HEVC Advance: What Do the Royalties Mean for Video Publishers?
This overly broad definition could be read to include revenue received from contracts closed due to marketing videos on a website, and many other areas of concern. I asked Moller about this, and he responded that while HEVC Advance wasn’t taking any potential markets off the table, the initial targets were pay-per-view and subscription services, along with video clearly supported by advertising or sponsorship revenue. He did mention that products built around HEVC, such as webinar systems and video conferencing, may become a target at some point, as might video within the enterprise.
When Do Royalties Start?
The royalty documents state that royalties are due upon the date of first sale, and I asked Moller when the licensing obligations begin. He responded that companies should have recognized that there might be additional royalties for HEVC when they created and started selling the products, and should have reserved for this. While cold, this is undoubtedly true. However, he also said that because the license wasn’t available prior to October 1, they might offer an incentive program (aka a substantial discount) to companies who “do the right thing” and pay right away.
There are several interesting issues to explore, starting with FRAND.
Is it FRAND?
I discussed FRAND pricing in my last article. The bottom line (as I see it) is that patent rights granted by IP owners who participate in a standard have to be fair, reasonable, and nondiscriminatory (FRAND). In figuring out what is FRAND, courts will consider rates assessed by other technology providers, like the $0.20 charged by MPEG LA, but also the relative “essentiality” of the patents. So if the technology represented in HEVC Advance’s patent pool was ten times more essential to HEVC technology as a whole than MPEG LA’s, its price might be reasonable. To be clear, I didn’t discuss this with Moller and he made no such assertion; I’m just pointing out that it’s not a strict mathematical comparison.
HEVC Advance is asking for up to $2.00 per TV, as compared to $0.20 for MPEG-LA, and about $1.00 max for a mobile phone ($0.80 plus the 25 percent compliance fee, plus 10 percent labelling discount). In addition, the MPEG LA HEVC pool is capped at $25 million annually, while the HEVC Advance pool has no cap. Obviously a cap is a component of the FRAND argument; MPEG LA is 20 cents compared to $2.00 for HEVC Advance, but it’s also $25 million as opposed to two or three times that for some manufacturers.
As an example, Apple will probably ship somewhere close to 150 million iPhone 6 models before the October 1, 2015 license availability. If Apple pays the full price, it will have to write a $120 million dollar check to HEVC Advance on October 1. That’s a big number any way you look at it.
Defending patent infringement claims is expensive, but it’s not that expensive. I checked, and it costs between $1 million and $1.75 million to defend a patent infringement suit when claimed royalties were in the $1 million to $25 million range. Overall, given the total royalties involved, I’d be shocked if hardware vendors didn’t raise a FRAND challenge.
With content-related royalties, the numbers are much smaller, which impacts the FRAND analysis, and also the licensee’s incentive to litigate. So whether major vendors contest these remains to be seen. My colleague Dan Rayburn discusses the potentially onerous reporting requirements the content royalty may engender, and he’s much more familiar with CDN reporting than I am. From a pure cost perspective though, the content-royalty doesn’t appear unreasonable.
Floors and Caps
MPEG LA uses de minimis exceptions to its royalty policies to exclude low-volume buyers. HEVC Advance didn’t choose this route, which will either force the group to ignore low-volume non-compliant users or set up an enforcement channel that won’t come close to paying for itself. I’m guessing it will opt for de minimis exceptions at or before the final announcement.
On the cap issue, while I can see why manufacturers like Apple or Samsung would want a cap, I can’t see why a technology developer that contributed to the pool would want one; it’s like telling Apple they can sell 25 million phones, but that’s it.
I asked MPEG LA about its caps, and learned that MPEG LA pools were largely comprised of vendors who profited from the standard by selling products that used the standard. The money earned from the pool was nice, but was ancillary revenue. Since many within the group were licensees subject to the royalty, such as Apple and Samsung, a cap made a lot of sense. None of the five vendors in the HEVC Advance group are in the retail mobile phone business, which will have the highest volume of any hardware item listed, so the lack of a cap isn’t a surprise.
Defining the Targets
HEVC Advance didn’t segment the target markets as definitively as we all would have liked, either within the royalty language itself or via examples like those shown above. I got the feeling talking with Moller that it was a monumental task simply getting the proposed terms defined, and I’m hoping it does a lot more work defining the targets before the final terms are released.
Migrating to VP9
One effect of the HEVC Advance royalty structure is to force web publishers to consider VP9 more strongly than they did in the past, which is an approach I’ve advocated for awhile (see The Case for VP9). As I explained in my last article about HEVC Advance, however, it’s unclear whether or not VP9 will ultimately be royalty-free:
On the other hand, back in March 2013, MPEG LA and Google entered into an agreement granting Google a license to technologies that may be essential to VP8 and VP9. By way of background, 11 of the 12 companies who participated in the license grant had responded to MPEG LA's call for patents essential to VP8. Though the twelfth company, Nokia, declined to join the MPEG LA group (and later sued Google in Germany) the threat of IP litigation from the MPEG LA camp was over.
Now that there is a separate group, it seems more likely that if VP9 makes serious inroads, additional IP owners will claim infringement. As mentioned in the Nokia article, Florian Mueller, owner of the authoritative Foss Patents blog, says the royalty picture for VP8 is far from clear. After the Google/MPEG LA agreement, Mueller wrote, "Claims that VP8 is now free from per-unit or per-implementer license fees are grossly exaggerated. There are simply too many video technology patents out there, and the backers of WebM/VP8 are primarily companies whose own patent portfolios are too weak to resolve patent infringement issues through royalty-free cross-licensing."
I tried to find out if any of the 5 founding companies in the HEVC Advance group were part of the 11 that licensed their technology to Google, but couldn’t track this down. I asked Moller about the likelihood of going after Google for VP9. He said that he hadn’t studied the issue yet, but that it was on his agenda. The bottom line is that the more successful VP9 gets, the more likely it is that HEVC Advance looks at it much more closely.
Overall, while many vendors will grumble about the content-royalty, it’s unclear whether it’s worth suing over. On the hardware side, HEVC Advance’s policy will almost certainly see one or more vendors litigating whether or not the proposed royalty is fair and reasonable.
After widespread criticism of its initial royalty proposal, HEVC Advance has reduced royalty rates and decoupled them from attributable revenue, as well as provided incentives for early adoption and a royalty credit for small publishers.
This year, HEVC has gone from the heir apparent to an unmanageable mess. Meanwhile, open source codec efforts are focused and progressing. What happened?
MLB Advanced Media says the proposed royalty isn't reasonable. With pressure mounting, the new patent pool promises an adjustment.
With HEVC licensing terms making it unsuitable for free software, Cisco sees the need for an advanced video codec unencumbered by fees.
Licensing costs for Netflix and Facebook would total over $100 million per year, and terms are retroactive. Group calls terms "fair and reasonable."
Details are sparse about HEVC Advance. The industry knew a second HEVC pool might form, but why do patent holders have a problem with MPEG LA?
The second HEVC patent pool promised to announce licensing terms and royalty rates last month, but its press conference was cancelled at the last minute, and no information has been made available.
Last week, HEVC Advance announced it is forming a new patent pool in addition to the one offered by MPEG LA. While the new group—which will likely include Dolby, GE, Mitsubishi Electric, Philips, and Technicolor—did not announce licensing terms, past precedent indicates that the royalties won't likely be onerous for most encoding and decoding vendors, though how it will affect content publishers is less clear.
There's no reason for Adobe to add HEVC support to Flash, and that keeps HEVC from becoming relevant for general purpose streaming.
Companies and Suppliers Mentioned