Standards are Failing the Streaming Industry

Imagine you’re walking through Heathrow Airport terminal, shopping for tchotchkes to bring home to spouse and children. You spot a handsome Cambridge sweatshirt, give it a quick rub and squeeze, and it feels great. What do you check next? For me, it’s price, because if it’s £30, it’s coming home, but if it’s £60, I’m buying a “Mind the Gap” or “Keep Calm and Carry On” T-shirt.

When you’re shopping for technology, it’s pretty much the same thing. Imagine considering an awesome new HP Zbook notebook (like the one I’m typing on) and hearing, “well, it could be $1,200, but it also could be $12,000. Will that be cash or charge?” I’m running, not walking away from the cash register.

The Shifting Cost of H.264

So it is with technology standards. For years, H.264 pricing was seemingly fixed at the rates set by MPEG LA, which is $0.20/unit with a sliding cap now at 9.75 million through 2020. In this case, the MPEG LA patent pool really did help commercialize the standard in the streaming industry, since the one-stop shopping and reasonable price convinced Adobe to integrate H.264 playback into Flash, while Apple and Google did the same for mobile. Very quickly, H.264 enabled a single encode that could play in almost all relevant streaming endpoints.

However, in 2010, Motorola sued Microsoft demanding between a $6-$8/unit royalty (2.25% of sale price) for the combined infringement of Wi-Fi and H.264 patents. The court disagreed, applying the Fair, Reasonable, and Non-Discriminatory (FRAND) model applicable to technologies essential to a technology standard, and concluding that Motorola’s H.264-related patents were worth a much more palatable $0.0055/unit. This was a great result for most H.264 users, since the numbers are so small that it would discourage further suits for unit quantities less than the hundreds of millions.

Then, in December 2016, Nokia sued Apple for H.264 infringement, with a novel argument that may exclude their H.264 patents from FRAND limitations. As I type on this keyboard, I have no idea how much deploying H.264 may cost, over eleven years after MPEG LA started charging royalties. Will it be $1,200 or $12,000? Nobody knows.

HEVC Advance (Pigs get Fat while Hogs Get Slaughtered)

MPEG LA was first out of the gate with an HEVC pool, with rates starting at $0.20/unit with an annual cap of $25 million, and now has over 150 licensees. In early 2015, HEVC Advance hit the market with rates as high as $1.95/unit for TVs, and initially no cap, and of course, their royalties are in addition to those charged by MPEG LA. In an article I wrote in 2016, I guestimated that Apple would owe HEVC Advance over $120 million if they simply paid list price. Again, does HEVC cost $25 million or $145 million?

HEVC’s pricing produced a completely appropriate hissy fit, and HEVC adoption stalled. Though HEVC Advance later set the cap at up to $47 million for content and devices, four years after the bitstream finalized, HEVC playback still isn’t available in any browser or on iOS devices. HEVC Advance lists only 17 licensees on its website. Other than use in high-end OTT applications, HEVC is dead in the water for streaming. Meanwhile, HEVC competitor VP9 has seen some uptake in applications like the JWPlayer OVP and Netflix, as well as in the Wowza Streaming Engine.

HEVC Advance is now offering software playback without royalty, but this only impacts their portfolio, not MPEG LA’s, or any other. Speaking of that, in addition to these two known pools, Technicolor also has HEVC essential patents that it’s licensing, and there are several other companies with HEVC IP that have yet to join a pool or announce their own strategy. So again, years after the release of HEVC we have absolutely no idea how much it costs to use it.

Surprise! MPEG DASH isn’t Free

The most irritating royalty relates to MPEG-DASH, which was supposed to a unifying format that would simplify the lives of all streaming producers. Back in 2011, when the standard was formulated, key technology contributors Microsoft and Cisco pledged that their contributions would be royalty free. At the same time Qualcomm pledged that they wouldn’t seek royalties on products that fully implement the DASH spec except for those capable of implementing the WWAN standard, in which case they’ll charge the standard upfront fee and royalty for WWAM, but no additional charges for DASH. Then, in 2015, MPEG LA announced they were forming a pool, which they launched in late 2016. Almost out of nowhere, simplicity cost a nickel a unit and up to $30 million per annum.

The DASH.if industry group claims they can’t organize a fight against the pool due to anti-trust issues. Microsoft, Qualcomm, and Cisco aren’t talking, most likely upon the advice of their attorneys, and face the Catch 22 alternative of revoking their IPR declarations and joining the pool (if that’s even possible), or honoring their word, and paying up to $30 million/annum. Imagine having that talk with your CFO. Or, they can fight when MPEG LA comes calling, another lovely conversation to have with your financial and legal people.

In the meantime, as with HEVC, the industry is searching for other alternatives to MPEG DASH, including extending the use of HLS. Is Apple immune from claims that the HLS infringes upon the same patents that are in the MPEG LA DASH pool? No, but Apple is clearly more formidable than DASH.IF or any single contributor or DASH user.

The Alliance Way

What’s the alternative to standards? The approach taken by the Alliance for Open Media (AOM), which is formulating the royalty free (and much delayed) AV1 codec. Anyone who contributes to the group must agree in advance that their contribution is royalty free. This makes sense, because all companies contributing/participating are operating companies that will use the codec integrally in their core businesses.

Is AOM immune from the types of patent claims that Apple and Microsoft suffered with H.264? No, but they can act as a group to fight these claims, or otherwise trade some of their patents to keep AV1 royalty free. This makes the group much more powerful than the toothless DASH.if or any single DASH contributor or user.

Now that MPEG is spinning up calls for the next generation video standard, it needs to find a way to address the cost and commercialization aspects of the codec as well as the technology. You know, something radical like a single known number on the price tag. Otherwise, they may end up launching a standard that’s totally ignored by the streaming industry.

(Note: These thoughts were triggered by research performed for a recent talk on royalties at the Streaming Forum in London. You can see a description of the talk and download the handout here).

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. I am a contributing editor to Streaming Media Magazine, writing about codecs and encoding tools. I have written multiple authoritative books on video encoding, including Video Encoding by the Numbers: Eliminate the Guesswork from your Streaming Video (https://amzn.to/3kV6R1j) and Learn to Produce Video with FFmpeg: In Thirty Minutes or Less (https://amzn.to/3ZJih7e). I have multiple courses relating to streaming media production, all available at https://bit.ly/slc_courses. I currently work as www.netint.com as a Senior Director in Marketing.

Check Also

Take the Bitmovin Video Developer Survey

Contribute to the one of the most valuable sources of industry data by completing the …

Speech-to-text In Premiere Pro – Fast, Easy, Accurate, and Free

This video tutorial teaches you how to convert speech-to-text in Premiere Pro. I’ve been using …

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

Leave a Reply

Your email address will not be published. Required fields are marked *