編譯好的ffmpeg資料夾下有ffplay和ffplay_g兩個可執行檔案,發現這兩個功能是一樣的,不知道到底什麼區別.

其他的可執行檔案,也是這個樣子的.

=======

直接輸入ffplay後

[email protected]:/home/zhangbin/Documents/ffmpeg# ffplay

FFplay version SVN-r26268, Copyright (c) 2003-2011 the FFmpeg developers

  built on Jan  8 2011 09:20:02 with gcc 4.4.3

編譯的配置資訊

  configuration: --enable-shared --enable-gpl --enable-nonfree --enable-pthreads --enable-libfaac --enable-libmp3lame --enable-libtheora --enable-libxvid --enable-x11grab --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libopencore-amrnb --enable-version3

用到的庫

  libavutil     50.36. 0 / 50.36. 0

  libavcore      0.16. 0 /  0.16. 0

  libavcodec    52.107. 0 / 52.107. 0

  libavformat   52.92. 0 / 52.92. 0

  libavdevice   52. 2. 3 / 52. 2. 3

  libavfilter    1.72. 0 /  1.72. 0

  libswscale     0.12. 0 /  0.12. 0

Simple media player

用法

usage: ffplay [options] input_file

An input file must be specified

Use -h to get full help or, even better, run 'man ffplay'

[email protected]:/home/zhangbin/Documents/ffmpeg# 

==========

ffplay的幫助文件

==========

FFPLAY(1)                                                            FFPLAY(1)

NAME

       ffplay - FFplay media player

SYNOPSIS

       ffplay [options] input_file

DESCRIPTION

       FFplay is a very simple and portable media player using the FFmpeg

       libraries and the SDL library. It is mostly used as a testbed for the

       various FFmpeg APIs.

OPTIONS

       All the numerical options, if not specified otherwise, accept in input

       a string representing a number, which may contain one of the

       International System number postfixes, for example 'K', 'M', 'G'.  If

       'i' is appended after the postfix, powers of 2 are used instead of

       powers of 10. The 'B' postfix multiplies the value for 8, and can be

       appended after another postfix or used alone. This allows using for

       example 'KB', 'MiB', 'G' and 'B' as postfix.

       Options which do not take arguments are boolean options, and set the

       corresponding value to true. They can be set to false by prefixing with

       "no" the option name, for example using "-nofoo" in the commandline

       will set to false the boolean option with name "foo".

-----------------------------------

   Generic options   通用選項

       These options are shared amongst the ff* tools.  ff*工具都公用這些選項

       -L  Show license.

       -h, -?, -help, --help

           Show help.

       -version

           Show version.

       -formats

           Show available formats.

           The fields preceding the format names have the following meanings:

           D   Decoding available

           E   Encoding available

       -codecs

           Show available codecs.

           The fields preceding the codec names have the following meanings:

           D   Decoding available

           E   Encoding available

           V/A/S

               Video/audio/subtitle codec

           S   Codec supports slices

           D   Codec supports direct rendering

           T   Codec can handle input truncated at random locations instead of

               only at frame boundaries

       -bsfs

           Show available bitstream filters.

       -protocols

           Show available protocols.

       -filters

           Show available libavfilter filters.

       -pix_fmts

           Show available pixel formats.

       -loglevel loglevel

           Set the logging level used by the library.  loglevel is a number or

           a string containing one of the following values:

           quiet

           panic

           fatal

           error

           warning

           info

           verbose

           debug

           By default the program logs to stderr, if coloring is supported by

           the terminal, colors are used to mark errors and warnings. Log

           coloring can be disabled setting the environment variable

           FFMPEG_FORCE_NOCOLOR or NO_COLOR, or can be forced setting the

           environment variable FFMPEG_FORCE_COLOR.  The use of the

           environment variable NO_COLOR is deprecated and will be dropped in

           a following FFmpeg version.

----------------------------------

   Main options   主要選項

       -x width

           Force displayed width.

       -y height

           Force displayed height.

       -s size

           Set frame size (WxH or abbreviation), needed for videos which don't

           contain a header with the frame size like raw YUV.

       -an Disable audio.

       -vn Disable video.

       -ss pos

           Seek to a given position in seconds.

       -t duration

           play <duration> seconds of audio/video

       -bytes

           Seek by bytes.

       -nodisp

           Disable graphical display.

       -f fmt

           Force format.

       -window_title title

           Set window title (default is the input filename).

       -loop number

           Loops movie playback <number> times. 0 means forever.

       -vf filter_graph

           filter_graph is a description of the filter graph to apply to the

           input video.  Use the option "-filters" to show all the available

           filters (including also sources and sinks).

-------------------------------------------------

   Advanced options  高階選項

       -pix_fmt format

           Set pixel format.

       -stats

           Show the stream duration, the codec parameters, the current

           position in the stream and the audio/video synchronisation drift.

       -debug

           Print specific debug info.

       -bug

           Work around bugs.

       -vismv

           Visualize motion vectors.

       -fast

           Non-spec-compliant optimizations.

       -genpts

           Generate pts.

       -rtp_tcp

           Force RTP/TCP protocol usage instead of RTP/UDP. It is only

           meaningful if you are streaming with the RTSP protocol.

