5 ms·

The post doesn't mention hardware or silicon implementations. That's a key requirement for modern market adoption: if a mobile device can't hardware-decode it (

The post doesn't mention hardware or silicon implementations. That's a key requirement for modern market adoption: if a mobile device can't hardware-decode it (which saves precious power!), that's going to be a significant blocker. I know Google provided a free hardware decode block for VP8, and I assume also for VP9 (since there appears to have been some, though not universal, hardware acceleration for that according the the Wikipedia article). I'm hoping they'll soon (or already) have the same for VP1. I see Apple, Nvidia, Intel, Allegro, AMD, and a bunch of other silicon vendors are also members of the Alliance for Open Media, so I expect to see hardware acceleration within the next 2-3 years. I'll be impressed if it's earlier. Edit: maybe hardware acceleration earlier. If the standard was released recently with the collaboration of the silicon industry, it's likely the current in-design/early-production silicon already has an implementation of the important parts of the codec, so we can expect consumer products with at least partial acceleration within 6 months.

djrogers 9y ago
> if a mobile device can't hardware-decode it (which saves precious power!), that's going to be a significant blocker. A mobile device needs to be able to hardware-encode it as well - countless hours of 4k/60 video are taken on phones.
- mistercow 9y ago
  I'm not sure that you need mobile hardware encoding to drive initial adoption. As long as streaming services and the like can start using it to save bandwidth, you have a clear win for a big segment of the market. Older codecs can serve just fine for mobile encoding until the hardware catches up. But if most existing phones can't decode it efficiently, it's going to be pretty stuck until that situation changes.
  - thomastjeffery 9y ago
    I'd argue that it's more important to get encoders working, since re-encoding lossy->lossy is always a bad idea.
    - brigade 9y ago
      There isn’t really a video hosting site that doesnt do a lossy transcode of all uploads, unless you count Dropbox. The bitrate used to encode real-time captures is generally two or more times greater than the bitrate you want to distribute.
    - isostatic 9y ago
      Lossless 4K is 12 gigabit a second (even then there's chroma compression) Nobody edits at that bitrate, compression is concattenated multiple times. How a codec copes with real world multi-compression is key.
      - Bromskloss 9y ago
        > Nobody edits at that bitrate Interesting. Not even as a final rendering step after everything has been done in a compressed "preview mode"?
        isostatic 9y ago
        Setting aside high end production houses. Even then things aiming for TV get compressed to a mezzanine format that's heavily compressed (in comparison), then compressed again for TX
      - simias 9y ago
        4K is common in TV production. In my experience lossy compression is a big no-no in TV and they do edit 4K losslessly, although in order to transfer it they generally split it as 4 times 3G links (either interleaving the pixels or generating a 2x2 mosaic). 12G links exist but 4x3G is still vastly more popular due to being backward compatible with existing equipment. Color correction, scaling, blending and compositing are all generally done with uncompressed video. Of course once mastered the video gets heavily compressed to be sent to the consumers.
        isostatic 9y ago
        Ok I shouldn't use the word nobody, I admonished someone for using that word a few days ago, but it's very rare, especially outside high end production houses. Even 12G sdi is compressed - we consider 4:2:2 and 10bit uncompressed, but 4:4:4 is less compressed than 4:2:2. Once you go into interlace or stupid things like 24p you're throwing away tons of data. 2022-6 seems to be the transmission method of choice so far, 2110 may well take over, but there are significant compatibility issues. Personally I'd prefer everything originated as 4:4:4 10 bit 150fps (no drop frames), resolution I'm less concerned with. I work in news though, broadcast quality is what we broadcast. 5mbit h264 PPP with a vbv of 200k is fine for a talking head internationally. 5 times that is plenty for an actual broadcast program with astons and stuff.
gsich 9y ago
Adoption is forced by pirated stuff. If scene groups start to use it, people will want to play it. And those groups don't care if you can play whole movies on your phone. Because thats not what a phone is mainly for.
- bri3d 9y ago
