MPEG LA VVC Licensing Terms: What Do They Mean for the Streaming Industry?
This article is a follow-up to our article, MPEG LA Announces VVC Licensing Terms. I wanted to get a sense of how many of the relevant patents were included in the two pools, and some measure of how well the Media Coding Industry Forum (MC-IF) performed in its attempt to "facilitate cross-industry discussions around the non-technical aspects of deployment of media coding standards, including patent licensing."
To accomplish this, I reached out to two well-known industry authorities, intellectual property attorney Robert J.L. Moore, an attorney at Caldwell Intellectual Property Law, and Tim Pohlmann, CEO and Founder at iPlytics. Robert has been kind enough to answer IP-related questions in the past, and I've attended several webinars presented by Tim regarding various IP rights. The two caught my eye with the article they co-authored, entitled "Who is leading the VVC technology race?"
Thanks to both of them for taking the time to respond to our questions.
After their contribution, I share some comments from the CEOs of the two patent pools.
- What's your best guess as to the percentage of relevant patents in the two pools?
It's exceptionally early to say. In fact, it will take at least a few years for VVC patents even to issue in large numbers, especially internationally, and without issuances to analyze, any estimates we might hazard here are going to fuzzy at best.
All that in mind, recognizing that this is useful information for prospective implementers right now, our best guess right now is that we think the pool licenses—including only current licensors—will get VVC implementers one-fourth to one-third of the patent rights they need. We arrive at this rough estimate by considering (i) who contributed to VVC; (ii) assignments of relevant patents from VVC contributors; and (iii) who has a significant HEVC portfolio.
As mentioned, VVC portfolios are young, with most still in the process of curation. Until a critical mass of VVC patents actually issued, contributions are the best available means of measuring portfolio strength. MPEG standards are, in certain respects, iterative, meaning that VVC builds on HEVC. This means that some patents essential to HEVC are also likely essential to HEVC, just as some AVC-essential patents are also essential to HEVC.
IPlytics is working on a detailed breakdown of this data, which we will publish in the near future, alongside our methodology. For now, we are comfortable providing the approximate range of pool coverage of VVC standard-essential patents (SEPs) indicated by the data. In the meantime, here is data we published last year on accepted contributions for VVC and HEVC.
At launch, the pool coverage for VVC, while well short of comprehensive, compares favorably with the proportion of rights a party implementing another ubiquitous standard, like Wi-Fi or 5G, might expect to get from dealing with only a couple of licensors. Implementers of cellular standards, for instance, should expect to deal with dozens of licensors without the transparency and ease of licensing through a pool with a significant share of applicable SEPs.
With the expectation that the percentages of VVC SEPs covered by the pools will continue to grow, the VVC pools are off to a promising start. That said, it is important for the pools' long-term success that they continue to add significant licensors.
- I notice that Qualcomm is the leader in VVC IP rights. What's the significance of Qualcomm not being in a pool?
No real significance. Pools license at the OEM level, and Qualcomm's sophisticated bilateral licensing program already deals with most—possibly every—major OEM that is a foreseeable VVC implementer. For instance, Qualcomm supplies application processors or systems on chips for about 25% of smartphones—the largest segment of VVC's addressable market—and more than half of all baseband processors. Devices that incorporate these chips will be licensed to Qualcomm's patent portfolio, obviating any need for a pool to help the OEMs obtain these patent rights efficiently.
- Much was made of the fostering process and MC-IF being able to avoid another HEVC-like royalty mess. Given the numbers you just laid out, was that a clear success, a clear failure, or too soon to tell?
We think MC-IF was and is a success, with qualifications. The true ambit of MC-IF was to streamline licensing for VVC and to encourage commercial reasonability in the overall patent cost of implementing the standard. Winnowing the field of prospective pools to one was also an explicit objective, one at which MC-IF did not succeed, but so long as a critical mass of licensors sign up for one of the pools, MC-IF achieved its larger aspirations. Past generations of codecs have launched with multiple pools and gone on to achieve ubiquity.
