FFmpeg FAQ

Table of Contents

1. General Questions

1.1 Why doesn’t FFmpeg support feature [xyz]?

Because no one has taken on that task yet. FFmpeg development is driven by the tasks that are important to the individual developers. If there is a feature that is important to you, the best way to get it implemented is to undertake the task yourself or sponsor a developer.

1.2 FFmpeg does not support codec XXX. Can you include a Windows DLL loader to support it?

No. Windows DLLs are not portable, bloated and often slow. Moreover FFmpeg strives to support all codecs natively. A DLL loader is not conducive to that goal.

1.3 I cannot read this file although this format seems to be supported by ffmpeg.

Even if ffmpeg can read the container format, it may not support all its codecs. Please consult the supported codec list in the ffmpeg documentation.

1.4 Which codecs are supported by Windows?

Windows does not support standard formats like MPEG very well, unless you install some additional codecs.

The following list of video codecs should work on most Windows systems:

msmpeg4v2

.avi/.asf

msmpeg4

.asf only

wmv1

.asf only

wmv2

.asf only

mpeg4

Only if you have some MPEG-4 codec like ffdshow or Xvid installed.

mpeg1video

.mpg only

Note, ASF files often have .wmv or .wma extensions in Windows. It should also be mentioned that Microsoft claims a patent on the ASF format, and may sue or threaten users who create ASF files with non-Microsoft software. It is strongly advised to avoid ASF where possible.

The following list of audio codecs should work on most Windows systems:

adpcm_ima_wav
adpcm_ms
pcm_s16le

always

libmp3lame

If some MP3 codec like LAME is installed.

2. Compilation

2.1 error: can't find a register in class 'GENERAL_REGS' while reloading 'asm'

This is a bug in gcc. Do not report it to us. Instead, please report it to the gcc developers. Note that we will not add workarounds for gcc bugs.

Also note that (some of) the gcc developers believe this is not a bug or not a bug they should fix: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11203. Then again, some of them do not know the difference between an undecidable problem and an NP-hard problem...

2.2 I have installed this library with my distro’s package manager. Why does configure not see it?

Distributions usually split libraries in several packages. The main package contains the files necessary to run programs using the library. The development package contains the files necessary to build programs using the library. Sometimes, docs and/or data are in a separate package too.

To build FFmpeg, you need to install the development package. It is usually called ‘libfoo-dev’ or ‘libfoo-devel’. You can remove it after the build is finished, but be sure to keep the main package.

3. Usage

3.1 ffmpeg does not work; what is wrong?

Try a make distclean in the ffmpeg source directory before the build. If this does not help see (http://ffmpeg.org/bugreports.html).

3.2 How do I encode single pictures into movies?

First, rename your pictures to follow a numerical sequence. For example, img1.jpg, img2.jpg, img3.jpg,... Then you may run:

 
ffmpeg -f image2 -i img%d.jpg /tmp/a.mpg

Notice that ‘%d’ is replaced by the image number.

img%03d.jpg’ means the sequence ‘img001.jpg’, ‘img002.jpg’, etc.

Use the ‘-start_number’ option to declare a starting number for the sequence. This is useful if your sequence does not start with ‘img001.jpg’ but is still in a numerical order. The following example will start with ‘img100.jpg’:

 
ffmpeg -f image2 -start_number 100 -i img%d.jpg /tmp/a.mpg

If you have large number of pictures to rename, you can use the following command to ease the burden. The command, using the bourne shell syntax, symbolically links all files in the current directory that match *jpg to the ‘/tmp’ directory in the sequence of ‘img001.jpg’, ‘img002.jpg’ and so on.

 
x=1; for i in *jpg; do counter=$(printf %03d $x); ln -s "$i" /tmp/img"$counter".jpg; x=$(($x+1)); done

If you want to sequence them by oldest modified first, substitute $(ls -r -t *jpg) in place of *jpg.

Then run:

 
ffmpeg -f image2 -i /tmp/img%03d.jpg /tmp/a.mpg

The same logic is used for any image format that ffmpeg reads.

You can also use cat to pipe images to ffmpeg:

 
cat *.jpg | ffmpeg -f image2pipe -c:v mjpeg -i - output.mpg

3.3 How do I encode movie to single pictures?

Use:

 
ffmpeg -i movie.mpg movie%d.jpg

The ‘movie.mpg’ used as input will be converted to ‘movie1.jpg’, ‘movie2.jpg’, etc...

Instead of relying on file format self-recognition, you may also use

-c:v ppm
-c:v png
-c:v mjpeg

to force the encoding.

Applying that to the previous example:

 
ffmpeg -i movie.mpg -f image2 -c:v mjpeg menu%d.jpg

Beware that there is no "jpeg" codec. Use "mjpeg" instead.

3.4 Why do I see a slight quality degradation with multithreaded MPEG* encoding?

For multithreaded MPEG* encoding, the encoded slices must be independent, otherwise thread n would practically have to wait for n-1 to finish, so it’s quite logical that there is a small reduction of quality. This is not a bug.

3.5 How can I read from the standard input or write to the standard output?

Use ‘-’ as file name.

3.6 -f jpeg doesn’t work.

Try ’-f image2 test%d.jpg’.

3.7 Why can I not change the frame rate?

Some codecs, like MPEG-1/2, only allow a small number of fixed frame rates. Choose a different codec with the -c:v command line option.

3.8 How do I encode Xvid or DivX video with ffmpeg?

Both Xvid and DivX (version 4+) are implementations of the ISO MPEG-4 standard (note that there are many other coding formats that use this same standard). Thus, use ’-c:v mpeg4’ to encode in these formats. The default fourcc stored in an MPEG-4-coded file will be ’FMP4’. If you want a different fourcc, use the ’-vtag’ option. E.g., ’-vtag xvid’ will force the fourcc ’xvid’ to be stored as the video fourcc rather than the default.

3.9 Which are good parameters for encoding high quality MPEG-4?

’-mbd rd -flags +mv4+aic -trellis 2 -cmp 2 -subcmp 2 -g 300 -pass 1/2’, things to try: ’-bf 2’, ’-flags qprd’, ’-flags mv0’, ’-flags skiprd’.

3.10 Which are good parameters for encoding high quality MPEG-1/MPEG-2?

’-mbd rd -trellis 2 -cmp 2 -subcmp 2 -g 100 -pass 1/2’ but beware the ’-g 100’ might cause problems with some decoders. Things to try: ’-bf 2’, ’-flags qprd’, ’-flags mv0’, ’-flags skiprd.

3.11 Interlaced video looks very bad when encoded with ffmpeg, what is wrong?

You should use ’-flags +ilme+ildct’ and maybe ’-flags +alt’ for interlaced material, and try ’-top 0/1’ if the result looks really messed-up.

3.12 How can I read DirectShow files?

If you have built FFmpeg with ./configure --enable-avisynth (only possible on MinGW/Cygwin platforms), then you may use any file that DirectShow can read as input.

Just create an "input.avs" text file with this single line ...

 
DirectShowSource("C:\path to your file\yourfile.asf")

... and then feed that text file to ffmpeg:

 
ffmpeg -i input.avs

For ANY other help on AviSynth, please visit the AviSynth homepage.

3.13 How can I join video files?

To "join" video files is quite ambiguous. The following list explains the different kinds of "joining" and points out how those are addressed in FFmpeg. To join video files may mean:

3.14 How can I concatenate video files?

There are several solutions, depending on the exact circumstances.

3.14.1 Concatenating using the concat filter

FFmpeg has a concat filter designed specifically for that, with examples in the documentation. This operation is recommended if you need to re-encode.

3.14.2 Concatenating using the concat demuxer

FFmpeg has a concat demuxer which you can use when you want to avoid a re-encode and your format doesn’t support file level concatenation.

3.14.3 Concatenating using the concat protocol (file level)

FFmpeg has a concat protocol designed specifically for that, with examples in the documentation.

A few multimedia containers (MPEG-1, MPEG-2 PS, DV) allow to concatenate video by merely concatenating the files containing them.

Hence you may concatenate your multimedia files by first transcoding them to these privileged formats, then using the humble cat command (or the equally humble copy under Windows), and finally transcoding back to your format of choice.

 
ffmpeg -i input1.avi -qscale:v 1 intermediate1.mpg
ffmpeg -i input2.avi -qscale:v 1 intermediate2.mpg
cat intermediate1.mpg intermediate2.mpg > intermediate_all.mpg
ffmpeg -i intermediate_all.mpg -qscale:v 2 output.avi

Additionally, you can use the concat protocol instead of cat or copy which will avoid creation of a potentially huge intermediate file.

 
ffmpeg -i input1.avi -qscale:v 1 intermediate1.mpg
ffmpeg -i input2.avi -qscale:v 1 intermediate2.mpg
ffmpeg -i concat:"intermediate1.mpg|intermediate2.mpg" -c copy intermediate_all.mpg
ffmpeg -i intermediate_all.mpg -qscale:v 2 output.avi

Note that you may need to escape the character "|" which is special for many shells.

Another option is usage of named pipes, should your platform support it:

 
mkfifo intermediate1.mpg
mkfifo intermediate2.mpg
ffmpeg -i input1.avi -qscale:v 1 -y intermediate1.mpg < /dev/null &
ffmpeg -i input2.avi -qscale:v 1 -y intermediate2.mpg < /dev/null &
cat intermediate1.mpg intermediate2.mpg |\
ffmpeg -f mpeg -i - -c:v mpeg4 -acodec libmp3lame output.avi

3.14.4 Concatenating using raw audio and video

Similarly, the yuv4mpegpipe format, and the raw video, raw audio codecs also allow concatenation, and the transcoding step is almost lossless. When using multiple yuv4mpegpipe(s), the first line needs to be discarded from all but the first stream. This can be accomplished by piping through tail as seen below. Note that when piping through tail you must use command grouping, { ;}, to background properly.

For example, let’s say we want to concatenate two FLV files into an output.flv file:

 
mkfifo temp1.a
mkfifo temp1.v
mkfifo temp2.a
mkfifo temp2.v
mkfifo all.a
mkfifo all.v
ffmpeg -i input1.flv -vn -f u16le -acodec pcm_s16le -ac 2 -ar 44100 - > temp1.a < /dev/null &
ffmpeg -i input2.flv -vn -f u16le -acodec pcm_s16le -ac 2 -ar 44100 - > temp2.a < /dev/null &
ffmpeg -i input1.flv -an -f yuv4mpegpipe - > temp1.v < /dev/null &
{ ffmpeg -i input2.flv -an -f yuv4mpegpipe - < /dev/null | tail -n +2 > temp2.v ; } &
cat temp1.a temp2.a > all.a &
cat temp1.v temp2.v > all.v &
ffmpeg -f u16le -acodec pcm_s16le -ac 2 -ar 44100 -i all.a \
       -f yuv4mpegpipe -i all.v \
       -y output.flv
rm temp[12].[av] all.[av]

3.15 Using ‘-f lavfi’, audio becomes mono for no apparent reason.

Use ‘-dumpgraph -’ to find out exactly where the channel layout is lost.

Most likely, it is through auto-inserted aresample. Try to understand why the converting filter was needed at that place.

Just before the output is a likely place, as ‘-f lavfi’ currently only support packed S16.

Then insert the correct aformat explicitly in the filtergraph, specifying the exact format.

 
aformat=sample_fmts=s16:channel_layouts=stereo

3.16 Why does FFmpeg not see the subtitles in my VOB file?

VOB and a few other formats do not have a global header that describes everything present in the file. Instead, applications are supposed to scan the file to see what it contains. Since VOB files are frequently large, only the beginning is scanned. If the subtitles happen only later in the file, they will not be initally detected.

Some applications, including the ffmpeg command-line tool, can only work with streams that were detected during the initial scan; streams that are detected later are ignored.

The size of the initial scan is controlled by two options: probesize (default ~5 Mo) and analyzeduration (default 5,000,000 µs = 5 s). For the subtitle stream to be detected, both values must be large enough.

3.17 Why was the ffmpeg-sameq’ option removed? What to use instead?

The ‘-sameq’ option meant "same quantizer", and made sense only in a very limited set of cases. Unfortunately, a lot of people mistook it for "same quality" and used it in places where it did not make sense: it had roughly the expected visible effect, but achieved it in a very inefficient way.

Each encoder has its own set of options to set the quality-vs-size balance, use the options for the encoder you are using to set the quality level to a point acceptable for your tastes. The most common options to do that are ‘-qscale’ and ‘-qmax’, but you should peruse the documentation of the encoder you chose.

4. Development

4.1 Are there examples illustrating how to use the FFmpeg libraries, particularly libavcodec and libavformat?

Yes. Check the ‘doc/examples’ directory in the source repository, also available online at: https://github.com/FFmpeg/FFmpeg/tree/master/doc/examples.

Examples are also installed by default, usually in $PREFIX/share/ffmpeg/examples.

Also you may read the Developers Guide of the FFmpeg documentation. Alternatively, examine the source code for one of the many open source projects that already incorporate FFmpeg at (projects.html).

4.2 Can you support my C compiler XXX?

It depends. If your compiler is C99-compliant, then patches to support it are likely to be welcome if they do not pollute the source code with #ifdefs related to the compiler.

4.3 Is Microsoft Visual C++ supported?

Yes. Please see the Microsoft Visual C++ section in the FFmpeg documentation.

4.4 Can you add automake, libtool or autoconf support?

No. These tools are too bloated and they complicate the build.

4.5 Why not rewrite FFmpeg in object-oriented C++?

FFmpeg is already organized in a highly modular manner and does not need to be rewritten in a formal object language. Further, many of the developers favor straight C; it works for them. For more arguments on this matter, read "Programming Religion".

4.6 Why are the ffmpeg programs devoid of debugging symbols?

The build process creates ffmpeg_g, ffplay_g, etc. which contain full debug information. Those binaries are stripped to create ffmpeg, ffplay, etc. If you need the debug information, use the *_g versions.

4.7 I do not like the LGPL, can I contribute code under the GPL instead?

Yes, as long as the code is optional and can easily and cleanly be placed under #if CONFIG_GPL without breaking anything. So, for example, a new codec or filter would be OK under GPL while a bug fix to LGPL code would not.

4.8 I’m using FFmpeg from within my C application but the linker complains about missing symbols from the libraries themselves.

FFmpeg builds static libraries by default. In static libraries, dependencies are not handled. That has two consequences. First, you must specify the libraries in dependency order: -lavdevice must come before -lavformat, -lavutil must come after everything else, etc. Second, external libraries that are used in FFmpeg have to be specified too.

An easy way to get the full list of required libraries in dependency order is to use pkg-config.

 
c99 -o program program.c $(pkg-config --cflags --libs libavformat libavcodec)

See ‘doc/example/Makefile’ and ‘doc/example/pc-uninstalled’ for more details.

4.9 I’m using FFmpeg from within my C++ application but the linker complains about missing symbols which seem to be available.

FFmpeg is a pure C project, so to use the libraries within your C++ application you need to explicitly state that you are using a C library. You can do this by encompassing your FFmpeg includes using extern "C".

See http://www.parashift.com/c++-faq-lite/mixing-c-and-cpp.html#faq-32.3

4.10 I’m using libavutil from within my C++ application but the compiler complains about ’UINT64_C’ was not declared in this scope

FFmpeg is a pure C project using C99 math features, in order to enable C++ to use them you have to append -D__STDC_CONSTANT_MACROS to your CXXFLAGS

4.11 I have a file in memory / a API different from *open/*read/ libc how do I use it with libavformat?

You have to create a custom AVIOContext using avio_alloc_context, see ‘libavformat/aviobuf.c’ in FFmpeg and ‘libmpdemux/demux_lavf.c’ in MPlayer or MPlayer2 sources.

4.12 Where is the documentation about ffv1, msmpeg4, asv1, 4xm?

see http://www.ffmpeg.org/~michael/

4.13 How do I feed H.263-RTP (and other codecs in RTP) to libavcodec?

Even if peculiar since it is network oriented, RTP is a container like any other. You have to demux RTP before feeding the payload to libavcodec. In this specific case please look at RFC 4629 to see how it should be done.

4.14 AVStream.r_frame_rate is wrong, it is much larger than the frame rate.

r_frame_rate is NOT the average frame rate, it is the smallest frame rate that can accurately represent all timestamps. So no, it is not wrong if it is larger than the average! For example, if you have mixed 25 and 30 fps content, then r_frame_rate will be 150 (it is the least common multiple). If you are looking for the average frame rate, see AVStream.avg_frame_rate.

4.15 Why is make fate not running all tests?

Make sure you have the fate-suite samples and the SAMPLES Make variable or FATE_SAMPLES environment variable or the --samples configure option is set to the right path.

4.16 Why is make fate not finding the samples?

Do you happen to have a ~ character in the samples path to indicate a home directory? The value is used in ways where the shell cannot expand it, causing FATE to not find files. Just replace ~ by the full path.