A few things - this is one step in a long, LONG path. AV2 is currently unusable in its current state (the encoder typically runs at around 1fps on good hardware), and likely will remain so til ~2028 when the first av2 hardware accelerated chips start dropping. Even then, I wouldn't expect AV2 streams to be common til 2030.
IMO, if it were just the efficiency gains on the table (which are substantial - ~20-30% over AV1), I'd say that AV2 isn't worth it. The biggest thing it does add though is multi-stream support, which will be a big win for VR and live sports. The other fun thing is you can send an alpha channel as a separate stream, which the file will then composite for proper transparent video support.
adgjlsfhk1 12 hours ago [-]
Based on AV1's trajectory, hardware encode isn't necessary (though it is nice). The current encoder is a reference encoder. Now that the spec is finalized, expect significant speed improvements from production encoders (realtime likely won't happen until we get it in hardware though)
Gigachad 8 hours ago [-]
Hardware encode is required if you want things like video calls, camera recording and such to use it.
It isn’t required for content distribution platforms which aren’t realtime and the cost of encode is easily made up by hundreds of thousands of streams.
Orphis 8 hours ago [-]
One of the interesting usage of AV1 was specifically for low bitrate calls, and software encoding was perfectly fine, even on mobile.
With low enough resolution, framerate and bitrate, you can get a quality stream without significant encoding artifacts compared to any other codec. It is in production right now and has been for a while.
The tradeoff CPU / bandwidth is quite advantageous in situations like this. And no, AV1 HW encoders cannot usually be used, they are not designed for a tight bitrate control or realtime communications like software encoding is usually.
jech 3 hours ago [-]
> One of the interesting usage of AV1 was specifically for low bitrate calls, and software encoding was perfectly fine, even on mobile.
You really want hardware decoding on mobile, otherwise you end up with 40 minutes battery life. Fortunately, for typical videoconference resolutions, VP8 and H.264 are just fine. AV1 is nice to have, though, due to excellent support for synthetic content (screen sharing), and for scalable video coding (a much more elegant solution than simulcast, IMHO).
In the world I live in, the general plan is to stick to VP8 and H.264 for the time being, and to skip to AV1 when it's universally available on mobile. I haven't seen any features of AV2 which would justify waiting for it.
ksec 6 hours ago [-]
Have you said this for Audio Codec I would have agreed. I do not know a single Smartphone Video Conferencing software that uses CPU encoding rather than hardware encoding. Neither WhatsApp or FaceTime, perhaps the largest of the two real time Video Call uses AV1.
dannyw 5 hours ago [-]
Yeah, no production or large scale VC system is running software AV1 encoders on smartphones. You will drain a full phone battery in 1-2 hours of calls.
It just doesn’t make sense and will result in extraordinary power/battery drainage at best, or output that’s worse than hardware encoding.
The only way you could get AV1 to software encode in realtime AND low latency on a mid-range Android chip is by disabling or skipping nearly all of the compression/encoding features that make it good at low bitrate.
tecleandor 3 hours ago [-]
> Yeah, no production or large scale VC system is running software AV1 encoders on smartphones. You will drain a full phone battery in 1-2 hours of calls.
Yeah but, half jokingly, Zoom does that (draining the battery in an hour) already :P
Andrex 4 hours ago [-]
So, status remains quo, the commons remain tragic, and glory to H.264 forever?
kergonath 47 minutes ago [-]
At least until a better codec has widespread enough hardware support, I think.
baybal2 11 minutes ago [-]
[dead]
diegocg 6 hours ago [-]
Anything running on a battery will need hardware acceleration
kalleboo 3 hours ago [-]
> The biggest thing it does add though is multi-stream support
I would have thought this would be a part of the container format rather than the video codec?
xmichael909 3 minutes ago [-]
Google H264 SVC
xbmcuser 8 hours ago [-]
The way things are going, we can pretty much forget about AV2 hardware encoders in PCs anytime soon. All the newest, best chip capacity is being completely hogged by Apple and AI companies.
Unless chipmakers port the AV2 design to older, cheaper nodes, it’s just not happening for average users. We’ll probably see some Chinese TV chip makers throw in an AV2 decoder just to check a box, but as an actual encoder? I wouldn't count on it anytime soon.
kryptiskt 5 hours ago [-]
I wouldn't be so pessimistic, Intel and AMD aren't going to stop making CPUs, and if their integrated graphics adds AV2 it will be motivation enough for others to follow.
Gigachad 5 hours ago [-]
PCs don’t need hardware encoders. There are no realtime jobs on these. You can let it encode in however long it takes.
