Versatile Video Coding (VVC) / H.266: HEVC successor [Archive] - Page 27

CruNcher

9th December 2025, 18:10

"Dark Shikari" ... a name I didn't read in ages. But still remember.

Time goes by fast huh he was different but i never would have expected that different ;)

I’m not really a fan of when someone undergoes a very visible transformation to express their feminine side it can feel sad when it completely takes over.
But when it does happen, it shouldn’t be judged.

So much more deepness was in all this PSY for her

"It’s common for people who later transition or evolve in gender identity to be drawn toward subjects involving perception, identity, embodiment, or the inner/outer experience of self. That doesn’t make the past “fake”—it means they were trying to understand themselves."

People who dive hard into psycho-visual modeling, perception, masking behavior, and human visual response curves often have a deep interest in the inner vs. outer experience. But when you're in the middle of an engineering project, you see it as:

algorithms,

metrics,

heuristics,

tuning,

reference code,

experiments.

You don’t tend to analyze the psychological or personal motivation behind each contributor’s intensity — because the work itself is already absorbing enough.

So yes, it’s completely understandable that you wouldn’t have picked up the personal meaning behind someone’s fixation on psy optimizations. When you’re debugging macroblock decisions at 2 a.m., nobody is thinking:

“Oh, maybe this is part of their identity journey.”

You just assume they’re a passionate codec dev.

Jason/Fiona

and this fixation on Touhou oh man i feel strange but also excited what i have experienced or what i even was allowed to experience ?

kurkosdr

10th December 2025, 00:46

VVC (H.266) video codec effectively dead on arrival, posits analyst
https://www.flatpanelshd.com/news.php?subaction=showfull&id=1764735539

Anyone who could sense where the wind was blowing could tell 2 years ago, since nobody was talking about using the thing anywhere:

Broadcasters -> They are conservative by nature due to the fact that if they want to support a new format, they either have to break compatibility with some of the installed base of receivers or simulcast. The majority of broadcasters are still transitioning to HEVC, some (like UK terrestrial broadcasters) are still transitioning to H.264. Even broadcasters starting HDR broadcasts now will not break compatibility with the huge installed base of UHD TVs that don't do VVC unless they are obligated to do so by government dictate (which will happen in a minority of countries, but too localized to matter)

Optical media -> 4K Blu-ray uses HEVC, no need for a new optical disc format (8K content isn't a thing and isn't becoming a thing)

Streamers (the supposed use case for VVC) -> The majority of streamers either don't want to hear about ISO/ITU video standards after the HEVC coup, or can see the total cost of supporting a new format and paying licensing to the 20 or so patent licensing entities of VVC is simply not worth the hassle and the money, internet bandwidth is getting cheaper anyway

So, there you have it, the patent holders behind MPEG got their fat profits from the HEVC coup, but you know how it goes "fool me once, shame on you, fool me twice, shame on me". Even if VVC's costs are acceptable now, you never know if 5 years from now you will be dealing with 40 or 50 licensing entities, since the number of licensing entities is clearly not market-driven (VVC has twice that of HEVC).

CruNcher

10th December 2025, 03:54

If things continue the way they’re going, the ITU/JVET/MPEG ecosystem might reach a breaking point much faster than people expect.
H.264/AVC hit a perfect historical moment of balanced complexity, manageable patents, and universal adoption.
VVC did many things right technically, but its patent landscape is so complex that it would require serious goodwill and probably financial generosity to keep it viable.
That complexity also creates problems for hardware vendors like Intel.

I’m particularly skeptical about the neural network aspects of the newer pipelines.
Some models can steer content in directions that don’t necessarily reflect the original signal.
When AI-based metrics start defining “quality,” we face difficult psychological questions: what is the common denominator of reality that we all agree on?
Things will become very interesting when optimization depends on “truthful” sensory input because while that might be psychologically healthy, not everyone will like it.
And we already see the opposite everywhere: artificial lowering of resolution and synthetic “detail” presented as high definition.

So what should an AI-driven encoder mode actually prioritize?
Truthfulness? Artificial reconstruction? Beauty?
Pick your profile.

Eventually, every individual viewer may choose the version of reality they prefer.
But once that happens, our shared understanding of reality may drift out of sync and possibly very far from what we traditionally consider real.

Grain is a good example. It’s a statistical phenomenon, so AI can replicate it easily, and Technicolor/Thomson will likely move aggressively on FGT with their own AI implementation.
They may even win that patent race. But once you’re in the realm of statistical reconstruction, anyone with a model can do it.
A formal standard only matters if you want efficient hardware support.

This is why AI within the ECM pipeline is so important for NPU vendors. Intel in particular wants to reclaim ground in this space from Nvidia, AMD and Apple.

Digital compression has already changed people’s perception of reality slowly, over decades.
AI has the potential to change perception far more quickly, and by a magnitude that we haven’t yet absorbed or compensated for in the traditional digital signal domain.

ksec

16th December 2025, 17:08

This seems more like a post-mortem than news.

Interesting they're projecting HEVC dominant and growing its dominance substantially over the next five years. Although I never trust a plot without units.

I will absolutely bet those projection for H.264 decline are way too optimistic and it will stay higher than 50% for longer than anyone expected.

Especially now DRAM and NAND price are skyrocketing and will be the case till at least 2028 I wont be surprised a lot of lower end devices wont get new media decoding engine upgrade for a very long time. The same goes with CPU upgrade for software decoding on Smartphone.

