HEVC Royalty Stacking and Uncertainty Threaten VVC Adoption

By Craig Thompson
August 13, 2021

“For VVC to capture market share among cellular device manufacturers, VVC’s royalty rates will have to be very attractive compared to the rates for AVC and AV1.”

The Versatile Video Coding standard (VVC), finalized in 2020, is now entering a fragmented, multi-codec market. However, VVC’s adoption is uncertain in the face of competitive video solutions that are subject to lower or no royalties. VVC owes thanks for this to the excessive royalties and licensing uncertainties that continue to plague VVC’s predecessor, HEVC.

Over the past five years, the formation of multiple pools for HEVC has led to licensing inefficiencies and royalty stacking that have hampered HEVC’s adoption. This has prompted the video industry to develop competitive, lower cost solutions, such as the Alliance for Open Media’s Advance Video 1 codec (AV1) and MPEG’s Essential Video Codec (EVC) and Low Complexity Enhancement Video Coding (LCEVC) standards. These video solutions, together with the already existing AVC and HEVC standards, provide a lot of choices to device manufacturers, streaming platforms, and content owners. VVC’s chances of success are further weakened by the dramatic decreases in broadband download, shown below, and video storage costs, which negate any gains in compression efficiency generated by VVC for most uses.

Economic Report

Unified Patents recently produced an independent economic study on VVC’s licensing value as a follow up to its 2019 study on the licensing value of HEVC. The report provides the economic analysis for long-run effective aggregate decoder rates for VVC over a 5-year period of $0.10 for streaming devices, $0.08 for connected TVs, and $0.05 mobile terminals when VVC’s performance is compared to HEVC. These long-run effective aggregate rates as shown below are more than 50% less than the corresponding rates for AVC.

Summary of adjustment factors and estimated VVC FRAND rates (5 – 10 year periods), using HEVC as performance benchmark

The report employs a hybrid top-down methodology using MPEG-LA’s AVC long-run effective royalty rates for three usage scenarios as a benchmark for fair, reasonable and non-discriminatory (FRAND) rates, and adjusts those rates with a factor derived from a comparison of the effective cost savings generated by the compression efficiencies of VVC over its predecessor HEVC against the effective costs savings generated by the compression efficiencies of AVC over its predecessor MPEG-2. The comparison between the compression efficiencies of VVC over HEVC and of AVC over MPEG-2 was used to justify a value for VVC that was comparable to the value of AVC as established by MPEG-LA’s pool.

HEVC Pool Rates are Un-FRANDly

Licensing by a single pool is generally viewed to lead to procompetitive effects, licensing efficiencies, and FRAND licensing rates. The implementation success of MPEG-LA’s AVC pool has been pointed out by courts in the United States and in Europe as being compliant with the FRAND obligations of those standard essential patent (SEP) holders involved with the development of the AVC standard.

But when licensing is conducted by several competing pools, the results can lead to royalty stacking and as such cease to be compliant with FRAND. Unified Patents’ report shows this to be the case with the licensing of HEVC. The multiple pools that arose to license HEVC SEPs were compelled to compete with each other for licensors by offering better returns on licensors’ SEPs. As is shown below, this competition has led to the long-run average aggregate royalty stack for HEVC to exceed the same royalty stack for AVC by more than 1,000%. Such a difference cannot be justified by HEVC’s compression improvements over AVC, which amount to only a 50% advantage for HEVC on a straightforward comparison. HEVC’s compression advantage over AVC disappears when the falling costs of bandwidth are accounted for.

HEVC’s royalty stacking problem is worsened by the fact that nearly 10% of the SEP families licensed in total between the Advance HEVC and MPEG-LA HEVC pools are licensed through both pools at the same time. While a reimbursement mechanism exists to correct the compounding effect of this double licensing, it is not known how much of the royalty stack is actually corrected. When adjusting for the full reimbursement from the Advance HEVC pool for patent families licensed twice, the long-run average aggregate royalty stack for HEVC exceeds the same royalty stack for AVC by more than 900%.