  Movie and music piracy has taken a prodigal nosedive to the point that I don't think this would be true at all anyway. In addition streaming video and short form content dwarfs film and TV consumption, making high mobile efficiency paramount.
  - weberc2 9y ago
    Prodigal? Proverbial?
    - B1FF_PSUVM 9y ago
      Prodigious, probably.
- whatever_dude 9y ago
  That's not really true. The data from YouTube or Netflix dwarfs what you'd get from a torrent distributor of a show or whatever. Some pirate/niche markets may adopt a codec solely for some arbitrary reason, but those reasons (say, maximum compression, good open source tools) are normally orthogonal to what a streaming business usually needs (compression, but also hardware/software compatibility for all sorts of form factors, and commercial unencumberment).
  - baddox 9y ago
    I bet you’re right. AFAIK pretty much all but the cheapest TVs sold now (or at least 4K TVs) support HEVC specifically to support builtin apps for Netflix and similar. Unfortunately YouTube uses the competing VP9, which my 2 year old Vizio 4K TV does not support, but apparently devices like the Chromecast Ultra and 4K Apple TV support it as well.
twotwotwo 9y ago
Yes, Google is developing a free AV1 HW decoder, as discussed near the end of this (interesting!) deck: https://parisvideotech.com/wp-content/uploads/2017/07/AOM-AV1-Video-Tech-meet-up.pdf https://parisvideotech.com/wp-content/uploads/2017/07/AOM-AV... For what it's worth, Intel shipped "hybrid" implementations of HEVC and VP9 for older products where the CPU does some work but the GPU's shader cores do shader-y bits (https://www.anandtech.com/show/10610/intel-announces-7th-gen-kaby-lake-14nm-plus-six-notebook-skus-desktop-coming-in-january/3 https://www.anandtech.com/show/10610/intel-announces-7th-gen...). I don't know how far that approach gets you, especially in mobile settings where you really care about power use. But you might see that before complete hardware decoders everywhere. A use case that might be interesting before full hardware acceleration is to provide an improved still image format. The graph in the post is a great demonstration that improving video encoding and improving image encoding are closely related these days. AV1 has backing and went through IP review that may help it avoid being mired in a patent mess like previous proposals to replace JPEG. (Apple is using .heic, based on HEVC intra coding. Android'll support .heic in the next major version, but the HEVC patent situation will likely constrain how many external tools, platforms, etc. pick it up.) There's work on defining an AVIF format (https://github.com/AOMediaCodec/av1-avif https://github.com/AOMediaCodec/av1-avif) and tech demos look great (https://people.xiph.org/~tdaede/av1stilldemo/ https://people.xiph.org/~tdaede/av1stilldemo/). So, maybe fewer ugly JPEG artifacts in our future, partly thanks to that deringing filter mentioned at the end of Monty's post. (I said more or less the stuff about stills above, with a couple more details and links, at https://news.ycombinator.com/item?id=16699000 https://news.ycombinator.com/item?id=16699000)
- asterius 9y ago
  There is some interesting Q&A at the end of Pascal's talk. My schoolboy French is not fast enough to do a proper translation unfortunately. He was asked about parallel (software) decoding and I think the gist was he thought it would be a bit difficult because of the dependencies between region images. https://parisvideotech.com/pvt-4-les-formats-et-codecs-du-futur-chez-mozilla/ https://parisvideotech.com/pvt-4-les-formats-et-codecs-du-fu...
lathiat 9y ago
It shouldn't surprise anyone to learn that a brand new just released video codec does not yet have hardware decode support.
- noobermin 9y ago
  Came here to comment this. Seriously, HN, relax. This is like the introductory blogpost, how is there supposed to be a meatspace hardware implementation yet?
  - wmf 9y ago
    They're just balancing out the "AV1 is here and Netflix is dropping H.265 tomorrow" posts.
    - omtose 9y ago
      Is Netflix even using H265?
      - Aissen 9y ago
        Maybe on platforms where it's the only hardware-enabled codec (they don't always impose their will, although it's increasingly looking like they do).
zamalek 9y ago
Hardware acceleration for encoding is also pretty important. Entry-level game streamers need it (Twitch recently proved that this isn't a negligible market[1]). Playing back AV1 on a phone implies receiving it - which means that users need to store their videos to AV1 directly. [1]: https://www.forbes.com/sites/insertcoin/2018/04/07/ninjas-new-fortnite-twitch-records-5-million-followers-250000-subs-875000-a-month/#47ee5a52478f https://www.forbes.com/sites/insertcoin/2018/04/07/ninjas-ne...

djrogers 9y ago

> if a mobile device can't hardware-decode it (which saves precious power!), that's going to be a significant blocker. A mobile device needs to be able to hardware-encode it as well - countless hours of 4k/60 video are taken on phones.

mistercow 9y ago
I'm not sure that you need mobile hardware encoding to drive initial adoption. As long as streaming services and the like can start using it to save bandwidth, you have a clear win for a big segment of the market. Older codecs can serve just fine for mobile encoding until the hardware catches up. But if most existing phones can't decode it efficiently, it's going to be pretty stuck until that situation changes.
- thomastjeffery 9y ago
  I'd argue that it's more important to get encoders working, since re-encoding lossy->lossy is always a bad idea.
  - brigade 9y ago
    There isn’t really a video hosting site that doesnt do a lossy transcode of all uploads, unless you count Dropbox. The bitrate used to encode real-time captures is generally two or more times greater than the bitrate you want to distribute.
  - isostatic 9y ago
    Lossless 4K is 12 gigabit a second (even then there's chroma compression) Nobody edits at that bitrate, compression is concattenated multiple times. How a codec copes with real world multi-compression is key.
    - Bromskloss 9y ago
      > Nobody edits at that bitrate Interesting. Not even as a final rendering step after everything has been done in a compressed "preview mode"?
      - isostatic 9y ago
        Setting aside high end production houses. Even then things aiming for TV get compressed to a mezzanine format that's heavily compressed (in comparison), then compressed again for TX
    - simias 9y ago
      4K is common in TV production. In my experience lossy compression is a big no-no in TV and they do edit 4K losslessly, although in order to transfer it they generally split it as 4 times 3G links (either interleaving the pixels or generating a 2x2 mosaic). 12G links exist but 4x3G is still vastly more popular due to being backward compatible with existing equipment. Color correction, scaling, blending and compositing are all generally done with uncompressed video. Of course once mastered the video gets heavily compressed to be sent to the consumers.
      - isostatic 9y ago
        Ok I shouldn't use the word nobody, I admonished someone for using that word a few days ago, but it's very rare, especially outside high end production houses. Even 12G sdi is compressed - we consider 4:2:2 and 10bit uncompressed, but 4:4:4 is less compressed than 4:2:2. Once you go into interlace or stupid things like 24p you're throwing away tons of data. 2022-6 seems to be the transmission method of choice so far, 2110 may well take over, but there are significant compatibility issues. Personally I'd prefer everything originated as 4:4:4 10 bit 150fps (no drop frames), resolution I'm less concerned with. I work in news though, broadcast quality is what we broadcast. 5mbit h264 PPP with a vbv of 200k is fine for a talking head internationally. 5 times that is plenty for an actual broadcast program with astons and stuff.

gsich 9y ago

Adoption is forced by pirated stuff. If scene groups start to use it, people will want to play it. And those groups don't care if you can play whole movies on your phone. Because thats not what a phone is mainly for.