Z2697

16th December 2025, 21:13

I will absolutely bet those projection for H.264 decline are way too optimistic and it will stay higher than 50% for longer than anyone expected.

Especially now DRAM and NAND price are skyrocketing and will be the case till at least 2028 I wont be surprised a lot of lower end devices wont get new media decoding engine upgrade for a very long time. The same goes with CPU upgrade for software decoding on Smartphone.

The "common" RAM and SSD size is more than enough to implement a new decoder, I don't understand the connection here.
And if storage is expensive, it's more likely that people (except videophiles) will prefer "more advanced" space-saving codecs.
But, nowadays the biggest incentive behind codec adoption is probably bandwidth, since most of the video are streamed, and they are not really saving space for the platform, they just add another re-encode option to the stack because of compatibility needs.

Z2697

16th December 2025, 21:51

The only "AI" feature currently in ECM is JVET-AJ0249: Neural network-based intra prediction with DIMD mode derivation.
Which sounds like harmless :)

Let's recap how intra prediction works:
With reconstructed pixels from single row / column of pixels from above and left,
it extrapolates pixels from neighbor pixels in the estimated direction (limited to some number of choices) to fill the current block (angular mode),
or use average of all neighbor pixels to fill the current block (DC mode),
or it interpolates each pixel's value in current block by coordinate, again from only two lines of neighbor pixel (planar mode).

The NN is just used to choose modes...

And then comes residual.
Prediction + residual = reconstruction

birdie

18th December 2025, 12:08

There’s an interesting (and somewhat depressing) real-world data point around VVC that seems worth sharing here, especially in light of the ongoing H.267/ECM discussion.

Oppo has recently filed what appears to be one of the first VVC SEP suits (https://ipfray.com/oppo-fires-back-at-asus-in-china-files-first-known-vvc-sep-suit/), targeting ASUS in China. What makes this notable is that ASUS doesn’t ship its own VVC implementation; any hypothetical VVC capability would come indirectly via Intel Lunar Lake. In other words, we’re already seeing litigation before VVC has any meaningful deployment or demand.

I raised this broader adoption/licensing concern directly with Vadim Seregin (Qualcomm / MPEG), along with a more technical question about VVC’s near-lossless behavior at high bitrates. His response was polite and technically correct, but also illustrative of a larger disconnect.

Near-lossless: VVC does have a lossless mode (VTM), but that doesn’t really address the practical observation that current VVC encoders (e.g. vvenc) tend to preserve fine detail worse than HEVC / AVC in high-bitrate, near-lossless regimes. This looks like a side effect of stronger RD-optimized tools and filtering rather than an encoder bug.

Adoption / licensing: The response was essentially that licensing issues are “not related to standards development.” That may be true from a committee perspective, but from an implementer’s point of view, licensing is the dominant constraint. A codec that cannot be shipped safely might as well not exist, regardless of its compression gains.

What worries me is that this same separation of concerns appears to be repeating for ECM / H.267: technical progress continuing in parallel with increasingly hostile or opaque licensing signals. The Oppo–ASUS case feels less like a meaningful dispute and more like an early warning of HEVC-style fragmentation — except this time there is already a viable, widely deployed royalty-free alternative.

I’m not claiming VVC is “dead” as a specification, but as a practical ecosystem choice it’s hard to see how these signals encourage adoption outside of narrow or captive niches.

benwaggoner

20th December 2025, 00:09

benwaggoner

20th December 2025, 00:19

The response was essentially that licensing issues are “not related to standards development.” That may be true from a committee perspective, but from an implementer’s point of view, licensing is the dominant constraint. A codec that cannot be shipped safely might as well not exist, regardless of its compression gains.

What worries me is that this same separation of concerns appears to be repeating for ECM / H.267: technical progress continuing in parallel with increasingly hostile or opaque licensing signals. The Oppo–ASUS case feels less like a meaningful dispute and more like an early warning of HEVC-style fragmentation — except this time there is already a viable, widely deployed royalty-free alternative.
Yep, 100%. Given the pretty exponential growth in the number of patents and unique patent holders each MPEG codec cycle, the IP risk is getting worse more than the compression efficiency is getting better. VVC could cost 3x more to license than H.264 did at launch and still have a great ROI for a variety of use cases. But a CFO could well look at VVC licensing and conclude that the actual net licensing cost will be retroactively determined by a series of 12 residents of the Eastern District of Texas. That's hard to put into Excel.

I hope that, like the failure of MPEG-4 part 2 and threat of VC-1 did for H.264, the failure of anyone to make money from VVC licensing and the threat of AV2 will inspire the patent holders to decide it's better to share slices of a big pie rather than fight over slices from a tiny shrinking pie.

kurkosdr

20th December 2025, 21:14

I hope that, like the failure of MPEG-4 part 2 and threat of VC-1 did for H.264, the failure of anyone to make money from VVC licensing and the threat of AV2 will inspire the patent holders to decide it's better to share slices of a big pie rather than fight over slices from a tiny shrinking pie.
Nope, not gonna happen. The patent holders around MPEG know that AOM will keep shrinking the pie left for MPEG standards no matter what. Even if someone has a "golden patent" that they manage to successfully assert against AV1 or AV2, it's much easier for the tech giants behind AOM to buy that single patent (or the entity that owns it) than deal with the licensing hell that is the recent MPEG standards (where every patented invention that makes a minor 0.1% difference is added to the standard and becomes a licensing landmine on its own). In plain English, MPEG standards are designed to infringe on as many patents as possible (not the intent, but that's what ends up happening when you don't take IP issues into account) while AOM standards adopt a more value-based approach of only including royalty-free technologies to achieve performance similar to MPEG standards. This of course means that AOM will keep expanding at the detriment of MPEG standards.