       -sync type

           Set the master clock to audio ("type=audio"), video ("type=video")

           or external ("type=ext"). Default is audio. The master clock is

           used to control audio-video synchronization. Most media players use

           audio as master clock, but in some cases (streaming or high quality

           broadcast) it is necessary to change that. This option is mainly

           used for debugging purposes.

       -threads count

           Set the thread count.

       -ast audio_stream_number

           Select the desired audio stream number, counting from 0. The number

           refers to the list of all the input audio streams. If it is greater

           than the number of audio streams minus one, then the last one is

           selected, if it is negative the audio playback is disabled.

       -vst video_stream_number

           Select the desired video stream number, counting from 0. The number

           refers to the list of all the input video streams. If it is greater

           than the number of video streams minus one, then the last one is

           selected, if it is negative the video playback is disabled.

       -sst subtitle_stream_number

           Select the desired subtitle stream number, counting from 0. The

           number refers to the list of all the input subtitle streams. If it

           is greater than the number of subtitle streams minus one, then the

           last one is selected, if it is negative the subtitle rendering is

           disabled.

       -autoexit

           Exit when video is done playing.

       -exitonkeydown

           Exit if any key is pressed.

       -exitonmousedown

           Exit if any mouse button is pressed.

-------------------------------------------------

   While playing  快捷鍵

       q, ESC

           Quit.

       f   Toggle full screen.

       p, SPC

           Pause.

       a   Cycle audio channel.

       v   Cycle video channel.

       t   Cycle subtitle channel.

       w   Show audio waves.

       left/right

           Seek backward/forward 10 seconds.

       down/up

           Seek backward/forward 1 minute.

       mouse click

           Seek to percentage in file corresponding to fraction of width.

-------------------------------------------------

EXPRESSION EVALUATION   表示式

       When evaluating an arithemetic expression, FFmpeg uses an internal

       formula evaluator, implemented through the libavutil/eval.h interface.

       An expression may contain unary, binary operators, constants, and

       functions.

       Two expressions expr1 and expr2 can be combined to form another

       expression "expr1;expr2".  expr1 and expr2 are evaluated in turn, and

       the new expression evaluates to the value of expr2.

       The following binary operators are available: "+", "-", "*", "/", "^".

       The following unary operators are available: "+", "-".

       The following functions are available:

       sinh(x)

       cosh(x)

       tanh(x)

       sin(x)

       cos(x)

       tan(x)

       atan(x)

       asin(x)

       acos(x)

       exp(x)

       log(x)

       abs(x)

       squish(x)

       gauss(x)

       isnan(x)

           Return 1.0 if x is NAN, 0.0 otherwise.

       mod(x, y)

       max(x, y)

       min(x, y)

       eq(x, y)

       gte(x, y)

       gt(x, y)

       lte(x, y)

       lt(x, y)

       st(var, expr)

           Allow to store the value of the expression expr in an internal

           variable. var specifies the number of the variable where to store

           the value, and it is a value ranging from 0 to 9. The function

           returns the value stored in the internal variable.

       ld(var)

           Allow to load the value of the internal variable with number var,

           which was previosly stored with st(var, expr).  The function

           returns the loaded value.

       while(cond, expr)

           Evaluate expression expr while the expression cond is non-zero, and

           returns the value of the last expr evaluation, or NAN if cond was

           always false.

       Note that:

       "*" works like AND

       "+" works like OR

       thus

               if A then B else C

       is equivalent to

               A*B + not(A)*C

       When A evaluates to either 1 or 0, that is the same as

               A*B + eq(A,0)*C

       In your C code, you can extend the list of unary and binary functions,

       and define recognized constants, so that they are available for your

       expressions.

       The evaluator also recognizes the International System number

       postfixes. If 'i' is appended after the postfix, powers of 2 are used

       instead of powers of 10. The 'B' postfix multiplies the value for 8,

       and can be appended after another postfix or used alone. This allows

       using for example 'KB', 'MiB', 'G' and 'B' as postfix.

       Follows the list of available International System postfixes, with

       indication of the corresponding powers of 10 and of 2.

       y   -24 / -80

       z   -21 / -70

       a   -18 / -60

       f   -15 / -50

       p   -12 / -40

       n   -9 / -30

       u   -6 / -20

       m   -3 / -10

       c   -2

       d   -1

       h   2

       k   3 / 10

       K   3 / 10

       M   6 / 20

       G   9 / 30

       T   12 / 40

       P   15 / 40

       E   18 / 50

       Z   21 / 60

       Y   24 / 70

-------------------------------------------------

INPUT DEVICES   輸入裝置

       Input devices are configured elements in FFmpeg which allow to access

       the data coming from a multimedia device attached to your system.

       When you configure your FFmpeg build, all the supported input devices

       are enabled by default. You can list all available ones using the

       configure option "--list-indevs".

       You can disable all the input devices using the configure option

       "--disable-indevs", and selectively enable an input device using the

       option "--enable-indev=INDEV", or you can disable a particular input

       device using the option "--disable-indev=INDEV".

       The option "-formats" of the ff* tools will display the list of

       supported input devices (amongst the demuxers).

       A description of the currently available input devices follows.

   alsa

       ALSA (Advanced Linux Sound Architecture) input device.

       To enable this input device during configuration you need libasound

       installed on your system.

       This device allows capturing from an ALSA device. The name of the

       device to capture has to be an ALSA card identifier.

