Announcement

Collapse
No announcement yet.

FFmpeg 3.3 Brings Native Opus Encoder, Support For Spherical Videos

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

  • #11
    Originally posted by LinAGKar View Post

    Or maybe it's gonna end up like their Vorbis encoder, so it's crap and they have to tell people not to use it.
    The guy writing the native opus encoder is the same one who revamped FFmpeg's internal AAC encoder and made it on par with fdk-aac quality wise, so there's that.

    Comment


    • #12
      https://www.ffmpeg.org/ffmpeg-codecs.html#opus says: “This is a native FFmpeg encoder for the Opus format. Currently its in development and only implements the CELT part of the codec. Its quality is usually worse and at best is equal to the libopus encoder.”

      Comment


      • #13
        monolithic is bad mmhhkay.
        Where is the standard for registering media encoder/decoders?

        Comment


        • #14
          Super happy about support for spherical. Linux support for spherical videos and photos is kind of... nonexistent. I'm aware of what exists, but it's not built into the standard viewers / players, nor does it exist as a standard package you can just apt-get.
          And yes, it's important. I've got spherical photos and videos from my Ricoh Theta S on some pretty important life events.

          Comment


          • #15
            Originally posted by cj.wijtmans View Post
            monolithic is bad mmhhkay.
            Where is the standard for registering media encoder/decoders?
            Bullshit. E.g. when the FFT is sped up, every codec using that benefits automatically.

            Comment


            • #16
              I would have consider this quite important to wet the appetite:

              DNxHR 444 and HQX encoding

              Comment


              • #17
                Originally posted by uid313 View Post
                The decoders, parsers and interpreters needs to be rewritten in Rust for safety against vulnerabilities and exploits.
                Spare us. The FFmpeg teams needs to implement quality bounds checking all the other necessary overrun checks to avoid it in C11. Enough of this ``Rust'' is the solution to all our lazy ways.

                Comment


                • #18
                  Originally posted by Marc Driftmeyer View Post
                  C11
                  c++17 is much more rewarding

                  Comment


                  • #19
                    Originally posted by Marc Driftmeyer View Post

                    Spare us. The FFmpeg teams needs to implement quality bounds checking all the other necessary overrun checks to avoid it in C11. Enough of this ``Rust'' is the solution to all our lazy ways.
                    Regardless of whether Rust in particular is an answer, "not being lazy with C" isn't one either. If "lazy ways" were the explanation for this type of bugs in low-level code, then all the ways of low-level programming are lazy (since those bugs seem to turn up everywhere) and "lazy" in that sense is a meaningless term. Also, laziness can be avoided on a individual basis but trying to avoid it in a population can be shown to be futile by using a simple statistical argument.
                    Last edited by tinko; 19 April 2017, 02:50 AM.

                    Comment


                    • #20
                      Originally posted by uid313 View Post
                      The decoders, parsers and interpreters needs to be rewritten in Rust for safety against vulnerabilities and exploits.
                      No bloating up, keep your rust away of FFmpeg. Linux world already has big mix of million languages that causes so much mess.

                      Comment

                      Working...
                      X