<?xml version='1.0' encoding="utf-8"?>
<!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.2//EN"
"docbook/dtd/xml/4.2/docbookx.dtd">
<book>
<title>The Music Player Daemon - User's Manual</title>
<chapter id="intro">
<title>Introduction</title>
<para>
This document is work in progress. Most of it may be incomplete
yet. Please help!
</para>
<para>
<application>MPD</application> (Music Player Daemon) is, as the
name suggests, a server software allowing you to remotely play
your music, handle playlists, deliver music (HTTP streams with
various sub-protocols) and organize playlists.
</para>
<para>
It has been written with minimal resource usage and stability in
mind! Infact, it runs fine on a Pentium 75, allowing you to use
your cheap old PC to create a stereo system!
</para>
<para>
<application>MPD</application> supports also gapless playback,
buffered audio output, and crossfading!
</para>
<para>
The separate client and server design allows users to choose a
user interface that best suites their tastes independently of
the underlying daemon, which actually plays music!
</para>
</chapter>
<chapter id="install">
<title>Installation</title>
<para>
We recommend that you use the software installation routines of
your distribution to install <application>MPD</application>.
Most operating systems have a <application>MPD</application>
package, which is very easy to install.
</para>
<section id="install_debian">
<title>Installing on Debian/Ubuntu</title>
<para>
Install the package <application>MPD</application> via APT:
</para>
<programlisting>apt-get install mpd</programlisting>
<para>
When installed this way, <application>MPD</application> by
default looks for music in
<filename>/var/lib/mpd/music/</filename>; this may not be
correct. Look at your <filename>/etc/mpd.conf</filename>
file...
</para>
</section>
<section id="install_source">
<title>Compiling from source</title>
<para>
Download the source tarball from <ulink
url="http://www.musicpd.org/download.html">the
<application>MPD</application> home page</ulink> and unpack
it:
</para>
<programlisting>tar xf mpd-version.tar.xz
cd mpd-version</programlisting>
<para>
Make sure that all the required libraries and build tools are
installed. The <filename>INSTALL</filename> file has a list.
</para>
<para>
For example, the following installs a fairly complete list of
build dependencies on Debian Wheezy:
</para>
<programlisting>
apt-get install g++ \
libmad0-dev libmpg123-dev libid3tag0-dev \
libflac-dev libvorbis-dev libopus-dev \
libadplug-dev libaudiofile-dev libsndfile1-dev libfaad-dev \
libfluidsynth-dev libgme-dev libmikmod2-dev libmodplug-dev \
libmpcdec-dev libwavpack-dev libwildmidi-dev \
libsidplay2-dev libsidutils-dev libresid-builder-dev \
libavcodec-dev libavformat-dev \
libmp3lame-dev \
libsamplerate0-dev libsoxr-dev \
libbz2-dev libcdio-paranoia-dev libiso9660-dev libmms-dev \
libzzip-dev \
libcurl4-gnutls-dev libyajl-dev libexpat-dev \
libasound2-dev libao-dev libjack-jackd2-dev libopenal-dev \
libpulse-dev libroar-dev libshout3-dev \
libmpdclient-dev \
libnfs-dev libsmbclient-dev \
libupnp-dev \
libavahi-client-dev \
libsqlite3-dev \
libsystemd-daemon-dev libwrap0-dev \
libcppunit-dev xmlto \
libboost-dev \
libicu-dev
</programlisting>
<para>
Now configure the source tree:
</para>
<programlisting>./configure</programlisting>
<para>
The <parameter>--help</parameter> argument shows a list of
compile-time options. When everything is ready and
configured, compile:
</para>
<programlisting>make</programlisting>
<para>
And install:
</para>
<programlisting>make install</programlisting>
</section>
<section id="systemd_socket">
<title><filename>systemd</filename> socket activation</title>
<para>
Using <filename>systemd</filename>, you can launch
<application>MPD</application> on demand when the first client
attempts to connect.
</para>
<para>
<application>MPD</application> comes with two
<application>systemd</application> unit files: a "service"
unit and a "socket" unit. These will only be installed when
<application>MPD</application> was configured with
<parameter>--with-systemdsystemunitdir=/lib/systemd</parameter>.
</para>
<para>
To enable socket activation, type:
</para>
<programlisting>systemctl enable mpd.socket
systemctl start mpd.socket</programlisting>
<para>
In this configuration, <application>MPD</application> will
ignore the <varname>bind_to_address</varname> and
<varname>port</varname> settings.
</para>
</section>
</chapter>
<chapter id="config">
<title>Configuration</title>
<section id="config_music_directory">
<title>Configuring the music directory</title>
<para>
When you play local files, you should organize them within a
directory called the "music directory". This is configured in
<application>MPD</application> with the
<varname>music_directory</varname> setting.