You need hardware encoders for things like cameras because they need to encode in real time since the buffer would quickly overflow otherwise.
circuit10 4 hours ago [-]
Use cases for hardware encoders in PCs:
- Video calls
- Screen/webcam recording
- Live streaming
- Real-time transcoding for media servers (don’t know much about this but I’ve heard it’s a thing)
- Game streaming
- Video editing (making exporting less frustrating)
sylware 3 hours ago [-]
Everything here is really niche, except video calls (and even that...).
In other words, unless on smart phones, don't expect broadly distributed AV2 encoding hardware.
If it does happen on PC, it will be most likely some courtesy of the hardware chip designers.
po1nt 5 hours ago [-]
Looking at how GPU development has been sidetracked for NPU, I worry that this is 2035 target at best. Manufacturers will push for maximising matrix operation silicon area. In the era of trillion dollar investments into datacenters, traffic cost is afterthought. The only benefitors might be YouTube, Netflix and such, but on their scale investment into ISP level caches might be cheaper.
nubinetwork 5 hours ago [-]
> enabling high-quality video delivery at significantly lower bitrates
> likely will remain so til ~2028 when the first av2 hardware accelerated chips start dropping
This might sound dumb, but whats the point if its intended for slower devices, but those slower devices don't even exist yet?
andersa 5 hours ago [-]
It's not for slower devices, it's for lower data transfer bills for providers like YouTube.
snek_case 3 hours ago [-]
It's a win for consumers as well if you can get better video quality or more reliable calls on a slower connection.
rovr138 5 hours ago [-]
so that new devices can adopt it
They can't adopt it if it doesn't exist.
nubinetwork 5 hours ago [-]
And what about old devices? I'm sure someone out there is still using an s5 as a daily driver... Future proofing is great and all, but 240p on modern devices looks like trash, even worse than tube tv.
Gigachad 5 hours ago [-]
240p for someone’s webcam thumbnail in Microsoft teams is perfectly sufficient though.
rovr138 5 hours ago [-]
They can keep using whatever they're currently using.
This doesn't take away anything. It's a new standard.
Based on your argument, one should add new safety standards to cars like seat belts, because old cars might not have them.
nubinetwork 3 hours ago [-]
To be fair, some municipalities do actually require old cars to be fitted with seatbelts... air bags, not so much, because apparently changing a steering column is too hard?
miohtama 8 hours ago [-]
In HW accelerated chips, what part of the calculations they usually accelerate? Could it be possible to repurpose old HW?
BillStrong 5 hours ago [-]
Essentially all of the processing of the video data, barring the container format which the CPU uses to know what part of the data to send to the GPU or the Audio chip or codec.
And HW acceleration is generally a preset baked in version of the encoder or decoder. These are mostly codec specific.
So, no using hardware from previous versions.
Now, you can see some software that tries to use the GPU itself, instead of the dedicated hardware acceleration, to decode, but that isn't the HW accelerated, and may not operate in real time.
At the same time, that will consume much more power, eliminating some of the advantages or the pure HW rendition, especially important for mobile.
I could see an argument being made for encoding, if it is 2x or faster than the CPU, but I haven't looked at any in a while, so don't know the speeds.
boriskourt 7 hours ago [-]
One of the biggest gains of having dedicated hardware is that the computation doesn’t happen on the general hardware.
This is what makes it viable on mobile devices where system responsiveness and power efficiency are high priority.
Generally these hardware decoders haven’t been retoolalble.
cyberax 4 hours ago [-]
Typically things like quantization and motion estimation.
Bombthecat 7 hours ago [-]
I feel like in 2030 it's more likely that we send 480p and just upscale with ai on the other end
hulitu 7 hours ago [-]
480p ? More like 200i. There's a race to the bottom driven by those "up to" codecs.
cyberax 4 hours ago [-]
> The biggest thing it does add though is multi-stream support
This was supported in H.264 MVC but only saw real use for 3D movies on physical BluRays. With almost no content available outside that.
shmerl 11 hours ago [-]
That's fine and not anything new for codecs, they always take a long time before mass adoption.
Take a look at AV1 itself, you can't even say it's really ubiquitous on all hardware. It's quite well along in adoption compared to early days, but some mobile devices are still lacking hardware acceleration for it.
BillStrong 5 hours ago [-]
It is still ironic to me that my Steam Deck has decode AV1 acceleration, on a really old CPU/GPU combo.
ksec 6 hours ago [-]
>........I'd say that AV2 isn't worth it.
Unless they have hardware encoder and decoder design done in parallel, otherwise it would be 2028 before a hardware block design is done and 2030 for the earliest product to ship with it. In reality I think 2031 or 2032 is more likely.
And I have been saying the same for quite some time that 20-30% for a generational codec improvement isn't worth it. I think they originally aimed at 50%, and then 40% and then 30%.
dgreensp 11 hours ago [-]
Where do you see information about the efficiency gains over AV1?
I feel like these encoders that require acceleration are a big reason why good hardware obsoletes so quickly.
PunchyHamster 6 hours ago [-]
> The biggest thing it does add though is multi-stream support, which will be a big win for VR and live sports.
AV1 supports it too ?
d--b 8 hours ago [-]
Does anyone know what is so costly to calculate in AV2?
ZeroGravitas 8 hours ago [-]
In general they just increase the numbers on everything when they go up a generation.
e.g. if you check in 4 directions to see if you can reuse a chunk then make it check in 8 or 16.
Faster encoders will have smart heuristics on when to use these new abilities and when to skip them but the reference encoder will try everything in a dumb way to eke out a tiny win to maximize a theoretical advantage and map out the extreme best case.
Dwedit 11 hours ago [-]
What I'm interested in is seeing how this will improve the AVIF image format. AVIF stomps the competition for low-bitrate still images (where chroma subsampling is used). For lossless images, not so much. Lossless JPEG XL and lossless WEBP make lossless AVIF look like a joke.
AVIF is for sure my favorite image format right now. No other format has the quadfecta of lossless, HDR, transparency, browser support. Plus as you said, for very compressed images it looks amazing. It blows my mind how small AVIF files can be. Also, unlike HEIC and Ultra HDR JPEG, it actually supports HDR natively as part of the file format rather than doing the hacky sidecar gain map trick. I know it doesn't matter to everyone, but I just love HDR and AVIF is the only format that I feel like really takes it seriously.
ksec 6 hours ago [-]
>but I just love HDR and AVIF is the only format that I feel like really takes it seriously.
JPEG XL would like a word.
gen2brain 5 hours ago [-]
He would like a word, but browsers said no?
Gigachad 5 hours ago [-]
Firefox is getting the flag to turn it on in the next release, chromium just added it back now there is a rust decoder. The wheels are turning again. Browser support for jpeg xl is very much in progress again.
erwan577 3 hours ago [-]
Do you think advanced users should consider AVIF/AVIF2 along JPEGXL for long term pictures archival ?
Which kind of encode settings do you suggest for conversion from high resolution RAWs or JPEGs ?
BoingBoomTschak 1 hours ago [-]
I really don't.
1. Lossless AVIF is a joke often beaten by WebP and even PNG. Even worse for grayscale.
2. Chroma subsampling remains a bad idea for still images unless the resolution is high enough to hide the artifacts.
I don't see AVIF being used for lossless, which is the largest reason I'd prefer JXL to win: one codec to rule them all sure is an alluring future.
theandrewbailey 3 hours ago [-]
Maybe libavif will pick up AV2 support, so all one needs to do is update it to v2.0.0 or something and bam! Extra efficiency!
wmf 9 hours ago [-]
Honestly AVIF2 is the last thing we need now. There are way too many minority image formats already.
Gigachad 7 hours ago [-]
There aren’t _that_ many. HEIF has been an unusually large pain in the ass just because it’s both patent encumbered and incredibly popular since the iPhone and many cameras use it.
JPEG is woefully outdated with the lack of HDR and modern compression, HEIF can’t be used without paying a license, webp was designed just for extremely efficient small images rather than local storage, avif I’ve never seen used ever, and JPEG XL is on track to be the next major format.
I agree we don’t need an avif2, but until jpeg xl there really weren’t any decent alternatives for jpeg.
Almost all of these are historic relics of no relevance today. FLIF for example was a research project that was later the basis of JPEG-XL.
Pretty much only the ones I listed previously are serious contenders in the space right now.
31 minutes ago [-]
thinkingQueen 8 hours ago [-]
AV1 is being actively claim-charted by a lot of companies right now, and lawsuits are almost certainly coming. The same process is already starting for AV2, but most players are waiting for the AV1 cases to mature first.
People keep calling the AV-family codecs “royalty free,” but in practice they increasingly look like a legal and financial gamble.
ZeroGravitas 8 hours ago [-]
People have been saying this for decades now.
I've never understood why some people seem to cheer this on like a corporation owning some maths was their local sports team.
For a while I assumed some people had put in a lot of effort on H.264 encoders and so the digital sharecroppers were angry and jealous that someone might be advocating for messy freedom.
But some people seem to just enjoy the thought of corporations putting a tax on video distribution.