       An ALSA identifier has the syntax:

               hw:<CARD>[,<DEV>[,<SUBDEV>]]

       where the DEV and SUBDEV components are optional.

       The three arguments (in order: CARD,DEV,SUBDEV) specify card number or

       identifier, device number and subdevice number (-1 means any).

       To see the list of cards currently recognized by your system check the

       files /proc/asound/cards and /proc/asound/devices.

       For example to capture with ffmpeg from an ALSA device with card id 0,

       you may run the command:

               ffmpeg -f alsa -i hw:0 alsaout.wav

       For more information see:

       <http://www.alsa-project.org/alsa-doc/alsa-lib/pcm.html>

   bktr

       BSD video input device.

   dv1394

       Linux DV 1394 input device.

   jack

       JACK input device.

       To enable this input device during configuration you need libjack

       installed on your system.

       A JACK input device creates one or more JACK writable clients, one for

       each audio channel, with name client_name:input_N, where client_name is

       the name provided by the application, and N is a number which

       identifies the channel.  Each writable client will send the acquired

       data to the FFmpeg input device.

       Once you have created one or more JACK readable clients, you need to

       connect them to one or more JACK writable clients.

       To connect or disconnect JACK clients you can use the jack_connect and

       jack_disconnect programs, or do it through a graphical interface, for

       example with qjackctl.

       To list the JACK clients and their properties you can invoke the

       command jack_lsp.

       Follows an example which shows how to capture a JACK readable client

       with ffmpeg.

               # Create a JACK writable client with name "ffmpeg".

               $ ffmpeg -f jack -i ffmpeg -y out.wav

               # Start the sample jack_metro readable client.

               $ jack_metro -b 120 -d 0.2 -f 4000

               # List the current JACK clients.

               $ jack_lsp -c

               system:capture_1

               system:capture_2

               system:playback_1

               system:playback_2

               ffmpeg:input_1

               metro:120_bpm

               # Connect metro to the ffmpeg writable client.

               $ jack_connect metro:120_bpm ffmpeg:input_1

       For more information read: <http://jackaudio.org/>

   libdc1394

       IIDC1394 input device, based on libdc1394 and libraw1394.

   oss

       Open Sound System input device.

       The filename to provide to the input device is the device node

       representing the OSS input device, and is usually set to /dev/dsp.

       For example to grab from /dev/dsp using ffmpeg use the command:

               ffmpeg -f oss -i /dev/dsp /tmp/oss.wav

       For more information about OSS see:

       <http://manuals.opensound.com/usersguide/dsp.html>

-------------------------------------------------

   video4linux and video4linux2

       Video4Linux and Video4Linux2 input video devices.

       The name of the device to grab is a file device node, usually Linux

       systems tend to automatically create such nodes when the device (e.g.

       an USB webcam) is plugged into the system, and has a name of the kind

       /dev/videoN, where N is a number associated to the device.

       Video4Linux and Video4Linux2 devices only support a limited set of

       widthxheight sizes and framerates. You can check which are supported

       for example with the command dov4l for Video4Linux devices and the

       command v4l-info for Video4Linux2 devices.

       If the size for the device is set to 0x0, the input device will try to

       autodetect the size to use.  Only for the video4linux2 device, if the

       frame rate is set to 0/0 the input device will use the frame rate value

       already set in the driver.

       Video4Linux support is deprecated since Linux 2.6.30, and will be

       dropped in later versions.

       Follow some usage examples of the video4linux devices with the ff*

       tools.

               # Grab and show the input of a video4linux device, frame rate is set

               # to the default of 25/1.

               ffplay -s 320x240 -f video4linux /dev/video0

               # Grab and show the input of a video4linux2 device, autoadjust size.

               ffplay -f video4linux2 /dev/video0

               # Grab and record the input of a video4linux2 device, autoadjust size,

               # frame rate value defaults to 0/0 so it is read from the video4linux2

               # driver.

               ffmpeg -f video4linux2 -i /dev/video0 out.mpeg

   vfwcap

       VfW (Video for Windows) capture input device.

       The filename passed as input is the capture driver number, ranging from

       0 to 9. You may use "list" as filename to print a list of drivers. Any

       other filename will be interpreted as device number 0.

   x11grab

       X11 video input device.

       This device allows to capture a region of an X11 display.

       The filename passed as input has the syntax:

               [<hostname>]:<display_number>.<screen_number>[+<x_offset>,<y_offset>]

       hostname:display_number.screen_number specifies the X11 display name of

       the screen to grab from. hostname can be ommitted, and defaults to

       "localhost". The environment variable DISPLAY contains the default

       display name.

       x_offset and y_offset specify the offsets of the grabbed area with

       respect to the top-left border of the X11 screen. They default to 0.

       Check the X11 documentation (e.g. man X) for more detailed information.

       Use the dpyinfo program for getting basic information about the

       properties of your X11 display (e.g. grep for "name" or "dimensions").

       For example to grab from :0.0 using ffmpeg:

               ffmpeg -f x11grab -r 25 -s cif -i :0.0 out.mpg

               # Grab at position 10,20.

               ffmpeg -f x11grab -25 -s cif -i :0.0+10,20 out.mpg

-------------------------------------------------

OUTPUT DEVICES   輸出裝置

       Output devices are configured elements in FFmpeg which allow to write

       multimedia data to an output device attached to your system.

