1 \input texinfo @c -*- texinfo -*-
3 @settitle FFmpeg Documentation
6 @center @titlefont{FFmpeg Documentation}
12 The generic syntax is:
16 ffmpeg [[infile options][@option{-i} @var{infile}]]... @{[outfile options] @var{outfile}@}...
21 @c man begin DESCRIPTION
23 FFmpeg is a very fast video and audio converter. It can also grab from
24 a live audio/video source.
26 The command line interface is designed to be intuitive, in the sense
27 that FFmpeg tries to figure out all parameters that can possibly be
28 derived automatically. You usually only have to specify the target
31 FFmpeg can also convert from any sample rate to any other, and resize
32 video on the fly with a high quality polyphase filter.
34 As a general rule, options are applied to the next specified
35 file. Therefore, order is important, and you can have the same
36 option on the command line multiple times. Each occurrence is
37 then applied to the next input or output file.
39 * To set the video bitrate of the output file to 64kbit/s:
41 ffmpeg -i input.avi -b 64k output.avi
44 * To force the frame rate of the output file to 24 fps:
46 ffmpeg -i input.avi -r 24 output.avi
49 * To force the frame rate of the input file (valid for raw formats only)
50 to 1 fps and the frame rate of the output file to 24 fps:
52 ffmpeg -r 1 -i input.m2v -r 24 output.avi
55 The format option may be needed for raw input files.
57 By default, FFmpeg tries to convert as losslessly as possible: It
58 uses the same audio and video parameters for the outputs as the one
59 specified for the inputs.
61 @c man end DESCRIPTION
66 @include fftools-common-opts.texi
75 @item -i @var{filename}
79 Overwrite output files.
81 @item -t @var{duration}
82 Restrict the transcoded/captured video sequence
83 to the duration specified in seconds.
84 @code{hh:mm:ss[.xxx]} syntax is also supported.
86 @item -fs @var{limit_size}
87 Set the file size limit.
89 @item -ss @var{position}
90 Seek to given time position in seconds.
91 @code{hh:mm:ss[.xxx]} syntax is also supported.
93 @item -itsoffset @var{offset}
94 Set the input time offset in seconds.
95 @code{[-]hh:mm:ss[.xxx]} syntax is also supported.
96 This option affects all the input files that follow it.
97 The offset is added to the timestamps of the input files.
98 Specifying a positive offset means that the corresponding
99 streams are delayed by 'offset' seconds.
101 @item -timestamp @var{time}
104 @item -metadata @var{key}=@var{value}
105 Set a metadata key/value pair.
107 For example, for setting the title in the output file:
109 ffmpeg -i in.avi -metadata title="my title" out.flv
112 @item -v @var{number}
113 Set the logging verbosity level.
115 @item -target @var{type}
116 Specify target file type ("vcd", "svcd", "dvd", "dv", "dv50", "pal-vcd",
117 "ntsc-svcd", ... ). All the format options (bitrate, codecs,
118 buffer sizes) are then set automatically. You can just type:
121 ffmpeg -i myfile.avi -target vcd /tmp/vcd.mpg
124 Nevertheless you can specify additional options as long as you know
125 they do not conflict with the standard, as in:
