HEVCAdvance Issues HEVC Royalty Schedule With Content Royalties

HEVCAdvance is the second HEVC patent group comprised of GE, Technicolor, Mitsubishi, Philips, and Dolby, which I wrote about in detail for Streaming Media in New HEVC Patent Pool: What Are the Implications? Last week, the group issues a proposed royalty schedule to much negative fanfare. Here’s the schedule for hardware for Region 1, which includes most developed countries. Region 2 is half the price. 


Built into the schedule is a 10% labeling discount; if you’re a TV vendor and don’t display the HEVCAdvance logo, your royalty increases 10% to $1.65. There’s also a 25% “compliance” discount; if you don’t voluntarily pay, the base price increases by 25%. Again, our TV vendor price increases from $1.50 to $1.875, plus the 10% labeling discount, since if you didn’t pay, you probably didn’t label. Royalties start at the date of first sale, even though you won’t be able to sign a license until October 1, 2015. 

These prices are about ten times higher than MPEG LA is charging and there is no cap; MPEG LA’s royalty is capped at $25 million annually. Guessing here, Apple will have sold perhaps 150 million HEVC-equipped iPhone 6 and 6 Plus phones by October 1, 2015, which means they’ll have to write a check for about $120 million on October 1. I’m thinking that Apple and other high volume manufacturers will challenge the royalty as too high under the Fair, Reasonable, and Non-Discriminatory (FRAND) pricing guidelines that control technologies licensed as part of a standard. Though the suit would cost $2-$4 million, it’s almost certainly worth it. 

Most TV and mobile device manufacturers should have been reserving for additional HEVC royalties since they started manufacturing products that used the technology, though I’m guessing few reserved in the price range actually set by HEVCAdvance. If not, these companies will have to add very significant additional reserves which translates to a short term financial hit. 

Note that HEVCAdvance is not pursuing royalties on professional encoders and decoders, hardware or software. Consumer decoders, like those contained in browsers, will be subject to the royalty at the $1.10 rate for other devices. With no cap, the royalty essentially means we’ll never see HEVC playback in browsers. 

Content Royalties

Though there are content royalties for MPEG-2 and H.264 in the MPEG LA patent pools, the HEVC group inexplicably decided not to pursue content royalties for HEVC. HEVCAdvance is charging .5%, plus the 25% non-compliance penalty for those who don’t pay voluntarily. The royalty applies to pay-per-view and subscription services, plus video that’s advertising supported like ESPN or CNN. 

I ran some numbers for a Streaming Media story on HEVCAdvance. For a $4.00 movie downloaded from Amazon Prime or M-Go, the royalty would be two cents, right in line with MPEG-2/H.264 content royalties. In a Netflix scenario, for a $10/month subscriber who watches 10% of video that uses HEVC, the royalty would only apply to 10% of the subscription price, so the royalty would be about half a penny ($0.005). Assuming the $10 subscriber watches 100% HEVC, the royalty would be a nickel. 

HEVCAdvance expects the royalty to be calculated on gross numbers, not on a per-subscriber basis. For an advertising supported site, if HEVC was 30% of all video distributed, the calculation would be 30% x total video-related advertising revenue x .005. In this scenario, if video-related advertising revenue was $1 billion, the royalty would be $1,000,000,000 x .3 x .005, or $1.5 million, a far cry from the $120 million Apple is staring at. 

My colleague Dan Rayburn pointed out that computing the content royalties could require data that web publishers may not currently have, along with several other major objections, some covered here, some not. All that said, the content royalties are more modest and better grounded in precedent than the hardware royalties, so a FRAND challenge doesn’t seem as imminent, though I have not spoken with any vendors who will be impacted by the royalty. 


From a hardware standpoint, the proposed rate and lack of a cap almost guarantees a FRAND challenge. Regarding content royalties, I expected MPEG LA to assess them and was surprised when they didn’t. In essence, H.264 was the foundational technology for services like Netflix and every content site on the web, and the H.264 patent holders got but a pittance. In retrospect, this is almost certainly one key reason that the HEVCAdvance members declined to join the MPEG LA patent pool. I personally don’t have an issue with HEVCAdvance claiming content royalties, though it appears that I’m in the minority here.

This is a key point. Most of the criticism surrounding the content royalty has focused on the cost, not the opportunity. HEVC either allows its users to enter new markets, like 4K TV, or reduce the bandwith cost of their streams. Netflix’s 4K content costs $2.00 more than their regular service, HEVCAdvance wants a nickel of this increment, or about 2.5%. Treated as a cost of goods and marked up by the usual 4x, this nickel translates to $0.20 extra at retail. 

As to the confusion caused by HEVCAdvance’s proposed royalties, I’m calling BS on that one as well. Now that the numbers are known, companies who integrate HEVC into their products can properly reserve for the expense, and adjust their pricing as necessary. Publishers seeking to earn revenues by displaying on 4K TVs know the cost, while publishers considering HEVC for bandwidth reduction do so as well. Apple is not going to leave the smartphone market, Samsung won’t stop making 4K TVs, and Netflix won’t bow out of the 4K market. This is not a seismic event or a game changer in any way, shape or form. 

The final point is this; no matter how much you dislike the terms offered by HEVCAdvance, dealing with the individual patent holders would have likely been more expensive, and certainly more complicated. IP rights are a reality, so like the T-shirt says, the market will keep calm and carry on. 

Here’s another link to the Streaming Media piece, which goes into all these topics in much more detail. In the meantime, expect CE and mobile manufacturers earnings to take a short term hit, and if you’re dating or married to a patent attorney, don’t expect them to be around much over the next three or four years. 

About Jan Ozer

Avatar photo
I help companies train new technical hires in streaming media-related positions; I also help companies optimize their codec selections and encoding stacks, and evaluate new encoders and codecs.

Check Also

Streaming Media 101: Training for App & Player Development/Testing Professionals