1 \input texinfo @c -*- texinfo -*-
2 @documentencoding UTF-8
4 @settitle ffmpeg Documentation
6 @center @titlefont{ffmpeg Documentation}
15 ffmpeg [@var{global_options}] @{[@var{input_file_options}] -i @file{input_url}@} ... @{[@var{output_file_options}] @file{output_url}@} ...
18 @c man begin DESCRIPTION
20 @command{ffmpeg} is a very fast video and audio converter that can also grab from
21 a live audio/video source. It can also convert between arbitrary sample
22 rates and resize video on the fly with a high quality polyphase filter.
24 @command{ffmpeg} reads from an arbitrary number of input "files" (which can be regular
25 files, pipes, network streams, grabbing devices, etc.), specified by the
26 @code{-i} option, and writes to an arbitrary number of output "files", which are
27 specified by a plain output url. Anything found on the command line which
28 cannot be interpreted as an option is considered to be an output url.
30 Each input or output url can, in principle, contain any number of streams of
31 different types (video/audio/subtitle/attachment/data). The allowed number and/or
32 types of streams may be limited by the container format. Selecting which
33 streams from which inputs will go into which output is either done automatically
34 or with the @code{-map} option (see the Stream selection chapter).
36 To refer to input files in options, you must use their indices (0-based). E.g.
37 the first input file is @code{0}, the second is @code{1}, etc. Similarly, streams
38 within a file are referred to by their indices. E.g. @code{2:3} refers to the
39 fourth stream in the third input file. Also see the Stream specifiers chapter.
41 As a general rule, options are applied to the next specified
42 file. Therefore, order is important, and you can have the same
43 option on the command line multiple times. Each occurrence is
44 then applied to the next input or output file.
45 Exceptions from this rule are the global options (e.g. verbosity level),
46 which should be specified first.
48 Do not mix input and output files -- first specify all input files, then all
49 output files. Also do not mix options which belong to different files. All
50 options apply ONLY to the next input or output file and are reset between files.
54 To set the video bitrate of the output file to 64 kbit/s:
56 ffmpeg -i input.avi -b:v 64k -bufsize 64k output.avi
60 To force the frame rate of the output file to 24 fps:
62 ffmpeg -i input.avi -r 24 output.avi
66 To force the frame rate of the input file (valid for raw formats only)
67 to 1 fps and the frame rate of the output file to 24 fps:
69 ffmpeg -r 1 -i input.m2v -r 24 output.avi
73 The format option may be needed for raw input files.
75 @c man end DESCRIPTION
77 @chapter Detailed description
78 @c man begin DETAILED DESCRIPTION
80 The transcoding process in @command{ffmpeg} for each output can be described by
81 the following diagram:
84 _______ ______________
86 | input | demuxer | encoded data | decoder
87 | file | ---------> | packets | -----+
88 |_______| |______________| |
95 ________ ______________ |
97 | output | <-------- | encoded data | <----+
98 | file | muxer | packets | encoder
99 |________| |______________|
104 @command{ffmpeg} calls the libavformat library (containing demuxers) to read
105 input files and get packets containing encoded data from them. When there are
106 multiple input files, @command{ffmpeg} tries to keep them synchronized by
107 tracking lowest timestamp on any active input stream.
109 Encoded packets are then passed to the decoder (unless streamcopy is selected
110 for the stream, see further for a description). The decoder produces
111 uncompressed frames (raw video/PCM audio/...) which can be processed further by
112 filtering (see next section). After filtering, the frames are passed to the
113 encoder, which encodes them and outputs encoded packets. Finally those are
114 passed to the muxer, which writes the encoded packets to the output file.
117 Before encoding, @command{ffmpeg} can process raw audio and video frames using
118 filters from the libavfilter library. Several chained filters form a filter
119 graph. @command{ffmpeg} distinguishes between two types of filtergraphs:
122 @subsection Simple filtergraphs
123 Simple filtergraphs are those that have exactly one input and output, both of
124 the same type. In the above diagram they can be represented by simply inserting
125 an additional step between decoding and encoding:
128 _________ ______________
130 | decoded | | encoded data |
131 | frames |\ _ | packets |
132 |_________| \ /||______________|
134 simple _\|| | / encoder
135 filtergraph | filtered |/
141 Simple filtergraphs are configured with the per-stream @option{-filter} option
142 (with @option{-vf} and @option{-af} aliases for video and audio respectively).
143 A simple filtergraph for video can look for example like this:
146 _______ _____________ _______ ________
148 | input | ---> | deinterlace | ---> | scale | ---> | output |
149 |_______| |_____________| |_______| |________|
153 Note that some filters change frame properties but not frame contents. E.g. the
154 @code{fps} filter in the example above changes number of frames, but does not
155 touch the frame contents. Another example is the @code{setpts} filter, which
156 only sets timestamps and otherwise passes the frames unchanged.
158 @subsection Complex filtergraphs
159 Complex filtergraphs are those which cannot be described as simply a linear
160 processing chain applied to one stream. This is the case, for example, when the graph has
161 more than one input and/or output, or when output stream type is different from
162 input. They can be represented with the following diagram:
167 | input 0 |\ __________
169 \ _________ /| output 0 |
171 _________ \| complex | /
173 | input 1 |---->| filter |\
174 |_________| | | \ __________
177 _________ / |_________| |__________|
184 Complex filtergraphs are configured with the @option{-filter_complex} option.
185 Note that this option is global, since a complex filtergraph, by its nature,
186 cannot be unambiguously associated with a single stream or file.
188 The @option{-lavfi} option is equivalent to @option{-filter_complex}.
190 A trivial example of a complex filtergraph is the @code{overlay} filter, which
191 has two video inputs and one video output, containing one video overlaid on top
192 of the other. Its audio counterpart is the @code{amix} filter.
195 Stream copy is a mode selected by supplying the @code{copy} parameter to the
196 @option{-codec} option. It makes @command{ffmpeg} omit the decoding and encoding
197 step for the specified stream, so it does only demuxing and muxing. It is useful
198 for changing the container format or modifying container-level metadata. The
199 diagram above will, in this case, simplify to this:
202 _______ ______________ ________
204 | input | demuxer | encoded data | muxer | output |
205 | file | ---------> | packets | -------> | file |
206 |_______| |______________| |________|
210 Since there is no decoding or encoding, it is very fast and there is no quality
211 loss. However, it might not work in some cases because of many factors. Applying
212 filters is obviously also impossible, since filters work on uncompressed data.
214 @c man end DETAILED DESCRIPTION
216 @chapter Stream selection
217 @c man begin STREAM SELECTION
219 By default, @command{ffmpeg} includes only one stream of each type (video, audio, subtitle)
220 present in the input files and adds them to each output file. It picks the
221 "best" of each based upon the following criteria: for video, it is the stream
222 with the highest resolution, for audio, it is the stream with the most channels, for
223 subtitles, it is the first subtitle stream. In the case where several streams of
224 the same type rate equally, the stream with the lowest index is chosen.
226 You can disable some of those defaults by using the @code{-vn/-an/-sn/-dn} options. For
227 full manual control, use the @code{-map} option, which disables the defaults just
230 @c man end STREAM SELECTION
235 @include fftools-common-opts.texi
237 @section Main options
241 @item -f @var{fmt} (@emph{input/output})
242 Force input or output file format. The format is normally auto detected for input
243 files and guessed from the file extension for output files, so this option is not
244 needed in most cases.
246 @item -i @var{url} (@emph{input})
249 @item -y (@emph{global})
250 Overwrite output files without asking.
252 @item -n (@emph{global})
253 Do not overwrite output files, and exit immediately if a specified
254 output file already exists.
256 @item -stream_loop @var{number} (@emph{input})
257 Set number of times input stream shall be looped. Loop 0 means no loop,
258 loop -1 means infinite loop.
260 @item -c[:@var{stream_specifier}] @var{codec} (@emph{input/output,per-stream})
261 @itemx -codec[:@var{stream_specifier}] @var{codec} (@emph{input/output,per-stream})
262 Select an encoder (when used before an output file) or a decoder (when used
263 before an input file) for one or more streams. @var{codec} is the name of a
264 decoder/encoder or a special value @code{copy} (output only) to indicate that
265 the stream is not to be re-encoded.
