Announcement

Collapse
No announcement yet.

Google's New VP8 Codec SDK Is Better, Faster

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Google's New VP8 Codec SDK Is Better, Faster

    Phoronix: Google's New VP8 Codec SDK Is Better, Faster

    Following Google opening up the VP8 video codec specification in May and launching the WebM container format, in July the developers behind FFmpeg created the ffvp8 decoder that was much faster than Google's own VP8 decoding library. Google has now, however, provided a new version of the VP8 Codec SDK that they have codenamed "Aylesbury" and it's designed to be better and much faster than their original release...

    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

  • #2
    FFMPEGs implementation still has its pros: it does not just use x86 optimizations, but also leverages existing DSP code for other platforms. For example I don't see any mention of ARM NEON optimizations done by Google, so they do not yet care about Android phones.

    And Google has yet to prove that WebM is any better on the patent front than H.264. How comes there have been no news on this since May?

    Comment


    • #3
      Originally posted by sturmflut View Post
      And Google has yet to prove that WebM is any better on the patent front than H.264. How comes there have been no news on this since May?
      What do you mean? The only way someone can PROVE there's no patent infringement is in a court of law. Obviously Google wouldn't release Webm if they thought it was a patent liability, so only way we will find out for sure is if someone (MPEGLA) tries to sue them and then having it decided in court.

      On the licencing side Webm is of course 100% better than h264 since Google gives everyone a worldwide, non-exclusive, no-charge, royalty-free patent license to use Webm.

      Great to see they're improving on the codec, a free web video codec evens the playing field and allows video content sites without huge backing capital to compete.

      Comment


      • #4
        Originally posted by sturmflut View Post
        And Google has yet to prove that WebM is any better on the patent front than H.264. How comes there have been no news on this since May?
        I'm guessing it's because the patent trolling was just a pile of FUD.

        Comment


        • #5
          Originally posted by XorEaxEax View Post
          On the licencing side Webm is of course 100% better than h264 since Google gives everyone a worldwide, non-exclusive, no-charge, royalty-free patent license to use Webm.
          Actually, the license was cooler than that.... it contained a single exception that overrode everything else -- if you look at it with the intention of, or use information pulled from it to attack google, then screw you, you have no right to even THINK about it and we'll sue your ass for license/copyright infringement until you're bankrupt.

          Now the real situation regarding patents and licensing is like this;
          Google it the one facing mpeg on this, effectively protecting everyone else. If mpeg attacks google, google is simply too big and no doubt has a bunch of patents in its portfolio that can PROBABLY be used to free up h.264. mpeg surely wouldn't attack google on this because if they did, google could probably force them to let h.264 out under GOOGLE'S licensing terms.

          Comment


          • #6
            So, Google's going to drop the ball sometime around the beginning of next year, huh? Sorry, had to say it...

            Anyway, that's good. I wonder what the ffmpeg people think about it.

            Comment


            • #7
              Originally posted by Nobu View Post
              So, Google's going to drop the ball sometime around the beginning of next year, huh? Sorry, had to say it...
              Drop what ball? They just made a new release 'Aylesbury' which contains speedups and better video quality, next year in Q1 they will make the next release 'Bali'. I saw nothing exceptional in the release plans for 'Bali', it will be focused on encoder speed (which is definately good since the current one is pretty slow imo), so what is this 'dropping the ball' talk?

              Comment


              • #8
                Originally posted by XorEaxEax View Post
                Drop what ball? They just made a new release 'Aylesbury' which contains speedups and better video quality, next year in Q1 they will make the next release 'Bali'. I saw nothing exceptional in the release plans for 'Bali', it will be focused on encoder speed (which is definately good since the current one is pretty slow imo), so what is this 'dropping the ball' talk?
                He was making a pun. 'Bali' is only 1 letter off from 'Ball'. To release a version of a product could in this case be called a code drop.

                So:
                Release the version codenamed 'Bali'
                Becomes:
                Drop the ball.

                Comment


                • #9
                  Originally posted by sturmflut View Post
                  FFMPEGs implementation still has its pros: it does not just use x86 optimizations, but also leverages existing DSP code for other platforms. For example I don't see any mention of ARM NEON optimizations done by Google, so they do not yet care about Android phones.

                  And Google has yet to prove that WebM is any better on the patent front than H.264. How comes there have been no news on this since May?
                  I'm not sure about what was in the release notes, but there's been work being done on the ARM front for WebM. I don't know what the current status of NEON support is, but there's run-time CPU detection in place for ARM at this time, and plenty of asm optimization in place for ARM machines.

                  I just don't have any of the relevant hardware, so I don't really keep up with it myself, but I thought there was some support for NEON present.

                  Comment


                  • #10
                    Yeah, actually I thought it's name was 'ball'. Guess it's time to look into a new pair of glasses. XD

                    Comment

                    Working...
                    X