</para>
<para>
By default, <application>MPD</application> follows symbolic
links in the music directory. This behavior can be switched
off: <varname>follow_outside_symlinks</varname> controls
whether <application>MPD</application> follows links pointing
to files outside of the music directory, and
<varname>follow_inside_symlinks</varname> lets you disable
symlinks to files inside the music directory.
</para>
<para>
Instead of using local files, you can use <link
linkend="storage_plugins">storage plugins</link> to access
files on a remote file server. For example, to use music from
the SMB/CIFS server "myfileserver" on the share called
"Music", configure the music directory
"<parameter>smb://myfileserver/Music</parameter>". For a
recipe, read the <link linkend="satellite">Satellite
MPD</link> section.
</para>
</section>
<section id="config_database_plugins">
<title>Configuring database plugins</title>
<para>
If a music directory is configured, one database plugin is
used. To configure this plugin, add a
<varname>database</varname> block to
<filename>mpd.conf</filename>:
</para>
<programlisting>database {
plugin "simple"
path "/var/lib/mpd/db"
}
</programlisting>
<para>
The following table lists the <varname>database</varname>
options valid for all plugins:
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>
Name
</entry>
<entry>
Description
</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>plugin</varname>
</entry>
<entry>
The name of the plugin.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
<para>
More information can be found in the <link
linkend="database_plugins">database plugin reference</link>.
</para>
</section>
<section id="config_neighbor_plugins">
<title>Configuring neighbor plugins</title>
<para>
All neighbor plugins are disabled by default to avoid unwanted
overhead. To enable (and configure) a plugin, add a
<varname>neighbor</varname> block to
<filename>mpd.conf</filename>:
</para>
<programlisting>neighbors {
plugin "smbclient"
}
</programlisting>
<para>
The following table lists the <varname>neighbor</varname>
options valid for all plugins:
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>
Name
</entry>
<entry>
Description
</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>plugin</varname>
</entry>
<entry>
The name of the plugin.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
<para>
More information can be found in the <link
linkend="neighbor_plugins">neighbor plugin reference</link>.
</para>
</section>
<section id="config_input_plugins">
<title>Configuring input plugins</title>
<para>
To configure an input plugin, add a <varname>input</varname>
block to <filename>mpd.conf</filename>:
</para>
<programlisting>input {
plugin "curl"
proxy "proxy.local"
}
</programlisting>
<para>
The following table lists the <varname>input</varname> options
valid for all plugins:
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>
Name
</entry>
<entry>
Description
</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>plugin</varname>
</entry>
<entry>
The name of the plugin.
</entry>
</row>
<row>
<entry>
<varname>enabled</varname>
<parameter>yes|no</parameter>
</entry>
<entry>
Allows you to disable a input plugin without
recompiling. By default, all plugins are enabled.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
<para>
More information can be found in the <link
linkend="input_plugins">input plugin reference</link>.
</para>
</section>
<section id="config_decoder_plugins">
<title>Configuring decoder plugins</title>
<para>
Most decoder plugins do not need any special configuration.
To configure a decoder, add a <varname>decoder</varname> block
to <filename>mpd.conf</filename>:
</para>
<programlisting>decoder {
plugin "wildmidi"
config_file "/etc/timidity/timidity.cfg"
}
</programlisting>
<para>
The following table lists the <varname>decoder</varname>
options valid for all plugins:
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>
Name
</entry>
<entry>
Description
</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>plugin</varname>
</entry>
<entry>
The name of the plugin.
</entry>
</row>
<row>
<entry>
<varname>enabled</varname>
<parameter>yes|no</parameter>
</entry>
<entry>
Allows you to disable a decoder plugin without
recompiling. By default, all plugins are enabled.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
<para>
More information can be found in the <link
linkend="decoder_plugins">decoder plugin reference</link>.
</para>
</section>
<section id="config_encoder_plugins">
<title>Configuring encoder plugins</title>
<para>
Encoders are used by some of the output plugins (such as <link
linkend="shout_output"><varname>shout</varname></link>). The
encoder settings are included in the
<varname>audio_output</varname> section. More information can
be found in the <link linkend="encoder_plugins">encoder plugin
reference</link>.
</para>
</section>
<section id="config_audio_outputs">
<title>Configuring audio outputs</title>
<para>
Audio outputs are devices which actually play the audio chunks
produced by <application>MPD</application>. You can configure
any number of audio output devices, but there must be at least
one. If none is configured, <application>MPD</application>
attempts to auto-detect. Usually, this works quite well with
ALSA, OSS and on Mac OS X.
