Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Before someone chimes in with that XKCD strip:

How does Opus compare to other codecs?

Opus is distinguished from most formats for high quality audio (AAC, Vorbis, MP3) by having low delay and it is distinguished from most low delay formats (G.711, GSM, Speex) by supporting high audio quality. It meets or exceeds existing codecs' quality across a wide range of bitrates, and it operates at lower delay than virtually any existing compressed format. Further, the Opus format itself and the reference implementation are available under liberal royalty-free licenses, making it easy to adopt, compatible with free software, and suitable for usage as part of the basic infrastructure of the Internet. See the Opus comparison page for more details.

http://wiki.xiph.org/OpusFAQ#How_does_Opus_compare_to_other_...

Which is to say, this isn't (okay, isn't just) a case of having "too many standards"; Opus is legitimately impressive from a technical standpoint. Oh, and it's unencumbered by patents. That's cool.

EDIT: And here's the coolest part of the article that completely took me by surprise:

"The IETF recently decided with “strong consensus” to adopt Opus as a mandatory-to-implement (MTI) codec for WebRTC, an upcoming standard for real-time communication on the web."



> And here's the coolest part of the article that completely took me by surprise: "The IETF recently decided with “strong consensus” to adopt Opus as a mandatory-to-implement (MTI) codec for WebRTC, an upcoming standard for real-time communication on the web."

Indeed. There may be interesting side consequences for that. The first article comment asks:

> Interesting! Any idea if this will become mandatory to implement for HTML5 audio as well? That’s another area in sore need of a single format that works everywhere.

This means that if a browser supports WebRTC, it has to support Opus, and it then happens that the codec will be available for <audio>. Now that does not mean that <audio> => Opus support, but it will have a (increasingly with time) high chance to be so.


I don't know which XKCD strip you're referring to; link?


http://xkcd.com/927/

Keep in mind that this strip is a strong contender for the title of "Most Frequently-Posted Link in HN Comment Threads."

That would be an interesting analysis, actually.


There is no such thing as "unencumbered by patents". Unless you have a technology that has already been tested thoroughly in the courts, which is the only way you can really determine whether something would be deemed to violate a patent or not, you simply can't know.


Even if it was tested by the courts, you can appeal. And even after the appeal, you can appeal to a higher court, all the way to the Supreme. And even if was tested there, congress can pass a law that allows it. And even if that happens the Supreme court can strike down the law. And even if that happens, the public can mount a revolution and tear down the legal system.

But, practically speaking, Opus has extremely good guarantees for being unencumbered, as much as is possible - the people working on it have given licenses to use it, and experts in the field have had a while to go over this stuff. Nothing is 100%, but this is as good as it gets.


Yeah. The public is going to mount a revolution over a music format. I don't think being contrary for its own sake is helpful here.

The whole point of "tested by the courts" is that it's already gone through the appeals process. As in, you have the final decision you're going to get out of the courts.


The courts determine the applicability of single patents, or a single parties patents (e.g. in cases of misconduct). There is no legal process that can prove something non-infringing, unless you expect something to go to court millions of times.

But this doesn't mean that something can't be designed in a way which avoids all known, and much unknown risk, and licensed in a way which makes frivolous attack less likely.

The obsessives here somes forget that when we talk about comparing things what matters is the relative merits. The commercial codes are absolutely and uncertainly encumbered. And the commercial codec providers have generally had a poor track record of creating codecs unencumbered by third party rights (mostly because they don't make an effort to: they erroneously assume the regular royalties will satisfy all comers). Opus doesn't have these issues and even compared to most software an effort was made to avoid infringement, but nothing can escape the impossibility of absolute certainty under the current patent system.


No matter how many times I'm downvoted, it's still not unencumbered. It's simply not in the nature of our patent system that something this complex can be unencumbered, unless it's with respect to an expired patent.

I know my position causes a lot of cognitive dissonance with people, because it implies that the system is fundamentally unfair. Well, you can either take the red or the blue pill.


> No matter how many times I'm downvoted, it's still not unencumbered.

Who said you were wrong? Quite the opposite, I strongly agreed with you - anything can be encumbered, and anything includes this.

But the point is that Opus is as safe as you can possibly be from being patent encumbered. Everything is potentially encumbered, but there are things with higher or lower likelihoods. And there isn't any lower than Opus. If this is still too worrying for you to use, then you can't use any software at all.