128 ffmpeg -i myfile.avi -target vcd -bf 2 /tmp/vcd.mpg
131 @item -dframes @var{number}
132 Set the number of data frames to record.
134 @item -scodec @var{codec}
135 Force subtitle codec ('copy' to copy stream).
138 Add a new subtitle stream to the current output stream.
140 @item -slang @var{code}
141 Set the ISO 639 language code (3 letters) of the current subtitle stream.
145 @section Video Options
148 @item -b @var{bitrate}
149 Set the video bitrate in bit/s (default = 200 kb/s).
150 @item -vframes @var{number}
151 Set the number of video frames to record.
153 Set frame rate (Hz value, fraction or abbreviation), (default = 25).
155 Set frame size. The format is @samp{wxh} (ffserver default = 160x128, ffmpeg default = same as source).
156 The following abbreviations are recognized:
218 @item -aspect @var{aspect}
219 Set aspect ratio (4:3, 16:9 or 1.3333, 1.7777).
220 @item -croptop @var{size}
221 Set top crop band size (in pixels).
222 @item -cropbottom @var{size}
223 Set bottom crop band size (in pixels).
224 @item -cropleft @var{size}
225 Set left crop band size (in pixels).
226 @item -cropright @var{size}
227 Set right crop band size (in pixels).
228 @item -padtop @var{size}
229 Set top pad band size (in pixels).
230 @item -padbottom @var{size}
231 Set bottom pad band size (in pixels).
232 @item -padleft @var{size}
233 Set left pad band size (in pixels).
234 @item -padright @var{size}
235 Set right pad band size (in pixels).
236 @item -padcolor @var{hex_color}
237 Set color of padded bands. The value for padcolor is expressed
238 as a six digit hexadecimal number where the first two digits
239 represent red, the middle two digits green and last two digits
240 blue (default = 000000 (black)).
242 Disable video recording.
243 @item -bt @var{tolerance}
244 Set video bitrate tolerance (in bits, default 4000k).
245 Has a minimum value of: (target_bitrate/target_framerate).
246 In 1-pass mode, bitrate tolerance specifies how far ratecontrol is
247 willing to deviate from the target average bitrate value. This is
248 not related to min/max bitrate. Lowering tolerance too much has
249 an adverse effect on quality.
250 @item -maxrate @var{bitrate}
251 Set max video bitrate (in bit/s).
252 Requires -bufsize to be set.
253 @item -minrate @var{bitrate}
254 Set min video bitrate (in bit/s).
255 Most useful in setting up a CBR encode:
257 ffmpeg -i myfile.avi -b 4000k -minrate 4000k -maxrate 4000k -bufsize 1835k out.m2v
259 It is of little use elsewise.
260 @item -bufsize @var{size}
261 Set video buffer verifier buffer size (in bits).
262 @item -vcodec @var{codec}
263 Force video codec to @var{codec}. Use the @code{copy} special value to
264 tell that the raw codec data must be copied as is.
266 Use same video quality as source (implies VBR).
269 Select the pass number (1 or 2). It is used to do two-pass
270 video encoding. The statistics of the video are recorded in the first
271 pass into a log file (see also the option -passlogfile),
272 and in the second pass that log file is used to generate the video
273 at the exact requested bitrate.
274 On pass 1, you may just deactivate audio and set output to null,
275 examples for Windows and Unix:
277 ffmpeg -i foo.mov -vcodec libxvid -pass 1 -an -f rawvideo -y NUL
278 ffmpeg -i foo.mov -vcodec libxvid -pass 1 -an -f rawvideo -y /dev/null
281 @item -passlogfile @var{prefix}
282 Set two-pass log file name prefix to @var{prefix}, the default file name
283 prefix is ``ffmpeg2pass''. The complete file name will be
284 @file{PREFIX-N.log}, where N is a number specific to the output
288 Add a new video stream to the current output stream.
290 @item -vlang @var{code}
291 Set the ISO 639 language code (3 letters) of the current video stream.
295 @section Advanced Video Options
298 @item -pix_fmt @var{format}
299 Set pixel format. Use 'list' as parameter to show all the supported
301 @item -sws_flags @var{flags}
303 @item -g @var{gop_size}
304 Set the group of pictures size.
306 Use only intra frames.
309 @item -qscale @var{q}
310 Use fixed video quantizer scale (VBR).
312 minimum video quantizer scale (VBR)
314 maximum video quantizer scale (VBR)
316 maximum difference between the quantizer scales (VBR)
317 @item -qblur @var{blur}
318 video quantizer scale blur (VBR) (range 0.0 - 1.0)
319 @item -qcomp @var{compression}
320 video quantizer scale compression (VBR) (default 0.5).
321 Constant of ratecontrol equation. Recommended range for default rc_eq: 0.0-1.0
323 @item -lmin @var{lambda}
324 minimum video lagrange factor (VBR)
325 @item -lmax @var{lambda}
326 max video lagrange factor (VBR)
327 @item -mblmin @var{lambda}
328 minimum macroblock quantizer scale (VBR)
329 @item -mblmax @var{lambda}
330 maximum macroblock quantizer scale (VBR)
332 These four options (lmin, lmax, mblmin, mblmax) use 'lambda' units,
333 but you may use the QP2LAMBDA constant to easily convert from 'q' units:
335 ffmpeg -i src.ext -lmax 21*QP2LAMBDA dst.ext
338 @item -rc_init_cplx @var{complexity}
339 initial complexity for single pass encoding
340 @item -b_qfactor @var{factor}
341 qp factor between P- and B-frames
342 @item -i_qfactor @var{factor}
343 qp factor between P- and I-frames
344 @item -b_qoffset @var{offset}
345 qp offset between P- and B-frames
346 @item -i_qoffset @var{offset}
347 qp offset between P- and I-frames
348 @item -rc_eq @var{equation}
349 Set rate control equation (@pxref{FFmpeg formula
350 evaluator}) (default = @code{tex^qComp}).
351 @item -rc_override @var{override}
352 rate control override for specific intervals
353 @item -me_method @var{method}
354 Set motion estimation method to @var{method}.
355 Available methods are (from lowest to best quality):
358 Try just the (0, 0) vector.
367 exhaustive search (slow and marginally better than epzs)
370 @item -dct_algo @var{algo}
371 Set DCT algorithm to @var{algo}. Available values are:
374 FF_DCT_AUTO (default)
387 @item -idct_algo @var{algo}
388 Set IDCT algorithm to @var{algo}. Available values are:
391 FF_IDCT_AUTO (default)
415 Set error resilience to @var{n}.
418 FF_ER_CAREFUL (default)
424 FF_ER_VERY_AGGRESSIVE
427 @item -ec @var{bit_mask}
428 Set error concealment to @var{bit_mask}. @var{bit_mask} is a bit mask of
429 the following values:
432 FF_EC_GUESS_MVS (default = enabled)
434 FF_EC_DEBLOCK (default = enabled)
437 @item -bf @var{frames}
438 Use 'frames' B-frames (supported for MPEG-1, MPEG-2 and MPEG-4).
439 @item -mbd @var{mode}
443 FF_MB_DECISION_SIMPLE: Use mb_cmp (cannot change it yet in FFmpeg).
445 FF_MB_DECISION_BITS: Choose the one which needs the fewest bits.
447 FF_MB_DECISION_RD: rate distortion
451 Use four motion vector by macroblock (MPEG-4 only).
453 Use data partitioning (MPEG-4 only).
454 @item -bug @var{param}
455 Work around encoder bugs that are not auto-detected.
456 @item -strict @var{strictness}
457 How strictly to follow the standards.
459 Enable Advanced intra coding (h263+).
461 Enable Unlimited Motion Vector (h263+)
464 Deinterlace pictures.
466 Force interlacing support in encoder (MPEG-2 and MPEG-4 only).
467 Use this option if your input file is interlaced and you want
468 to keep the interlaced format for minimum losses.
469 The alternative is to deinterlace the input stream with
470 @option{-deinterlace}, but deinterlacing introduces losses.
472 Calculate PSNR of compressed frames.
474 Dump video coding statistics to @file{vstats_HHMMSS.log}.
475 @item -vstats_file @var{file}
476 Dump video coding statistics to @var{file}.
478 top=1/bottom=0/auto=-1 field first
479 @item -dc @var{precision}
481 @item -vtag @var{fourcc/tag}
482 Force video tag/fourcc.
485 @item -vbsf @var{bitstream_filter}
486 Bitstream filters available are "dump_extra", "remove_extra", "noise", "h264_mp4toannexb", "imxdump", "mjpegadump".
488 ffmpeg -i h264.mp4 -vcodec copy -vbsf h264_mp4toannexb -an out.h264
492 @section Audio Options
495 @item -aframes @var{number}
496 Set the number of audio frames to record.
498 Set the audio sampling frequency (default = 44100 Hz).
499 @item -ab @var{bitrate}
500 Set the audio bitrate in bit/s (default = 64k).
502 Set the audio quality (codec-specific, VBR).
503 @item -ac @var{channels}
504 Set the number of audio channels (default = 1).
506 Disable audio recording.
507 @item -acodec @var{codec}
508 Force audio codec to @var{codec}. Use the @code{copy} special value to
509 specify that the raw codec data must be copied as is.
511 Add a new audio track to the output file. If you want to specify parameters,
512 do so before @code{-newaudio} (@code{-acodec}, @code{-ab}, etc..).
514 Mapping will be done automatically, if the number of output streams is equal to
515 the number of input streams, else it will pick the first one that matches. You
516 can override the mapping using @code{-map} as usual.