</para>
<para>
To configure an audio output manually, add an
<varname>audio_output</varname> block to
<filename>mpd.conf</filename>:
</para>
<programlisting>audio_output {
type "alsa"
name "my ALSA device"
device "hw:0"
}
</programlisting>
<para>
The following table lists the <varname>audio_output</varname>
options valid for all plugins:
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>
Name
</entry>
<entry>
Description
</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>type</varname>
</entry>
<entry>
The name of the plugin.
</entry>
</row>
<row>
<entry>
<varname>name</varname>
</entry>
<entry>
The name of the audio output. It is visible to the
client. Some plugins also use it internally, e.g. as
a name registered in the PULSE server.
</entry>
</row>
<row id="ao_format">
<entry>
<varname>format</varname>
</entry>
<entry>
<para>
Always open the audio output with the specified audio
format (samplerate:bits:channels), regardless of the
format of the input file. This is optional for most
plugins.
</para>
<para>
Any of the three attributes may be an asterisk to
specify that this attribute should not be enforced,
example: <parameter>48000:16:*</parameter>.
<parameter>*:*:*</parameter> is equal to not having
a <varname>format</varname> specification.
</para>
<para>
The following values are valid for
<varname>bits</varname>: <varname>8</varname>
(signed 8 bit integer samples),
<varname>16</varname>, <varname>24</varname> (signed
24 bit integer samples padded to 32 bit),
<varname>32</varname> (signed 32 bit integer
samples), <varname>f</varname> (32 bit floating
point, -1.0 to 1.0).
</para>
</entry>
</row>
<row>
<entry>
<varname>enabled</varname>
<parameter>yes|no</parameter>
</entry>
<entry>
Specifies whether this audio output is enabled when
<application>MPD</application> is started. By
default, all audio outputs are enabled.
</entry>
</row>
<row>
<entry>
<varname>tags</varname>
<parameter>yes|no</parameter>
</entry>
<entry>
If set to <parameter>no</parameter>, then
<application>MPD</application> will not send tags to
this output. This is only useful for output plugins
that can receive tags, for example the <link
linkend="httpd_output"><varname>httpd</varname></link>
output plugin.
</entry>
</row>
<row>
<entry>
<varname>always_on</varname>
<parameter>yes|no</parameter>
</entry>
<entry>
If set to <parameter>yes</parameter>, then
<application>MPD</application> attempts to keep this
audio output always open. This may be useful for
streaming servers, when you don't want to disconnect
all listeners even when playback is accidentally
stopped.
</entry>
</row>
<row>
<entry>
<varname>mixer_type</varname>
<parameter>hardware|software|null|none</parameter>
</entry>
<entry>
Specifies which mixer should be used for this audio
output: the hardware mixer (available for <link
linkend="alsa_output">ALSA</link>, <link
linkend="oss_output">OSS</link> and <link
linkend="pulse_output">PulseAudio</link>), the
software mixer, the "null" mixer
(<parameter>null</parameter>; allows setting the
volume, but with no effect) or no mixer
(<parameter>none</parameter>). By default, the
hardware mixer is used for devices which support it,
and none for the others.
</entry>
</row>
<row>
<entry>
<varname>replay_gain_handler</varname>
<parameter>software|mixer|none</parameter>
</entry>
<entry>
Specifies how replay gain is applied. The default is
<parameter>software</parameter>, which uses an
internal software volume control.
<parameter>mixer</parameter> uses the configured
(hardware) mixer control. <parameter>none</parameter>
disables replay gain on this audio output.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section id="config_filters">
<title>Configuring filters</title>
<para>
Filters are plugins which modify an audio stream.
</para>
<para>
To configure a filter, add a <varname>filter</varname> block
to <filename>mpd.conf</filename>:
</para>
<programlisting>filter {
plugin "volume"
name "software volume"
}
</programlisting>
<para>
The following table lists the <varname>filter</varname>
options valid for all plugins:
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>
Name
</entry>
<entry>
Description
</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>plugin</varname>
</entry>
<entry>
The name of the plugin.
</entry>
</row>
<row>
<entry>
<varname>name</varname>
</entry>
<entry>
The name of the filter.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section id="config_playlist_plugins">
<title>Configuring playlist plugins</title>
<para>
Playlist plugins are used to load remote playlists. This is
not related to <application>MPD</application>'s playlist
directory.
</para>
<para>
To configure a playlist plugin, add a
<varname>playlist_plugin</varname> block to
<filename>mpd.conf</filename>:
</para>
<programlisting>playlist_plugin {
name "m3u"
enabled "true"
}
</programlisting>
<para>
The following table lists the
<varname>playlist_plugin</varname> options valid for all
plugins:
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>
Name
</entry>
<entry>
Description
</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>name</varname>
</entry>
<entry>
The name of the plugin.