Come on, it's only 45kloc. Do you go around insisting every non-trivial piece of software on hacker news is encumbered?

And in most those cases people were not making a special effort to avoid being effectively encumbered, while Opus was.


if its a non-trivial peice of software, it almost certainly is encumbered, don't forget about idiotic patents like 1 click, there are likely millions.


I believe the system is fundamentally unfair, but also think that you are the one that's bought into their propaganda to the point that you're not thinking straight.

Particularly as you naively think that an expired patent is enough to protect you from the system.


"Particularly as you naively think that an expired patent is enough to protect you from the system."

Note that I said with respect to...


It is unencumbered, unless someone proves that it is encumbered.


It is unencumbered, unless there's a ton of money to be made by suing them. Same goes for Theora. When some deep-pocketed hardware manufacturer dares put one of these "unencumbered" algorithms into its hardware, then we will see. But I doubt they'll dare.


Let them present their patents and try to sue, then you can say something is encumbered. Until now they engaged in empty threats and demagogy. They make tons of money by dominating the market, and when some potential completion shows up they try to prevent its adoption by spreading empty FUD. So it has nothing to do with patent encumbrance, it's just about their greed and thirst for market domination.


They won't present their patents until there's a big haul of cash to be extracted. There won't be a big haul of cash to be extracted until some deep-pocketed manufacturer implements the algorithms. Which they likely won't do because they understand the same thing I do -- the system is rigged.


They won't present their patents until there's a big haul of cash to be extracted.

Algorithms are already implemented and software which uses them is distributed broadly. Where are their court cases? Until those patent trolls (trolls as in spreading FUD) actually show what those patents are, they can't say that anything is encumbered. And they themselves have no idea about what they are. So whatever lies they spread now, even if it comes from Jobs, is simply irrelevant.


Show me an entity, any entity, that could be sued for lots and lots of cash, that is using these "unencumbered" algorithms.


You can easily find VP8 used on Youtube. Not big enough? Those predators wouldn't hesitate to sue, if they'd have something to bite with. But they have no teeth.


It's way too early to be suing with merely YouTube, no network effects. Google can easily switch away.


You can come up with any kind of arguments or guesses about why they don't sue, but the bottom line - they didn't provide any proofs, so their FUD doesn't worth anyone's time, and these codecs can be considered unencumbered and free to use (and that's what's happening to their displeasure).


I think you're engaged in wishful thinking. I wish you were right too.


According to another comment further down the page, Tegra 3 and OMAP4 have "hardware" VP8 decoding support. I've also heard that a patent holder can't sue for damages that occur long after they were made aware of infringement (it would be nice if one of HN's regular lawyers would jump in and define estoppel).

There are far more interesting and HN-worthy things to discuss than the non-Opus-specific fact that the patent system is broken, like its potential applications in WebRTC, the hybrid design that merges Celt and Silk, the fact that having a conversation with someone 50 feet away using Opus can have lower latency than just shouting at them (as demonstrated by Mumble's use of Celt beating the speed of sound down a hallway), etc.



No updates for 18 months. Did they find any patents?


The mere threat is enough to keep people from investing much in these "alternatives." E.g., Steve Jobs wrote:

"All video codecs are covered by patents," Jobs wrote from his iPad. "A patent pool is being assembled to go after Theora and other 'open source' codecs now. Unfortunately, just because something is open source, it doesn’t mean or guarantee that it doesn’t infringe on others patents. An open standard is different from being royalty free or open source."

You can bet that if Theora ever got traction and seriously interrupted MPLA's revenue stream, there would be suits. Before that time it'd be pointless for them to sue.


That's the whole point. They have no proof to show, because really those codecs are not violating any patents. But they want to create fear to ban competition.



The citation above was about video codecs, FUD about which is spread by MPEG-LA, who never showed any patents as a proof.

Opus is another story, since Qualcomm and Huawei listed supposedly violated patents, so Opus creators were able to review them and to confirm that those claims are bogus.


Right. This is the advantage of the IETF's patent policy, and it's one of the reasons we wanted to do this work in the context of the IETF.


It means to the best of authors' knowledge.


> There is no such thing as "unencumbered by patents".

When do the lawsuits over PNG begin?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: