Slashdot Mirror

← Back to Stories (view on slashdot.org)

VP8 Codec Coming To FFmpeg

Posted by kdawson on from the small-fast dept.

Jim Buzbee writes "Interested in Google's VP8 codec? Well, so were the FFmpeg guys, so they went ahead and wrote their own native decoder in only 1,400 lines of unique code. They were able to keep the line-count low by relying on heavy reuse from the existing H.264 codebase."

23 of 218 comments (clear)

  1. Hmmm... by Anonymous Coward · · Score: 5, Interesting

    Is anyone else worried by..

    They were able to keep the line-count low by relying on heavy reuse from the existing H.264 codebase."

    I bet the MPEG-LA will see that as proof that it violates their patents.

    1. Re:Hmmm... by Nerdfest · · Score: 3, Informative

      Just that is enough. That alone is enough in some cases to even outweigh some disadvantages.

    2. Re:Hmmm... by poetmatt · · Score: 3, Interesting

      so you're saying an x264 development blog by an x264 developer is going to be biased against vp8, has been quoted a million times, and has no real world tests (there are real world tests out there). color me surprised! /sarcasm.

      Here a real article, trollop.

      Saying that H264 is better or worse than vp8 shows straight up ignorance because they both have specific scenarios which they cater to. To avoid recognizing that is a lie.

      In the real world, studies have shown the two perform quite similarly, actually. Also, at the rate VP8 adoption is going MPEG is going to have to sue a lot of people, and they're going to lose in public image among other things.

    3. Re:Hmmm... by TheKidWho · · Score: 3, Insightful

      Is it? The vast majority of mobile phones including Apples iPhone/iPod/iPad devices have hardware decoding of H.264. Can the same be said of VP8?

    4. Re:Hmmm... by The+New+Andy · · Score: 4, Insightful
      ... or alternatively, it means that Google has found that it owns patents to bits inside H.264. So then as soon as someone sues Google (or "any entity") for stuff in VP8 they lose the right to use the bits of H.264 which are covered by patents that Google acquired when they purchased on2.

      I wouldn't be surprised if the ace up Google's sleeve is a patent on something which is key in both H.264 and VP8.

    5. Re:Hmmm... by MrHanky · · Score: 5, Insightful

      No, and it never will. Apple's support for "open standards" is limited to only support for such standards when they depend on proprietary formats like AAC, mp3, h.264, etc. No support for Vorbis, Theora, VP8 or anything that can be implemented freely without a patent license. You wouldn't want free software to be able to compete, would you?

    6. Re:Hmmm... by Anonymous Coward · · Score: 4, Insightful

      They were able to keep the line-count low by relying on heavy reuse from the existing H.264 codebase."

      I bet the MPEG-LA will see that as proof that it violates their patents.

      They might, but that's not how patents work. MPEG relies on old math, when MPEG got together to write the spec they were deciding what the constants should be. For example, say they decided to use 20 taps in some filter, then one of the guys in the room went back to his company and said "patent using 20 taps in filter Y, stat!"* To avoid that patent all you need to do is use a different number of taps, say lets say 24 for concreteness.

      ffmpeg decodes many different formats, so they use reusable code. It's not surprising that they have a function that just implements the age old function and takes a parameter for the number of taps you want.

      *In some cases the constant was already patented, but because the MPEG process is "patent agnostic" there is no incentive to use some other non-patented constant and the guys owning a patent on that constant are in the room so they have every incentive to argue for using that constant.

    7. Re:Hmmm... by TheKidWho · · Score: 3, Insightful

      Free software should be able to compete and it does, unfortunately most free software is usually around 5 years behind the state of the art.

    8. Re:Hmmm... by BZ · · Score: 5, Informative

      > Can the same be said of VP8?

      Not yet. Of course the format is less than 2 months old, and there _were_ several hardware companies who committed to implementing support for it at launch.

      Also, note that mobile phones don't support "hardware decoding of H.264". They support hardware-acceleration of operations needed to decode a particular profile of H.264: the Basic profile. The one that has lower quality output than VP8 does.

      So if you start bringing the hardware accel issue into the picture, then your quality metrics are suddenly in VP8's favor...

      All of which is to say that the situation is complicated. ;)

    9. Re:Hmmm... by tepples · · Score: 4, Informative

      "Hardware decoding" on these ARM-based devices usually doesn't mean H.264 is implemented directly in silicon. A lot of these codecs with "hardware support" are implemented either on a secondary CPU optimized for digital signal processing, which might even be a GPU running shaders. It isn't expected to be too difficult to port VP8 decoders to these DSPs because, as the article points out, VP8 is so similar to H.264's baseline profile that it has been called H.264 with the patent numbers filed off.