bri3d 9y ago
Movie and music piracy has taken a prodigal nosedive to the point that I don't think this would be true at all anyway. In addition streaming video and short form content dwarfs film and TV consumption, making high mobile efficiency paramount.
- weberc2 9y ago
  Prodigal? Proverbial?
  - B1FF_PSUVM 9y ago
    Prodigious, probably.
whatever_dude 9y ago
That's not really true. The data from YouTube or Netflix dwarfs what you'd get from a torrent distributor of a show or whatever. Some pirate/niche markets may adopt a codec solely for some arbitrary reason, but those reasons (say, maximum compression, good open source tools) are normally orthogonal to what a streaming business usually needs (compression, but also hardware/software compatibility for all sorts of form factors, and commercial unencumberment).
- baddox 9y ago
  I bet you’re right. AFAIK pretty much all but the cheapest TVs sold now (or at least 4K TVs) support HEVC specifically to support builtin apps for Netflix and similar. Unfortunately YouTube uses the competing VP9, which my 2 year old Vizio 4K TV does not support, but apparently devices like the Chromecast Ultra and 4K Apple TV support it as well.

twotwotwo 9y ago

Yes, Google is developing a free AV1 HW decoder, as discussed near the end of this (interesting!) deck: https://parisvideotech.com/wp-content/uploads/2017/07/AOM-AV1-Video-Tech-meet-up.pdf https://parisvideotech.com/wp-content/uploads/2017/07/AOM-AV... For what it's worth, Intel shipped "hybrid" implementations of HEVC and VP9 for older products where the CPU does some work but the GPU's shader cores do shader-y bits (https://www.anandtech.com/show/10610/intel-announces-7th-gen-kaby-lake-14nm-plus-six-notebook-skus-desktop-coming-in-january/3 https://www.anandtech.com/show/10610/intel-announces-7th-gen...). I don't know how far that approach gets you, especially in mobile settings where you really care about power use. But you might see that before complete hardware decoders everywhere. A use case that might be interesting before full hardware acceleration is to provide an improved still image format. The graph in the post is a great demonstration that improving video encoding and improving image encoding are closely related these days. AV1 has backing and went through IP review that may help it avoid being mired in a patent mess like previous proposals to replace JPEG. (Apple is using .heic, based on HEVC intra coding. Android'll support .heic in the next major version, but the HEVC patent situation will likely constrain how many external tools, platforms, etc. pick it up.) There's work on defining an AVIF format (https://github.com/AOMediaCodec/av1-avif https://github.com/AOMediaCodec/av1-avif) and tech demos look great (https://people.xiph.org/~tdaede/av1stilldemo/ https://people.xiph.org/~tdaede/av1stilldemo/). So, maybe fewer ugly JPEG artifacts in our future, partly thanks to that deringing filter mentioned at the end of Monty's post. (I said more or less the stuff about stills above, with a couple more details and links, at https://news.ycombinator.com/item?id=16699000 https://news.ycombinator.com/item?id=16699000)

asterius 9y ago
There is some interesting Q&A at the end of Pascal's talk. My schoolboy French is not fast enough to do a proper translation unfortunately. He was asked about parallel (software) decoding and I think the gist was he thought it would be a bit difficult because of the dependencies between region images. https://parisvideotech.com/pvt-4-les-formats-et-codecs-du-futur-chez-mozilla/ https://parisvideotech.com/pvt-4-les-formats-et-codecs-du-fu...

lathiat 9y ago

It shouldn't surprise anyone to learn that a brand new just released video codec does not yet have hardware decode support.

noobermin 9y ago
Came here to comment this. Seriously, HN, relax. This is like the introductory blogpost, how is there supposed to be a meatspace hardware implementation yet?
- wmf 9y ago
  They're just balancing out the "AV1 is here and Netflix is dropping H.265 tomorrow" posts.
  - omtose 9y ago
    Is Netflix even using H265?
    - Aissen 9y ago
      Maybe on platforms where it's the only hardware-enabled codec (they don't always impose their will, although it's increasingly looking like they do).

zamalek 9y ago

Hardware acceleration for encoding is also pretty important. Entry-level game streamers need it (Twitch recently proved that this isn't a negligible market[1]). Playing back AV1 on a phone implies receiving it - which means that users need to store their videos to AV1 directly. [1]: https://www.forbes.com/sites/insertcoin/2018/04/07/ninjas-new-fortnite-twitch-records-5-million-followers-250000-subs-875000-a-month/#47ee5a52478f https://www.forbes.com/sites/insertcoin/2018/04/07/ninjas-ne...