Luckily those greedy corporations have repeatedly shot themselves on the foot and so their influence is waning.
Klaus23 8 hours ago [-]
How long has it been since AV1 was released? About eight years, and there's still no credible patent holder. The vultures are always circling around compression standards. You shouldn't take that too seriously. Even if a lawsuit is filed, there's a legal defence fund to protect against baseless claims.
Klaus23 5 hours ago [-]
Most importantly, there is no alternative. As kasabali said, the patent situation for the other codecs is a mess. Additionally, they could be hit by the same "No FRAND" problem. If someone comes forward with a patent that was used unintentionally, the situation is the same for all codecs.
throw0101a 3 hours ago [-]
> People keep calling the AV-family codecs “royalty free,” but in practice they increasingly look like a legal and financial gamble.
And the alternative is… ?
For H.265 there are two HEVC licensing pools you have to sign with plus at least two non-pool companies:
Going with a non-AVx codec is no less complicated and fraught with lawsuit risk AFAICT.
kasabali 6 hours ago [-]
> in practice they increasingly look like a legal and financial gamble
As opposed to what, like HEVC? Where you need to pay 3 different patent pools to be sure (which all has different terms), then there's still other patent holders that aren't in any pools and come and hit you with loyalty requests any time under terms however they like to?
amelius 51 minutes ago [-]
It should be not possible to patent communication standards. The opportunity for abuse through lock-in effects is just too big.
3 hours ago [-]
ethin 9 hours ago [-]
And how long will it take before someone implements this standard and gets sued because Adobe or Dolby or whoever wanted to get slapped down? My knowledge may be out of date but if this is as "open" as AV1, I'm very skeptical that the individual companies will actually allow that. Greed and all that.
zamadatix 9 hours ago [-]
It took 7 years for the first patent assertion claim against AV1 to go to the courts and it will probably take a while for that case to resolve. Funnily enough, it wasn't from the pool constantly putting itself in the news about it over the years. I.e. it can take quite a while before attempts are made.
throawayonthe 5 hours ago [-]
i'll be surprised if anything comes of it because like... https://aomedia.org/about/members/ if the legal departments of all these corporations haven't made them drop av1 in all these years there's probably not much there?
zamadatix 4 hours ago [-]
I'd like to think so as well (well, I don't know they'd necessarilymake them "drop" it as much as license those particular patents/accept the risk they may have to pay in the future) but on the other hand history has shown VP8 had a lot of companies behind it too before Google signed to sublicense patents from the MPEG LA & other holders for it.
ZeroGravitas 2 hours ago [-]
VP8 was developed by a small company and bought and open sourced by Google on a fairly short timescale because the proprietary codec group had tried to start exerting their control.
So it's not accurate to say that had a lot of companies behind it. That usage came later and even then it was mostly in odd corners of the video industry.
zamadatix 1 hours ago [-]
Who originally developed the codec is only half the story. In 2011, 17 companies created a CCL agreement around VP8+WebM. In 2013 Google signed the sub-licensing agreement to keep it free. Any of the legal teams at those 17 companies had the chance to catch it before then. That there were a lot of legal team reviews by the big tech companies involved did not catch everything. The system isn't designed in a way to make that a solid guarantee like it should be.
buran77 6 hours ago [-]
> It took 7 years for the first patent assertion claim against AV1
This might just mean that if the claim is found valid, there's seven years' worth of inertia slowing down any effort to move on. Seven years in which HW and SW manufacturers worked to build in the support, and you the user developed your processes or workflows around assumptions specifically tied to that solution. I'd rather know on day one if I should go that way or not.
Patent trolls are nasty. How long will it take for one to get the full support of those shaking the independence of the US judiciary for their own gain? Let's hope the rot stops before that.
ParadisoShlee 10 hours ago [-]
Mostly a joke... I've been waiting for the AV1 Apple TV, so now I'm just waiting for AV2 support as Apple TV as well now.
breve 10 hours ago [-]
My 10 year old iPhone 7 can play 1080p AV1 video in software for more than 200 minutes with VLC. The iPhone 7 was released a year and a half before AV1 was.
