My desire for royalty-free "open" codecs has nothing to do with that, though. I know it has a lot to do with why Google and Netflix care about royalty-free codecs, but it has nothing to do with why I do, or why I think they are very important to an open Internet.

Case in point, consider Wikipedia.

May be it should be patent free video codec rather than royalty free video codec?

Patent-unencumbered is the term that I should've probably used, but as you will see in the neighboring reply, there is no reason to believe anyone would pull on this thread in good faith. This sucks, because I'd rather assume good faith, but.

Can you name one “patent unencumbered” codec?

And “Words Mean Things”. There is a huge difference between “open source implementations” and “patent unencumbered” codecs

Can you accept that I'm not sincerely engaging with this line of thinking? I know where it leads. I'm not stupid.

What line of thinking? If you said you want to ride a unicorn, I would ask you where can you find one

I strongly, strongly dislike what you are doing. You know exactly what I mean when I say it, but are pretending not to, even explicitly suggesting 'ah, just because the codec's open source doesn't mean it's not patent encumbered' without explicitly saying what you really want to say, just for the sake of allowing me to activate a "trap card" wherein I say "Ahh, no, I don't mean like x264, I mean like VP9!" and then you can flip around and say "Ah HAH! Actually, VP9 ISN'T patent-unencumbered: check out all of this propaganda from a literal patent cartel about how if you use VP9, they will bust your ass sideways, unless you license it using their patent pool!"

And I don't care. I'll care when Wikipedia gets sued for using Opus, or AV1, or when Mozilla gets sued for shipping it in their browser, or something. Until then, as far as I'm concerned, it's empty threats from dog shit liars.

But this aside, I don't understand why you can't just say what you mean right out the gates since I know you know what I am talking about, I know you knew from reply 1, and I'm sure you've already played through this conversation before. I could've said Theora, Quite OK Audio or perhaps even MPEG Layer 3 just to fuck around, but I really would like to have an honest conversation, and not just mess around with technicalities.

I'm not interested in debating the legal system, or how AOM's license agreement works, or any of that crap. In practice, almost everyone can use "open", patent-unencumbered codecs without fear of lawsuits. The companies big enough to have to worry about patent lawsuits virtually no matter what have basically been using patent agreements as a weapon against the traditional patent cartels. This is of course, deeply unfortunate, since it just perpetuates the terribly broken U.S. patent system and the situation it has with video codecs.

I understand that you have Opinions:tm: regarding this, I already caught on. But what you have to understand is that to most people, including plenty of lawyers, these opinions are essentially fringe. If they weren't, AV1 wouldn't be shipping in Firefox and Wikipedia wouldn't be using them. And not only that, it has nothing to do with the concept of patent-unencumbered codecs. There's always going to be codecs where the known patents have simply expired. At least in the U.S., I'm pretty sure that even includes MPEG2, but it'd suck if open source software was limited to just MPEG2. (And of course, if we do find new patents not covered under previous patent pools, that happen to apply, you have to realize that this liability existed for everyone already, including people who paid royalties anyways.)

I mean that if I’m a business the size of Netflix, I very much care about the difference between “patent encumbered with an open source” implementation and “patent unencumbered”. In the first case, a patent troll can come after me.

On the other hand, if I use H.264 as a large business and I choose to modify the open source implementation and pay the royalties, I’m mostly covered.

Theora was last updated over a decade ago and currently no processor has built in support for it making it a non startee on mobile devices where battery life is at a premium.

And if AV1 meets your definition of “open” why are you complaining about Apple not supporting an “open codec” since Safari supports AV1?

While you may not be interested in talking about the legal system, I can guarantee you that every major company is before they re-encode their library using a codec or add die to the chip to optimize decoding.

Big companies can’t use their patents against small patent trolls who are just holding companies built just to sue large companies.

In other words, it is financially dumb for any large company to support an “open” protocol based on 20 year old technology (when patents expire) or newer supposed “open” codecs that are just as patent encumbered.

I said iPhone/iPad not Safari. The difference is imperative, since Safari on iOS does NOT support AV1. I have a fully patched iPad, I'm absolutely 100% certain.

Anyway, if it's good enough for Mozilla and Wikipedia and thousands of smaller companies, it's good enough for me. You do you.