    10. Re:Hmmm... by Virak · · Score: 5, Insightful

      The post the parent linked to goes into extensive detail about the technical aspects of the codec, has a real world comparison, a proper one, and is overall an excellent article. In contrast, the article you linked to uses poor quality source videos, JPEG for their comparison images, and by their own admission didn't even manage to use the same frame for both codecs in the images, among other problems. If you're calling that a "real article", you are in no position to be calling someone else a troll.

      And enough of these fucking asinine claims about the x264 developers being out to get your poor, precious VP8 that crop up every time someone posts that link. They don't work for MPEG. They don't make obscene mounts of money off of all the people using their free (as in both sense of the word) open source software. They're not secret Chinese agents working to destroy the West from within through the patent system. There is absolutely no motive for them to lie about this sort of thing. VP8 is simply not as good of a codec, and no amount of baseless accusations will change this.

    11. Re:Hmmm... by MrHanky · · Score: 4, Interesting

      Vorbis isn't any worse than the competition. Certainly better than mp3, and AFAIK with better encoders than AAC. Quality isn't the reason why Apple refuses to support it. Denying competition is.

    12. Re:Hmmm... by LWATCDR · · Score: 3, Insightful

      No it isn't
      Right now H.264 is free as in beer to every Windows, Mac, iPhone, iPod, iPad, Android, WinMo, and Palm WebOS users on the planet. And probably a good deal more people that I am leaving out.
      For the vast majority of them that is free enough.
      To make this work Google is going to have to get VP8 out on a lot of devices convince a lot of developers to produce video in that format.
      Will it happen? Maybe one can really hope.
      The other solution will be if we can ever get software patents overturned.
      BTW which I am all for.
      But to even state that just being free is good enough goes 100% counter to history.
      MP3 isn't free Vorbis Ogg is. You just don't see much music in Ogg format do you?
      You have to have the support in devices to make a format fly.

      --
      See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
    13. Re:Hmmm... by MrHanky · · Score: 4, Insightful

      Calling the MPEG-LA a cartel would be more accurate than calling it "an international standards body".

    14. Re:Hmmm... by Neil · · Score: 4, Insightful

      But, if the original poster's speculation were true, it would put Google in the traditional role of a technology patent holder who holds a defensive arsenal of patents: if MPEG-LA makes a fuss about aspects of VP8 which they claim infringe MPEG-LA patents, then Google can threaten to retaliate by suing everyone in the world who is currently shipping an implementation of H.264 for infringement of the On2/VP8 patents (and so publicly demonstrate the fact that being an licensee of the MPEG-LA H.264 pool doesn't protect one from all patent claims, and provides no insurance or indemnity).

      Stalemate. Mutually-assured-destruction stand-off. Result: VP8 available for royalty-free for use, without MPEG-LA interference.

      But only if Google really have inherited some killer On2 patents as part of their acquisition. I hope they have - it would make sense of their strategy and confidence in VP8 if this kind of thing were going on in the background.

    15. Re:Hmmm... by Svartalf · · Score: 4, Insightful

      And do you know WHY it should be able to (I know you do because you brought it up- this is for the crowd that don't get what I'm about to mention here...)?

      This is because not a single one of the mobile devices out there with an A8 SoC really use "dedicated hardware" to decode h.264 or any of the other codecs you care to mention for video or audio.

      What do they use?

      A high-performance DSP chip. Not. Dedicated. Hardware.

      The same goes for anything with a Snapdragon or similar SoC. In fact, most devices don't use dedicated anything because you'd ned a bunch of special silicon for MPEG 1/2/4, MP3, WMA, etc. When you think about it, throwing a bunch of DSP muscle at the problem is cheaper than the dedicated hardware for all but a narrow range of applications.

      All one has to do write a DSP program for the codec in question and go for most devices.

      The "dedicated hardware" line is less of a real argument and more of a straw man argument that keeps getting trotted out every time some competing codec comes along.