So I think it's a safe bet the current Apple TV devices are capable of playing AV1 video in software. There's a VLC release for Apple TV:
Not especially relevant, as the obvious use of AV1 on the AppleTV is streaming, and the OS frameworks don't request AV1 without hardware decoding. Services which provide their own video decoding (are there any?) don't seem interested providing their own software decoder for the ATV, despite the bandwidth savings.
breve 9 hours ago [-]
Six years ago Apple TV added support for playing 4K YouTube video:
The evergreen prediction in the Apple TV world :p. IIRC, Mark Gruman initially predicted the next model would be out H1 of 2024
londons_explore 10 hours ago [-]
Outside the apple ecosystem, AV1 is supported nearly everywhere.
techpression 8 hours ago [-]
They’ve had hardware decoding since M3 and equivalent A cpus. So I’d say it’s pretty well supported.
abacadaba 27 minutes ago [-]
Is there an AV1 2.0? I'm not using this codec if they can't do basic semantic versioning right.
angelmanuel 5 hours ago [-]
I'm not a expert on video encoding
But i wonder if the future could depend less on fixed-function compression methods and more on AI networks that recreate the video but weight much less that a compressed video.
Neural codecs such as github.com/Orange-OpenSource/Cool-Chic
jech 3 hours ago [-]
It will probably depend on whether NPUs are universally available in smartphones, and whether we get a standard API for accessing NPUs. But I don't know whether AI-based codecs can have battery usage competitive with fixed-function hardware.
sarah-robiin 7 hours ago [-]
AV1 already was a big leap toward efficient and open video formats. I'm awaiting AV2 since a long time.
Sure it'll take a while since it's implemented in chips and hardware so we got efficient and fast hardware encoding/decoding.
But a ~25% higher efficiency sounds very promising in times of increasing storage prices and chip crises.
mmastrac 11 hours ago [-]
Dav2d doesn't have the same nice ring to it. I hope there's someone with a decent repo-name punning skill who'll contribute before that.
avi2ude? av2go?
jbk 7 hours ago [-]
It was difficult to find a nice name, with av2 :(
It works in French d2vid (Deuvid)
Dwedit 11 hours ago [-]
At least it's not D4vd.
toast0 11 hours ago [-]
2av2quit?
Retr0id 4 hours ago [-]
daviid - or trim to davii and pronounce it "davey". But tbh I quite like dav2d.
zamadatix 9 hours ago [-]
I like it - not as punny as the first but pretty straightforward.
themafia 6 hours ago [-]
e2ed
Telaneo 8 hours ago [-]
Looking forward to a decently speedy encoder coming around. The reference one for AV1 is really not that great, and the same is true here. But as soon as we get SVT-AV2 or whatever, I'll be a very happy camper.
perching_aix 5 hours ago [-]
anyone performed a encoding and decoding benchmarking with the reference codec yet? I'd expect encoding to be dreadful, but maybe the decode is already usable
maxloh 9 hours ago [-]
It takes a few years for vendors to support hardware decoding for a new standard, so we won't see it in widespread use anytime soon.
shmerl 11 hours ago [-]
Congrats!
How is the case of fighting off Dolby's patent racketeering going? They tried to attack Snapchat for using AV1.
mmastrac 11 hours ago [-]
Last update seems to be "lawsuit was filed" with zero updates since then. That stuff tends to move slowly.
shmerl 11 hours ago [-]
Hopefully their patents will be busted and preferably Dolby will be also forced to pay damages for filing invalid lawsuits. That's the only way to teach patent trolls proper lessons.
necovek 8 hours ago [-]
While I agree with the general sentiment, I would not classify Dolby as a "patent troll" — they still do invest in research and developing products, right?
kasabali 6 hours ago [-]
Patent mob seems more fitting.
Another example is Qualcomm
lofaszvanitt 8 hours ago [-]
How is Dolby a patent troll? :D
cyberax 4 hours ago [-]
Dolby's patents are garbage, and they know it. They managed to patent entropy coding somehow. This will not stand to challenges, and they managed to poke a well-resourced company.
basilgohar 11 hours ago [-]
This will always happen. There are just some entities that can't stand not seeking rent.
ethin 9 hours ago [-]
It may always happen but it would happen less if we updated patent laws to fine people who filed invalid patents or enforced some kind of similar punishment. If you file a patent, it's up to you to verify that your patent is actually valid, and the courts shouldn't have to do that legwork for you. It also doesn't help that the patent office/components of governments don't review patents as thoroughly as they used to. Same with trademarks.
zamadatix 8 hours ago [-]
I generally don't like the current patent law but it sounds a bit off to pay the government & wait for them to review your patent claim and then get fined by the government when both of you were wrong about it. There are already processes to additionally fine a company bringing about a truly frivolous patent lawsuit, it's just rare because usually it's not so cut and dry as we'd like it to be.
ethin 8 hours ago [-]
I mean my idea isn't the only one in that solution space. My reasoning was to ensure that the government actually reviewed the patent and ensured it was valid instead of rubber stamping it. Or, even better, the filer of the patent application would do that. Although the best is probably to make software unpatentable anyway.
zamadatix 5 hours ago [-]
Yeah, I wouldn't mind getting rid of software patents. Or, at least, making the patent reviews more rigorous before assignment.
shmerl 8 hours ago [-]
Even better, software patents should not be allowed in the first place.
lofaszvanitt 8 hours ago [-]
Why patents is a problem, please explain. If you build something that has been patented, then, well, you pay the per piece fee on it.
