Slashdot Mirror

← Back to Stories (view on slashdot.org)

Jury Rules That H.264 is Not Patented

Posted by Zonk on from the yay-for-tivos-i-mean-broadcom dept.

Dr Kool, PhD writes "According to Bloomberg, a jury ruled against Qualcomm in their patent lawsuit against Broadcom. Qualcomm had sought $8.3 million in damages for patent infringement stemming from Broadcom's H.264 encoder/decoder chips. From the article: 'The patents, covering a way to compress high-definition video, are unenforceable in part because Qualcomm withheld information from the U.S. Patent and Trademark Office, jurors in San Diego said today after deliberating less than six hours.' This ruling clears the way for H.264 to become a widely adopted open standard."

14 of 111 comments (clear)

  1. Free ... of which patents? by rzei · · Score: 5, Interesting

    Doesn't this make H.264 only free of the two patents held by Qualcomm? There has to be dozens and dozens of other patents used as AFAIK H.264 is just a profile (AVC) of MPEG-4?

    And afaik again, MPEG-4 is very far from being patent encumbered.

    1. Re:Free ... of which patents? by mbone · · Score: 4, Interesting

      You let out the crucial internet portions

      Over-the-air free broadcast - There are no royalties for over-the-air free broadcast AVC video to markets of 100,000 or fewer households. For over-the-air free broadcast AVC video to markets of greater than 100,000 households, royalties are $10,000 per year per local market service (by a transmitter or transmitter simultaneously with repeaters, e.g., multiple transmitters serving one station).

      Internet broadcast (non-subscription, not title-by-title) - Since this market is still developing, no royalties will be payable for internet broadcast services (non-subscription, not title-by-title) during the initial term of the license (which runs through December 31, 2010) and then shall not exceed the over-the-air free broadcast TV encoding fee during the renewal term.

      So, nothing is owed between now and 2010 on the Intenet. However, the fee could be $ 10K per channel after then; if that's the case, then there will be trouble in 2011. Also note that it is unclear if the VOD is per download (in which case it is quite high) or per title offered (in which case, quite low).

  2. Not really by SanityInAnarchy · · Score: 3, Interesting

    The only truly, intentionally open standard I know of is Theora, and I really haven't heard much about it.

    For that matter, I haven't heard any measurements lately of AAC vs Vorbis, but it seems to me that unless Vorbis is actually better, the best way to encode a video would be h.264+aac, probably wrapped in ogm or mkv, but could also work as avi or mov.

    Of course, I often just keep the original DVD stream around, which means -- what -- mpeg2+aac?

    --
    Don't thank God, thank a doctor!
    1. Re:Not really by RalphBNumbers · · Score: 3, Interesting
      Well, IANAL, but looking at the summary of the AVC license here, specifically the portion quoted below, it seems like royalties are only required to be paid by "end product manufacturers". You could certainly argue that source code is not the end product, and thus you could distribute it without limit. And if you want to distribute object code as well, the only limit would be that no single person who builds it should distribute more than 100,000 compiled copies unless they want to pay royalties.
      I seem to recall that some existing OSS MPEG-4 related projects distribute source code only for that sort of reason.

      Royalties to be paid by end product manufacturers for an encoder, a decoder or both ("unit") begin at US $0.20 per unit after the first 100,000 units each year. There are no royalties on the first 100,000 units each year. Above 5 million units per year, the royalty is US $0.10 per unit.
      --
      "The worst tyrannies were the ones where a governance required its own logic on every embedded node." - Vernor Vinge
    2. Re:Not really by Jah-Wren+Ryel · · Score: 4, Interesting

      I seem to recall that some existing OSS MPEG-4 related projects distribute source code only for that sort of reason.

      They distribute source code because the courts (in the USA at least) have ruled that source code is speech, as in "freedom of" and binaries are not. Thus they are a lot better protected from claims of patent infringment if they stay away from the binaries.

      --
      When information is power, privacy is freedom.
  3. NO! There are ones in development though... by Anonymous Coward · · Score: 5, Interesting

    You have two major ones that can beat out H.264 in quality and file sizes... potentially.

    They are Snow, which is a experimental codec being developed out of the FFMPEG project, and there is Dirac which is being developed by the BBC as a open standard for web-based HD content.

    Both of these are based on 'wavelett' style technology which is something that is fairly unique about them. The downside though is that Snow, while being much simplier then Dirac, suffers from a lack of development and stability (not crash-iness, but change-iness). Dirac is not mature enough for use. Both of them still use WAY to much CPU to be usefull currently, but both offer possibilities of compression and quality that surpass even H.264.

    Theora is completely open, having the benifit from patent donated to open source by a corporation for their codecs, but it suffers from high CPU utilization and a very serious lack of visual quality.

    It's not like with Ogg vs MP3 or Flac vs whatever were those guys offer good compression, quality, and lower cpu usage as well as being open source. With Theora vs Mpeg4-related stuff (Xvid/Divx, h.264. AVC, etc) it is not realy in the same ballpack. It is more closely related to Mpeg1 in quality.

    And when I mean 'quality' I mean the ability to provide high quality image at high compression, which is the whole point behind things like Theora and H.264.