Access Advance launched a VVC pool that leverages its position as the leading HEVC pool to offer implementers a low marginal cost of adding a VVC license, and MPEG-LA's pool offers per-unit rates the market has accepted for codecs in the past. For most smartphones sold in the U.S., for example, the OEM would pay $0.30 to license both pools, subject to royalty caps—$0.10 for Access Advance (the marginal cost for an HEVC pool licensee to add a VVC license) and $0.20 for MPEG-LA. This assumes that, like most smartphones aside those sold by Apple and Xiaomi, the OEM in this example already has a license to Access Advance's HEVC pool.
Past experience suggests that the industry will accept $0.30 per unit for most of the patent rights a codec implementer needs. MPEG-LA launched its AVC pool in 2004, requiring $0.20 per unit in royalties. Today, that pool has over 1,500 members, indicating that its terms are commercially reasonable. Adjusted for inflation, $0.20 in 2004 U.S. dollars is $0.30 in 2022.
Comments from the VVC Patent Pool CEOs
If you run the numbers, Access Advance enjoys a much larger share of known patents than MPEG LA. I asked MPEG LA CEO Larry Horn whether Access Advance had an advantage because it launched with more patent owners, and he commented, "VVC is still in its infancy, and many large patent holders are apparently waiting to see how the market develops. We are pleased by the initial support our pool has received from patent holders. They provide a solid foundation."
Continuing, he stated, "With participation by a large number of patent holders yet to be determined, a few patent owners can make a big difference. Whatever its size, we are confident that our license will meet implementers' needs for a reasonable license providing coverage with the patent holders who are represented, creating a profitable pool for them."
I also asked Horn about pool pricing, particularly why MPEG LA's $0.20 royalty was so much lower than the high end of Access Advance's royalty structure (max royalty $1.50). He explained that "a pool should be structured to solve a market problem, not create one. That means providing reasonable access to foster technology acceptance that strikes a balance between patent owners and implementers. A pool established for the purpose of revenue generation ahead of addressing the market's need for technology adoption will do neither."
To be fair, royalties are set, at least in part, upon the IP delivered by the pool license. If Access Advance maintains its numerical IP advantage over MPEG LA going forward, the current price differential appears to be more than justified.
Speaking of that, IP professionals with their ear to the ground may have heard that a District Court in Dusseldorf Germany ruled that the Access Advance HEVC royalty was not FRAND, or not fair, reasonable and non-discriminatory, which makes it unenforceable. Since the VVC pool is structured and priced very similarly, this might raise concerns that the VVC royalty is also non-FRAND.
I asked Access Advance CEO Pete Moller about this. He pointed out that the same court has previously found Access Advance's royalty pricing to be FRAND, and that this ruling involves how Access Advance manages reimbursements relating to patents included in multiple pools.
To explain, a patent owner can include patents in multiple pools, but if a licensee pays the royalty to both pools, they are entitled to a reimbursement relating to the patents in both pools. Moller explained that the court decision involved Access Advance's policies relating to these reimbursements and that he expects to release a new policy he believes will be FRAND by the end of February.
I then asked Moller about the share of VVC-related IP he felt was in the Access Advance pool. He responded, "if you include essential HEVC patents that will also apply to VVC, 25%, plus or minus 5%." Moller was also still hopeful that "at the end of the day there will be effectively one pool." He continued, "we are working very hard to bring in all the patent owners on the sidelines. If everything stopped right now, our pool would be very viable, and we're hopeful that licensors on the sideline will join Access Advance and try hard to get one-stop pool."
With H.264 usage finally beginning to decline and several newer codecs ascendant, this 2022 codec update reports on the most significant announcements from the last year relating to H.264, VP9, HEVC, AV1, VVC, LCEVC, and EVC.
The terms are out—20 cents per unit for hardware/paid software and 5 cents per unit for free software—but there's lots of fine print and plenty of questions remain, including those about the contributors that aren't in either the MPEG LA pool or the Access Advance pool.
VVC today can be both useful and usable; let's hope that VVC IP owners can formulate a royalty policy that delivers the same.
The old realities that used to dictate codec adoption no longer apply. Opening up new markets now matters more than reducing operating expenses. How are HEVC, AV1, and VVC positioned for the future?
BBC R&D finds that AV1 produces better low-bitrate quality than HEVC, but the codec picture will get even muddier in 2020 as MPEG fast tracks VVC, MPEG-5 EVC, and LCEVC