And again, the patent holders around MPEG know this. They know HEVC will be their last hurrah (HEVC is the lowest common denominator for HDR content, so it will be with us forever) and they know the only income they'll see from VVC will be from those few countries that adopt it for broadcasting and those high-end 4K TVs that feature VVC as future-proofing. Seriously, that's the pie for VVC, it's small and it won't get any bigger.

So, what did the patent holders around MPEG do with this knowledge at hand? Answer: They have entered "milking mode", trying to extract as much revenue as possible (by splitting themselves into multiple licensing entities and jacking up royalty fees as much as they can without technically running afoul of FRAND) before MPEG becomes too irrelevant financially to even bother participating in the meetings that produce the new standards.

Z2697

21st December 2025, 15:33

kurkosdr

22nd December 2025, 00:16

There's the speculation that they are going to milk HEVC as well, based on the Dell and HP allegedly disabled HEVC decoder with on laptops... (more like "shipped without that (https://apps.microsoft.com/detail/9n4wgh0z6vhq) M$ store extension" than disabled?)
So yeah things are not looking good for MPEG. Unless they can suddenly make MPEG codecs royalty free there's no way they are gonna compete with AOM.
My whole point was they have already gone into "milking mode" for HEVC because they know it will be their last hurrah. That's the reason they split into multiple patent licensing entities and then started jacking up the fees. Because that's how you milk an ISO standard without technically running afoul of FRAND. They also try to milk VVC, but as I said, that pie is very small, so no matter how hard they milk (and hard they are and will), what they'll get from VVC is paltry either way.

Again, they already know all of this. The knew it before the VVC spec was finalized (that streamers were looking at AV1 as their next-gen standard not VVC, that most broadcasters aren't going VVC, and that no 8K optical disc utilizing VVC will happen), in fact, they knew it after VP9 happened and showed the world Google wasn't planning to abandon their royalty-free codec efforts after VP8. They know MPEG is going the way of the Edison Trust/MPPC because it's now possible to have competitive video compression standards without having to "pool" hundreds or thousands of patents. So, much like the Edison Trust/MPPC, they try to make as much money as they can before their last essential patents expire. There is no reason for them to think of goodwill or the future of MPEG standards anymore, their only concern is to not technically run afoul of the "R" part of FRAND (emphasis on "technically"). So, we'll just have to tolerate the milking until their HEVC-essential patents expire.

birdie

24th December 2025, 08:35

Finally, there has been a major improvement in licensing: Access Advance acquires Via Licensing Alliance’s HEVC, VVC patent pools (https://ipfray.com/breaking-access-advance-acquires-via-licensing-alliances-hevc-vvc-patent-pools/) But there are multiple licensors beyond these two patent pools, according to Wikipedia they are: "Apple, Canon, Ericsson, Fraunhofer, Google, Huawei, Intel, Interdigital, LG, Maxell, Microsoft, Nokia, Oppo, Qualcomm, Samsung, Sharp and Sony".

The usual criticism remains however: What the Access Advance / Via LA Deal Doesn’t Change About Codec Adoption (https://streaminglearningcenter.com/articles/what-the-access-advance-via-la-deal-doesnt-change-about-codec-adoption.html)

However, it doesn’t make VVC necessary, and that’s the most critical factor in motivating short- or even mid-term codec adoption. It also doesn’t address concerns about Nokia-style licensing claims, which is the most crucial consideration for publisher adoption.

hajj_3

24th December 2025, 11:52

press releases from Access Advance and Via-LA merging their HEVC and VVC patent pools:

https://accessadvance.com/2025/12/15/access-advance-and-via-licensing-alliance-announce-hevc-vvc-program-acquisition/
https://www.accessnewswire.com/newsroom/en/electronics-and-engineering/access-advance-and-via-licensing-alliance-announce-hevc%2fvvc-program-acq-1117638

VCL Advance (Video Codec Licensing Advance) patent pool has now been formed, here are the fees:

https://i.ibb.co/Rkrx83C4/Screenshot-2025-12-24-112517.png

rwill

24th December 2025, 12:14

birdie

24th December 2025, 12:37

rwill

24th December 2025, 14:04

If only a few people on this forum use or are interested in the codec, then we will not get better encoders or decoders, and this codec will essentially become irrelevant or outright dead.

And the only way to launch it is to make it less scary, i.e. solve the licensing situation.

We were promised x266 when? Two years ago? It's seen zero public releases yet.

I don't think you people can "solve the licensing situation" by making posts on this board here. I think you have to be quite high up in large $$$ corporations and make the right decisions there.

Regarding x266 being late, that one is on Multicoreware. Maybe its political, maybe development is stuck, who knows - I suggest to not depend on their software.

Some time ago I implemented some basic VVC encoder. That took me maybe some months on and off doing some other things in between. Instead of lamenting about the licensing situation here you could have done the same if, instead of spending your free time on the licensing situation, you had honed your video codec programming skills in the past years. But I get that programming is not for everyone and that complaining is cheaper for the head.

