I don't think it's exactly fair to call AV1 "technically poor". Even if HEVC is better (and I don't know enough to say if it is or not), that would be a bit like calling gzip technically poor because zstd exists.
The latest consensus seems to be that VP9 is roughly on par with HEVC, mostly being held back by the libvpx encoder's poor rate control.
It's no surprise then that AV1 (which is basically VP10) uses more resources, but compresses more, and that Netflix and Intel are keen to replace the libvpx derived encoder with their own implementation of the standard.
gzip absolutely without doubt is technically poor in this day and age. It's an algorithm from 1993. That's 26 years of research into data compression it is missing out on.
I would be careful about rooting for a codec that has strengths/weaknesses that are suspiciously tailored to extremely large media-serving companies. By which I mean, Netflix/Google don't care so much about encode time since their scale is so big, and they certainly don't care as much about decode time as users with their precious battery life do.
Just because something is open does not mean it is meant to serve consumers' interests.
Netflix and Google definitely care about encoding time precisely because their scale is so big. A 1% increase in encoding time could mean a difference of a tremendous amount of computing time and thus money.
Google and Twitch also care about encoding time, but in a different way - they have strong use cases for real time encoding being possible.
Everyone cares about decoding efficiency - Google wants you to be able to watch more videos and thus more ads. If your battery is dead, you can't watch any ads. Netflix wants you to get hooked on more series. etc.
> Just because something is open does not mean it is meant to serve consumers' interests.
This is true in general, but I really don't see how it applies to this case at all. The big companies have incentives that actually do line up pretty directly with user's interests.
"We will be satisfied with 20% efficiency improvement over HEVC when measured across a diverse set of content and would consider a 3-5x increase in computational complexity reasonable."
Netflix encodes and serves a lot of video. So obviously both encode time and low filesizes are important to them. That doesn't change my point that their priorities may be skewed towards lowering filesize moreso than other entities, since they serve more videos per encode.
> and they certainly don't care as much about decode time as users with their precious battery life do.
Of course they do. Google wouldn't sell many Pixels if you could only watch half as much YouTube or Netflix on a single charge compared to, say, an iPhone. Netflix wouldn't sell many subscriptions if you could only watch half as much on a single charge as Hulu or Prime Video either. Your point about encoding is valid, but Google and Netflix are definitely incentivised towards low decoding complexity on the client.
Huh? Have you noticed the stats of how much video is uploaded to YouTube? I won't provide any numbers, due to working there, but a back of envelope calculation should answer the question "how much money does it cost for Google to encode at twice the compute cost".
I think it used to be that Google encoded every video in H.264, but also encoded some videos in VP9: the popular ones. That made a lot of sense--no need to dedicate a lot of encoding power for every video when quick-and-dirty x264 will get the job done fine for >99% of them. And I assume the same will be true for AV1, where only a small percentage of videos will actually be encoded in it. So my envelope calculations aren't going to help me gain much insight.
This is a large and old topic but I'll attempt to summarize.
1. The difference between $0 and even the smallest license fee is essentially infinite; just the friction to track and collect license fees would basically kill off the open source ecosystem. Also, it really rubs some people the wrong way that you can get a complete operating system and hundreds of apps for free but a single modern video codec used to cost money.
2. Codec patent licenses are sometimes kind of trolly; some codecs want to charge fees per minute of video instead of per encoder/decoder which is an accounting nightmare and feels abusive.
There are plenty of legal open source H.264, MP3, etc implementations. AFAIK source code license and patents are orthogonal concepts. You can use any source code, proprietary, open source or written by yourself, but you should have permission from patent owners, because you're likely using their patented algorithms.
So, to produce a legal open-source (as in "free to reuse and fork") implementation, the implementor would need to obtain a permission from the patent holders, and make sure it covers all the possible derived work?
It's not about writing code, it's about using algorithms. If you're using MP3 algorithms to make money selling phones, you have to pay a fee. If you're using MP3 algorithms to make money selling your WinAMP player, you have to pay a fee. Whether you found implementation on github or read patents and implemented it yourself, does not matter. I don't think that you have to obtain a permission just to write implementation out of curiosity, but I may be wrong. And I have no idea about status of open source projects which include patented algorithms, but distributed free of charge. Probably they can't do that, because Linux distributions did not distribute mp3 codecs by default.
Also all of this is applicable to countries where those patents are working, of course.
