Lawsuits, licensing, and royalties are complicating 4K video support in gadgets

Jeff S

Ars Legatus Legionis
11,158
Subscriptor++
What a shitshow. If something is purchased with capability X, such as X being the ability to decode HEVC in hardware, then it should have the capability for the entire life of the product, and not be subject to the whims of... 🤷‍♂️ ... because they're too cheap or too lazy or scared or, whatever. Frankly I don't give a shit what their reason is, they shouldn't be allowed to yank out the rug from under users - and I suspect in Australia such a thing would be considered misleading and deceptive conduct and unlawful.
I'm not entirely sure, but I don't think HP, Dell, etc are removing the functionality from people who already bought products while HEVC was an advertised feature (or even if they did advertise it as a feature at all - when you look at adverts for laptops, I don't think you usually see a list of supported codecs)?

Wouldn't whether it's deceptive depend on whether there was advertising to the effect that when you buy this product, you get this as a feature? I suppose also if you checked and it was enabled, e.g. on an in-store demo unit, that would count as advertising, since the in-store demo features should match what's actually sold to customers, else it's a bit of a bait-and-switch.

As for HP and Dell, if I understand correctly, it's that future laptops they sell of those low-end models will not advertise nor come with HEVC support. I think, I could be wrong, but I think if you have one of those laptops that already supports HEVC, it will continue to do so?
 
Upvote
-8 (7 / -15)

Ryan.Switzer

Smack-Fu Master, in training
81
And the bigger worry with AV1....is that in 10 years if/when AV1 supplants HEVC--we could end up right back here. Because AV1 is a pet project of US-centric tech monopoly interests like Microsoft and Google. Personally, as someone in the USA, I think that hitching mid/long-term tech licensing (and therefore dependence) to US companies is an insanely bad idea these days.

Ideally things that are "standards" should be open rather than patent infested.
But most of the patent holders for HEVC are in Europe or Asia…
 
Upvote
16 (16 / 0)
And the bigger worry with AV1....is that in 10 years if/when AV1 supplants HEVC--we could end up right back here. Because AV1 is a pet project of US-centric tech monopoly interests like Microsoft and Google. Personally, as someone in the USA, I think that hitching mid/long-term tech licensing (and therefore dependence) to US companies is an insanely bad idea these days.

Ideally things that are "standards" should be open rather than patent infested.

It's not just a worry, it's reality. The usual suspects are coming after it already with patent lawsuits. With HEVC licensing issues at play, the next best (read: cheapest) alternative is AV1, and do you think the patent trolls haven't noticed?
 
Upvote
9 (14 / -5)

Jeff S

Ars Legatus Legionis
11,158
Subscriptor++
It sounds nice but the reality is very different. Usually once a R&D company accumulates enough patents, someone in the management eventually realizes, that extracting more value from existing patents is much easier and reliable than doing research, which is expensive and naturally very chancy and fortuitous.
Well, one limiting factor to that is the limited time duration of patents. They do expire eventually. Although in the computer industry, they do last long enough that usually by expiration, the patents are thoroughly obsolete anyhow.
 
Upvote
13 (15 / -2)

Jeff S

Ars Legatus Legionis
11,158
Subscriptor++
It's not just a worry, it's reality. The usual suspects are coming after it already with patent lawsuits. With HEVC licensing issues at play, the next best (read: cheapest) alternative is AV1, and do you think the patent trolls haven't noticed?
The other risk, as is becoming pretty clear with this current maladministration, is that the US government may decide to put export controls or punitive duties on any US-held patents.

So far Trump has been more focused on tariffs on imports, than on duties on exports, but I keep waiting for us all to wake up one day and see that Trump decided to slap a 100% duty on exports to some group of countries because he's throwing a tantrum. Which of course would just hurt US companies trying to export. Cue the "curb your enthusiasm" music.
 
Upvote
0 (12 / -12)
So I realize that this doesn't do Netflix (et. all) any good, but maybe it's time to just rely on H.264 until the wheels well and truly fall off. H.265 becomes patent free in 2031, so maybe we just use more bandwidth until then and go without? AV1 doesn't become patent free until the 2040s, but many of the leeches trying to attach to it may have their patents expire in the next few years.