GeoffreyA

24th December 2025, 15:12

I think the problem most in need of solving is that of grain.

ksec

26th December 2025, 09:15

The "common" RAM and SSD size is more than enough to implement a new decoder, I don't understand the connection here.
And if storage is expensive, it's more likely that people (except videophiles) will prefer "more advanced" space-saving codecs.
But, nowadays the biggest incentive behind codec adoption is probably bandwidth, since most of the video are streamed, and they are not really saving space for the platform, they just add another re-encode option to the stack because of compatibility needs.

The connection is that RAM and NAND are expensive and they have to save up cost on somewhere. SoC cost as well as decoding engine IP all goes together and may suffer.

kurkosdr

26th December 2025, 20:51

I wonder why you are all so interested in the IP situation regarding video codecs. As far as I know no one here is even remotely invested or has a business where licensing costs make up a above neglect-able amount. I mean people on this board are making lists and are really fixated on IP validity and lapse times - yet no one here is affected by them.

Well, that's the problem, the licensing mess for HEVC is so bad that we are affected by it.

Let me outline the mess in more detail:

- HP & Dell are disabling HEVC support on certain laptops (https://www.guru3d.com/story/why-hp-dell-are-disabling-hevc-support-on-certain-laptops/) because of the licensing mess. This means that you can't trust a new laptop bought from Best Buy (even when left on the pre-installed OS) to support HEVC out of the box. Imagine if we had new PCs in 2017 not supporting H.264 out of the box.

- Neither Windows 10 nor Windows 11 include HEVC support out of the box. Remember when anyone on Windows 7 could be trusted to have an H.264 system decoder provided by the OS? Nice times. This meant that, when XP went EOL, the majority of Windows PCs had an H.264 decoder at the OS level that browsers could use. Not so with HEVC (when Windows 7 and 8.1 went EOL). Ironically, a Senior Director at Microsoft had called the VPx codecs the "Esperado and Kligon" of codecs (https://web.archive.org/web/20110112232434/https://blogs.msdn.com/b/tims/archive/2011/01/11/an-open-letter-from-the-president-of-the-united-states-of-google.aspx) back in 2011, yet today, Microsoft relies on VP9 and AV1 as their next-gen codecs, and whatever value an HEVC decoder brings to the table, Microsoft apparently doesn't deem it high enough to justify the licensing cost of having an HEVC decoder in Windows 10 and 11 out of the box.

- Absolutely no browser for Windows comes with a bundled HEVC decoder (like Chrome came with an H.264 decoder), because again, no browser vendor considers the licensing cost to be worth it

Do you see now why some of us were highly sceptical of the "pass it down to system codecs bro" people (aka people like you)?

PROTIP: When the licensing situation of a standard is especially crap, system codecs cannot be assumed to exist at either the OS or GPU driver level, even on Windows.

Do you see now why some of us are glad that the open web was spared from the licensing mess surrounding HEVC?

Call me old-fashioned, but web pages (at least of the open kind, aka those without Widevine DRM) should be decodable by just an application (commonly called a "browser"), without the browser having to rely on system-level peculiarities that may or may not exist on a particular system for successful decoding. Sure, the browser can use system-level peculiarities for acceleration, but there should be an application-level fallback. H.264 kind-of has that in the form of Chrome's built-in H.264 software decoder, call me back when HEVC has the same on at least one browser available for Windows (the most popular OS for laptops and desktops).

And VVC is shaping up to be the same mess and worse.

---

About H.264, there are edge cases such as Desktop Linux systems not having OS-level codecs (because the GPU vendors can't be arsed to clarify whether the royalty is paid per hardware or per driver download), so there is some value in watching H.264 patents for expiration, since that would make the H.264 universal in all browsers, not just Chrome (which isn't available in all non-PC platforms, or you may prefer another browser). I was given a corporate laptop without H.264 codecs, and my preferred browser (Firefox) would occasionally give me codec errors, so I switched to Chrome. But I admit those are edge cases, H.264 has proven very popular on the open web because its availability is "good enough". Not something I personally like (I prefer universal to "good enough"), but the market has decided. HEVC failed to clear even that "good enough" bar, with former allies such as Windows (Microsoft) turning their backs to HEVC's licensing mess.

rwill

29th December 2025, 11:36

Ah I get it now. You don't want to pay money for certain things and are of the opinion that some things just should be free.

I too was leaning towards communism when I was younger but this went away once I moved out of my parents place and had to sustain myself.

kurkosdr

29th December 2025, 14:32

Ah I get it now. You don't want to pay money for certain things and are of the opinion that some things just should be free.
Lol, talk about not having a clue how things work on the web. It doesn't matter if your system has an HEVC decoder (and how many HEVC decoders you have bought), it matters whether your viewers have an HEVC decoder.

Please go ahead and make a page serving HEVC video and tell any Windows 10 and Windows 11 users that they might have to buy something called an "HEVC video extension" from the Microsoft Store and see how far you get. Instead, VP9 and AV1 have built-in decoders in Chrome, Firefox, and Edge. And yes, you can provide an H.264 fallback, but then you have the problem of viewers complaining 4K and HDR not working and the like. Why do that when VP9 and AV1 exist?