269 ffmpeg -i INPUT -map 0 -c:v libx264 -c:a copy OUTPUT
271 encodes all video streams with libx264 and copies all audio streams.
273 For each stream, the last matching @code{c} option is applied, so
275 ffmpeg -i INPUT -map 0 -c copy -c:v:1 libx264 -c:a:137 libvorbis OUTPUT
277 will copy all the streams except the second video, which will be encoded with
278 libx264, and the 138th audio, which will be encoded with libvorbis.
280 @item -t @var{duration} (@emph{input/output})
281 When used as an input option (before @code{-i}), limit the @var{duration} of
282 data read from the input file.
284 When used as an output option (before an output url), stop writing the
285 output after its duration reaches @var{duration}.
287 @var{duration} must be a time duration specification,
288 see @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}.
290 -to and -t are mutually exclusive and -t has priority.
292 @item -to @var{position} (@emph{input/output})
293 Stop writing the output or reading the input at @var{position}.
294 @var{position} must be a time duration specification,
295 see @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}.
297 -to and -t are mutually exclusive and -t has priority.
299 @item -fs @var{limit_size} (@emph{output})
300 Set the file size limit, expressed in bytes. No further chunk of bytes is written
301 after the limit is exceeded. The size of the output file is slightly more than the
304 @item -ss @var{position} (@emph{input/output})
305 When used as an input option (before @code{-i}), seeks in this input file to
306 @var{position}. Note that in most formats it is not possible to seek exactly,
307 so @command{ffmpeg} will seek to the closest seek point before @var{position}.
308 When transcoding and @option{-accurate_seek} is enabled (the default), this
309 extra segment between the seek point and @var{position} will be decoded and
310 discarded. When doing stream copy or when @option{-noaccurate_seek} is used, it
313 When used as an output option (before an output url), decodes but discards
314 input until the timestamps reach @var{position}.
316 @var{position} must be a time duration specification,
317 see @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}.
319 @item -sseof @var{position} (@emph{input/output})
321 Like the @code{-ss} option but relative to the "end of file". That is negative
322 values are earlier in the file, 0 is at EOF.
324 @item -itsoffset @var{offset} (@emph{input})
325 Set the input time offset.
327 @var{offset} must be a time duration specification,
328 see @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}.
330 The offset is added to the timestamps of the input files. Specifying
331 a positive offset means that the corresponding streams are delayed by
332 the time duration specified in @var{offset}.
334 @item -timestamp @var{date} (@emph{output})
335 Set the recording timestamp in the container.
337 @var{date} must be a date specification,
338 see @ref{date syntax,,the Date section in the ffmpeg-utils(1) manual,ffmpeg-utils}.
340 @item -metadata[:metadata_specifier] @var{key}=@var{value} (@emph{output,per-metadata})
341 Set a metadata key/value pair.
343 An optional @var{metadata_specifier} may be given to set metadata
344 on streams, chapters or programs. See @code{-map_metadata}
345 documentation for details.
347 This option overrides metadata set with @code{-map_metadata}. It is
348 also possible to delete metadata by using an empty value.
350 For example, for setting the title in the output file:
352 ffmpeg -i in.avi -metadata title="my title" out.flv
355 To set the language of the first audio stream:
357 ffmpeg -i INPUT -metadata:s:a:0 language=eng OUTPUT
360 @item -disposition[:stream_specifier] @var{value} (@emph{output,per-stream})
361 Sets the disposition for a stream.
363 This option overrides the disposition copied from the input stream. It is also
364 possible to delete the disposition by setting it to 0.
366 The following dispositions are recognized:
375 @item hearing_impaired
376 @item visual_impaired
383 For example, to make the second audio stream the default stream:
385 ffmpeg -i in.mkv -disposition:a:1 default out.mkv
388 To make the second subtitle stream the default stream and remove the default
389 disposition from the first subtitle stream:
391 ffmpeg -i INPUT -disposition:s:0 0 -disposition:s:1 default OUTPUT
394 @item -program [title=@var{title}:][program_num=@var{program_num}:]st=@var{stream}[:st=@var{stream}...] (@emph{output})
396 Creates a program with the specified @var{title}, @var{program_num} and adds the specified
397 @var{stream}(s) to it.
399 @item -target @var{type} (@emph{output})
400 Specify target file type (@code{vcd}, @code{svcd}, @code{dvd}, @code{dv},
401 @code{dv50}). @var{type} may be prefixed with @code{pal-}, @code{ntsc-} or
402 @code{film-} to use the corresponding standard. All the format options
403 (bitrate, codecs, buffer sizes) are then set automatically. You can just type:
406 ffmpeg -i myfile.avi -target vcd /tmp/vcd.mpg
409 Nevertheless you can specify additional options as long as you know
410 they do not conflict with the standard, as in:
413 ffmpeg -i myfile.avi -target vcd -bf 2 /tmp/vcd.mpg
416 @item -dframes @var{number} (@emph{output})
417 Set the number of data frames to output. This is an obsolete alias for
418 @code{-frames:d}, which you should use instead.
420 @item -frames[:@var{stream_specifier}] @var{framecount} (@emph{output,per-stream})
421 Stop writing to the stream after @var{framecount} frames.
423 @item -q[:@var{stream_specifier}] @var{q} (@emph{output,per-stream})
424 @itemx -qscale[:@var{stream_specifier}] @var{q} (@emph{output,per-stream})
425 Use fixed quality scale (VBR). The meaning of @var{q}/@var{qscale} is
427 If @var{qscale} is used without a @var{stream_specifier} then it applies only
428 to the video stream, this is to maintain compatibility with previous behavior
429 and as specifying the same codec specific value to 2 different codecs that is
430 audio and video generally is not what is intended when no stream_specifier is
433 @anchor{filter_option}
434 @item -filter[:@var{stream_specifier}] @var{filtergraph} (@emph{output,per-stream})
435 Create the filtergraph specified by @var{filtergraph} and use it to
438 @var{filtergraph} is a description of the filtergraph to apply to
439 the stream, and must have a single input and a single output of the
440 same type of the stream. In the filtergraph, the input is associated
441 to the label @code{in}, and the output to the label @code{out}. See
442 the ffmpeg-filters manual for more information about the filtergraph
445 See the @ref{filter_complex_option,,-filter_complex option} if you
446 want to create filtergraphs with multiple inputs and/or outputs.
448 @item -filter_script[:@var{stream_specifier}] @var{filename} (@emph{output,per-stream})
449 This option is similar to @option{-filter}, the only difference is that its
450 argument is the name of the file from which a filtergraph description is to be
453 @item -filter_threads @var{nb_threads} (@emph{global})
454 Defines how many threads are used to process a filter pipeline. Each pipeline
455 will produce a thread pool with this many threads available for parallel processing.
456 The default is the number of available CPUs.
458 @item -pre[:@var{stream_specifier}] @var{preset_name} (@emph{output,per-stream})
459 Specify the preset for matching stream(s).
461 @item -stats (@emph{global})
462 Print encoding progress/statistics. It is on by default, to explicitly
463 disable it you need to specify @code{-nostats}.
465 @item -progress @var{url} (@emph{global})
466 Send program-friendly progress information to @var{url}.
468 Progress information is written approximately every second and at the end of
469 the encoding process. It is made of "@var{key}=@var{value}" lines. @var{key}
470 consists of only alphanumeric characters. The last key of a sequence of
471 progress information is always "progress".
473 @anchor{stdin option}
475 Enable interaction on standard input. On by default unless standard input is
476 used as an input. To explicitly disable interaction you need to specify
479 Disabling interaction on standard input is useful, for example, if
480 ffmpeg is in the background process group. Roughly the same result can
481 be achieved with @code{ffmpeg ... < /dev/null} but it requires a
484 @item -debug_ts (@emph{global})
485 Print timestamp information. It is off by default. This option is
486 mostly useful for testing and debugging purposes, and the output
487 format may change from one version to another, so it should not be
488 employed by portable scripts.