I feel like these battles were a lot more impactful years ago when companies were fighting over MPEG-2 and MP3, because there really was no alternative. At this point, however, MPEG2 and MP3 are free and clear of any patents, and many later standards (e.g. h.264) are either clear or almost clear of patents as well.

It's a mess of a situation, but unlike most messes, this one will actually get better in time, around 5 years. Just paying the bandwidth bill might be better than dealing with all the leeches.
 
Upvote
40 (41 / -1)
Post content hidden for low score. Show…

Ryan.Switzer

Smack-Fu Master, in training
81
If one wanted to create a full ogg video with audio, does that mean only the video would be encoded with Theora, and the Audio would be compressed as Vorbis, and the player software would basically decode them separately and synchronize the playback?
This is how all codecs work. HEVC and AV1 mentioned in the article are both video codecs. Audio codecs include things like Dolby Digital or Advanced Audio Codec (AAC). Video and audio tracks use these separate encodings and they’re packaged together into a file format called a “container.”
 
Upvote
22 (22 / 0)
Hot take: Algorithms should not be patentable. When you get right down to it, it's math. Very complicated math, sure. But math it is. Copyright your specific implementation of it? Sure. Patent the math itself? No way. Does this lead to the death of software patents in general? Dunno, but would that be a bad thing? I'm thinking not.
I feel like there needs to be a generalized POSA (Person of Ordinary Skill in the Arts) challenge process for patents generally.

You should be able to invalidate a patent by finding a dozen ordinary engineers of the appropriate type, sitting them down with the goals of the patent but not its implementation, and providing them with info on every previous patents (perhaps carefully cherry-picked to encourage certain lines of thought). Given them access to any information that was publicly available before the filing date of the target patent, again, possibly cherry-picked to encourage certain lines of reasoning.

Give them some time, maybe a week or two, and see what they come up with. If they generate any "patent" filings that the target patent would have been found to infringe on its inception, then the resulting patent is not "non-obvious" and should be invalidated. If the POSA crew fails, then try again in a few years.

And in all cases, any "patent" filings they make should be incorporated into the patent database as expired patents on the day they generate them, essentially additional prior art to help invalidate future patents.

(edit, acronym, DYAC)
 
Upvote
6 (12 / -6)

elboku

Wise, Aged Ars Veteran
141
Subscriptor
So, if I read the defense here of patent law: they encourage research and development? Prior to 1474 when the first patent laws went into effect, no research and development happened. Or were patents designed to protect- surprise!- generally rich people who COULD do research and development? Are we to believe that without patenty law, no further progress in human history would have occurred?
 
Last edited:
Upvote
16 (19 / -3)

balthazarr

Ars Tribunus Angusticlavius
6,901
Subscriptor++
I'm not entirely sure, but I don't think HP, Dell, etc are removing the functionality from people who already bought products while HEVC was an advertised feature (or even if they did advertise it as a feature at all - when you look at adverts for laptops, I don't think you usually see a list of supported codecs)?

Wouldn't whether it's deceptive depend on whether there was advertising to the effect that when you buy this product, you get this as a feature? I suppose also if you checked and it was enabled, e.g. on an in-store demo unit, that would count as advertising, since the in-store demo features should match what's actually sold to customers, else it's a bit of a bait-and-switch.

As for HP and Dell, if I understand correctly, it's that future laptops they sell of those low-end models will not advertise nor come with HEVC support. I think, I could be wrong, but I think if you have one of those laptops that already supports HEVC, it will continue to do so?
Oh, right. I got the impression that they were removing functionality from existing, purchased devices via an update.

Isn't the advertising also in the product specs? If the specs say it is capable of X, and it isn't actually (or is updated to remove it) - that's bad. It's like the PS3 and dual booting that they removed (and got sued for).

Removing functionality that was previously available sucks, but if it's clear that's what you're getting - then it's just part of the overall enshittification trend like everything else. You get to pay more for less... Yay, I guess. (/s)
 
Upvote
3 (5 / -2)

ikjadoon

Ars Tribunus Militum
1,626
Subscriptor++
AV1 is a perfectly cromulent codec and all, but....I have a household full of devices and only two of them have AV1-support in hardware. HEVC? I'm not sure I have anything that doesn't support it ...
That tracks. You're right it'll take a decade.

HEVC was published in 2013. It has been 13 years.
AV1 was published in 2018. It has been 8 years, slightly over half of HEVC's lifetime.

//

And the bigger worry with AV1....is that in 10 years if/when AV1 supplants HEVC--we could end up right back here. Because AV1 is a pet project of US-centric tech monopoly interests like Microsoft and Google. Personally, as someone in the USA, I think that hitching mid/long-term tech licensing (and therefore dependence) to US companies is an insanely bad idea these days.

Ideally things that are "standards" should be open rather than patent infested.

AV1 is a codec released by AOMedia, a legally-recognized non-profit.

AOMedia's parent organisation is the Linux Foundation.

AV1 is open : aom - Git at Google

Skepticism is always necessary, but it soon needs actual evidence to keep itself relevant.

//

They want to save money...for themselves, not for us. It would attract litigation...that wouldn't go anywhere and be anything but a sternly worded letter to Big Tech.

How would that work? How has any codec directly and literally saved money for consumers? Codecs are a product. You use a product and you get value from it: lower bitrates, less buffering, etc.

This train of thought has nothing to do specifically with AV1. Of course companies paying bandwidth fees want to lower their fees.

The existence of Big Tech in a project does not mean it is owned by Big Tech.
 
Upvote
64 (64 / 0)

AlbatrossMoss

Wise, Aged Ars Veteran
102
Subscriptor
What you call Business Leaches, other people might call R&D companies.
No, definitely not. R&D companies dig into the unknown. It's hard as fuck. Maybe what they find is trivial or boring, maybe it's brilliant and game-changing. Either way, it's valuable and worth every penny (and more).

But I wasn't talking about R&D companies. Instead, I'm talking about companies that have effectively become patent trolls, like Ericsson and Nokia (both legendary in their time, for very good reasons, now hollowed out and zombified, existing just to "own and license the IP").

Of course, the IP is theirs, and cannot be taken from them. But they shouldn't be able to force the industry to reinvent wheels just because "we own the IP even if there's nobody in the whole company that understands the IP".

I have no solutions, but this doesn't feel right.
 
Upvote
45 (46 / -1)
Post content hidden for low score. Show…

tuffy

Ars Scholae Palatinae
946
I'm starting to think it would be worthwhile for the chip makers to reverse-engineer the codec in a well documented clean environment to create their OWN decoding algorithm just so they can sidestep the licensing issue entirely.
Reverse-engineering is a way to avoid infringing on someone else's copyright, but it's not going to avoid infringing on a software patent. No matter how the codec is implemented, any patents on its implementation will still apply.
 
Upvote
68 (70 / -2)

mdw

Wise, Aged Ars Veteran
177
Subscriptor
I'm starting to think it would be worthwhile for the chip makers to reverse-engineer the codec in a well documented clean environment to create their OWN decoding algorithm just so they can sidestep the licensing issue entirely.
There's nothing to reverse engineer, the specification is open. And reverse engineering does not avoid patents anyway.
 
Upvote
39 (39 / 0)
Hot take: Algorithms should not be patentable. When you get right down to it, it's math. Very complicated math, sure. But math it is. Copyright your specific implementation of it? Sure. Patent the math itself? No way. Does this lead to the death of software patents in general? Dunno, but would that be a bad thing? I'm thinking not.
And if we allow them to be patentable, then the term limit should be 5 years maximum. I can understand why in traditional industries 20 years can make sense (hardware is risky in terms of material and money), but not in software.

Software doesn’t need patents to spur innovation and 20 years is beyond a lifetime in software. Consider how far Windows and MacOS have come in 20 years as an example.

The lack of patents in software is what motivates business to monetise and move forward quickly.

Short lifetime patents could be a compromise, but the issue is a patent is meant to describe an implementation, but we’ve seen people being sued for code with similar behaviour, but a different implementation. For this reason I’m generally skeptical of the effectiveness of patents in this context.
 
Upvote
32 (35 / -3)

DrewW

Ars Tribunus Militum
2,010
Subscriptor++
Actually you can, you just can't do it using Netflix. Crap like this is exactly why the high seas have such a rich bounty. Not only do we not own anything, we have to pay for a license to decode it.
It’s funny you mention that. I vividly remember a big part of 90s/00s tech culture was getting the right codec packs to make an AVI or MKV play. We’re just going in circles.
 