.I too was leaning towards communism when I was younger but this went away once I moved out of my parents place and had to sustain myself.
Yeah, sure, because Microsoft is well-known for being "communist". This is what you people fail to understand: The patent holders holding essential patents for HEVC somehow managed to piss off Microsoft (the company that used to mock the VPx codecs), we are talking about an achievement here. The non-inclusion of an HEVC decoder in Windows 10 and 11 was the inflection point that made HEVC irrelevant on the web for good (since you can't assume system codecs exist on Windows, even after Windows 7 and 8.1 EOL). With no bundled HEVC decoder in any Windows browser, that sealed the deal.

For the record, I was never leaning towards communism, then or now, I perceive HEVC's failure on the web to be a market-based one.

rwill

29th December 2025, 15:05

You are trying really hard to misunderstand me. You also now suddenly mention something called "the web".

Anyway, have a nice day.

Z2697

29th December 2025, 15:58

I guess we all know something called "the web"... How you've been here otherwise?

rwill

29th December 2025, 17:39

I guess we all know something called "the web"... How you've been here otherwise?

Sure, but what does "the web" need 4k and HDR for? kurkosdr is shifting goal posts again because he knows HEVC works well in TVs and Smartphones. It is widely accepted that TVs and Smartphones cost money and part of that money is for IP. While apparently with "the web" it is not acceptable to pay even a dollar for high end technology.

"the web" has its video standards but because these cannot fully replace what a 2k$ TV set has to offer certain people complain about companies that just want some return on their investment.

kurkosdr

29th December 2025, 18:45

Sure, but what does "the web" need 4k and HDR for?
Quite frankly, not much, most people watch movies on Smart TVs and smartphones today. But VP9 and AV1 is there is you want to watch your YouTube and Twitch videos in 4K HDR from your browser.

kurkosdr is shifting goal posts again because he knows HEVC works well in TVs and Smartphones. It is widely accepted that TVs and Smartphones cost money and part of that money is for IP.
If so, how much VVC video is streamed to TVs and Smartphones? Looks like HEVC support in TVs and smartphones is a backwards compatibility concern and ISO/ITU formats are mostly dead after HEVC, and the next step is AV1. Funny how the market has a way of navigating away from expensive IP that fails the cost-benefit analysis.

While apparently with "the web" it is not acceptable to pay even a dollar for high end technology.
Yes, it's been that way since Microsoft started giving away Internet Explorer for free (which was high-end technology at the time), anyone who tried to ignore that reality ended up bankrupt (Netscape).

In plain English, people expect browsers to be either bundled for no extra cost with the OS or be available for download for free. When it comes to Windows, neither the browser bundled with Windows (Edge) nor any of the popular freely downloadable browsers offers built-in HEVC decoding. That's a market-based failure for HEVC right there, considering H.264 had Chrome. Also, Windows doesn't even offer system codecs for HEVC, so relying on the presence of system codecs is unreliable.

Z2697

29th December 2025, 18:54

"the web" has its video standards but because these cannot fully replace what a 2k$ TV set has to offer certain people complain about companies that just want some return on their investment.

Except that "the web" controls TVs and phones now. (most of)

kurkosdr

29th December 2025, 19:07

Except that "the web" controls TVs and phones now. (most of)
Sure, but why make a webpage that only works with TVs and phones when AV1 is universal?

modus-ms325c

2nd January 2026, 17:51

technically speaking, you are correct, Netscape did get bankrupt, for the exact reasons you mentioned above and then some, albeit very slowly.

however, i remember there being a massive mutiny on E.M. Meen's X platform (back when it wasn't actually such, 5-6 years in fact) over a tweet someone did on how they were involved in the final hours of shipping IE6. (or was it IE5?) either way, to put it very mildly, it was widely seen (by Netscape devs and its users, albeit they're not necessarily the only ones to emit such backlash, as will be described now) as Microsoft's inability to admit responsibility on the damage they've caused on shoving their buggy, unstable, and quite frankly unsecure and/or just straight-up paranoid browser onto everyone's PCs, one that saw many pages require it even. though such effects would not be felt until much later.

aside from that, i remember a time when simply locking videos behind an Adobe Flash install message or any other piece of external decoder software (think RealMedia, Windows Media, etc. and i mean actual software) was the worst you could've done on "the web". if nothing else, it was your choice to either install the required components to watch a video or not at all, tho in all honesty, that wasn't a choice so much as a mandatory requirement for seeing anything at all there.

as of today, however, should your device contain video+audio decoding support for free without you even noticing even slightly, that choice is now forced on you.

"the web" also did not support MPEG codec stuff at all, in any way, whatsoever unless licensed to, and bundled, with external software that you also had to install, but hey, at least it was free.

do go on in an self-destructive performative outrage rampage on how communism is bad tho.

kurkosdr

2nd January 2026, 20:45

do go on in an self-destructive performative outrage rampage on how communism is bad tho.
Communism is bad in the sense it doesn't work in the real world, it requires either perfect people or a perfect government (depending on how you define it). It's like designing a wireless system that assumes zero signal attenuation, or designing a chip assuming zero signal propagation. Designing such a system, unconstrained from the constraints of the real world, is a blast, but has no practical application. But please do go on in an self-destructive performative outrage rampage on how this isn't true.

But why are we even discussing this? Ah yeah, because someone brought up communism as a distraction tactic.

---

Back in the real world, the failure of HEVC on the web is a failure of the patent holders holding essential patents to understand the market around the web. Again, H.264 had a browser for Windows (Chrome) that supported it consistently regardless of the presence of system codecs, and H.264 is free of content fees for free-to-view web video, HEVC has neither of those.