You may have a “fully patched iPad”. I assume you don’t have an iPhone 15 Pro or Pro Max with hardware decoding of AV1…

https://bitmovin.com/apple-av1-support/

> The WebKit blog provided more information about Safari 17.0, confirming support for the AV1 video codec was added on devices with hardware decoding support

And if you don’t believe the blog, maybe you believe the caniuse site.

https://caniuse.com/av1

That is genuinely great news but:

- Only iPhone 15 Pro? not even an iPhone 15? This means almost nothing. You can't just be like, "Oh my site only works on every device except iPhones except iPhone 15 Pro.

- Still no Opus...

I will fully admit that I am out of date on this, by like a couple months. But the reason why I say it like this is because people have been making this claim for years even though it has apparently been true for like two months. Apple is half-assing this every step of the way. It will be like, 2025 before a site can just use AV1 without needing multiple encodings, and at that point you probably still won't be able to use Opus.

Thanks, Apple?

Because only the iPhone 15 Pto series chip has hardware decoding support for AV1. You’ve had to have code for fallback support for other codecs for years.

And Opus support has been around since iOS 11

https://caniuse.com/opus

There is no such thing as an “open” codecs. There are open source implementations of patent encumbered codecs.

Uh huh.

Name one.

No thanks.

Software patents are only valid in jurisdictions that believe in them. Fortunately the EU still doesn't, and that's why VLC is at all possible.

So as long as you don’t care about the US market, you as a business can ignore software patents…

Yep, and that's a few billion people. I think it's alright.

Yes if only money and disposable income was equally distributed

https://www.forbesindia.com/amp/article/explainers/top-10-la...

This seems correct, but probably leaves most readers with the wrong impression. Theora is a codec that is free and open, and the patents that cover it do not encumber it (since they are also freely licensed), so I'd consider it patent unencumbered.

https://www.theora.org/benefits/

So do you think it would be a good use of die space on a processor to support a substandard by today’s standard codec that was last used a decade ago?

It is a complete non starter for Apple or any mobile device maker to build in support for a non hardware optimized codec on a mobile device.

How does that benefit users besides the ones who bemoan not being able to run Linux on the Apple Pencil?

> So do you think it would be a good use of die space on a processor to support a substandard by today’s standard codec that was last used a decade ago?

There's a specific reason why Theora was last used a decade ago... Apple. Apple is the reason. There was some other opposition, but in 2009, Apple drove the hardest against Theora being included in the HTML5 spec, and they ultimately won, leaving us with no standard codec everyone could actually agree on. Because the web is supposed to be open, and h264 is not.

It being excluded from the spec, you could not rely on Theora support in the browser.

> It is a complete non starter for Apple or any mobile device maker to build in support for a non hardware optimized codec on a mobile device.

Given that a RPi 5 takes less than a full core to decode h264, I'm not even sure this is a real concern. In my tests, ogv.js is able to decode VP9 at modern resolutions on an iPad without even really making the device warm; it's undoutedly inefficient, but the fact that it can even be done using WebAssembly gives me a feeling that this isn't really the huge problem it's made out to be.

For Theora... I mean. It's just not an issue. You can just do it in software. Maybe not in 2009, but certainly today, right now, on any hardware made in the past few years, on one core, without even driving the clocks up.

For AV1, yes, sure, this is a real issue; but Apple has been a part of the AOM for years, so I have no idea why they are building machines without hardware AV1 acceleration today. That's a question that only Apple can answer, given that it has real adoption already.

> How does that benefit users besides the ones who bemoan not being able to run Linux on the Apple Pencil?

Yes sure, let's take a random unrelated detour to make fun of tinkerers who like to be able to reads notes run their own software on hardware they purchased. I'm disappointed, although unfortunately not surprised at all.

> There's a specific reason why Theora was last used a decade ago... Apple. Apple is the reason

The last stable release was in 2009. The iPhone was still a bit player then, only available via a few carriers and still behind RIM in market share. No one cared about Apple’s opinion in 2009.

> For AV1, yes, sure, this is a real issue; but Apple has been a part of the AOM for years, so I have no idea why they are building machines without hardware AV1 acceleration today. That's a question that only Apple can answer, given that it has real adoption already.

https://bitmovin.com/av1-playback-support

> Yes sure, let's take a random unrelated detour to make fun of tinkerers who like to be able to reads notes run their own software on hardware they purchased. I'm disappointed, although unfortunately not surprised at all.

How is not supporting an “open standard” stopping anyone from “tinkering”? There are open source implementations of the encoders and they will be decoded in hardware once you install whatever you are tinkering with on your phone.

I was simply responding to the query about a non-patent-encumbered codec.

for the Raspberry Pi, they made paying for the codecs optional. So you had the option to purchase a separate license just for that single device you buy.