Not surprised you are downvoted. On one hand HN complains about IP being stolen on the other hand they want all IPs to be free.
I don't have a problem with paying license fees, it is just most of those companies don't agrees on it.
I could only wish we make the standard as free for Software Encode and Decode.
All Hardware Accelerated Encoder and Decode will be $0.5 per unit. And $0.3 for only Encoder Or Decoder. With no Caps.
For Mobile alone that is anywhere between $360M to $600M alone assuming all devices supports it. And if we include PC, Tablet, Console, All other accessories it is up to $1B per year, and for the life time of the Codec easily $10B+ in patents total split across all companies.
The consumer will be paying for it, and we all enjoy better video quality with smaller downloads. Unfortunately most of the sweet spot for newer codec tends to be in 4K or even 8K. I wish they could put a lot more focus on 1080P at 1 / 2Mbps. Where the vast majority of video on Internet could settle on.
> Unified Patents is a membership organization that has aggressively contested patent abuse and recently challenged the validity of 29% of the HEVC-related patents in the Velos Media HEVC pool. Commenting on the Sisvel announcement, Unified Patents CEO Kevin Jakel said, "We continue to be concerned with the lack of transparency in these licensing programs. Sisvel publishes its pricing for VP9 and AV1, but not how it came about with this price. They also have not provided a public list or even number of patents which they consider essential. It makes it very hard for anyone to ascertain how relevant the patents are and how to value them. This creates uncertainty for companies deploying technology which we think is negative for everyone."
To my eyes, there's a pretty decent chance that the Sisvel pool is an HEVC industry ploy to propagate FUD around AV1 and VP9 (suspiciously both, the latter seeming to have been the subject of no lawsuits, despite being deployed widely for about six years), to stop the bleeding w.r.t. HEVC licensing (which is still an absolute steaming dump in the lap of your legal department).
I'll believe it when I see the receipts, and even then, it'd probably be less of a minefield than HEVC. And to be fair Sisvel has not, in other industries, seemed to be a particularly insidious actor; but given the number of big fish who are obviously using VP9 encoders to do tens of billions of dollars of business without licensing anything from Sisvel, I can't see it as anything other than bluster.
HEVC Advance lists 681 US patents in their patent list.
It is completely unlikely that AV1 infringes on none of them or any other patents.
However, there's a big giant gap between a valid patent and actually expecting to get royalties without even revealing which patents are supposedly being infringed on even now that AV1 is standardised and in production usage.
So do I think that AV1 is patent-free when all is said and done from a legal standpoint? No. Do I think it'll be royalty-free anyway? Yes.
Also:
> 1.3. Defensive Termination. If any Licensee, its Affiliates, or its agents initiates patent litigation or files, maintains, or voluntarily participates in a lawsuit against another entity or any person asserting that any Implementation infringes Necessary Claims, any patent licenses granted under this License directly to the Licensee are immediately terminated as of the date of the initiation of action unless 1) that suit was in response to a corresponding suit regarding an Implementation first brought against an initiating entity, or 2) that suit was brought to enforce the terms of this License (including intervention in a third-party action by a Licensee).
If they are intending to initiate any patent litigation (which is going to be a very long time yet not least of all because they won't even reveal which patents) I think they're going to be at a losing end very very quickly having waited this long for VP9/AV1 to establish themselves.
It occurs to me that the easiest way to flush out the risk is for AV1 to launch a defamation case against any patent licensing pool which claims to hold a relevant patent. If it’s a lie, that is straight up defamation.
(Or a similar tort/breach. In Australia it would constitute a case of misleading or deceptive conduct under the ACL.)
This particular form of suit may be difficult in the U.S, owing to a somewhat different set of tradeoffs. It might be simple commercial fraud, if you could prove that they had misrepresented their certainty that their pool contained patents which they reasonably believed to be essential to implementing an AV1/VP9 encoder/decoder; though I think it may be very very hard to prove something like that.
And if it's not a lie, and they're going to find some judge in East Texas who'll see infer some infringement among one some detail in one of the bazillion patents in the pool, you're screwed.
Patent cases of this sort seem to broadly favour the litigant so why stir the hornet's nest.
It wouldn’t be a patent case, it would be a tort. Either you win and the problem goes away, or you lose and you know which patent you need to rewrite around.
For the cost of some legal fees, a loss seems like a great value.