       When you configure your FFmpeg build, all the supported output devices

       are enabled by default. You can list all available ones using the

       configure option "--list-outdevs".

       You can disable all the output devices using the configure option

       "--disable-outdevs", and selectively enable an output device using the

       option "--enable-outdev=OUTDEV", or you can disable a particular input

       device using the option "--disable-outdev=OUTDEV".

       The option "-formats" of the ff* tools will display the list of enabled

       output devices (amongst the muxers).

       A description of the currently available output devices follows.

   alsa

       ALSA (Advanced Linux Sound Architecture) output device.

   oss

       OSS (Open Sound System) output device.

-------------------------------------------------

PROTOCOLS    協議

       Protocols are configured elements in FFmpeg which allow to access

       resources which require the use of a particular protocol.

       When you configure your FFmpeg build, all the supported protocols are

       enabled by default. You can list all available ones using the configure

       option "--list-protocols".

       You can disable all the protocols using the configure option

       "--disable-protocols", and selectively enable a protocol using the

       option "--enable-protocol=PROTOCOL", or you can disable a particular

       protocol using the option "--disable-protocol=PROTOCOL".

       The option "-protocols" of the ff* tools will display the list of

       supported protocols.

       A description of the currently available protocols follows.

-------------------------------------------------

   concat   

       Physical concatenation protocol.

       Allow to read and seek from many resource in sequence as if they were a

       unique resource.

       A URL accepted by this protocol has the syntax:

               concat:<URL1>|<URL2>|...|<URLN>

       where URL1, URL2, ..., URLN are the urls of the resource to be

       concatenated, each one possibly specifying a distinct protocol.

       For example to read a sequence of files split1.mpeg, split2.mpeg,

       split3.mpeg with ffplay use the command:

               ffplay concat:split1.mpeg/|split2.mpeg/|split3.mpeg

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

       many shells.

-------------------------------------------------

   file  檔案訪問協議

       File access protocol.

       Allow to read from or read to a file.

       For example to read from a file input.mpeg with ffmpeg use the command:

               ffmpeg -i file:input.mpeg output.mpeg

       The ff* tools default to the file protocol, that is a resource

       specified with the name "FILE.mpeg" is interpreted as the URL

       "file:FILE.mpeg".

   gopher

       Gopher protocol.

   http

       HTTP (Hyper Text Transfer Protocol).

   mmst

       MMS (Microsoft Media Server) protocol over TCP.

   mmsh

       MMS (Microsoft Media Server) protocol over HTTP.

       The required syntax is:

               mmsh://<server>[:<port>][/<app>][/<playpath>]

   md5

       MD5 output protocol.

       Computes the MD5 hash of the data to be written, and on close writes

       this to the designated output or stdout if none is specified. It can be

       used to test muxers without writing an actual file.

       Some examples follow.

               # Write the MD5 hash of the encoded AVI file to the file output.avi.md5.

               ffmpeg -i input.flv -f avi -y md5:output.avi.md5

               # Write the MD5 hash of the encoded AVI file to stdout.

               ffmpeg -i input.flv -f avi -y md5:

       Note that some formats (typically MOV) require the output protocol to

       be seekable, so they will fail with the MD5 output protocol.

-------------------------------------------------

   pipe   unix的管道訪問協議

       UNIX pipe access protocol.

       Allow to read and write from UNIX pipes.

       The accepted syntax is:

               pipe:[<number>]

       number is the number corresponding to the file descriptor of the pipe

       (e.g. 0 for stdin, 1 for stdout, 2 for stderr).  If number is not

       specified, by default the stdout file descriptor will be used for

       writing, stdin for reading.

       For example to read from stdin with ffmpeg:

               cat test.wav | ffmpeg -i pipe:0

               # ...this is the same as...

               cat test.wav | ffmpeg -i pipe:

       For writing to stdout with ffmpeg:

               ffmpeg -i test.wav -f avi pipe:1 | cat > test.avi

               # ...this is the same as...

               ffmpeg -i test.wav -f avi pipe: | cat > test.avi

       Note that some formats (typically MOV), require the output protocol to

       be seekable, so they will fail with the pipe output protocol.

   rtmp

       Real-Time Messaging Protocol.

       The Real-Time Messaging Protocol (RTMP) is used for streaming

       multimeaXX dia content across a TCP/IP network.

       The required syntax is:

               rtmp://<server>[:<port>][/<app>][/<playpath>]

       The accepted parameters are:

       server

           The address of the RTMP server.

       port

           The number of the TCP port to use (by default is 1935).

       app It is the name of the application to access. It usually corresponds

           to the path where the application is installed on the RTMP server

           (e.g. /ondemand/, /flash/live/, etc.).

       playpath

           It is the path or name of the resource to play with reference to

           the application specified in app, may be prefixed by "mp4:".

       For example to read with ffplay a multimedia resource named "sample"

       from the application "vod" from an RTMP server "myserver":

               ffplay rtmp://myserver/vod/sample

   rtmp, rtmpe, rtmps, rtmpt, rtmpte

       Real-Time Messaging Protocol and its variants supported through

       librtmp.

       Requires the presence of the librtmp headers and library during

       configuration. You need to explicitely configure the build with

       "--enable-librtmp". If enabled this will replace the native RTMP

       protocol.

