NAB 2019, AV1, and the Future of Video Codecs

Subscribe To Download This Insight

2Q 2019 | IN-5492

Leading up to NAB 2019 there was mounting hype around AOMedia Video 1 (AV1), including announcements from Samsung and Intel—Samsung joined the Alliance for Open Media, and Intel along with Netflix announced Scalable Video Technology for AV1 (SVT-AV1). This was in contrast to NAB 2018 where most companies had ascribed to a wait-and-see viewpoint for AV1. Although general consensus was cautiously optimistic, many suspected it would be difficult for AV1 to supplant MPEG in the premium video space. Before discussing what changed in one year, it’s important to discuss how the market arrived at this potential crossroads.

Registered users can unlock up to five pieces of premium content each month.

Log in or register to unlock this Insight.

 

NAB 2019: Less Talk around AV1 Than Expected, but More Optimism

NEWS


Leading up to NAB 2019 there was mounting hype around AOMedia Video 1 (AV1), including announcements from Samsung and Intel—Samsung joined the Alliance for Open Media, and Intel along with Netflix announced Scalable Video Technology for AV1 (SVT-AV1). This was in contrast to NAB 2018 where most companies had ascribed to a wait-and-see viewpoint for AV1. Although general consensus was cautiously optimistic, many suspected it would be difficult for AV1 to supplant MPEG in the premium video space. Before discussing what changed in one year, it’s important to discuss how the market arrived at this potential crossroads.   

The Alliance for Open Media (AOMedia)—launched by Google, Amazon, Cisco, Microsoft, Mozilla, and Netflix—now lists other heavy hitters among its “founding members,” including Apple, ARM, Facebook, IBM, Intel, NVIDIA, and now Samsung. These companies comprise the most significant players within the mobile, computing, and social media markets. They have led some to believe that AV1 is poised to take the codec crown from MPEG and succeed H.264 as the next industry leader. In addition, it is currently (and is expected to remain) unencumbered by royalties, among its greatest assets when comparing it to High Efficiency Video Coding (HEVC) and its three patent pools (MPEG LA, HEVC Advance, Velos Media). The royalty issues has led to delayed deployments of HEVC, helping to explain why H.264/AVC remains the leading codec. This isn’t to say that HEVC isn’t being deployed; it remains tightly coupled with Ultra High Definition (UHD) launches and is of course supported by a large number of end-user devices, but its rate of diffusion has been slowed due to the issue with patent pools. 

Jumping ahead to 2019, there were less overt displays of support for AV1 than expected. Feedback provided on the show floor felt at times reminiscent of 2018, although more companies expressed higher levels of optimism overall. There was affirmation that some operators have held off on HEVC investments (due to uncertainties with royalties) and are heavily considering AV1. In these cases, the arrival of hardware decoders (e.g., for set-top boxes, Broadcom is a promoter-member) is viewed as a key turning point for AV1, while others still believe AV1 will best serve the streaming markets, leaving HEVC to continue to gain traction through UHD deployments. Lingering concerns such as encoding time and cost were cited, although AV1 has made significant strides in the past year to address many of these early issues.     

So what does this mean for the market?

Still HVEC's Market to Lose, but It's Doing a Decent Job of Giving AV1 an Open Window of Opportunity

IMPACT


Unless the patent pool and royalty situation resolves itself, it appears HEVC will remain tied to UHD. For AV1 this is certainly an opportunity not only to compete head-to-head with HEVC but to capture share from Advanced Video Coding (AVC) as well. This is a prime reason AV1 is often pitted against AVC instead of HEVC. Once devices begin to support AV1 decoding, the floodgates could open and the market could see a domino effect in support of the new codec. For a time AV1 seemed somewhat reminiscent of VP9—albeit with stronger backers—and this might have remained the case if not for HEVC’s ongoing patent pool problem.

If AV1 becomes the top codec, this could change the motivating factors behind investments in video compression technology. Without a patent pool and royalties, one could argue the industry might see less innovation and investment into video compression technology and codecs. While the financial incentive (tied to licensing) would certainly evaporate, the move to royalty-free codecs would not alter the other motivating factors to improve video compression technology. End users will continue to watch increasing amounts of digital content, and the costs associated with serving and storing this content will certainly provide incentive to keep development moving forward, particularly as the market makes transitions to higher resolutions and more immersive content.

Video platform companies and services that have not yet bet on AV1 might seem at a disadvantage, but with the move to the cloud and virtual services, the transition to AV1 for these companies is not expected to be an issue. In this regard many companies are simply agnostic when it comes to codecs and will support whatever their clients are demanding.

The Industry Should Make the Move to AV1

RECOMMENDATIONS


Assuming the AV1 codec can perform at least on par with HEVC (regarding cost for encoding and overall codec performance), the shift to the royalty-free option—on paper, at least—would seem to be the best plan of action. The patent pool issues have hampered HEVC, and the industry needs to move past this. AV1 is shaping up to be the answer to do so; however, when or if AV1 does become the leading codec, it is possible that companies may issue patent claims to certain aspects of the codec that could bring the industry back to where it is now. Although most believe that this is not a likely outcome, some still cite this as a potential threat to AV1’s future. While early adopters often reap rewards when these bets are correct, the wait-and-see approach is not a poor option in this case, as the market will not transition overnight, leaving plenty of time for the market to add AV1 to its solutions.