    Already Linux and Free software people have a good H.264 implimentation thanks to the FFmpeg people. Their mpeg4 Divx-stuff is already very high quality.. much better then anything from Xvid or Divx, they have the beginnings of very good H.264 support and have decoding and encoding speeds that rival the best propriatory codecs aviable. They need to fill out some of the H.264 features, but if this is true that H.264 is truly usable in Free software environment, then I expect that development will very quickly take off as the people become aware of this and Linux distros will want to jump on the opportunity to provide world-class HD support!

    This should also pave the way for future adoption of Dirac and maybe Snow since then the use of ffmpeg libs should increase in both Linux and Windows-land. Once people get used to it and programs start shipping with ffmpeg libs then this will make it easier for these projects to gain acceptance as ffmpeg is multi-codec and will include these open source technologies as they come out.

    1. Re:NO! There are ones in development though... by kripkenstein · · Score: 2, Interesting

      Both of these are based on 'wavelett' style technology

      Actually 'wavelet' is the correct spelling (sorry to be pedantic). But you are right that wavelet applications are an interesting topic.

      In short, wavelets are like Fourier transforms, but they have a location, not just a frequency. Like with the FT, you can represent spatial data by wavelets, and the localization aspect turns out to be useful in practice, in particular for codecs (but it is also useful from a theoretical aspect, wavelets were - perhaps still are - somewhat in fashion among in statistical circles).

    2. Re:NO! There are ones in development though... by Mawbid · · Score: 2, Interesting

      I thought everyone ran away from wavelets because a couple of companies locked up the whole field with broad patents (without ever delivering the technology). Is that just an old myth?

      --
      Fuck the system? Nah, you might catch something.
    3. Re:NO! There are ones in development though... by pyite · · Score: 3, Interesting

      If JPEG-2000 is any sort of indication, then this is part of the problem. JPEG-2000 incorporates some very cool exploitations of the fact that there are redundancies in the location information in wavelet subbands, and can offer better compression because of it. There are two issues. 1) Is the quality/space savings enough to warrant a change? 2) Patents. It's hard to say which is the primary reason for JPEG-2000 not being adopted, but I'll go with #1 being the primary reason and #2 being part of the problem, but not as relevant because of #1. Video is still pretty big and at least in the US, it doesn't seem like there wil be any universal "order of magnitude" jumps in bandwidth anytime soon. So maybe there will be more of an impetus to use wavelet methods for video.

      --

      "Nature doesn't care how smart you are. You can still be wrong." - Richard Feynman

    4. Re:NO! There are ones in development though... by tim90402 · · Score: 4, Interesting

      Wavelets work well for a single image, but I don't think anyone has figured out how to improve on the block based motion compensation techniques used to exploit temporarl redundancy in most video coding. And once you are doing block based motion prediction, then the residuals tend to have a block structure that is better compressed with a block based transform, rather than wavelets.

  4. Slashdot editors: Only pretending to be editors? by Futurepower(R) · · Score: 3, Interesting

    Story Errors: I would have thought that, after all these years, Slashdot editors would have learned to be editors. Often Slashdot stories are posted that show not even the simplest examination, such as this one, that references an article that does not support was said in the Slashdot story.

    This is more of the real story Broadcom sees win for 'H.264' industry (January 27, 2007). However, the article does NOT say that the patents were invalidated; they have not been invalidated.

    This statement from the Slashdot story is incorrect: "This ruling clears the way for H.264 to become a widely adopted open standard." If that were true, it would be important, but it is not true, for three reasons: 1) The patents have not been invalidated (yet). 2) There can be an appeal. 3) There are other patents.

  5. Another reason patents don't make a lot of sense by jmichaelg · · Score: 2, Interesting

    I have no idea whether Qualcomm's idea rose to the ideal patent standard but I'd bet dollars to donuts the jury didn't either. Given the time constraints, they can't possibly learn enough to understand the technology to determine whether Qualcomm had a lousy patent or Broadcom was infringing. Patent enforcement decisions make about as much sense as flipping a coin.

    Patents are designed by and implemented by attorneys. They're the beneficiaries of this system, not the public nor the inventor. The inventors and public just end up getting screwed.

  6. Did anyone else notice this? by hasbeard · · Score: 2, Interesting

    Did anyone notice that one of the spokesmen for the companies had the title of Vice-President for Intellectual Property Litigation? I don't know that I want to do away with software patents altogether (maybe, I'm not sure), but it bothers me when a company has a department, evidently important enough to be headed by a vice president, dedicated to litigation. Here's another vote for some serious reform in the patent system.

  7. What Qualcomm Wants by Anonymous Coward · · Score: 1, Interesting

    Qualcomm wants to control video on mobile devices.
    http://www.qualcomm.com/mediaflo/index.shtml

    That's why they spent so much time and money ..er..'convincing' regulators
    to allow them to take over part of the spectrum for mobile video transmission.
    Verizon and other carriers want this so they can move video off of their
    digital voice lines and on to something parallel with a different infrastructure.
    An infrastructure that, no doubt, the carriers will recieve loads of federal funding
    to complete (even though it won't be opened up to benefit anyone but Qualcomm financially).

    I don't think this loss is much of a blow to them really. They have many other chip
    monopolies to exploit.