Recherche avancée

Sur d’autres sites (8088)

• aaccoder : rewrite PNS implementation

  6 septembre 2015, par Rostislav Pehlivanov

  aaccoder : rewrite PNS implementation
  This commit rewrites the PNS implementation and significantly
  
  improves sonic quality.
  The previous implementation marked an incredibly big amount
  
  of SFBs to predict when there was no need for this and this
  
  resulted in quite a large amount of artifacts. Also the
  
  quantization was incorrect (av_clip(4+log2f(...))) which
  
  led to 3x the intensity for PNS values leading to even more
  
  artifacts.
  This commit rewrites the PNS search function and introduces
  
  a major change : the PNS values are synthesized and are compared
  
  to the current coefficients in addition to passing through
  
  the revised checks to see whether PNS can be used.
  This decreases distortions and makes the current PNS implementation
  
  mainly focused on replacing any low-power non-zero bands as well
  
  as adding any zeroed bands back.
  The current encoder’s performance is enough (especially with
  
  IS) so PNS isn’t really required except to fill in the occasional
  
  few bands as well as extend any zeroed high frequency, so this
  
  combination which is already enabled by default works
  
  to get as much quality as it can within the bits allowed.
  Signed-off-by : Rostislav Pehlivanov <atomnuker@gmail.com>
  

  • [DH] libavcodec/aaccoder.c

• ffmpeg itsoffset doesn't work with pcm audio and raw 264 video

  28 janvier 2019, par Danny

  I need to create an MP4 container with data from a hardware encoder. The encoder outputs PCM 16-bit signed audio and raw H.264 ES video frames.

  This ffmpeg command line I’ve got works but the audio and video are not sync’d.

  From other posts I know that itsoffset only works with video and probably doesn’t work with -v copy

  I’ve confirmed that applying an itsoffset has no effect.

  Here’s the command line. Any suggestions ?

  One post suggested itsoffset works if you re-encode the video. But doing that needs CPU power and adds latency. (And what’s the point of a hardware encoder then ?)

  ffmpeg -f s16le -ar 44.1k -ac 2      -i Audio_20190110-165736.pcm 
  
         -fflags +genpts -itsoffset -5 -i Video_20190110-165736.264
  
         -c:v copy -c:a aac -b:a 128k 
  
         -f mp4 -movflags +faststart  output.mp4

  EDIT I

  Here is a link to the audio/video input files referenced in the above command.

• Texture bound to texture unit 0 is not renderable

  2 décembre 2018, par Trying_To_Understand

  I followed this tutorial https://github.com/phoboslab/jsmpeg. I have an open wesocket that gets the data. Sometimes, it’s take a long time to the ffmpeg (on a remote computer) to get the data from my ip camera, so I wrote a setTimeout function that waites for 10 sec to be "sure" that the ffmpeg getting the data from the ip camera. If I remove this setTimeout function this error will show up :

  [.WebGL-0000020CFA5C04D0]RENDER WARNING : texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering.

  This is my code for showing the stream to the client :

  this.dataService.getDataByParam(camera.CameraId.toString())
  
    .subscribe(
  
    (result: any) => {
  
  
  
      setTimeout(() => {
  
        this.ws = new WebSocket('ws://' + result);
  
        $(document).ready(() => {
  
  
  
          this.player = new jsmpeg(this.ws, {
  
            canvas: document.getElementById("canvas"),
  
            autoplay: true,
  
            audio: true,
  
            pauseWhenHidden: false,
  
          });
  
  
  
        })
  
      }, 10000);
  
    },
  
    (error: Error) => {
  
      this.streamErrorMsg = "Problem with the server, please try again after some time."
  
    });
  
   }

  How can i know for sure that the ffmpeg has connected successfully to the camera and started to convert the data on the client ? i want to avoid writting the setTimeout function.