Finally, the rates offered by the HEVC pools are only aspirational and thus have very little to do with the actual value offered by the licensed SEPs. This becomes evident when one considers that MPEG-LA did not lower its HEVC pool rate when over 200 SEP families from Samsung, NTT Docomo, ETRI, and KBS were removed from the scope of its license at the end of 2020.

Multi-Codec Threat

The report outlines the factors threatening VVC’s adoption and how VVC compares to some of its competitors. Those factors include compression efficiency, coding complexity, release timing, access to new markets, and adoption by big players. Below is a summary of VVC’s competitiveness with respect to AV1 and EVC.

VVC holds an edge in terms of compression efficiency over competing codecs as it is capable of compressing a video by as much as 40% over HEVC to achieve the same quality playback. This comes of course at the cost of coding complexity, which will require at least initially hardware updates and hence delay VVC’s adoption. This compression superiority could be eroded by LCEVC, which is a software enhancement layer designed to be used in combination with a capable codec like AV1. Such a combination promises to outperform VVC.

VVC’s most serious threat is AV1, which has not wasted its three-year head start over VVC. Currently, AV1 is supported by the majority of the top 5 streaming platforms, browsers, cellular device manufacturers, and TV manufacturers. It will take VVC likely a few years before it can attract such support.

VVC is also entering a market where its use cases are under scrutiny. VVC excels in 8K video compression and is thus well-suited for broadcast TV. Unfortunately, 8K TV ownership is forecast to reach only 3.9% of North American HD TV households in 2023 and broadcast TV is on the decline. There are also low industry expectations that VVC will be widely adopted for use in cellular devices as the benefits of HD video are not discernable on small screen devices. For VVC to capture market share among cellular device manufacturers, VVC’s royalty rates will have to be very attractive compared to the rates for AVC and AV1.

Unified Patents’ VVC study does examine AV1’s potential effect on VVC’s royalty rate. The calculation in the study uses the same method discussed earlier but instead of calculating the cost savings generated by VVC’s compression efficiency over HEVC, the study calculates the cost savings generated by VVC’s compression efficiency over AV1. The chart below shows that a successful AV1 codec could lower VVC’s royalty rates by as much as 50% compared to the situation where AV1 did not exist.

Options for Staying Competitive

VVC caps almost three decades of video codec development. Today, VVC’s compression capabilities answer the technical requirements of 8K video distribution and high-definition broadcast TV. Unfortunately, though, VVC’s success is questionable because of the fragmented licensing market and unreasonably high royalties that have plagued HEVC’s adoption. These licensing challenges have given rise to the development of competitive video coding solutions that come with a much lower or even zero royalty burden. If VVC is to compete with these solutions or even with its predecessors, Unified Patents’ study shows that the royalty rates for VVC will need to be significantly less than the FRAND rates for AVC. If VVC’s royalty rates are set higher than the rates calculated in the study, it is possible that device manufacturers and content distributors will continue to adopt AV1 as the mainstay video coding solution. In such case, VVC could be relegated to marginal use cases.

 

The Author

Craig Thompson

Craig Thompson is the General Manager and COO for Unified Consulting. He has decades of experience in the IP field focused on strategy, patent and technology licensing, and IP-driven acquisitions. Craig co-founded TnT IP, LLC, a global IP consulting partnership, and headed up the IP Group at Alcatel Lucent and Bell Labs and was on the Senior Leadership Team of Nokia’s Mobile Phones Group with responsibility for the Group’s IP and legal matters. Craig started his legal career as an associate and later as a partner with Finland’s largest law firm Roschier where he practiced IP, technology, energy, data protection, competition, communications, and M&A law for 10 years. Craig is a law graduate of the University of Helsinki (Masters of Law, 1994) and of the Columbia University Law School (LLM, 1997).

Warning & Disclaimer: The pages, articles and comments on IPWatchdog.com do not constitute legal advice, nor do they create any attorney-client relationship. The articles published express the personal opinion and views of the author as of the time of publication and should not be attributed to the author’s employer, clients or the sponsors of IPWatchdog.com. Read more.

Discuss this

There are currently No Comments comments. Join the discussion.

Post a Comment

Respectfully add to the discussion.

Name *
Email *
Website