chiariglione, Chairman, MPEG (Moving Picture Experts Group)
(Blog Post – A Crisis, the Causes, and a Solution)
VP9 – Google’s Alternative to HEVC
When VP9 was launched, Google was fighting against the odds to push its own royalty-free codec VP9. It made total sense for the likes of Google to work on its own video codec technology, considering the scale of video consumption it was driving (Chrome, YouTube, Android) and the costs involved. Apart from being the primary consumer and biggest beneficiary of this technology, it aimed to enhance the experience on YouTube and Chrome. Google started pushing a royalty-free codec VP9, which offered compression ratios similar to that of HEVC (which was largely doubted by the industry). However, it didn’t gather much support outside its own ecosystem. Firms were quite apprehensive about VP9, as it was competing against the already set standards of the industry which are governed by key licensing firms. Later, it joined the Alliance of Open Media to supplement the development of the AV1 codec.
AOM Challenges HEVC Licensing Pools
The royalty model was shaky. Steam was building up against the royalty-based model due to the high royalties charged by licensing firms and the payment complexities. Key firms in the codec value chain joined hands to form an alliance – AOM, to develop a royalty-free AV1 codec. The list of these firms included Amazon, Cisco, Intel, NVIDIA, Google, Microsoft, Mozilla, Netflix and many others. Obviously, these firms were also the largest beneficiaries of an open royalty-free codec standard. AOM leveraged the features of existing codecs such as Google’s VP9 (for code), Mozilla’s Daala and Cisco’s Thor, and finally released AV1 in early 2018.
The shift favoring AV1 became more prominent when Apple also joined the AOM camp in earlier in 2018. Meanwhile, media owners and distributors are still running tests. Various industry reports suggest that the AV1’s performance is comparable (or at par) with HEVC. On top of that, content aggregators such as Netflix, Amazon, and Hulu (apart from Google/YouTube) are already in support of AV1.
The Future Outlook
As the industry seems inclined to pledge support for the new codec tech, the anticipated success of new royalty-free codecs has exerted more pressure on the video codec licensors. Having said that, HEVC is far more mature in terms of hardware optimizations and software support. The industry also awaits the outcome of existing patent infringement legal battles against the AV1 codec by HEVC flag bearers.
Therefore, HEVC is expected receive priority over VP9 for now. Companies will just sit and watch for a while to weigh the options. They will analyze and compare HEVC and AV1, on lines of encode/decode requirements, hardware/software support, cost, IP risk, HDR support, etc.
It may be worrisome for the industry if it takes the royalty-free path. It will certainly give much needed stability to the video codec market, but this will come at the cost of reduced technical progress, as there will no incentive for companies to develop new video compression technologies. The next two years will be key to these developments. Nevertheless, the market dynamics has given the licensing firms a chance to rethink their strategy, sort their issues and untangle the licensing model mystery.