520 ffmpeg -i file.mpg -vcodec copy -acodec ac3 -ab 384k test.mpg -acodec mp2 -ab 192k -newaudio
522 @item -alang @var{code}
523 Set the ISO 639 language code (3 letters) of the current audio stream.
526 @section Advanced Audio options:
529 @item -atag @var{fourcc/tag}
530 Force audio tag/fourcc.
531 @item -absf @var{bitstream_filter}
532 Bitstream filters available are "dump_extra", "remove_extra", "noise", "mp3comp", "mp3decomp".
535 @section Subtitle options:
538 @item -scodec @var{codec}
539 Force subtitle codec ('copy' to copy stream).
541 Add a new subtitle stream to the current output stream.
542 @item -slang @var{code}
543 Set the ISO 639 language code (3 letters) of the current subtitle stream.
545 Disable subtitle recording.
546 @item -sbsf @var{bitstream_filter}
547 Bitstream filters available are "mov2textsub", "text2movsub".
549 ffmpeg -i file.mov -an -vn -sbsf mov2textsub -scodec copy -f rawvideo sub.txt
553 @section Audio/Video grab options
556 @item -vc @var{channel}
557 Set video grab channel (DV1394 only).
558 @item -tvstd @var{standard}
559 Set television standard (NTSC, PAL (SECAM)).
561 Synchronize read on input.
564 @section Advanced options
567 @item -map @var{input_stream_id}[:@var{sync_stream_id}]
568 Set stream mapping from input streams to output streams.
569 Just enumerate the input streams in the order you want them in the output.
570 @var{sync_stream_id} if specified sets the input stream to sync
572 @item -map_meta_data @var{outfile}:@var{infile}
573 Set meta data information of @var{outfile} from @var{infile}.
575 Print specific debug info.
577 Show benchmarking information at the end of an encode.
578 Shows CPU time used and maximum memory consumption.
579 Maximum memory consumption is not supported on all systems,
580 it will usually display as 0 if not supported.
582 Dump each input packet.
584 When dumping packets, also dump the payload.
586 Only use bit exact algorithms (for codec testing).
588 Set RTP payload size in bytes.
590 Read input at native frame rate. Mainly used to simulate a grab device.
592 Loop over the input stream. Currently it works only for image
593 streams. This option is used for automatic FFserver testing.
594 @item -loop_output @var{number_of_times}
595 Repeatedly loop output for formats that support looping such as animated GIF
596 (0 will loop the output infinitely).
597 @item -threads @var{count}
599 @item -vsync @var{parameter}
601 0 Each frame is passed with its timestamp from the demuxer to the muxer
602 1 Frames will be duplicated and dropped to achieve exactly the requested
604 2 Frames are passed through with their timestamp or dropped so as to prevent
605 2 frames from having the same timestamp
606 -1 Chooses between 1 and 2 depending on muxer capabilities. This is the default method.
608 With -map you can select from
609 which stream the timestamps should be taken. You can leave either video or
610 audio unchanged and sync the remaining stream(s) to the unchanged one.
611 @item -async @var{samples_per_second}
612 Audio sync method. "Stretches/squeezes" the audio stream to match the timestamps,
613 the parameter is the maximum samples per second by which the audio is changed.
614 -async 1 is a special case where only the start of the audio stream is corrected
615 without any later correction.
617 Copy timestamps from input to output.
619 Finish encoding when the shortest input stream ends.
620 @item -dts_delta_threshold
621 Timestamp discontinuity delta threshold.
622 @item -muxdelay @var{seconds}
623 Set the maximum demux-decode delay.
624 @item -muxpreload @var{seconds}
625 Set the initial demux-decode delay.
628 @section Preset files
630 A preset file contains a sequence of @var{option}=@var{value} pairs,
631 one for each line, specifying a sequence of options which would be
632 awkward to specify on the command line. Lines starting with the hash
633 ('#') character are ignored and are used to provide comments. Check
634 the @file{ffpresets} directory in the FFmpeg source tree for examples.
