What better way to kick off our 2021 coverage of mobile video than by picking up on a narrative that defined and shaped the video entertainment ecosystem during 2020, and was covered in depth by Wireless Watch’s sister service, Faultline, which provides weekly analysis of digital video technologies and business models? This theme was video codecs, or more specifically, patent royalties – the perpetual thorn in the side of the video streaming industry.
Eyes generally roll back into heads whenever codec royalties are mentioned in conversation – an unfortunate side-effect of some five years of division and chaos. The video codec scene has become so politically charged in recent years that it has driven the world’s greatest codec minds to the brink, namely MPEG founding father Leonardo Chiariglione, who turned his back on the codec group he helped create last year. Such volatility is indeed testament to technology that is so fundamentally integral to keeping video ticking, but at the cost of crushing value. The time is now for change.
Faultline therefore expects a serious shake up of the codec royalties scene starting in 2021 – one that has been a long time coming.
This will, in part, be triggered by the soon-to-be finalized licensing model for LCEVC (low complexity enhancement video codec). We feel like the misrepresentation of royalty-free has been discussed to death, but it is worth reiterating that while there are obviously hidden fees in any so-called royalty-free model, a completely royalty-free video codec world is simply not achievable. It is merely a pipedream for larger organizations, one that we understand is even forbidden by antitrust laws given that royalty-free everything would generate a microeconomic deadweight loss.
There is simply no business logic here. It’s almost as if the founding members of AOMedia, the group behind the AV1 codec, have forgotten the countless hours of painstaking research conducted by tens of thousands of dedicated R&D personnel dating back decades, many of whom have worked for these very founding companies – Amazon, Cisco, Google, Intel, Microsoft, Mozilla, and Netflix.
If you are Google, throwing a few hundred researchers at a problem with little-to-no ROI is fine. A few million dollars are filed under ‘miscellaneous’ in the financial reports alongside autonomous cars and the loss is forgotten about. Except that this isn’t a multi-million dollar market, it is one that underpins trillions of dollars in US GDP.
Despite this, the royalty-free mantra should not be discarded. Royalty-free advocates have their hearts in the right place, for piling up intellectual property costs on top of general-purpose devices is not sustainable long term. Services running on these devices must continue to be free (or almost free) in order for the likes of Google to continue getting rich from advertising, yet devices and platforms are the ones paying patent royalty fees, despite them not immediately benefitting from more efficient compression.
The services (apps and websites) are the beneficiaries of advancements in video codec technologies – from higher quality of experience and lower costs reflected in improved revenues.
Approaching this problem like LCEVC looks set to, although we have not yet seen the finalized licensing model, by operating a fair licensing scheme (or “cheaper than royalty free” model as V-Nova has previously described its strategy) aims to set a bar for other IP technologies. The goal is to price video codec licensing in a fair and acceptable way by charging nothing for general purpose enablement and charging very little for usage.
This should please the device makers and platform developers, chiefly browsers, who don’t have to pay for enablement and without being required to embed LCEVC, although devices and platforms might embed the technology anyway which would be boon for V-Nova. The apps and service providers, who are the ones actively benefiting from improved compression, then will supposedly only have to pay low and capped royalty fees for LCEVC. Whether or not these fees will creep up over time, or rise sharply in say a year’s time, is still unclear, as is which services would exceed the cap.
But this does not come without its risks. Companies could abuse this fair licensing model and therefore any potential to make money from LCEVC could be hampered by excuses to simply not pay for usage, because LCEVC is firstly not a standalone codec, but also as its developers are so eager to avoid any legal bloodshed that has overshadowed this industry for so long.
V-Nova will be busy figuring out these technicalities with a crack legal team as we write – walking an incredibly thin tightrope between money and uptake.