      --
      I am not merely a "consumer" or a "taxpayer". I am a Citizen of the State of Texas
    16. Re:Hmmm... by fuzzyfuzzyfungus · · Score: 4, Informative

      It should be noted that the MPEG-LA specifically does not "take on the responsibility and risk for future patent infringement lawsuits". Your MPEG-LA license covers only MPEG-LA patent pool patents, it does not constitute any sort of promise that these patents are the only ones required to implement the spec, nor does it offer indemnification in the event that you are sued over your implementation of the standard for which you purchased an MPEG-LA license.

      Paying your protection money to them does ensure than none of the MPEG-LA members will sue you(at least over any patent that they have contributed to the MPEG-LA pool for the technology in question); but it confers no protection against any nonmember with a patent that they believe is being infringed upon(given that this group includes a little old mom 'n pop operation they call "AT&T" this isn't exactly a theoretical risk)...

    17. Re:Hmmm... by GameboyRMH · · Score: 5, Insightful

      5 years behind? In what way? Maybe PC gaming?

      Free browsers are leading the pack. Free media players are at least on par with their commercial equivalents (I'm being very generous here, I have yet to see any commercial player like VLC or Mplayer). Free OSes are now comparable in terms of usability to commercial ones, and in technical terms are years AHEAD of commercial OSes. Vorbis and Theora are comparable to their best closed counterparts. VP8/WebM has totally closed the gap with H.264, for those who like to split hairs about Theora.

      I use all free software on all of my computers (apart from most of my games, the OS on my gaming PC, and some of the stock apps on my PDA) and I sure don't feel like I'm missing anything.

      --
      "When information is power, privacy is freedom" - Jah-Wren Ryel
    18. Re:Hmmm... by MrNaz · · Score: 5, Insightful

      Perhaps in some, but lets not forget that some FOSS projects are either the best, or very close to the best in their class. A few that come to mind: Apache, PostgreSQL, Linux, BSD, Asterisk, Firefox, Thunderbird, OpenSSH... and that's just a 10 second brainstorm. All of those are either clearly the best in their breed, or at least comparable to the top end product.

      --
      I hate printers.
  2. Re:Good idea? by Anonymous Coward · · Score: 4, Insightful

    No, it's better to hide it and hope the problem goes away.

  3. It gets worse by Chrisq · · Score: 4, Interesting

    Is anyone else worried by..

    They were able to keep the line-count low by relying on heavy reuse from the existing H.264 codebase."

    I bet the MPEG-LA will see that as proof that it violates their patents.

    From tfa:
    since H.264 (the current industry standard video codec) and VP8 are highly similar, we can share code (and more importantly: optimizations) between FFmpeg’s H.264 and VP8 decoders (e.g. intra prediction).

  4. Re:What does that tell you about the patent trolls by fuzzyfuzzyfungus · · Score: 5, Insightful

    I think that it has been well understood, for some time, that VP8 is, by design, largely H.264-esque. Based on that "technical analysis of VP8 by an x264 developer" article that ran on slashdot shortly after Google's announcement, it would appear that the development strategy went more or less like this:

    1. Examine H.264
    2. Where the technique in question is not patent-encumbered, or patent encumbrances can be worked around, implement like H.264 did. Unless you have good reason to believe the contrary, your brilliant innovation probably isn't, and the guys who build decode silicon/write DSP firmware are not handing out prizes for novelty for its own sake.
    3. Where the technique in question is patent-encumbered, and the encumbrance cannot be compatibly worked around, implement the least-worst alternative.
    4. Get purchased by Google.

    Obviously, from a standpoint of legal defense and market acceptance, a codec of breathtaking novelty and power, looking like an algorithmic refugee from the comp-sci genocides of the 32nd century, would be preferable. Unfortunately, such isn't available by any known means. H.264 more or less represents the present consensus on best available technique in the field; but is heavily patent encumbered. The only real reason to deviate from it is to avoid patents. Assuming that they did, in fact, perform steps 2 and 3 correctly, they will have achieved approximately the best available result at the lowest possible cost.

  5. Re:What is the problem? by Abcd1234 · · Score: 4, Insightful

    Oy vay... how the heck did that get modded interesting?

    Yes. VP8 is supposed to be free. And the code Google released is free. But the issues surrounding VP8 have absolutely nothing, zero, nada, to do with copyright law.

    The question is: Does VP8 include technology/methods covered by patents contributed to the MPEG-LA H.264 patent pool? The fact that a huge amount of H.264-related code could be reused in their VP8 decoder strongly suggests that, at minimum, VP8 and H.264 are very similar, and that greatly increases the odds that this is the case, and that any codec implementing VP8 would violate one or more of those patents.

    That's bad.