ZeroGravitas 8 hours ago [-]
> In the 1980s, when IBM accused Sun of violating seven patents, Sun examined the patents and argued that IBM didn't have a case. The reply of IBM's lawyers was "maybe you don't infringe these seven patents. But we have 10,000 U.S. patents. Do you really want us to go back to Armonk [IBM headquarters in New York] and find seven patents you do infringe? Or do you want to make this easy and just pay us $20 million?" And Sun paid out.[4]
BoingBoomTschak 4 hours ago [-]
Now the real question: will The Industry™ again need for nerds and digital privateers to actually write and graft all the psy coding tools that makes their encoders usable (and I mean the word, using x264 as benchmark) for the quality-conscious and not just CPU-efficient VoD blurred to death with marketing yelling "PSNR! PSNR! PSNR!" at the top of their lungs? Will FGS be more usable?
I’m curious how much AV2 will actually help older hardware in practice.
I’m on a 2019 Intel MacBook Pro: 2.6 GHz 6-core i7, 64 GB RAM. The machine is still more than powerful enough for normal desktop work and software dev, but YouTube in Chrome has become borderline unusable for me. My internet is fine, Safari plays the same videos smoothly, and YouTube “Stats for nerds” shows plenty of buffer but the decoding makes youtube unusable in chrome for me.
kasabali 6 hours ago [-]
Has nothing to do with video codec.
Download the video with yt-dlp & play it in mpv you'll see it even flies on a potato.
Play the same in browser and it'll be dropping frames left and right.
ZeroGravitas 8 hours ago [-]
I use Firefox but YouTube has recently started giving me a pop-up occasionally telling me that they are intentionally slowing down the site because they don't like some of the browser extensions I use.
Telaneo 8 hours ago [-]
Sound like a Chrome/Youtube problem. My 2012 Macbook Pro plays 1080p AV1 just fine in VLC (pretty sure Youtube works fine too in Firefox, but I didn't check whether or not it was AV1 or H264).
Telaneo 8 hours ago [-]
For reference: dav1d 0.5 can decode 143 FPS of a 1080p 8-bit video on a third gen core i7.[1] I doubt there's been much in the way of regressions since then. 10-bit and 4k is obviously a lot more heavy, but not really relevant to older devices.
use the enhanced h264-ify to block the av1 stream, av1 takes a lot of cpu
wmf 9 hours ago [-]
Unfortunately for you, newer codecs use more CPU than older ones so AV2 would probably be even worse.
alephnil 6 hours ago [-]
Newer codecs has generally introduced more ways a video can be encoded, so that the encoder need to work much harder to encode a video, so that it actually achieve the gains that the newer codec allows much more processing will be required. Decoding on the other hand, will mostly stay the same or increase only slightly. It's not likely do decrease though, so if you struggle playing av1 today, you will also struggle with av2.
For encoding, you can always write a simple encoder that use only the features that were present in mpeg2, and it will be about as efficient as mpeg2 as well. Newer codecs has more features that allows more efficient encoding, at the cost of more processing.
drob518 8 hours ago [-]
I gave up using Chrome a decade ago. It’s a power sucking pig. Safari has its own issues, but at least it’s usable. When I need something that isn’t Safari, I use Firefox.
LunaSea 6 hours ago [-]
Safari has a terrible developer experience and has been behind in implementing the various browser API for years, including AV1 support.
lII1lIlI11ll 2 hours ago [-]
Safari has had AV1 support for a long time. Absolute majority of "various browser API" that they don't support are all kinds of "Web Bluetooth"-like crap that I don't need and which only introduces attack and tracking surface when supported.
Rendered at 14:54:22 GMT+0000 (Coordinated Universal Time) with Vercel.
IMO, if it were just the efficiency gains on the table (which are substantial - ~20-30% over AV1), I'd say that AV2 isn't worth it. The biggest thing it does add though is multi-stream support, which will be a big win for VR and live sports. The other fun thing is you can send an alpha channel as a separate stream, which the file will then composite for proper transparent video support.
It isn’t required for content distribution platforms which aren’t realtime and the cost of encode is easily made up by hundreds of thousands of streams.