Upvote
32 (34 / -2)
Post content hidden for low score. Show…

cyberfunk

Ars Scholae Palatinae
1,415
you can’t, for example, download a 4K episode of a show from Netflix and play it in VLC

The hell you cant. You absolutely can. As with many things there are uncomfortably free solutions to red-tape problems like this. Certainly if you've already paid netflix, however, I'm not sure you should feel that bad.
 
Upvote
12 (15 / -3)
Post content hidden for low score. Show…
Post content hidden for low score. Show…

wiak

Wise, Aged Ars Veteran
126
i really hope AOMedia or their members will start to sue or even buy up whoever that says they have AV1 patents. that way they can remove the trash before its to late, with all the hevc/avc cartels just making everything more expensive and their monopoly control over codecs, i have ran into this bullshit issue as well hevc codec missing due to patent

might well just outlaw software patents
 
Upvote
-1 (2 / -3)

Kushan

Smack-Fu Master, in training
84
You do realize that the patents in question aren't software, but rather algorithms, right? They just so happen to be implemented in software. They are in no way equivalent to the JAVA debacle when it comes to software patents.
That's because software patents aren't allowed in the EU, so companies work around it by patenting the process instead. It's still utter nonsense.
 
Upvote
21 (22 / -1)

numerobis

Ars Tribunus Angusticlavius
50,859
Subscriptor
Yeah because there's absolutely no way that open source projects get malware injected into them. Grow the fuck u p
Dear god, it’s 2026 and you’re still acting out the open source denier troll persona?

Next you’re going to argue Y2K is overblown and we shouldn’t try to manage it.
 
Upvote
28 (31 / -3)

Jeff S

Ars Legatus Legionis
11,158
Subscriptor++
This is how all codecs work. HEVC and AV1 mentioned in the article are both video codecs. Audio codecs include things like Dolby Digital or Advanced Audio Codec (AAC). Video and audio tracks use these separate encodings and they’re packaged together into a file format called a “container.”
Right, what I meant more specifically is I wasn't sure if Theora included a newer audio codec that was normally used with it, or if it explicitly just uses the older Vorbis codec for audio ( I suppose you could pair other audio codecs in the container if you really wanted to force the issue, but I presume normally Ogg Theora would be paired with some Ogg audio codec, and I guess that would usually be Vorbis)?
 
Upvote
0 (0 / 0)

Nalyd

Ars Praefectus
3,057
Subscriptor
And if we allow them to be patentable, then the term limit should be 5 years maximum. I can understand why in traditional industries 20 years can make sense (hardware is risky in terms of material and money), but not in software.

Software doesn’t need patents to spur innovation and 20 years is beyond a lifetime in software. Consider how far Windows and MacOS have come in 20 years as an example.

The lack of patents in software is what motivates business to monetise and move forward quickly.

Short lifetime patents could be a compromise, but the issue is a patent is meant to describe an implementation, but we’ve seen people being sued for code with similar behaviour, but a different implementation. For this reason I’m generally skeptical of the effectiveness of patents in this context.
20? Damn I must be old. I thought it was 7.
 
Upvote
1 (1 / 0)
Post content hidden for low score. Show…

numerobis

Ars Tribunus Angusticlavius
50,859
Subscriptor
Usually R&D companies GET THEIR MONEY from what you call Patent Trolls. That's how they make money. If the R&D companies, as you say, are worth every penny, because they are doing hard work, which I agree with that premise, then that means you need to accept the "patent trolls" because they are who pays the R&D companies.

And if a company like Ericsson or Nokia stop doing R&D? Well, so what. You aren't paying for current R&D when you license a patent. You are paying for past R&D which was already invested in, and payment in the present is part of the ROI for those companies on their past research.

If they stop doing research, soon enough they won't be making any money, because patents expire, or become obsoleted by new tech, and then nobody is licensing anymore (well, in the case of a codec, licensing will probably continue right up until patent expiration, simply for backwards compatibility so people can continue to play, transcode, etc old videos that people made years ago).
Nokia and Ericsson made bank from the telephony equipment they sold, which were top notch thanks to their R&D.
 
Upvote
21 (22 / -1)