636 Preset files are specified with the @code{vpre}, @code{apre},
637 @code{spre}, and @code{fpre} options. The @code{fpre} option takes the
638 filename of the preset instead of a preset name as input and can be
639 used for any kind of codec. For the @code{vpre}, @code{apre}, and
640 @code{spre} options, the options specified in a preset file are
641 applied to the currently selected codec of the same type as the preset
644 The argument passed to the @code{vpre}, @code{apre}, and @code{spre}
645 preset options identifies the preset file to use according to the
648 First ffmpeg searches for a file named @var{arg}.ffpreset in the
649 directories @file{$FFMPEG_DATADIR} (if set), and @file{$HOME/.ffmpeg}, and in
650 the datadir defined at configuration time (usually @file{PREFIX/share/ffmpeg})
651 in that order. For example, if the argument is @code{libx264-max}, it will
652 search for the file @file{libx264-max.ffpreset}.
654 If no such file is found, then ffmpeg will search for a file named
655 @var{codec_name}-@var{arg}.ffpreset in the above-mentioned
656 directories, where @var{codec_name} is the name of the codec to which
657 the preset file options will be applied. For example, if you select
658 the video codec with @code{-vcodec libx264} and use @code{-vpre max},
659 then it will search for the file @file{libx264-max.ffpreset}.
661 @anchor{FFmpeg formula evaluator}
662 @section FFmpeg formula evaluator
664 When evaluating a rate control string, FFmpeg uses an internal formula
667 The following binary operators are available: @code{+}, @code{-},
668 @code{*}, @code{/}, @code{^}.
670 The following unary operators are available: @code{+}, @code{-},
673 The following statements are available: @code{ld}, @code{st},
676 The following functions are available:
704 The following constants are available:
732 The file name can be @file{-} to read from standard input or to write
735 FFmpeg also handles many protocols specified with an URL syntax.
737 Use 'ffmpeg -protocols' to see a list of the supported protocols.
739 The protocol @code{http:} is currently used only to communicate with
740 FFserver (see the FFserver documentation). When FFmpeg will be a
741 video player it will also be used for streaming :-)
748 For streaming at very low bitrate application, use a low frame rate
749 and a small GOP size. This is especially true for RealVideo where
750 the Linux player does not seem to be very fast, so it can miss
751 frames. An example is:
754 ffmpeg -g 3 -r 3 -t 10 -b 50k -s qcif -f rv10 /tmp/b.rm
758 The parameter 'q' which is displayed while encoding is the current
759 quantizer. The value 1 indicates that a very good quality could
760 be achieved. The value 31 indicates the worst quality. If q=31 appears
761 too often, it means that the encoder cannot compress enough to meet
762 your bitrate. You must either increase the bitrate, decrease the
763 frame rate or decrease the frame size.
766 If your computer is not fast enough, you can speed up the
767 compression at the expense of the compression ratio. You can use
768 '-me zero' to speed up motion estimation, and '-intra' to disable
769 motion estimation completely (you have only I-frames, which means it
770 is about as good as JPEG compression).
773 To have very low audio bitrates, reduce the sampling frequency
774 (down to 22050 Hz for MPEG audio, 22050 or 11025 for AC-3).
777 To have a constant quality (but a variable bitrate), use the option
778 '-qscale n' when 'n' is between 1 (excellent quality) and 31 (worst
782 When converting video files, you can use the '-sameq' option which
783 uses the same quality factor in the encoder as in the decoder.
784 It allows almost lossless encoding.
790 @c man begin EXAMPLES
792 @section Video and Audio grabbing
794 FFmpeg can grab video and audio from devices given that you specify the input
798 ffmpeg -f oss -i /dev/dsp -f video4linux2 -i /dev/video0 /tmp/out.mpg
801 Note that you must activate the right video source and channel before
802 launching FFmpeg with any TV viewer such as xawtv
803 (@url{http://linux.bytesex.org/xawtv/}) by Gerd Knorr. You also
804 have to set the audio recording levels correctly with a
807 @section X11 grabbing
809 FFmpeg can grab the X11 display.
812 ffmpeg -f x11grab -s cif -i :0.0 /tmp/out.mpg
815 0.0 is display.screen number of your X11 server, same as
816 the DISPLAY environment variable.
819 ffmpeg -f x11grab -s cif -i :0.0+10,20 /tmp/out.mpg
822 0.0 is display.screen number of your X11 server, same as the DISPLAY environment
823 variable. 10 is the x-offset and 20 the y-offset for the grabbing.