       This protocol provides most client functions and a few server functions

       needed to support RTMP, RTMP tunneled in HTTP (RTMPT), encrypted RTMP

       (RTMPE), RTMP over SSL/TLS (RTMPS) and tunneled variants of these

       encrypted types (RTMPTE, RTMPTS).

       The required syntax is:

               <rtmp_proto>://<server>[:<port>][/<app>][/<playpath>] <options>

       where rtmp_proto is one of the strings "rtmp", "rtmpt", "rtmpe",

       "rtmps", "rtmpte", "rtmpts" corresponding to each RTMP variant, and

       server, port, app and playpath have the same meaning as specified for

       the RTMP native protocol.  options contains a list of space-separated

       options of the form key=val.

       See the librtmp manual page (man 3 librtmp) for more information.

       For example, to stream a file in real-time to an RTMP server using

       ffmpeg:

               ffmpeg -re -i myfile -f flv rtmp://myserver/live/mystream

       To play the same stream using ffplay:

               ffplay "rtmp://myserver/live/mystream live=1"

   rtp

       Real-Time Protocol.

   rtsp

       RTSP is not technically a protocol handler in libavformat, it is a

       demuxer and muxer. The demuxer supports both normal RTSP (with data

       transferred over RTP; this is used by e.g. Apple and Microsoft) and

       Real-RTSP (with data transferred over RDT).

       The muxer can be used to send a stream using RTSP ANNOUNCE to a server

       supporting it (currently Darwin Streaming Server and Mischa

       Spiegelmock's RTSP server, <http://github.com/revmischa/rtsp-server>).

       The required syntax for a RTSP url is:

               rtsp://<hostname>[:<port>]/<path>[?<options>]

       options is a "&"-separated list. The following options are supported:

       udp Use UDP as lower transport protocol.

       tcp Use TCP (interleaving within the RTSP control channel) as lower

           transport protocol.

       multicast

           Use UDP multicast as lower transport protocol.

       http

           Use HTTP tunneling as lower transport protocol, which is useful for

           passing proxies.

       filter_src

           Accept packets only from negotiated peer address and port.

       Multiple lower transport protocols may be specified, in that case they

       are tried one at a time (if the setup of one fails, the next one is

       tried).  For the muxer, only the "tcp" and "udp" options are supported.

       When receiving data over UDP, the demuxer tries to reorder received

       packets (since they may arrive out of order, or packets may get lost

       totally). In order for this to be enabled, a maximum delay must be

       specified in the "max_delay" field of AVFormatContext.

       When watching multi-bitrate Real-RTSP streams with ffplay, the streams

       to display can be chosen with "-vst" n and "-ast" n for video and audio

       respectively, and can be switched on the fly by pressing "v" and "a".

       Example command lines:

       To watch a stream over UDP, with a max reordering delay of 0.5 seconds:

               ffplay -max_delay 500000 rtsp://server/video.mp4?udp

       To watch a stream tunneled over HTTP:

               ffplay rtsp://server/video.mp4?http

       To send a stream in realtime to a RTSP server, for others to watch:

               ffmpeg -re -i <input> -f rtsp -muxdelay 0.1 rtsp://server/live.sdp

   sap

       Session Announcement Protocol (RFC 2974). This is not technically a

       protocol handler in libavformat, it is a muxer and demuxer.  It is used

       for signalling of RTP streams, by announcing the SDP for the streams

       regularly on a separate port.

-------------------------------------------------

       Muxer

       The syntax for a SAP url given to the muxer is:

               sap://<destination>[:<port>][?<options>]

       The RTP packets are sent to destination on port port, or to port 5004

       if no port is specified.  options is a "&"-separated list. The

       following options are supported:

       announce_addr=address

           Specify the destination IP address for sending the announcements

           to.  If omitted, the announcements are sent to the commonly used

           SAP announcement multicast address 224.2.127.254 (sap.mcast.net),

           or ff0e::2:7ffe if destination is an IPv6 address.

       announce_port=port

           Specify the port to send the announcements on, defaults to 9875 if

           not specified.

       ttl=ttl

           Specify the time to live value for the announcements and RTP

           packets, defaults to 255.

       same_port=0|1

           If set to 1, send all RTP streams on the same port pair. If zero

           (the default), all streams are sent on unique ports, with each

           stream on a port 2 numbers higher than the previous.  VLC/Live555

           requires this to be set to 1, to be able to receive the stream.

           The RTP stack in libavformat for receiving requires all streams to

           be sent on unique ports.

       Example command lines follow.

       To broadcast a stream on the local subnet, for watching in VLC:

               ffmpeg -re -i <input> -f sap sap://224.0.0.255?same_port=1

       Similarly, for watching in ffplay:

               ffmpeg -re -i <input> -f sap sap://224.0.0.255

       And for watching in ffplay, over IPv6:

               ffmpeg -re -i <input> -f sap sap://[ff0e::1:2:3:4]

       Demuxer

       The syntax for a SAP url given to the demuxer is:

               sap://[<address>][:<port>]

       address is the multicast address to listen for announcements on, if

       omitted, the default 224.2.127.254 (sap.mcast.net) is used. port is the

       port that is listened on, 9875 if omitted.

       The demuxers listens for announcements on the given address and port.

       Once an announcement is received, it tries to receive that particular

       stream.

       Example command lines follow.

       To play back the first stream announced on the normal SAP multicast

       address:

               ffplay sap://

       To play back the first stream announced on one the default IPv6 SAP

       multicast address:

               ffplay sap://[ff0e::2:7ffe]

   tcp

       Trasmission Control Protocol.

   udp

       User Datagram Protocol.

       The required syntax for a UDP url is:

               udp://<hostname>:<port>[?<options>]

       options contains a list of &-seperated options of the form key=val.

       Follow the list of supported options.

       buffer_size=size

           set the UDP buffer size in bytes

       localport=port

           override the local UDP port to bind with

       pkt_size=size

           set the size in bytes of UDP packets

       reuse=1|0

           explicitly allow or disallow reusing UDP sockets

       ttl=ttl

           set the time to live value (for multicast only)

       connect=1|0

           Initialize the UDP socket with "connect()". In this case, the

           destination address can't be changed with udp_set_remote_url later.

           If the destination address isn't known at the start, this option

           can be specified in udp_set_remote_url, too.  This allows finding

           out the source address for the packets with getsockname, and makes

           writes return with AVERROR(ECONNREFUSED) if "destination

           unreachable" is received.  For receiving, this gives the benefit of

           only receiving packets from the specified peer address/port.

       Some usage examples of the udp protocol with ffmpeg follow.

       To stream over UDP to a remote endpoint:

               ffmpeg -i <input> -f <format> udp://<hostname>:<port>

       To stream in mpegts format over UDP using 188 sized UDP packets, using

       a large input buffer:

               ffmpeg -i <input> -f mpegts udp://<hostname>:<port>?pkt_size=188&buffer_size=65535

       To receive over UDP from a remote endpoint:

               ffmpeg -i udp://[<multicast-address>]:<port>

-------------------------------------------------

FILTERGRAPH DESCRIPTION

       A filtergraph is a directed graph of connected filters. It can contain

       cycles, and there can be multiple links between a pair of filters. Each

       link has one input pad on one side connecting it to one filter from

       which it takes its input, and one output pad on the other side

       connecting it to the one filter accepting its output.

       Each filter in a filtergraph is an instance of a filter class

       registered in the application, which defines the features and the

       number of input and output pads of the filter.

       A filter with no input pads is called a "source", a filter with no

       output pads is called a "sink".

   Filtergraph syntax

       A filtergraph can be represented using a textual representation, which

       is recognized by the "-vf" and "-af" options of the ff* tools, and by

       the "av_parse_graph()" function defined in libavfilter/avfiltergraph.

       A filterchain consists of a sequence of connected filters, each one

       connected to the previous one in the sequence. A filterchain is

       represented by a list of ","-separated filter descriptions.

       A filtergraph consists of a sequence of filterchains. A sequence of

       filterchains is represented by a list of ";"-separated filterchain

       descriptions.

       A filter is represented by a string of the form:

       [in_link_1]...[in_link_N]filter_name=arguments[out_link_1]...[out_link_M]

       filter_name is the name of the filter class of which the described

       filter is an instance of, and has to be the name of one of the filter

       classes registered in the program.  The name of the filter class is

       optionally followed by a string "=arguments".

       arguments is a string which contains the parameters used to initialize

       the filter instance, and are described in the filter descriptions

       below.

       The list of arguments can be quoted using the character "'" as initial

       and ending mark, and the character '/' for escaping the characters

       within the quoted text; otherwise the argument string is considered

       terminated when the next special character (belonging to the set

       "[]=;,") is encountered.

       The name and arguments of the filter are optionally preceded and

       followed by a list of link labels.  A link label allows to name a link

       and associate it to a filter output or input pad. The preceding labels

       in_link_1 ... in_link_N, are associated to the filter input pads, the

       following labels out_link_1 ... out_link_M, are associated to the

       output pads.

       When two link labels with the same name are found in the filtergraph, a

       link between the corresponding input and output pad is created.

       If an output pad is not labelled, it is linked by default to the first

       unlabelled input pad of the next filter in the filterchain.  For

       example in the filterchain:

               nullsrc, split[L1], [L2]overlay, nullsink

       the split filter instance has two output pads, and the overlay filter

       instance two input pads. The first output pad of split is labelled

       "L1", the first input pad of overlay is labelled "L2", and the second

       output pad of split is linked to the second input pad of overlay, which

       are both unlabelled.

       In a complete filterchain all the unlabelled filter input and output

       pads must be connected. A filtergraph is considered valid if all the

       filter input and output pads of all the filterchains are connected.

       Follows a BNF description for the filtergraph syntax:

               <NAME>             ::= sequence of alphanumeric characters and '_'

               <LINKLABEL>        ::= "[" <NAME> "]"

               <LINKLABELS>       ::= <LINKLABEL> [<LINKLABELS>]

               <FILTER_ARGUMENTS> ::= sequence of chars (eventually quoted)

               <FILTER>           ::= [<LINKNAMES>] <NAME> ["=" <ARGUMENTS>] [<LINKNAMES>]

               <FILTERCHAIN>      ::= <FILTER> [,<FILTERCHAIN>]

               <FILTERGRAPH>      ::= <FILTERCHAIN> [;<FILTERGRAPH>]

-------------------------------------------------

AUDIO FILTERS

       When you configure your FFmpeg build, you can disable any of the

       existing filters using --disable-filters.  The configure output will

       show the audio filters included in your build.

       Below is a description of the currently available audio filters.

   anull

       Pass the audio source unchanged to the output.

AUDIO SOURCES

       Below is a description of the currently available audio sources.

   anullsrc

       Null audio source, never return audio frames. It is mainly useful as a

       template and to be employed in analysis / debugging tools.

       It accepts as optional parameter a string of the form

       sample_rate:channel_layout.

       sample_rate specify the sample rate, and defaults to 44100.

       channel_layout specify the channel layout, and can be either an integer

       or a string representing a channel layout. The default value of

       channel_layout is 3, which corresponds to CH_LAYOUT_STEREO.

       Check the channel_layout_map definition in libavcodec/audioconvert.c

       for the mapping between strings and channel layout values.

       Follow some examples:

               #  set the sample rate to 48000 Hz and the channel layout to CH_LAYOUT_MONO.

               anullsrc=48000:4

               # same as

               anullsrc=48000:mono

AUDIO SINKS

       Below is a description of the currently available audio sinks.

   anullsink

       Null audio sink, do absolutely nothing with the input audio. It is

       mainly useful as a template and to be employed in analysis / debugging

       tools.

VIDEO FILTERS

       When you configure your FFmpeg build, you can disable any of the

       existing filters using --disable-filters.  The configure output will

       show the video filters included in your build.

       Below is a description of the currently available video filters.

   blackframe

       Detect frames that are (almost) completely black. Can be useful to

       detect chapter transitions or commercials. Output lines consist of the

       frame number of the detected frame, the percentage of blackness, the

       position in the file if known or -1 and the timestamp in seconds.

       In order to display the output lines, you need to set the loglevel at

       least to the AV_LOG_INFO value.

       The filter accepts the syntax:

               blackframe[=<amount>:[<threshold>]]

       amount is the percentage of the pixels that have to be below the

       threshold, and defaults to 98.

       threshold is the threshold below which a pixel value is considered

       black, and defaults to 32.

   copy

       Copy the input source unchanged to the output. Mainly useful for

       testing purposes.

   crop

       Crop the input video to out_w:out_h:x:y.

       The parameters are expressions containing the following constants:

       E, PI, PHI

           the corresponding mathematical approximated values for e (euler

           number), pi (greek PI), PHI (golden ratio)

       x, y

           the computed values for x and y. They are evaluated for each new

           frame.

       in_w, in_h

           the input width and heigth

       iw, ih

           same as in_w and in_h

       out_w, out_h

           the output (cropped) width and heigth

       ow, oh

           same as out_w and out_h

       n   the number of input frame, starting from 0

       pos the position in the file of the input frame, NAN if unknown

       t   timestamp expressed in seconds, NAN if the input timestamp is

           unknown

       The out_w and out_h parameters specify the expressions for the width

       and height of the output (cropped) video. They are evaluated just at

       the configuration of the filter.

       The default value of out_w is "in_w", and the default value of out_h is

       "in_h".

       The expression for out_w may depend on the value of out_h, and the

       expression for out_h may depend on out_w, but they cannot depend on x

       and y, as x and y are evaluated after out_w and out_h.

       The x and y parameters specify the expressions for the position of the

       top-left corner of the output (non-cropped) area. They are evaluated

       for each frame. If the evaluated value is not valid, it is approximated

       to the nearest valid value.

       The default value of x is "(in_w-out_w)/2", and the default value for y

       is "(in_h-out_h)/2", which set the cropped area at the center of the

       input image.

       The expression for x may depend on y, and the expression for y may

       depend on x.

       Follow some examples:

               # crop the central input area with size 100x100

               crop=100:100

               # crop the central input area with size 2/3 of the input video

               "crop=2/3*in_w:2/3*in_h"

               # crop the input video central square

               crop=in_h

               # delimit the rectangle with the top-left corner placed at position

               # 100:100 and the right-bottom corner corresponding to the right-bottom

               # corner of the input image.

               crop=in_w-100:in_h-100:100:100

               # crop 10 pixels from the lefth and right borders, and 20 pixels from

               # the top and bottom borders

               "crop=in_w-2*10:in_h-2*20"

               # keep only the bottom right quarter of the input image

               "crop=in_w/2:in_h/2:in_w/2:in_h/2"

               # crop height for getting Greek harmony

               "crop=in_w:1/PHI*in_w"

               # trembling effect

               "crop=in_w/2:in_h/2:(in_w-out_w)/2+((in_w-out_w)/2)*sin(n/10):(in_h-out_h)/2 +((in_h-out_h)/2)*sin(n/7)"

               # erratic camera effect depending on timestamp and position

               "crop=in_w/2:in_h/2:(in_w-out_w)/2+((in_w-out_w)/2)*sin(t*10):(in_h-out_h)/2 +((in_h-out_h)/2)*sin(t*13)"

               # set x depending on the value of y

               "crop=in_w/2:in_h/2:y:10+10*sin(n/10)"

   cropdetect

       Auto-detect crop size.

       Calculate necessary cropping parameters and prints the recommended

       parameters through the logging system. The detected dimensions

       correspond to the non-black area of the input video.

       It accepts the syntax:

               cropdetect[=<limit>[:<round>[:<reset>]]]

       limit

           Threshold, which can be optionally specified from nothing (0) to

           everything (255), defaults to 24.

       round

           Value which the width/height should be divisible by, defaults to

           16. The offset is automatically adjusted to center the video. Use 2

           to get only even dimensions (needed for 4:2:2 video). 16 is best

           when encoding to most video codecs.

       reset

           Counter that determines after how many frames cropdetect will reset

           the previously detected largest video area and start over to detect

           the current optimal crop area. Defaults to 0.

           This can be useful when channel logos distort the video area. 0

           indicates never reset and return the largest area encountered

           during playback.

   drawbox

       Draw a colored box on the input image.

       It accepts the syntax:

               drawbox=<x>:<y>:<width>:<height>:<color>

       x, y

           Specify the top left corner coordinates of the box. Default to 0.

       width, height

           Specify the width and height of the box, if 0 they are interpreted

           as the input width and height. Default to 0.

       color

           Specify the color of the box to write, it can be the name of a

           color (case insensitive match) or a 0xRRGGBB[AA] sequence.

       Follow some examples:

               # draw a black box around the edge of the input image

               drawbox

               # draw a box with color red and an opacity of 50%

               drawbox=10:20:200:60:[email protected]"

   fifo

       Buffer input images and send them when they are requested.

       This filter is mainly useful when auto-inserted by the libavfilter

       framework.

       The filter does not take parameters.

   format

       Convert the input video to one of the specified pixel formats.

       Libavfilter will try to pick one that is supported for the input to the

       next filter.

       The filter accepts a list of pixel format names, separated by ":", for

       example "yuv420p:monow:rgb24".

       Some examples follow:

               # convert the input video to the format "yuv420p"

               format=yuv420p

               # convert the input video to any of the formats in the list

               format=yuv420p:yuv444p:yuv410p

   frei0r

       Apply a frei0r effect to the input video.

       To enable compilation of this filter you need to install the frei0r

       header and configure FFmpeg with --enable-frei0r.

       The filter supports the syntax:

               <filter_name>[{:|=}<param1>:<param2>:...:<paramN>]

       filter_name is the name to the frei0r effect to load. If the

       environment variable FREI0R_PATH is defined, the frei0r effect is

       searched in each one of the directories specified by the colon

       separated list in FREIOR_PATH, otherwise in the standard frei0r paths,

       which are in this order: HOME/.frei0r-1/lib/, /usr/local/lib/frei0r-1/,

       /usr/lib/frei0r-1/.

       param1, param2, ... , paramN specify the parameters for the frei0r

       effect.

       A frei0r effect parameter can be a boolean (whose values are specified

       with "y" and "n"), a double, a color (specified by the syntax R/G/B, R,

       G, and B being float numbers from 0.0 to 1.0) or by an

       "av_parse_color()" color description), a position (specified by the

       syntax X/Y, X and Y being float numbers) and a string.

       The number and kind of parameters depend on the loaded effect. If an

       effect parameter is not specified the default value is set.

       Some examples follow:

               # apply the distort0r effect, set the first two double parameters

               frei0r=distort0r:0.5:0.01

               # apply the colordistance effect, takes a color as first parameter

               frei0r=colordistance:0.2/0.3/0.4

               frei0r=colordistance:violet

               frei0r=colordistance:0x112233

               # apply the perspective effect, specify the top left and top right

               # image positions

               frei0r=perspective:0.2/0.2:0.8/0.2

       For more information see: <http://piksel.org/frei0r>

   gradfun

       Fix the banding artifacts that are sometimes introduced into nearly

       flat regions by truncation to 8bit colordepth.  Interpolate the

       gradients that should go where the bands are, and dither them.

       The filter takes two optional parameters, separated by ':':

       strength:radius

       strength is the maximum amount by which the filter will change any one

       pixel. Also the threshold for detecting nearly flat regions. Acceptable

       values range from .51 to 255, default value is 1.2, out-of-range values

       will be clipped to the valid range.

       radius is the neighborhood to fit the gradient to. A larger radius

       makes for smoother gradients, but also prevents the filter from

       modifying the pixels near detailed regions. Acceptable values are 8-32,

       default value is 16, out-of-range values will be clipped to the valid

       range.

               # default parameters

               gradfun=1.2:16

               # omitting radius

               gradfun=1.2

   hflip

       Flip the input video horizontally.

       For example to horizontally flip the video in input with ffmpeg:

               ffmpeg -i in.avi -vf "hflip" out.avi

   hqdn3d

       High precision/quality 3d denoise filter. This filter aims to reduce

       image noise producing smooth images and making still images really

       still. It should enhance compressibility.

       It accepts the following optional parameters:

       luma_spatial:chroma_spatial:luma_tmp:chroma_tmp

       luma_spatial

           a non-negative float number which specifies spatial luma strength,

           defaults to 4.0

       chroma_spatial

           a non-negative float number which specifies spatial chroma

           strength, defaults to 3.0*luma_spatial/4.0

       luma_tmp

           a float number which specifies luma temporal strength, defaults to

           6.0*luma_spatial/4.0

       chroma_tmp

           a float number which specifies chroma temporal strength, defaults

           to luma_tmp*chroma_spatial/luma_spatial