With low enough resolution, framerate and bitrate, you can get a quality stream without significant encoding artifacts compared to any other codec. It is in production right now and has been for a while.
The tradeoff CPU / bandwidth is quite advantageous in situations like this. And no, AV1 HW encoders cannot usually be used, they are not designed for a tight bitrate control or realtime communications like software encoding is usually.
You really want hardware decoding on mobile, otherwise you end up with 40 minutes battery life. Fortunately, for typical videoconference resolutions, VP8 and H.264 are just fine. AV1 is nice to have, though, due to excellent support for synthetic content (screen sharing), and for scalable video coding (a much more elegant solution than simulcast, IMHO).
In the world I live in, the general plan is to stick to VP8 and H.264 for the time being, and to skip to AV1 when it's universally available on mobile. I haven't seen any features of AV2 which would justify waiting for it.
It just doesn’t make sense and will result in extraordinary power/battery drainage at best, or output that’s worse than hardware encoding.
The only way you could get AV1 to software encode in realtime AND low latency on a mid-range Android chip is by disabling or skipping nearly all of the compression/encoding features that make it good at low bitrate.
Yeah but, half jokingly, Zoom does that (draining the battery in an hour) already :P
I would have thought this would be a part of the container format rather than the video codec?
Unless chipmakers port the AV2 design to older, cheaper nodes, it’s just not happening for average users. We’ll probably see some Chinese TV chip makers throw in an AV2 decoder just to check a box, but as an actual encoder? I wouldn't count on it anytime soon.
You need hardware encoders for things like cameras because they need to encode in real time since the buffer would quickly overflow otherwise.
- Video calls
- Screen/webcam recording
- Live streaming
- Real-time transcoding for media servers (don’t know much about this but I’ve heard it’s a thing)
- Game streaming
- Video editing (making exporting less frustrating)
In other words, unless on smart phones, don't expect broadly distributed AV2 encoding hardware.
If it does happen on PC, it will be most likely some courtesy of the hardware chip designers.
> likely will remain so til ~2028 when the first av2 hardware accelerated chips start dropping
This might sound dumb, but whats the point if its intended for slower devices, but those slower devices don't even exist yet?
They can't adopt it if it doesn't exist.
This doesn't take away anything. It's a new standard.
Based on your argument, one should add new safety standards to cars like seat belts, because old cars might not have them.
And HW acceleration is generally a preset baked in version of the encoder or decoder. These are mostly codec specific.
So, no using hardware from previous versions.
Now, you can see some software that tries to use the GPU itself, instead of the dedicated hardware acceleration, to decode, but that isn't the HW accelerated, and may not operate in real time.
At the same time, that will consume much more power, eliminating some of the advantages or the pure HW rendition, especially important for mobile.
I could see an argument being made for encoding, if it is 2x or faster than the CPU, but I haven't looked at any in a while, so don't know the speeds.
This is what makes it viable on mobile devices where system responsiveness and power efficiency are high priority.
Generally these hardware decoders haven’t been retoolalble.
This was supported in H.264 MVC but only saw real use for 3D movies on physical BluRays. With almost no content available outside that.
Take a look at AV1 itself, you can't even say it's really ubiquitous on all hardware. It's quite well along in adoption compared to early days, but some mobile devices are still lacking hardware acceleration for it.
Unless they have hardware encoder and decoder design done in parallel, otherwise it would be 2028 before a hardware block design is done and 2030 for the earliest product to ship with it. In reality I think 2031 or 2032 is more likely.
And I have been saying the same for quite some time that 20-30% for a generational codec improvement isn't worth it. I think they originally aimed at 50%, and then 40% and then 30%.
https://www.youtube.com/watch?v=Se8E_SUlU3w&t=1480
AV1 supports it too ?
e.g. if you check in 4 directions to see if you can reuse a chunk then make it check in 8 or 16.
Faster encoders will have smart heuristics on when to use these new abilities and when to skip them but the reference encoder will try everything in a dumb way to eke out a tiny win to maximize a theoretical advantage and map out the extreme best case.
JPEG XL would like a word.
Which kind of encode settings do you suggest for conversion from high resolution RAWs or JPEGs ?
1. Lossless AVIF is a joke often beaten by WebP and even PNG. Even worse for grayscale.
2. Chroma subsampling remains a bad idea for still images unless the resolution is high enough to hide the artifacts.
3. Tooling is the worst part, AV1 encoders are basically focused 99% on video and leave a measly 1% to image; unlike JXL, of course. SVT-AV1 still doesn't do YUV444 and libaom was unusable. Fortunately, the unpaid enthusiasts were here: https://giannirosato.com/blog/post/the-multimedia-renaissanc... (and more recently https://giannirosato.com/blog/post/oavif/)
I don't see AVIF being used for lossless, which is the largest reason I'd prefer JXL to win: one codec to rule them all sure is an alluring future.