825 @section Video and Audio file format conversion
827 * FFmpeg can use any supported file format and protocol as input:
831 * You can use YUV files as input:
834 ffmpeg -i /tmp/test%d.Y /tmp/out.mpg
837 It will use the files:
839 /tmp/test0.Y, /tmp/test0.U, /tmp/test0.V,
840 /tmp/test1.Y, /tmp/test1.U, /tmp/test1.V, etc...
843 The Y files use twice the resolution of the U and V files. They are
844 raw files, without header. They can be generated by all decent video
845 decoders. You must specify the size of the image with the @option{-s} option
846 if FFmpeg cannot guess it.
848 * You can input from a raw YUV420P file:
851 ffmpeg -i /tmp/test.yuv /tmp/out.avi
854 test.yuv is a file containing raw YUV planar data. Each frame is composed
855 of the Y plane followed by the U and V planes at half vertical and
856 horizontal resolution.
858 * You can output to a raw YUV420P file:
861 ffmpeg -i mydivx.avi hugefile.yuv
864 * You can set several input files and output files:
867 ffmpeg -i /tmp/a.wav -s 640x480 -i /tmp/a.yuv /tmp/a.mpg
870 Converts the audio file a.wav and the raw YUV video file a.yuv
873 * You can also do audio and video conversions at the same time:
876 ffmpeg -i /tmp/a.wav -ar 22050 /tmp/a.mp2
879 Converts a.wav to MPEG audio at 22050 Hz sample rate.
881 * You can encode to several formats at the same time and define a
882 mapping from input stream to output streams:
885 ffmpeg -i /tmp/a.wav -ab 64k /tmp/a.mp2 -ab 128k /tmp/b.mp2 -map 0:0 -map 0:0
888 Converts a.wav to a.mp2 at 64 kbits and to b.mp2 at 128 kbits. '-map
889 file:index' specifies which input stream is used for each output
890 stream, in the order of the definition of output streams.
892 * You can transcode decrypted VOBs:
895 ffmpeg -i snatch_1.vob -f avi -vcodec mpeg4 -b 800k -g 300 -bf 2 -acodec libmp3lame -ab 128k snatch.avi
898 This is a typical DVD ripping example; the input is a VOB file, the
899 output an AVI file with MPEG-4 video and MP3 audio. Note that in this
900 command we use B-frames so the MPEG-4 stream is DivX5 compatible, and
901 GOP size is 300 which means one intra frame every 10 seconds for 29.97fps
902 input video. Furthermore, the audio stream is MP3-encoded so you need
903 to enable LAME support by passing @code{--enable-libmp3lame} to configure.
904 The mapping is particularly useful for DVD transcoding
905 to get the desired audio language.
907 NOTE: To see the supported input formats, use @code{ffmpeg -formats}.
909 * You can extract images from a video, or create a video from many images:
911 For extracting images from a video:
913 ffmpeg -i foo.avi -r 1 -s WxH -f image2 foo-%03d.jpeg
916 This will extract one video frame per second from the video and will
917 output them in files named @file{foo-001.jpeg}, @file{foo-002.jpeg},
918 etc. Images will be rescaled to fit the new WxH values.
920 If you want to extract just a limited number of frames, you can use the
921 above command in combination with the -vframes or -t option, or in
922 combination with -ss to start extracting from a certain point in time.
924 For creating a video from many images:
926 ffmpeg -f image2 -i foo-%03d.jpeg -r 12 -s WxH foo.avi
929 The syntax @code{foo-%03d.jpeg} specifies to use a decimal number
930 composed of three digits padded with zeroes to express the sequence
931 number. It is the same syntax supported by the C printf function, but
932 only formats accepting a normal integer are suitable.
934 * You can put many streams of the same type in the output:
937 ffmpeg -i test1.avi -i test2.avi -vcodec copy -acodec copy -vcodec copy -acodec copy test12.avi -newvideo -newaudio
940 In addition to the first video and audio streams, the resulting
941 output file @file{test12.avi} will contain the second video
942 and the second audio stream found in the input streams list.
944 The @code{-newvideo}, @code{-newaudio} and @code{-newsubtitle}
945 options have to be specified immediately after the name of the output
946 file to which you want to add them.
952 @settitle FFmpeg video converter
955 ffplay(1), ffprobe(1), ffserver(1) and the FFmpeg HTML documentation
959 The FFmpeg developers