490 See also the option @code{-fdebug ts}.
492 @item -attach @var{filename} (@emph{output})
493 Add an attachment to the output file. This is supported by a few formats
494 like Matroska for e.g. fonts used in rendering subtitles. Attachments
495 are implemented as a specific type of stream, so this option will add
496 a new stream to the file. It is then possible to use per-stream options
497 on this stream in the usual way. Attachment streams created with this
498 option will be created after all the other streams (i.e. those created
499 with @code{-map} or automatic mappings).
501 Note that for Matroska you also have to set the mimetype metadata tag:
503 ffmpeg -i INPUT -attach DejaVuSans.ttf -metadata:s:2 mimetype=application/x-truetype-font out.mkv
505 (assuming that the attachment stream will be third in the output file).
507 @item -dump_attachment[:@var{stream_specifier}] @var{filename} (@emph{input,per-stream})
508 Extract the matching attachment stream into a file named @var{filename}. If
509 @var{filename} is empty, then the value of the @code{filename} metadata tag
512 E.g. to extract the first attachment to a file named 'out.ttf':
514 ffmpeg -dump_attachment:t:0 out.ttf -i INPUT
516 To extract all attachments to files determined by the @code{filename} tag:
518 ffmpeg -dump_attachment:t "" -i INPUT
521 Technical note -- attachments are implemented as codec extradata, so this
522 option can actually be used to extract extradata from any stream, not just
526 Disable automatically rotating video based on file metadata.
530 @section Video Options
533 @item -vframes @var{number} (@emph{output})
534 Set the number of video frames to output. This is an obsolete alias for
535 @code{-frames:v}, which you should use instead.
536 @item -r[:@var{stream_specifier}] @var{fps} (@emph{input/output,per-stream})
537 Set frame rate (Hz value, fraction or abbreviation).
539 As an input option, ignore any timestamps stored in the file and instead
540 generate timestamps assuming constant frame rate @var{fps}.
541 This is not the same as the @option{-framerate} option used for some input formats
542 like image2 or v4l2 (it used to be the same in older versions of FFmpeg).
543 If in doubt use @option{-framerate} instead of the input option @option{-r}.
545 As an output option, duplicate or drop input frames to achieve constant output
546 frame rate @var{fps}.
548 @item -s[:@var{stream_specifier}] @var{size} (@emph{input/output,per-stream})
551 As an input option, this is a shortcut for the @option{video_size} private
552 option, recognized by some demuxers for which the frame size is either not
553 stored in the file or is configurable -- e.g. raw video or video grabbers.
555 As an output option, this inserts the @code{scale} video filter to the
556 @emph{end} of the corresponding filtergraph. Please use the @code{scale} filter
557 directly to insert it at the beginning or some other place.
559 The format is @samp{wxh} (default - same as source).
561 @item -aspect[:@var{stream_specifier}] @var{aspect} (@emph{output,per-stream})
562 Set the video display aspect ratio specified by @var{aspect}.
564 @var{aspect} can be a floating point number string, or a string of the
565 form @var{num}:@var{den}, where @var{num} and @var{den} are the
566 numerator and denominator of the aspect ratio. For example "4:3",
567 "16:9", "1.3333", and "1.7777" are valid argument values.
569 If used together with @option{-vcodec copy}, it will affect the aspect ratio
570 stored at container level, but not the aspect ratio stored in encoded
571 frames, if it exists.
573 @item -vn (@emph{output})
574 Disable video recording.
576 @item -vcodec @var{codec} (@emph{output})
577 Set the video codec. This is an alias for @code{-codec:v}.
579 @item -pass[:@var{stream_specifier}] @var{n} (@emph{output,per-stream})
580 Select the pass number (1 or 2). It is used to do two-pass
581 video encoding. The statistics of the video are recorded in the first
582 pass into a log file (see also the option -passlogfile),
583 and in the second pass that log file is used to generate the video
584 at the exact requested bitrate.
585 On pass 1, you may just deactivate audio and set output to null,
586 examples for Windows and Unix:
588 ffmpeg -i foo.mov -c:v libxvid -pass 1 -an -f rawvideo -y NUL
589 ffmpeg -i foo.mov -c:v libxvid -pass 1 -an -f rawvideo -y /dev/null
592 @item -passlogfile[:@var{stream_specifier}] @var{prefix} (@emph{output,per-stream})
593 Set two-pass log file name prefix to @var{prefix}, the default file name
594 prefix is ``ffmpeg2pass''. The complete file name will be
595 @file{PREFIX-N.log}, where N is a number specific to the output
598 @item -vf @var{filtergraph} (@emph{output})
599 Create the filtergraph specified by @var{filtergraph} and use it to
602 This is an alias for @code{-filter:v}, see the @ref{filter_option,,-filter option}.
605 @section Advanced Video options
608 @item -pix_fmt[:@var{stream_specifier}] @var{format} (@emph{input/output,per-stream})
609 Set pixel format. Use @code{-pix_fmts} to show all the supported
611 If the selected pixel format can not be selected, ffmpeg will print a
612 warning and select the best pixel format supported by the encoder.
613 If @var{pix_fmt} is prefixed by a @code{+}, ffmpeg will exit with an error
614 if the requested pixel format can not be selected, and automatic conversions
615 inside filtergraphs are disabled.
616 If @var{pix_fmt} is a single @code{+}, ffmpeg selects the same pixel format
617 as the input (or graph output) and automatic conversions are disabled.
619 @item -sws_flags @var{flags} (@emph{input/output})
624 @item -rc_override[:@var{stream_specifier}] @var{override} (@emph{output,per-stream})
625 Rate control override for specific intervals, formatted as "int,int,int"
626 list separated with slashes. Two first values are the beginning and
627 end frame numbers, last one is quantizer to use if positive, or quality
631 Force interlacing support in encoder (MPEG-2 and MPEG-4 only).
632 Use this option if your input file is interlaced and you want
633 to keep the interlaced format for minimum losses.
634 The alternative is to deinterlace the input stream with
635 @option{-deinterlace}, but deinterlacing introduces losses.
637 Calculate PSNR of compressed frames.
639 Dump video coding statistics to @file{vstats_HHMMSS.log}.
640 @item -vstats_file @var{file}
641 Dump video coding statistics to @var{file}.
642 @item -vstats_version @var{file}
643 Specifies which version of the vstats format to use. Default is 2.
647 @code{frame= %5d q= %2.1f PSNR= %6.2f f_size= %6d s_size= %8.0fkB time= %0.3f br= %7.1fkbits/s avg_br= %7.1fkbits/s}
651 @code{out= %2d st= %2d frame= %5d q= %2.1f PSNR= %6.2f f_size= %6d s_size= %8.0fkB time= %0.3f br= %7.1fkbits/s avg_br= %7.1fkbits/s}
652 @item -top[:@var{stream_specifier}] @var{n} (@emph{output,per-stream})
653 top=1/bottom=0/auto=-1 field first
654 @item -dc @var{precision}
656 @item -vtag @var{fourcc/tag} (@emph{output})
657 Force video tag/fourcc. This is an alias for @code{-tag:v}.
658 @item -qphist (@emph{global})
660 @item -vbsf @var{bitstream_filter}
663 @item -force_key_frames[:@var{stream_specifier}] @var{time}[,@var{time}...] (@emph{output,per-stream})
664 @item -force_key_frames[:@var{stream_specifier}] expr:@var{expr} (@emph{output,per-stream})
665 Force key frames at the specified timestamps, more precisely at the first
666 frames after each specified time.
668 If the argument is prefixed with @code{expr:}, the string @var{expr}
669 is interpreted like an expression and is evaluated for each frame. A
670 key frame is forced in case the evaluation is non-zero.
672 If one of the times is "@code{chapters}[@var{delta}]", it is expanded into
673 the time of the beginning of all chapters in the file, shifted by
674 @var{delta}, expressed as a time in seconds.
675 This option can be useful to ensure that a seek point is present at a
676 chapter mark or any other designated place in the output file.