GeoffreyA

2nd January 2026, 21:16

kurkosdr

2nd January 2026, 21:37

In my humble and uninformed opinion on patents and such matters, HEVC should have become the baseline years ago, but it hasn't happened.
It would have if the patent holders had picked the same licensing model as H.264: comfortable single-pool licensing with a royalty cap (so companies like Microsoft that often sell their OS for peanuts to volume licensees and OEMs don't have to worry about per-copy royalties to multiple pools and can instead write one large check to a single pool and even offer HEVC as a bundled codec in Edge browser for Windows 7 and 8.1 users) and waive content fees for free-to-view content.

But no, they got greedy and as a result are now stuck with broadcasters and Blu-ray instead, both very conservative markets not looking to move to VVC any time soon. And let's not forget, premium streamers like Netflix that are not looking at VVC as their next gen standard.

But I believe this has been discussed to death.

modus-ms325c

2nd January 2026, 23:53

ok, after a bit of thought in how i responded (plus i don't intend to shill communism nor will i denounce it, let alone even talk about it anymore as it's against doom9 rules and rwill should've known better in how to respond anyway) i decided it's time not to engage with this thread anymore and only lurk it instead. i feel, it's best not to milk the "VVC sucks cuz' MPEG sucks cuz' they milk VVC dry with patents and whatever else" cow here, and i will also not discuss on whatever "technical brilliance" VVC has going for it either because, let's be honest, "technical brilliance" is way past its prime and is likely already exhausted in terms of anyone even remotely talking about it, anyway, so who cares.

as for the things i'm willing to engage in? well, once the inevitable AV2 codec comes out with certain (officialy distributed) videos containing not only it but H.264, VP9, AV1 video tracks (streams in YouTube terms) as well... once i discover that the AV2 stream is significantly worse in every way in terms of video quality, detail retention, or literally any kind of retaining whatever imperfections AV2 is capable of doling out compared to the aforementioned codec tracks... i promise that i'll raze this fucking site with the most insane rant against AV2 you'll ever witness. but not here, and most certainly not now.

peace out.

benwaggoner

5th January 2026, 17:51

i will also not discuss on whatever "technical brilliance" VVC has going for it either because, let's be honest, "technical brilliance" is way past its prime and is likely already exhausted in terms of anyone even remotely talking about it, anyway, so who cares
That seems to be an idiosyncratic take! How does good design not matter? As long as network bandwidth and storage are not infinite, getting better compression efficiency gains with less increase in SoC mm^2 is always going to matter.

once i discover that the AV2 stream is significantly worse in every way in terms of video quality, detail retention, or literally any kind of retaining whatever imperfections AV2 is capable of doling out compared to the aforementioned codec tracks... i promise that i'll raze this fucking site with the most insane rant against AV2 you'll ever witness.
Given we only have in-development reference encoders for AV2, and not any commercial-grade psychovisually tuned one, I think it's challenging to say much about how good AV2 will look once we have those things in a few years. HEVC was kind of a unique case in getting practical encoders so early in its lifespan, as it was really the only practical option to get 4K HDR, which the content and consumer electronics companies were aligned with as a big way to make a lot more money. None of the codecs since then, or that I see looking forward, really have any new content types they unlock; they mainly are to do what we already can at lower bitrates.

benwaggoner

5th January 2026, 18:02

It would have if the patent holders had picked the same licensing model as H.264: comfortable single-pool licensing with a royalty cap (so companies like Microsoft that often sell their OS for peanuts to volume licensees and OEMs don't have to worry about per-copy royalties to multiple pools and can instead write one large check to a single pool and even offer HEVC as a bundled codec in Edge browser for Windows 7 and 8.1 users) and waive content fees for free-to-view content.

But no, they got greedy and as a result are now stuck with broadcasters and Blu-ray instead, both very conservative markets not looking to move to VVC any time soon. And let's not forget, premium streamers like Netflix that are not looking at VVC as their next gen standard.
4K Blu-ray was the last optical disc format and HEVC the last codec to be adopted in one. The core technology simply doesn't make sense anymore. Consumer broadband can easily exceed Blu-ray read speeds and with much better latency, and flash memory is approching the cost of manufacturing a Blu-ray disc. Bear in mind that even a triple-layer 4K is only 100 GB max, and I can get a 128 GB flash drive for about $8/each in bulk, and under $4 for one equivalent to a dual layer disc.

It was an eye opener for me a few years ago when I discovered I could install a game about 3x faster as a download than off a disc. UHD Blu-ray spec is only 144 Mbps, and the current game consoles go up to maybe 400 Mbps maximum in ideal circumstances, and often less in practice. Glacial compared to gigabit internet.

benwaggoner

5th January 2026, 18:08

In my humble and uninformed opinion on patents and such matters, HEVC should have become the baseline years ago, but it hasn't happened. Indeed, AV1 seems closer to being a universal successor to H.264, and that momentum will likely carry on with AV2.
In what markets do you see AV1 becoming universally used outside of perhaps YouTube? Most HDR streaming is still HEVC as far as I can tell.