JPEG is woefully outdated with the lack of HDR and modern compression, HEIF can’t be used without paying a license, webp was designed just for extremely efficient small images rather than local storage, avif I’ve never seen used ever, and JPEG XL is on track to be the next major format.
I agree we don’t need an avif2, but until jpeg xl there really weren’t any decent alternatives for jpeg.
https://en.wikipedia.org/wiki/Comparison_of_graphics_file_fo...
Pretty much only the ones I listed previously are serious contenders in the space right now.
People keep calling the AV-family codecs “royalty free,” but in practice they increasingly look like a legal and financial gamble.
I've never understood why some people seem to cheer this on like a corporation owning some maths was their local sports team.
For a while I assumed some people had put in a lot of effort on H.264 encoders and so the digital sharecroppers were angry and jealous that someone might be advocating for messy freedom.
But some people seem to just enjoy the thought of corporations putting a tax on video distribution.
Luckily those greedy corporations have repeatedly shot themselves on the foot and so their influence is waning.
And the alternative is… ?
For H.265 there are two HEVC licensing pools you have to sign with plus at least two non-pool companies:
* https://en.wikipedia.org/wiki/High_Efficiency_Video_Coding#P...
Going with a non-AVx codec is no less complicated and fraught with lawsuit risk AFAICT.
As opposed to what, like HEVC? Where you need to pay 3 different patent pools to be sure (which all has different terms), then there's still other patent holders that aren't in any pools and come and hit you with loyalty requests any time under terms however they like to?
So it's not accurate to say that had a lot of companies behind it. That usage came later and even then it was mostly in odd corners of the video industry.
This might just mean that if the claim is found valid, there's seven years' worth of inertia slowing down any effort to move on. Seven years in which HW and SW manufacturers worked to build in the support, and you the user developed your processes or workflows around assumptions specifically tied to that solution. I'd rather know on day one if I should go that way or not.
So I think it's a safe bet the current Apple TV devices are capable of playing AV1 video in software. There's a VLC release for Apple TV:
https://www.videolan.org/vlc/download-appletv.html
https://apps.apple.com/us/app/vlc-media-player/id650377962?p...
https://9to5mac.com/2020/06/22/tvos-14-brings-support-for-st...
YouTube's 4K videos are only available in VP9 and AV1.
So the YouTube app on tvOS has supported at least VP9 for six years and I wouldn't be surprised if it supports AV1 today.
https://en.wikipedia.org/wiki/Apple_TV_(device)#4K_(3rd_gene...
There may be a new Apple TV released this year.
The evergreen prediction in the Apple TV world :p. IIRC, Mark Gruman initially predicted the next model would be out H1 of 2024
But i wonder if the future could depend less on fixed-function compression methods and more on AI networks that recreate the video but weight much less that a compressed video.
Neural codecs such as github.com/Orange-OpenSource/Cool-Chic
Sure it'll take a while since it's implemented in chips and hardware so we got efficient and fast hardware encoding/decoding.
But a ~25% higher efficiency sounds very promising in times of increasing storage prices and chip crises.
avi2ude? av2go?
It works in French d2vid (Deuvid)
How is the case of fighting off Dolby's patent racketeering going? They tried to attack Snapchat for using AV1.
Another example is Qualcomm
Take a look at https://gitlab.com/AOMediaCodec/SVT-AV1/-/work_items/2269 for details (PS: SVT-AV1 claims to be suitable for AVIF yet doesn't support YUV444, lel)
I’m on a 2019 Intel MacBook Pro: 2.6 GHz 6-core i7, 64 GB RAM. The machine is still more than powerful enough for normal desktop work and software dev, but YouTube in Chrome has become borderline unusable for me. My internet is fine, Safari plays the same videos smoothly, and YouTube “Stats for nerds” shows plenty of buffer but the decoding makes youtube unusable in chrome for me.
Download the video with yt-dlp & play it in mpv you'll see it even flies on a potato.
Play the same in browser and it'll be dropping frames left and right.
[1] https://www.phoronix.com/news/dav1d-0.5 (it's mislabled as a core i3, but the 3770K is a core i7).
For encoding, you can always write a simple encoder that use only the features that were present in mpeg2, and it will be about as efficient as mpeg2 as well. Newer codecs has more features that allows more efficient encoding, at the cost of more processing.