use 3 pts predictors for vorbis, slightly more efficient

Originally committed as revision 2960 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2959 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2958 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2957 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2956 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2955 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2954 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2953 to svn://svn.ffmpeg.org/ffmpeg/trunk

fix vorbis in ogg decoding

Originally committed as revision 2952 to svn://svn.ffmpeg.org/ffmpeg/trunk

 conform to latest spec
 remove get/put_s() not needed anymore
 update_packetheader() fixed (wasnt always working)
 store/load extradata & aspect ratio
 memleak fix
 dont use vorbis by default yet
 use frame_size from the codec instead of hardcoding it in the source

Originally committed as revision 2951 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2950 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2949 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2948 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2947 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2946 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2945 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2944 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2943 to svn://svn.ffmpeg.org/ffmpeg/trunk

split stream into valid mp3 frames, at least flv & nut absolutely need this, but probably most other formats too

Originally committed as revision 2942 to svn://svn.ffmpeg.org/ffmpeg/trunk

      because of the sample_size == 1 and MINOLTA hack relying on
      the information. So in a way, it's a hack of a hack.

      btw, if somebody knows why in the world even Apple's software
      thinks that for PCM 16bit sample_size == 1 please let me know.
      It clearly isn't documented that way.

Originally committed as revision 2941 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2940 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2939 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2938 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2937 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2936 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2935 to svn://svn.ffmpeg.org/ffmpeg/trunk

popular request

Originally committed as revision 2934 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2933 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2932 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2931 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2930 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2929 to svn://svn.ffmpeg.org/ffmpeg/trunk

	 * fixing a bug in movenc.c: for QT files hdrl should contain 'mhlr'

Originally committed as revision 2928 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2927 to svn://svn.ffmpeg.org/ffmpeg/trunk

     simple now, take a look for yourself).

   * additional optimizations of the decoder. It runs at 55fps now
     on my desktop and it used to be ~45fps.

Originally committed as revision 2926 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2925 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2924 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2923 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2922 to svn://svn.ffmpeg.org/ffmpeg/trunk

Originally committed as revision 2921 to svn://svn.ffmpeg.org/ffmpeg/trunk