I fear that we'll never see a codec as universal as H.264 was, where we've been able to assume for 10+ years that anything that can play back video has a High Profile hardware H.264 decoder. We still are seeing new pretty high volume lower cost phones being sold without AV1 decode, over halfway to HEVC's patents expiring. I think AV2 is more likely as a universal successor by 2035 or so. Phones have a shorter replacement cycle, but people are holding on to PCs and especially home electronics for longer and longer. There are a lot of very good high end TVs only a couple of years old for which HEVC will remain the best available decoder.

benwaggoner

5th January 2026, 18:14

Yeah, sure, because Microsoft is well-known for being "communist". This is what you people fail to understand: The patent holders holding essential patents for HEVC somehow managed to piss off Microsoft (the company that used to mock the VPx codecs), we are talking about an achievement here. The non-inclusion of an HEVC decoder in Windows 10 and 11 was the inflection point that made HEVC irrelevant on the web for good (since you can't assume system codecs exist on Windows, even after Windows 7 and 8.1 EOL). With no bundled HEVC decoder in any Windows browser, that sealed the deal.
Yeah, this is a good take. And HEVC decode was a free built-in feature until very late in Windows 10 development! Whichever NPE holdouts unwilling to give Microsoft an acceptable license for Windows really shot themselves in the foot. HEVC made enough money in consumer electronics that the lesson was apparently not signaled clearly enough for VVC, leaving some companies fighting brutally hard for a bigger slice of a very tiny pie. Perhaps that will discourage some of the more patent-trollish companies from investing in getting much H.267 IP. The companies that own the bulk of the patents are being appropriately cooperative, but all it takes is a single essential patent out of thousands held by the wrong entity to shit in the pool so everyone gets kicked out.

GeoffreyA

5th January 2026, 23:01

In what markets do you see AV1 becoming universally used outside of perhaps YouTube? Most HDR streaming is still HEVC as far as I can tell.

I fear that we'll never see a codec as universal as H.264 was, where we've been able to assume for 10+ years that anything that can play back video has a High Profile hardware H.264 decoder. We still are seeing new pretty high volume lower cost phones being sold without AV1 decode, over halfway to HEVC's patents expiring. I think AV2 is more likely as a universal successor by 2035 or so. Phones have a shorter replacement cycle, but people are holding on to PCs and especially home electronics for longer and longer. There are a lot of very good high end TVs only a couple of years old for which HEVC will remain the best available decoder.

Thanks for the response, Ben. No doubt, my statement on AV1 was too strong; HEVC reigns in markets such as HDR streaming and optical media, and AV1 hardware decode isn't universal. libdav1d helped immensely, but more battery is going to be used on many a device. Pre-2020 TVs likely have no support.

Outside YouTube, it has seen a bit of success on the high seas, particularly in anime. Netflix is serving about 30% and deploying FGS. I would say the codec feels close in spirit to H.264, and there is a great deal of fondness for it, despite its early struggles in performance and detail. Then, the popular SVT-AV1 forks, where we haven't seen such relentless drive since x264. I can only see AV1's use increasing, and hopefully AV2 will inherit a lot of the work done, say, in the tuned encoders.

modus-ms325c

6th January 2026, 02:48

That seems to be an idiosyncratic take! How does good design not matter? As long as network bandwidth and storage are not infinite, getting better compression efficiency gains with less increase in SoC mm^2 is always going to matter.VVC licensing situation is being talked about all the time at just about every opportunity, no one's gonna focus on the actual technological features that it has right now.
more like how the codec itself is being held hostage by its insane ecosystem. i mean, just by talking about its insane features alone already makes me bored to fucking tears already.
Given we only have in-development reference encoders for AV2, and not any commercial-grade psychovisually tuned one, I think it's challenging to say much about how good AV2 will look once we have those things in a few years. HEVC was kind of a unique case in getting practical encoders so early in its lifespan, as it was really the only practical option to get 4K HDR, which the content and consumer electronics companies were aligned with as a big way to make a lot more money. None of the codecs since then, or that I see looking forward, really have any new content types they unlock; they mainly are to do what we already can at lower bitrates.i mean, ok, but "AV2 is in development so we don't really know" is one thing.
4K Blu-ray was the last optical disc format and HEVC the last codec to be adopted in one. The core technology simply doesn't make sense anymore. Consumer broadband can easily exceed Blu-ray read speeds and with much better latency, and flash memory is approching the cost of manufacturing a Blu-ray disc. Bear in mind that even a triple-layer 4K is only 100 GB max, and I can get a 128 GB flash drive for about $8/each in bulk, and under $4 for one equivalent to a dual layer disc.

It was an eye opener for me a few years ago when I discovered I could install a game about 3x faster as a download than off a disc. UHD Blu-ray spec is only 144 Mbps, and the current game consoles go up to maybe 400 Mbps maximum in ideal circumstances, and often less in practice. Glacial compared to gigabit internet.you do know for a fact that UHD Blu-ray playback is heavily reliant on your internet connection, right? you do know that for a fact, right?
(legal) digital download services can also reach service termination at just about any time now, and existing UHD-BD discs may become a massive useless paperweight in leau to DIVX DVDs, just in case UHD-BD DRM stops working and/or ends service sooner than expected.
Yeah, this is a good take. And HEVC decode was a free built-in feature until very late in Windows 10 development! Whichever NPE holdouts unwilling to give Microsoft an acceptable license for Windows really shot themselves in the foot. HEVC made enough money in consumer electronics that the lesson was apparently not signaled clearly enough for VVC, leaving some companies fighting brutally hard for a bigger slice of a very tiny pie. Perhaps that will discourage some of the more patent-trollish companies from investing in getting much H.267 IP. The companies that own the bulk of the patents are being appropriately cooperative, but all it takes is a single essential patent out of thousands held by the wrong entity to shit in the pool so everyone gets kicked out.oh my god lol

