Recherche avancée

Sur d’autres sites (3858)

• ffmpeg : how to convert iphone4 720p video into similar quality Flash video ?

  5 septembre 2011, par Sajee

  What are the proper arguments to ffmpeg to covert an iphone4 720p video into the similar quality Flash video ?

  I tried the following which resulted in a low quality Flash video :

  ffmpeg movie.mov movie.flv

  What are the right switches ?

• How Much H.264 In Each Encoder ?

  8 septembre 2010, par Multimedia Mike — General

  Thanks to my recent experiments with code coverage tools, I have a powerful new — admittedly somewhat specious — method of comparing programs. For example, I am certain that I have read on more than one occasion that Apple’s H.264 encoder sucks compared to x264 due, at least in part, to the Apple encoder’s alleged inability to exercise all of H.264′s features. I wonder how to test that claim ?

  Experiment
  Use code coverage tools to determine which H.264 encoder uses the most features.

  Assumptions

  • Movie trailers hosted by Apple will all be encoded with the same settings using Apple’s encoder.
  • Similarly, Yahoo’s movie trailers will be encoded with consistent settings using an unknown encoder.
  • Encoding a video using FFmpeg’s libx264-slow setting will necessarily throw a bunch of H.264′s features into the mix (I really don’t think this assumption holds much water, but I also don’t know what “standard” x264 settings are).

  Methodology

  • Grab a random Apple-hosted 1080p movie trailer and random Yahoo-hosted 1080p movie trailer from Dave’s Trailer Page.
  • Use libx264/FFmpeg with the ‘slow’ preset to encode Big Buck Bunny 1080p from raw PNG files.
  • Build FFmpeg with code coverage enabled.
  • Decode each file to raw YUV, ignore audio decoding, generate code coverage statistics using gcovr, reset stats after each run by deleting *.gcda files.

  Results

  • x264 1080p video : 9968 / 134203 lines
  • Apple 1080p trailer : 9968 / 134203 lines
  • Yahoo 1080p trailer : 9914 / 134203 lines

  I also ran this old x264-encoded file (ImperishableNightStage6Low.mp4) through the same test. It demonstrated the most code coverage with 10671 / 134203 lines.

  Conclusions
  Conclusions ? Ha ! Go ahead and jump all over this test. I’m already fairly confident that it’s impossible (or maybe just very difficult) to build a single H.264-encoded video that exercises every feature that FFmpeg’s decoder supports. For example, is it possible for a file to use both CABAC and CAVLC entropy methods ? If it’s possible, does any current encoder do that ?

• Sorenson Video 3 For Posterity

  12 août 2010, par Multimedia Mike — Software Museum

  On a recent dumpster dive, I procured a complete, never-been-opened copy of Sorenson Video 3 Professional Edition Compressor for Windows. And because I’m me, I thought it would be interesting to document it here :

  
  

  
  

  Not opened until now, anyway :

  
  
  

  I chuckled at the thought that I might open the user manual and find the complete data stream format listed therein. As expected, the guide offers a broad overview of video compression concepts (lossy vs. lossless, inter/intraframes, that kind of thing) and goes on to describe general guidelines for compressing different types of data. Then there is the feature reference. There are standard features and professional features (the latter includes things like bidirectional prediction, masking, and watermarking).

  I was hoping to figure out how to encode some video with this software. But I think I need full Quicktime Pro in order to do that. If you’re interested, here’s the user manual in PDF format : Sorenson Video 3 User Guide (626 Kbytes). Here’s an interesting claim from the chapter on audio compression :

  Most movies are made up of two parts, video and audio. Historically, the video portion of a digital movie was so large that the audio was only a minor piece of the puzzle. However, with Sorenson Video’s excellent compression capabilities it is possible to create a file where audio is the largest portion.

  I know it’s possible to do that, but is it really recommended ? I’m sure I have some samples in my vast repository where this is the case but it still doesn’t strike me as optimal for network delivery.