678 For example, to insert a key frame at 5 minutes, plus key frames 0.1 second
679 before the beginning of every chapter:
681 -force_key_frames 0:05:00,chapters-0.1
684 The expression in @var{expr} can contain the following constants:
687 the number of current processed frame, starting from 0
689 the number of forced frames
691 the number of the previous forced frame, it is @code{NAN} when no
692 keyframe was forced yet
694 the time of the previous forced frame, it is @code{NAN} when no
695 keyframe was forced yet
697 the time of the current processed frame
700 For example to force a key frame every 5 seconds, you can specify:
702 -force_key_frames expr:gte(t,n_forced*5)
705 To force a key frame 5 seconds after the time of the last forced one,
706 starting from second 13:
708 -force_key_frames expr:if(isnan(prev_forced_t),gte(t,13),gte(t,prev_forced_t+5))
711 Note that forcing too many keyframes is very harmful for the lookahead
712 algorithms of certain encoders: using fixed-GOP options or similar
713 would be more efficient.
715 @item -copyinkf[:@var{stream_specifier}] (@emph{output,per-stream})
716 When doing stream copy, copy also non-key frames found at the
719 @item -init_hw_device @var{type}[=@var{name}][:@var{device}[,@var{key=value}...]]
720 Initialise a new hardware device of type @var{type} called @var{name}, using the
721 given device parameters.
722 If no name is specified it will receive a default name of the form "@var{type}%d".
724 The meaning of @var{device} and the following arguments depends on the
729 @var{device} is the number of the CUDA device.
732 @var{device} is the number of the Direct3D 9 display adapter.
735 @var{device} is either an X11 display name or a DRM render node.
736 If not specified, it will attempt to open the default X11 display (@emph{$DISPLAY})
737 and then the first DRM render node (@emph{/dev/dri/renderD128}).
740 @var{device} is an X11 display name.
741 If not specified, it will attempt to open the default X11 display (@emph{$DISPLAY}).
744 @var{device} selects a value in @samp{MFX_IMPL_*}. Allowed values are:
755 If not specified, @samp{auto_any} is used.
756 (Note that it may be easier to achieve the desired result for QSV by creating the
757 platform-appropriate subdevice (@samp{dxva2} or @samp{vaapi}) and then deriving a
758 QSV device from that.)
761 @var{device} selects the platform and device as @emph{platform_index.device_index}.
763 The set of devices can also be filtered using the key-value pairs to find only
764 devices matching particular platform or device strings.
766 The strings usable as filters are:
768 @item platform_profile
769 @item platform_version
771 @item platform_vendor
772 @item platform_extensions
778 @item device_extensions
782 The indices and filters must together uniquely select a device.
786 @item -init_hw_device opencl:0.1
787 Choose the second device on the first platform.
789 @item -init_hw_device opencl:,device_name=Foo9000
790 Choose the device with a name containing the string @emph{Foo9000}.
792 @item -init_hw_device opencl:1,device_type=gpu,device_extensions=cl_khr_fp16
793 Choose the GPU device on the second platform supporting the @emph{cl_khr_fp16}
799 @item -init_hw_device @var{type}[=@var{name}]@@@var{source}
800 Initialise a new hardware device of type @var{type} called @var{name},
801 deriving it from the existing device with the name @var{source}.
803 @item -init_hw_device list
804 List all hardware device types supported in this build of ffmpeg.
806 @item -filter_hw_device @var{name}
807 Pass the hardware device called @var{name} to all filters in any filter graph.
808 This can be used to set the device to upload to with the @code{hwupload} filter,
809 or the device to map to with the @code{hwmap} filter. Other filters may also
810 make use of this parameter when they require a hardware device. Note that this
811 is typically only required when the input is not already in hardware frames -
812 when it is, filters will derive the device they require from the context of the
813 frames they receive as input.
815 This is a global setting, so all filters will receive the same device.
817 @item -hwaccel[:@var{stream_specifier}] @var{hwaccel} (@emph{input,per-stream})
818 Use hardware acceleration to decode the matching stream(s). The allowed values
819 of @var{hwaccel} are:
822 Do not use any hardware acceleration (the default).
825 Automatically select the hardware acceleration method.
828 Use VDPAU (Video Decode and Presentation API for Unix) hardware acceleration.
831 Use DXVA2 (DirectX Video Acceleration) hardware acceleration.
834 Use VAAPI (Video Acceleration API) hardware acceleration.
837 Use the Intel QuickSync Video acceleration for video transcoding.
839 Unlike most other values, this option does not enable accelerated decoding (that
840 is used automatically whenever a qsv decoder is selected), but accelerated
841 transcoding, without copying the frames into the system memory.
843 For it to work, both the decoder and the encoder must support QSV acceleration
844 and no filters must be used.
847 This option has no effect if the selected hwaccel is not available or not
848 supported by the chosen decoder.
850 Note that most acceleration methods are intended for playback and will not be
851 faster than software decoding on modern CPUs. Additionally, @command{ffmpeg}
852 will usually need to copy the decoded frames from the GPU memory into the system
853 memory, resulting in further performance loss. This option is thus mainly
856 @item -hwaccel_device[:@var{stream_specifier}] @var{hwaccel_device} (@emph{input,per-stream})
857 Select a device to use for hardware acceleration.
859 This option only makes sense when the @option{-hwaccel} option is also specified.
860 It can either refer to an existing device created with @option{-init_hw_device}
861 by name, or it can create a new device as if
862 @samp{-init_hw_device} @var{type}:@var{hwaccel_device}
863 were called immediately before.
866 List all hardware acceleration methods supported in this build of ffmpeg.
870 @section Audio Options
873 @item -aframes @var{number} (@emph{output})
874 Set the number of audio frames to output. This is an obsolete alias for
875 @code{-frames:a}, which you should use instead.
876 @item -ar[:@var{stream_specifier}] @var{freq} (@emph{input/output,per-stream})
877 Set the audio sampling frequency. For output streams it is set by
878 default to the frequency of the corresponding input stream. For input
879 streams this option only makes sense for audio grabbing devices and raw
880 demuxers and is mapped to the corresponding demuxer options.
881 @item -aq @var{q} (@emph{output})
882 Set the audio quality (codec-specific, VBR). This is an alias for -q:a.
883 @item -ac[:@var{stream_specifier}] @var{channels} (@emph{input/output,per-stream})
884 Set the number of audio channels. For output streams it is set by
885 default to the number of input audio channels. For input streams
886 this option only makes sense for audio grabbing devices and raw demuxers
887 and is mapped to the corresponding demuxer options.
888 @item -an (@emph{output})
889 Disable audio recording.
890 @item -acodec @var{codec} (@emph{input/output})
891 Set the audio codec. This is an alias for @code{-codec:a}.
892 @item -sample_fmt[:@var{stream_specifier}] @var{sample_fmt} (@emph{output,per-stream})
893 Set the audio sample format. Use @code{-sample_fmts} to get a list
894 of supported sample formats.
896 @item -af @var{filtergraph} (@emph{output})
897 Create the filtergraph specified by @var{filtergraph} and use it to
900 This is an alias for @code{-filter:a}, see the @ref{filter_option,,-filter option}.
903 @section Advanced Audio options
906 @item -atag @var{fourcc/tag} (@emph{output})
907 Force audio tag/fourcc. This is an alias for @code{-tag:a}.
908 @item -absf @var{bitstream_filter}
910 @item -guess_layout_max @var{channels} (@emph{input,per-stream})
911 If some input channel layout is not known, try to guess only if it
912 corresponds to at most the specified number of channels. For example, 2
913 tells to @command{ffmpeg} to recognize 1 channel as mono and 2 channels as
914 stereo but not 6 channels as 5.1. The default is to always try to guess. Use
915 0 to disable all guessing.
918 @section Subtitle options
921 @item -scodec @var{codec} (@emph{input/output})
922 Set the subtitle codec. This is an alias for @code{-codec:s}.
923 @item -sn (@emph{output})
924 Disable subtitle recording.
925 @item -sbsf @var{bitstream_filter}
929 @section Advanced Subtitle options
933 @item -fix_sub_duration
934 Fix subtitles durations. For each subtitle, wait for the next packet in the
935 same stream and adjust the duration of the first to avoid overlap. This is
936 necessary with some subtitles codecs, especially DVB subtitles, because the
937 duration in the original packet is only a rough estimate and the end is
938 actually marked by an empty subtitle frame. Failing to use this option when
939 necessary can result in exaggerated durations or muxing failures due to
940 non-monotonic timestamps.
942 Note that this option will delay the output of all data until the next
943 subtitle packet is decoded: it may increase memory consumption and latency a
946 @item -canvas_size @var{size}
947 Set the size of the canvas used to render subtitles.
951 @section Advanced options
954 @item -map [-]@var{input_file_id}[:@var{stream_specifier}][?][,@var{sync_file_id}[:@var{stream_specifier}]] | @var{[linklabel]} (@emph{output})
956 Designate one or more input streams as a source for the output file. Each input
957 stream is identified by the input file index @var{input_file_id} and
958 the input stream index @var{input_stream_id} within the input
959 file. Both indices start at 0. If specified,
960 @var{sync_file_id}:@var{stream_specifier} sets which input stream
961 is used as a presentation sync reference.
963 The first @code{-map} option on the command line specifies the
964 source for output stream 0, the second @code{-map} option specifies
965 the source for output stream 1, etc.
967 A @code{-} character before the stream identifier creates a "negative" mapping.
968 It disables matching streams from already created mappings.
970 A trailing @code{?} after the stream index will allow the map to be
971 optional: if the map matches no streams the map will be ignored instead
972 of failing. Note the map will still fail if an invalid input file index
973 is used; such as if the map refers to a non-existent input.
975 An alternative @var{[linklabel]} form will map outputs from complex filter
976 graphs (see the @option{-filter_complex} option) to the output file.
977 @var{linklabel} must correspond to a defined output link label in the graph.
979 For example, to map ALL streams from the first input file to output
981 ffmpeg -i INPUT -map 0 output
984 For example, if you have two audio streams in the first input file,
985 these streams are identified by "0:0" and "0:1". You can use
986 @code{-map} to select which streams to place in an output file. For
989 ffmpeg -i INPUT -map 0:1 out.wav
991 will map the input stream in @file{INPUT} identified by "0:1" to
992 the (single) output stream in @file{out.wav}.
994 For example, to select the stream with index 2 from input file
995 @file{a.mov} (specified by the identifier "0:2"), and stream with
996 index 6 from input @file{b.mov} (specified by the identifier "1:6"),
997 and copy them to the output file @file{out.mov}:
999 ffmpeg -i a.mov -i b.mov -c copy -map 0:2 -map 1:6 out.mov
1002 To select all video and the third audio stream from an input file:
1004 ffmpeg -i INPUT -map 0:v -map 0:a:2 OUTPUT
1007 To map all the streams except the second audio, use negative mappings
1009 ffmpeg -i INPUT -map 0 -map -0:a:1 OUTPUT
1012 To map the video and audio streams from the first input, and using the
1013 trailing @code{?}, ignore the audio mapping if no audio streams exist in
1016 ffmpeg -i INPUT -map 0:v -map 0:a? OUTPUT
1019 To pick the English audio stream:
1021 ffmpeg -i INPUT -map 0:m:language:eng OUTPUT
1024 Note that using this option disables the default mappings for this output file.
1026 @item -ignore_unknown
1027 Ignore input streams with unknown type instead of failing if copying
1028 such streams is attempted.
1031 Allow input streams with unknown type to be copied instead of failing if copying
1032 such streams is attempted.
1034 @item -map_channel [@var{input_file_id}.@var{stream_specifier}.@var{channel_id}|-1][?][:@var{output_file_id}.@var{stream_specifier}]
1035 Map an audio channel from a given input to an output. If
1036 @var{output_file_id}.@var{stream_specifier} is not set, the audio channel will
1037 be mapped on all the audio streams.
1039 Using "-1" instead of
1040 @var{input_file_id}.@var{stream_specifier}.@var{channel_id} will map a muted
1043 A trailing @code{?} will allow the map_channel to be
1044 optional: if the map_channel matches no channel the map_channel will be ignored instead
1047 For example, assuming @var{INPUT} is a stereo audio file, you can switch the
1048 two audio channels with the following command:
1050 ffmpeg -i INPUT -map_channel 0.0.1 -map_channel 0.0.0 OUTPUT
1053 If you want to mute the first channel and keep the second:
1055 ffmpeg -i INPUT -map_channel -1 -map_channel 0.0.1 OUTPUT
1058 The order of the "-map_channel" option specifies the order of the channels in
1059 the output stream. The output channel layout is guessed from the number of
1060 channels mapped (mono if one "-map_channel", stereo if two, etc.). Using "-ac"
1061 in combination of "-map_channel" makes the channel gain levels to be updated if
1062 input and output channel layouts don't match (for instance two "-map_channel"
1063 options and "-ac 6").
1065 You can also extract each channel of an input to specific outputs; the following
1066 command extracts two channels of the @var{INPUT} audio stream (file 0, stream 0)
1067 to the respective @var{OUTPUT_CH0} and @var{OUTPUT_CH1} outputs:
1069 ffmpeg -i INPUT -map_channel 0.0.0 OUTPUT_CH0 -map_channel 0.0.1 OUTPUT_CH1
1072 The following example splits the channels of a stereo input into two separate
1073 streams, which are put into the same output file:
1075 ffmpeg -i stereo.wav -map 0:0 -map 0:0 -map_channel 0.0.0:0.0 -map_channel 0.0.1:0.1 -y out.ogg
1078 Note that currently each output stream can only contain channels from a single
1079 input stream; you can't for example use "-map_channel" to pick multiple input
1080 audio channels contained in different streams (from the same or different files)
1081 and merge them into a single output stream. It is therefore not currently
1082 possible, for example, to turn two separate mono streams into a single stereo
1083 stream. However splitting a stereo stream into two single channel mono streams
1086 If you need this feature, a possible workaround is to use the @emph{amerge}
1087 filter. For example, if you need to merge a media (here @file{input.mkv}) with 2
1088 mono audio streams into one single stereo channel audio stream (and keep the
1089 video stream), you can use the following command:
1091 ffmpeg -i input.mkv -filter_complex "[0:1] [0:2] amerge" -c:a pcm_s16le -c:v copy output.mkv
1094 To map the first two audio channels from the first input, and using the
1095 trailing @code{?}, ignore the audio channel mapping if the first input is
1096 mono instead of stereo:
1098 ffmpeg -i INPUT -map_channel 0.0.0 -map_channel 0.0.1? OUTPUT
1101 @item -map_metadata[:@var{metadata_spec_out}] @var{infile}[:@var{metadata_spec_in}] (@emph{output,per-metadata})
1102 Set metadata information of the next output file from @var{infile}. Note that
1103 those are file indices (zero-based), not filenames.
1104 Optional @var{metadata_spec_in/out} parameters specify, which metadata to copy.
1105 A metadata specifier can have the following forms:
1108 global metadata, i.e. metadata that applies to the whole file
1110 @item @var{s}[:@var{stream_spec}]
1111 per-stream metadata. @var{stream_spec} is a stream specifier as described
1112 in the @ref{Stream specifiers} chapter. In an input metadata specifier, the first
1113 matching stream is copied from. In an output metadata specifier, all matching
1114 streams are copied to.
1116 @item @var{c}:@var{chapter_index}
1117 per-chapter metadata. @var{chapter_index} is the zero-based chapter index.
1119 @item @var{p}:@var{program_index}
1120 per-program metadata. @var{program_index} is the zero-based program index.
1122 If metadata specifier is omitted, it defaults to global.
1124 By default, global metadata is copied from the first input file,
1125 per-stream and per-chapter metadata is copied along with streams/chapters. These
1126 default mappings are disabled by creating any mapping of the relevant type. A negative
1127 file index can be used to create a dummy mapping that just disables automatic copying.
1129 For example to copy metadata from the first stream of the input file to global metadata
1132 ffmpeg -i in.ogg -map_metadata 0:s:0 out.mp3
1135 To do the reverse, i.e. copy global metadata to all audio streams:
1137 ffmpeg -i in.mkv -map_metadata:s:a 0:g out.mkv
1139 Note that simple @code{0} would work as well in this example, since global
1140 metadata is assumed by default.
1142 @item -map_chapters @var{input_file_index} (@emph{output})
1143 Copy chapters from input file with index @var{input_file_index} to the next
1144 output file. If no chapter mapping is specified, then chapters are copied from
1145 the first input file with at least one chapter. Use a negative file index to
1146 disable any chapter copying.
1148 @item -benchmark (@emph{global})
1149 Show benchmarking information at the end of an encode.
1150 Shows CPU time used and maximum memory consumption.
1151 Maximum memory consumption is not supported on all systems,
1152 it will usually display as 0 if not supported.
1153 @item -benchmark_all (@emph{global})
1154 Show benchmarking information during the encode.
1155 Shows CPU time used in various steps (audio/video encode/decode).
1156 @item -timelimit @var{duration} (@emph{global})
1157 Exit after ffmpeg has been running for @var{duration} seconds.
1158 @item -dump (@emph{global})
1159 Dump each input packet to stderr.
1160 @item -hex (@emph{global})
1161 When dumping packets, also dump the payload.
1162 @item -re (@emph{input})
1163 Read input at native frame rate. Mainly used to simulate a grab device,
1164 or live input stream (e.g. when reading from a file). Should not be used
1165 with actual grab devices or live input streams (where it can cause packet
1167 By default @command{ffmpeg} attempts to read the input(s) as fast as possible.
1168 This option will slow down the reading of the input(s) to the native frame rate
1169 of the input(s). It is useful for real-time output (e.g. live streaming).
1170 @item -loop_output @var{number_of_times}
1171 Repeatedly loop output for formats that support looping such as animated GIF
1172 (0 will loop the output infinitely).
1173 This option is deprecated, use -loop.
1174 @item -vsync @var{parameter}
1176 For compatibility reasons old values can be specified as numbers.
1177 Newly added values will have to be specified as strings always.
1180 @item 0, passthrough
1181 Each frame is passed with its timestamp from the demuxer to the muxer.
1183 Frames will be duplicated and dropped to achieve exactly the requested
1184 constant frame rate.
1186 Frames are passed through with their timestamp or dropped so as to
1187 prevent 2 frames from having the same timestamp.
1189 As passthrough but destroys all timestamps, making the muxer generate
1190 fresh timestamps based on frame-rate.
1192 Chooses between 1 and 2 depending on muxer capabilities. This is the
1196 Note that the timestamps may be further modified by the muxer, after this.
1197 For example, in the case that the format option @option{avoid_negative_ts}
1200 With -map you can select from which stream the timestamps should be
1201 taken. You can leave either video or audio unchanged and sync the
1202 remaining stream(s) to the unchanged one.
1204 @item -frame_drop_threshold @var{parameter}
1205 Frame drop threshold, which specifies how much behind video frames can
1206 be before they are dropped. In frame rate units, so 1.0 is one frame.
1207 The default is -1.1. One possible usecase is to avoid framedrops in case
1208 of noisy timestamps or to increase frame drop precision in case of exact
1211 @item -async @var{samples_per_second}
1212 Audio sync method. "Stretches/squeezes" the audio stream to match the timestamps,
1213 the parameter is the maximum samples per second by which the audio is changed.
1214 -async 1 is a special case where only the start of the audio stream is corrected
1215 without any later correction.
1217 Note that the timestamps may be further modified by the muxer, after this.
1218 For example, in the case that the format option @option{avoid_negative_ts}
1221 This option has been deprecated. Use the @code{aresample} audio filter instead.
1224 Do not process input timestamps, but keep their values without trying
1225 to sanitize them. In particular, do not remove the initial start time
1228 Note that, depending on the @option{vsync} option or on specific muxer
1229 processing (e.g. in case the format option @option{avoid_negative_ts}
1230 is enabled) the output timestamps may mismatch with the input
1231 timestamps even when this option is selected.
1233 @item -start_at_zero
1234 When used with @option{copyts}, shift input timestamps so they start at zero.
1236 This means that using e.g. @code{-ss 50} will make output timestamps start at
1237 50 seconds, regardless of what timestamp the input file started at.
1239 @item -copytb @var{mode}
1240 Specify how to set the encoder timebase when stream copying. @var{mode} is an
1241 integer numeric value, and can assume one of the following values:
1245 Use the demuxer timebase.
1247 The time base is copied to the output encoder from the corresponding input
1248 demuxer. This is sometimes required to avoid non monotonically increasing
1249 timestamps when copying video streams with variable frame rate.
1252 Use the decoder timebase.
1254 The time base is copied to the output encoder from the corresponding input
1258 Try to make the choice automatically, in order to generate a sane output.
1261 Default value is -1.
1263 @item -enc_time_base[:@var{stream_specifier}] @var{timebase} (@emph{output,per-stream})
1264 Set the encoder timebase. @var{timebase} is a floating point number,
1265 and can assume one of the following values:
1269 Assign a default value according to the media type.
1271 For video - use 1/framerate, for audio - use 1/samplerate.
1274 Use the input stream timebase when possible.
1276 If an input stream is not available, the default timebase will be used.
1279 Use the provided number as the timebase.
1281 This field can be provided as a ratio of two integers (e.g. 1:24, 1:48000)
1282 or as a floating point number (e.g. 0.04166, 2.0833e-5)
1287 @item -bitexact (@emph{input/output})
1288 Enable bitexact mode for (de)muxer and (de/en)coder
1289 @item -shortest (@emph{output})
1290 Finish encoding when the shortest input stream ends.
1291 @item -dts_delta_threshold
1292 Timestamp discontinuity delta threshold.
1293 @item -muxdelay @var{seconds} (@emph{input})
1294 Set the maximum demux-decode delay.
1295 @item -muxpreload @var{seconds} (@emph{input})
1296 Set the initial demux-decode delay.
1297 @item -streamid @var{output-stream-index}:@var{new-value} (@emph{output})
1298 Assign a new stream-id value to an output stream. This option should be
1299 specified prior to the output filename to which it applies.
1300 For the situation where multiple output files exist, a streamid
1301 may be reassigned to a different value.
1303 For example, to set the stream 0 PID to 33 and the stream 1 PID to 36 for
1304 an output mpegts file:
1306 ffmpeg -i inurl -streamid 0:33 -streamid 1:36 out.ts
1309 @item -bsf[:@var{stream_specifier}] @var{bitstream_filters} (@emph{output,per-stream})
1310 Set bitstream filters for matching streams. @var{bitstream_filters} is
1311 a comma-separated list of bitstream filters. Use the @code{-bsfs} option
1312 to get the list of bitstream filters.
1314 ffmpeg -i h264.mp4 -c:v copy -bsf:v h264_mp4toannexb -an out.h264
1317 ffmpeg -i file.mov -an -vn -bsf:s mov2textsub -c:s copy -f rawvideo sub.txt
1320 @item -tag[:@var{stream_specifier}] @var{codec_tag} (@emph{input/output,per-stream})
1321 Force a tag/fourcc for matching streams.
1323 @item -timecode @var{hh}:@var{mm}:@var{ss}SEP@var{ff}
1324 Specify Timecode for writing. @var{SEP} is ':' for non drop timecode and ';'
1327 ffmpeg -i input.mpg -timecode 01:02:03.04 -r 30000/1001 -s ntsc output.mpg
1330 @anchor{filter_complex_option}
1331 @item -filter_complex @var{filtergraph} (@emph{global})
1332 Define a complex filtergraph, i.e. one with arbitrary number of inputs and/or
1333 outputs. For simple graphs -- those with one input and one output of the same
1334 type -- see the @option{-filter} options. @var{filtergraph} is a description of
1335 the filtergraph, as described in the ``Filtergraph syntax'' section of the
1336 ffmpeg-filters manual.
1338 Input link labels must refer to input streams using the
1339 @code{[file_index:stream_specifier]} syntax (i.e. the same as @option{-map}
1340 uses). If @var{stream_specifier} matches multiple streams, the first one will be
1341 used. An unlabeled input will be connected to the first unused input stream of
1344 Output link labels are referred to with @option{-map}. Unlabeled outputs are
1345 added to the first output file.
1347 Note that with this option it is possible to use only lavfi sources without
1350 For example, to overlay an image over video
1352 ffmpeg -i video.mkv -i image.png -filter_complex '[0:v][1:v]overlay[out]' -map
1355 Here @code{[0:v]} refers to the first video stream in the first input file,
1356 which is linked to the first (main) input of the overlay filter. Similarly the
1357 first video stream in the second input is linked to the second (overlay) input
1360 Assuming there is only one video stream in each input file, we can omit input
1361 labels, so the above is equivalent to
1363 ffmpeg -i video.mkv -i image.png -filter_complex 'overlay[out]' -map
1367 Furthermore we can omit the output label and the single output from the filter
1368 graph will be added to the output file automatically, so we can simply write
1370 ffmpeg -i video.mkv -i image.png -filter_complex 'overlay' out.mkv
1373 To generate 5 seconds of pure red video using lavfi @code{color} source:
1375 ffmpeg -filter_complex 'color=c=red' -t 5 out.mkv
1378 @item -filter_complex_threads @var{nb_threads} (@emph{global})
1379 Defines how many threads are used to process a filter_complex graph.
1380 Similar to filter_threads but used for @code{-filter_complex} graphs only.
1381 The default is the number of available CPUs.
1383 @item -lavfi @var{filtergraph} (@emph{global})
1384 Define a complex filtergraph, i.e. one with arbitrary number of inputs and/or
1385 outputs. Equivalent to @option{-filter_complex}.
1387 @item -filter_complex_script @var{filename} (@emph{global})
1388 This option is similar to @option{-filter_complex}, the only difference is that
1389 its argument is the name of the file from which a complex filtergraph
1390 description is to be read.
1392 @item -accurate_seek (@emph{input})
1393 This option enables or disables accurate seeking in input files with the
1394 @option{-ss} option. It is enabled by default, so seeking is accurate when
1395 transcoding. Use @option{-noaccurate_seek} to disable it, which may be useful
1396 e.g. when copying some streams and transcoding the others.
1398 @item -seek_timestamp (@emph{input})
1399 This option enables or disables seeking by timestamp in input files with the
1400 @option{-ss} option. It is disabled by default. If enabled, the argument
1401 to the @option{-ss} option is considered an actual timestamp, and is not
1402 offset by the start time of the file. This matters only for files which do
1403 not start from timestamp 0, such as transport streams.
1405 @item -thread_queue_size @var{size} (@emph{input})
1406 This option sets the maximum number of queued packets when reading from the
1407 file or device. With low latency / high rate live streams, packets may be
1408 discarded if they are not read in a timely manner; raising this value can
1411 @item -sdp_file @var{file} (@emph{global})
1412 Print sdp information for an output stream to @var{file}.
1413 This allows dumping sdp information when at least one output isn't an
1414 rtp stream. (Requires at least one of the output formats to be rtp).
1416 @item -discard (@emph{input})
1417 Allows discarding specific streams or frames of streams at the demuxer.
1418 Not all demuxers support this.
1425 Default, which discards no frames.
1428 Discard all non-reference frames.
1431 Discard all bidirectional frames.
1434 Discard all frames excepts keyframes.
1440 @item -abort_on @var{flags} (@emph{global})
1441 Stop and abort on various conditions. The following flags are available:
1445 No packets were passed to the muxer, the output is empty.
1448 @item -xerror (@emph{global})
1449 Stop and exit on error
1451 @item -max_muxing_queue_size @var{packets} (@emph{output,per-stream})
1452 When transcoding audio and/or video streams, ffmpeg will not begin writing into
1453 the output until it has one packet for each such stream. While waiting for that
1454 to happen, packets for other streams are buffered. This option sets the size of
1455 this buffer, in packets, for the matching output stream.
1457 The default value of this option should be high enough for most uses, so only
1458 touch this option if you are sure that you need it.
1462 As a special exception, you can use a bitmap subtitle stream as input: it
1463 will be converted into a video with the same size as the largest video in
1464 the file, or 720x576 if no video is present. Note that this is an
1465 experimental and temporary solution. It will be removed once libavfilter has
1466 proper support for subtitles.
1468 For example, to hardcode subtitles on top of a DVB-T recording stored in
1469 MPEG-TS format, delaying the subtitles by 1 second:
1471 ffmpeg -i input.ts -filter_complex \
1472 '[#0x2ef] setpts=PTS+1/TB [sub] ; [#0x2d0] [sub] overlay' \
1473 -sn -map '#0x2dc' output.mkv
1475 (0x2d0, 0x2dc and 0x2ef are the MPEG-TS PIDs of respectively the video,
1476 audio and subtitles streams; 0:0, 0:3 and 0:7 would have worked too)
1478 @section Preset files
1479 A preset file contains a sequence of @var{option}=@var{value} pairs,
1480 one for each line, specifying a sequence of options which would be
1481 awkward to specify on the command line. Lines starting with the hash
1482 ('#') character are ignored and are used to provide comments. Check
1483 the @file{presets} directory in the FFmpeg source tree for examples.
1485 There are two types of preset files: ffpreset and avpreset files.
1487 @subsection ffpreset files
1488 ffpreset files are specified with the @code{vpre}, @code{apre},
1489 @code{spre}, and @code{fpre} options. The @code{fpre} option takes the
1490 filename of the preset instead of a preset name as input and can be
1491 used for any kind of codec. For the @code{vpre}, @code{apre}, and
1492 @code{spre} options, the options specified in a preset file are
1493 applied to the currently selected codec of the same type as the preset
1496 The argument passed to the @code{vpre}, @code{apre}, and @code{spre}
1497 preset options identifies the preset file to use according to the
1500 First ffmpeg searches for a file named @var{arg}.ffpreset in the
1501 directories @file{$FFMPEG_DATADIR} (if set), and @file{$HOME/.ffmpeg}, and in
1502 the datadir defined at configuration time (usually @file{PREFIX/share/ffmpeg})
1503 or in a @file{ffpresets} folder along the executable on win32,
1504 in that order. For example, if the argument is @code{libvpx-1080p}, it will
1505 search for the file @file{libvpx-1080p.ffpreset}.
1507 If no such file is found, then ffmpeg will search for a file named
1508 @var{codec_name}-@var{arg}.ffpreset in the above-mentioned
1509 directories, where @var{codec_name} is the name of the codec to which
1510 the preset file options will be applied. For example, if you select
1511 the video codec with @code{-vcodec libvpx} and use @code{-vpre 1080p},
1512 then it will search for the file @file{libvpx-1080p.ffpreset}.
1514 @subsection avpreset files
1515 avpreset files are specified with the @code{pre} option. They work similar to
1516 ffpreset files, but they only allow encoder- specific options. Therefore, an
1517 @var{option}=@var{value} pair specifying an encoder cannot be used.
1519 When the @code{pre} option is specified, ffmpeg will look for files with the
1520 suffix .avpreset in the directories @file{$AVCONV_DATADIR} (if set), and
1521 @file{$HOME/.avconv}, and in the datadir defined at configuration time (usually
1522 @file{PREFIX/share/ffmpeg}), in that order.
1524 First ffmpeg searches for a file named @var{codec_name}-@var{arg}.avpreset in
1525 the above-mentioned directories, where @var{codec_name} is the name of the codec
1526 to which the preset file options will be applied. For example, if you select the
1527 video codec with @code{-vcodec libvpx} and use @code{-pre 1080p}, then it will
1528 search for the file @file{libvpx-1080p.avpreset}.
1530 If no such file is found, then ffmpeg will search for a file named
1531 @var{arg}.avpreset in the same directories.
1536 @c man begin EXAMPLES
1538 @section Video and Audio grabbing
1540 If you specify the input format and device then ffmpeg can grab video
1544 ffmpeg -f oss -i /dev/dsp -f video4linux2 -i /dev/video0 /tmp/out.mpg
1547 Or with an ALSA audio source (mono input, card id 1) instead of OSS:
1549 ffmpeg -f alsa -ac 1 -i hw:1 -f video4linux2 -i /dev/video0 /tmp/out.mpg
1552 Note that you must activate the right video source and channel before
1553 launching ffmpeg with any TV viewer such as
1554 @uref{http://linux.bytesex.org/xawtv/, xawtv} by Gerd Knorr. You also
1555 have to set the audio recording levels correctly with a
1558 @section X11 grabbing
1560 Grab the X11 display with ffmpeg via
1563 ffmpeg -f x11grab -video_size cif -framerate 25 -i :0.0 /tmp/out.mpg
1566 0.0 is display.screen number of your X11 server, same as
1567 the DISPLAY environment variable.
1570 ffmpeg -f x11grab -video_size cif -framerate 25 -i :0.0+10,20 /tmp/out.mpg
1573 0.0 is display.screen number of your X11 server, same as the DISPLAY environment
1574 variable. 10 is the x-offset and 20 the y-offset for the grabbing.
1576 @section Video and Audio file format conversion
1578 Any supported file format and protocol can serve as input to ffmpeg:
1583 You can use YUV files as input:
1586 ffmpeg -i /tmp/test%d.Y /tmp/out.mpg
1589 It will use the files:
1591 /tmp/test0.Y, /tmp/test0.U, /tmp/test0.V,
1592 /tmp/test1.Y, /tmp/test1.U, /tmp/test1.V, etc...
1595 The Y files use twice the resolution of the U and V files. They are
1596 raw files, without header. They can be generated by all decent video
1597 decoders. You must specify the size of the image with the @option{-s} option
1598 if ffmpeg cannot guess it.
1601 You can input from a raw YUV420P file:
1604 ffmpeg -i /tmp/test.yuv /tmp/out.avi
1607 test.yuv is a file containing raw YUV planar data. Each frame is composed
1608 of the Y plane followed by the U and V planes at half vertical and
1609 horizontal resolution.
1612 You can output to a raw YUV420P file:
1615 ffmpeg -i mydivx.avi hugefile.yuv
1619 You can set several input files and output files:
1622 ffmpeg -i /tmp/a.wav -s 640x480 -i /tmp/a.yuv /tmp/a.mpg
1625 Converts the audio file a.wav and the raw YUV video file a.yuv
1629 You can also do audio and video conversions at the same time:
1632 ffmpeg -i /tmp/a.wav -ar 22050 /tmp/a.mp2
1635 Converts a.wav to MPEG audio at 22050 Hz sample rate.
1638 You can encode to several formats at the same time and define a
1639 mapping from input stream to output streams:
1642 ffmpeg -i /tmp/a.wav -map 0:a -b:a 64k /tmp/a.mp2 -map 0:a -b:a 128k /tmp/b.mp2
1645 Converts a.wav to a.mp2 at 64 kbits and to b.mp2 at 128 kbits. '-map
1646 file:index' specifies which input stream is used for each output
1647 stream, in the order of the definition of output streams.
1650 You can transcode decrypted VOBs:
1653 ffmpeg -i snatch_1.vob -f avi -c:v mpeg4 -b:v 800k -g 300 -bf 2 -c:a libmp3lame -b:a 128k snatch.avi
1656 This is a typical DVD ripping example; the input is a VOB file, the
1657 output an AVI file with MPEG-4 video and MP3 audio. Note that in this
1658 command we use B-frames so the MPEG-4 stream is DivX5 compatible, and
1659 GOP size is 300 which means one intra frame every 10 seconds for 29.97fps
1660 input video. Furthermore, the audio stream is MP3-encoded so you need
1661 to enable LAME support by passing @code{--enable-libmp3lame} to configure.
1662 The mapping is particularly useful for DVD transcoding
1663 to get the desired audio language.
1665 NOTE: To see the supported input formats, use @code{ffmpeg -demuxers}.
1668 You can extract images from a video, or create a video from many images:
1670 For extracting images from a video:
1672 ffmpeg -i foo.avi -r 1 -s WxH -f image2 foo-%03d.jpeg
1675 This will extract one video frame per second from the video and will
1676 output them in files named @file{foo-001.jpeg}, @file{foo-002.jpeg},
1677 etc. Images will be rescaled to fit the new WxH values.
1679 If you want to extract just a limited number of frames, you can use the
1680 above command in combination with the @code{-frames:v} or @code{-t} option,
1681 or in combination with -ss to start extracting from a certain point in time.
1683 For creating a video from many images:
1685 ffmpeg -f image2 -framerate 12 -i foo-%03d.jpeg -s WxH foo.avi
1688 The syntax @code{foo-%03d.jpeg} specifies to use a decimal number
1689 composed of three digits padded with zeroes to express the sequence
1690 number. It is the same syntax supported by the C printf function, but
1691 only formats accepting a normal integer are suitable.
1693 When importing an image sequence, -i also supports expanding
1694 shell-like wildcard patterns (globbing) internally, by selecting the
1695 image2-specific @code{-pattern_type glob} option.
1697 For example, for creating a video from filenames matching the glob pattern
1700 ffmpeg -f image2 -pattern_type glob -framerate 12 -i 'foo-*.jpeg' -s WxH foo.avi
1704 You can put many streams of the same type in the output:
1707 ffmpeg -i test1.avi -i test2.avi -map 1:1 -map 1:0 -map 0:1 -map 0:0 -c copy -y test12.nut
1710 The resulting output file @file{test12.nut} will contain the first four streams
1711 from the input files in reverse order.
1714 To force CBR video output:
1716 ffmpeg -i myfile.avi -b 4000k -minrate 4000k -maxrate 4000k -bufsize 1835k out.m2v
1720 The four options lmin, lmax, mblmin and mblmax use 'lambda' units,
1721 but you may use the QP2LAMBDA constant to easily convert from 'q' units:
1723 ffmpeg -i src.ext -lmax 21*QP2LAMBDA dst.ext
1729 @include config.texi
1731 @ifset config-avutil
1734 @ifset config-avcodec
1735 @include codecs.texi
1736 @include bitstream_filters.texi
1738 @ifset config-avformat
1739 @include formats.texi
1740 @include protocols.texi
1742 @ifset config-avdevice
1743 @include devices.texi
1745 @ifset config-swresample
1746 @include resampler.texi
1748 @ifset config-swscale
1749 @include scaler.texi
1751 @ifset config-avfilter
1752 @include filters.texi
1760 @url{ffmpeg.html,ffmpeg}
1762 @ifset config-not-all
1763 @url{ffmpeg-all.html,ffmpeg-all},
1765 @url{ffplay.html,ffplay}, @url{ffprobe.html,ffprobe},
1766 @url{ffmpeg-utils.html,ffmpeg-utils},
1767 @url{ffmpeg-scaler.html,ffmpeg-scaler},
1768 @url{ffmpeg-resampler.html,ffmpeg-resampler},
1769 @url{ffmpeg-codecs.html,ffmpeg-codecs},
1770 @url{ffmpeg-bitstream-filters.html,ffmpeg-bitstream-filters},
1771 @url{ffmpeg-formats.html,ffmpeg-formats},
1772 @url{ffmpeg-devices.html,ffmpeg-devices},
1773 @url{ffmpeg-protocols.html,ffmpeg-protocols},
1774 @url{ffmpeg-filters.html,ffmpeg-filters}
1781 @ifset config-not-all
1784 ffplay(1), ffprobe(1),
1785 ffmpeg-utils(1), ffmpeg-scaler(1), ffmpeg-resampler(1),
1786 ffmpeg-codecs(1), ffmpeg-bitstream-filters(1), ffmpeg-formats(1),
1787 ffmpeg-devices(1), ffmpeg-protocols(1), ffmpeg-filters(1)
1790 @include authors.texi
1795 @settitle ffmpeg video converter