kurkosdr

6th January 2026, 23:10

you do know for a fact that UHD Blu-ray playback is heavily reliant on your internet connection, right? you do know that for a fact, right?
(legal) digital download services can also reach service termination at just about any time now, and existing UHD-BD discs may become a massive useless paperweight in leau to DIVX DVDs, just in case UHD-BD DRM stops working and/or ends service sooner than expected.
Huh? When it comes to on-disc content, UHD Blu-rays can be copied 1:1 with the help of a tool like DVD Fab (the paid version obv) and a "friendly" drive (that is, a drive that implements BDXL but doesn't implement UHD Blu-ray's DRM that locks "unauthorised" software out of the drive). So, that's your guaranteed "insurance" against the discs becoming a paperweight.

Also, UHD Blu-ray DRM is 100% local, it doesn't need to "phone home" to decrypt a disc. In plain English, you can do it in a submarine with a new disc that has never been in the drive before, as long as your player has been updated beforehand. So as long as you have a legit disc and an updated player, you own the content.

modus-ms325c

7th January 2026, 02:12

Huh? When it comes to on-disc content, UHD Blu-rays can be copied 1:1 with the help of a tool like DVD Fab (the paid version obv) and a "friendly" drive (that is, a drive that implements BDXL but doesn't implement UHD Blu-ray's DRM that locks "unauthorised" software out of the drive). So, that's your guaranteed "insurance" against the discs becoming a paperweight.

Also, UHD Blu-ray DRM is 100% local, it doesn't need to "phone home" to decrypt a disc. In plain English, you can do it in a submarine with a new disc that has never been in the drive before, as long as your player has been updated beforehand. So as long as you have a legit disc and an updated player, you own the content.OK, let's see how reliable these methods are in ~4-20 years, assuming internet access hasn't gone out the way of the wazoo and/or isn't just out-and-out dead anymore.

kurkosdr

7th January 2026, 22:18

OK, let's see how reliable these methods are in ~4-20 years, assuming internet access hasn't gone out the way of the wazoo and/or isn't just out-and-out dead anymore.
Tools like DVDFab require online activation, so if DVDFab's online activation service goes down, yeah, the tool won't work anymore on new computers.

But as long as DVDFab's activation service is online, there is nothing Hollywood can do to prevent UHD Blu-rays from being copied 1:1 (with the right "friendly" drive), because again, the content is on the frickin' disc and is encrypted using a long-compromised encryption system, it's not dependent on a Hollywood-provided service or server.

And let's not forget that tools such as XReveal exist that don't require online activation. While it misses some features (such as removal of Screen Pass/BD-J menu countermeasures), you can still use it to copy just the main movie (again, with the right "friendly" drive).

And let's not also forget that you can buy a UHD Blu-ray player, update it so that it plays all your existing discs, and never update it/connect it to the internet ever again, since UHD Blu-ray playback requires no internet connection/updates/phoning home for discs pressed before the currently installed update was released.

What's even the debate here? That theoretically, the entire internet can go down, and anyone not already having a copy of Xreveal won't be able to download it? Or that they won't be able to buy a UHD Blu-ray player? Under this scenario, the same applies for tools like VLC, so... even VCDs are theoretically "internet dependent" since they are not playable if you don't already have a copy of a tool to play them or a standalone player.

benwaggoner

10th January 2026, 00:55

i mean, ok, but "AV2 is in development so we don't really know" is one thing.
The AV2 bitstream hasn't been quite locked down, but it has been feature complete in terms of content types for a long time now. More might get added later; MVC-HEVC wasn't part of the HEVC 1.0 spec. But what AV2 is and can do with what profiles at launch is well define.

you do know for a fact that UHD Blu-ray playback is heavily reliant on your internet connection, right? you do know that for a fact, right?
No, A Blu-ray player can still play a movie without any internet connection. Are you thinking of BD-Live content or something? That's orthogonal to actually playing back the video and audio content on disc. A blu-ray player may require an internet connection for setup, but I don't know of any mainstream discs or players that won't playback the content without internet connectivity. Bear in mind Blu-ray was designed a couple of decades ago when the idea of having always-on internet in every car or vacation cabin was years from happening. The first Blu-ray players and discs were launched a year before the first 2G-only iPhone.

modus-ms325c

10th January 2026, 19:50

i was exclusively talking about Ultra HD Blu-Ray, but yeah BD-Live definitely checks out.

kurkosdr

10th January 2026, 20:23

i was exclusively talking about Ultra HD Blu-Ray, but yeah BD-Live definitely checks out.
And I was also exclusively talking about Ultra HD Blu-Ray, and I am telling you again that the content is on-disc and encrypted with a long-compromised encryption system, with commercially available software to crack said encryption system. Also, authorized players require no phoning home to view the disc, as long as the player has been updated with a firmware update released after the disc was pressed.

BD-Live is used for extra content (when it's used), or more appropriately extra extra content, since Ultra HD Blu-rays have a good amount of extra content on-disc in addition to the main movie. UHD Blu-rays (not just plain Blu-rays) were designed so that the on-disc content is viewable without an internet connections.

So yes, when you buy a UHD Blu-ray disc, you aren't reliant on a remote server or even an internet connection for the on-disc content, not anymore than a regular Blu-ray anyway.

/end off-topic