</entry>
</row>
<row>
<entry>
<varname>enabled</varname>
<parameter>yes|no</parameter>
</entry>
<entry>
Allows you to disable a input plugin without
recompiling. By default, all plugins are enabled.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
<para>
More information can be found in the <link
linkend="playlist_plugins">playlist plugin reference</link>.
</para>
</section>
<section id="config_audio_format">
<title>Audio Format Settings</title>
<section id="config_global_audio_format">
<title>Global Audio Format</title>
<para>
The setting <varname>audio_output_format</varname> forces
<application>MPD</application> to use one audio format for
all outputs. Doing that is usually not a good idea. The
values are the same as in <link
linkend="ao_format"><varname>format</varname> in the <link
linkend="config_audio_outputs"><varname>audio_output</varname></link>
section</link>.
</para>
</section>
<section>
<title>Resampler</title>
<para>
Sometimes, music needs to be resampled before it can be
played; for example, CDs use a sample rate of 44,100 Hz
while many cheap audio chips can only handle 48,000 Hz.
Resampling reduces the quality and consumes a lot of CPU.
There are different options, some of them optimized for high
quality and others for low CPU usage, but you can't have
both at the same time. Often, the resampler is the
component that is responsible for most of
<application>MPD</application>'s CPU usage. Since
<application>MPD</application> comes with high quality
defaults, it may appear that <application>MPD</application>
consumes more CPU than other software.
</para>
<para>
Check the <link linkend="resampler_plugins">resampler plugin
reference</link> for a list of resamplers and how to
configure them.
</para>
</section>
</section>
<section id="config_other">
<title>Other Settings</title>
<section>
<title>The State File</title>
<para>
The <emphasis>state file</emphasis> is a file where
<application>MPD</application> saves and restores its state
(play queue, playback position etc.) to keep it persistent
across restarts and reboots. It is an optional setting.
</para>
<para>
<application>MPD</application> will attempt to load the
state file during startup, and will save it when shutting
down the daemon. Additionally, the state file is refreshed
every two minutes (after each state change).
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>state_file</varname>
<parameter>PATH</parameter>
</entry>
<entry>
Specify the state file location. The parent
directory must be writable by the
<application>MPD</application> user
(<parameter>+wx</parameter>).
</entry>
</row>
<row>
<entry>
<varname>state_file_interval</varname>
<parameter>SECONDS</parameter>
</entry>
<entry>
Auto-save the state file this number of seconds
after each state change. Defaults to
<parameter>120</parameter> (2 minutes).
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section>
<title>Resource Limitations</title>
<para>
These settings are various limitations to prevent
<application>MPD</application> from using too many
resources (denial of service).
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>connection_timeout</varname>
<parameter>SECONDS</parameter>
</entry>
<entry>
If a client does not send any new data in this time
period, the connection is closed. Clients waiting
in "idle" mode are excluded from this. Default is
<parameter>60</parameter>.
</entry>
</row>
<row>
<entry>
<varname>max_connections</varname>
<parameter>NUMBER</parameter>
</entry>
<entry>
This specifies the maximum number of clients that
can be connected to <application>MPD</application>
at the same time. Default is
<parameter>5</parameter>.
</entry>
</row>
<row>
<entry>
<varname>max_playlist_length</varname>
<parameter>NUMBER</parameter>
</entry>
<entry>
The maximum number of songs that can be in the
playlist. Default is <parameter>16384</parameter>.
</entry>
</row>
<row>
<entry>
<varname>max_command_list_size</varname>
<parameter>KBYTES</parameter>
</entry>
<entry>
The maximum size a command list. Default is
<parameter>2048</parameter> (2 MiB).
</entry>
</row>
<row>
<entry>
<varname>max_output_buffer_size</varname>
<parameter>KBYTES</parameter>
</entry>
<entry>
The maximum size of the output buffer to a client
(maximum response size). Default is
<parameter>8192</parameter> (8 MiB).
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section>
<title>Buffer Settings</title>
<para>
Do not change these unless you know what you are doing.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>audio_buffer_size</varname>
<parameter>KBYTES</parameter>
</entry>
<entry>
Adjust the size of the internal audio buffer.
Default is <parameter>4096</parameter> (4 MiB).
</entry>
</row>
<row>
<entry>
<varname>buffer_before_play</varname>
<parameter>PERCENT</parameter>
</entry>
<entry>
Control the percentage of the buffer which is filled
before beginning to play. Increasing this reduces
the chance of audio file skipping, at the cost of
increased time prior to audio playback. Default is
<parameter>10%</parameter>.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
</section>
</chapter>
<chapter id="advanced_config">
<title>Advanced configuration</title>
<section id="satellite">
<title>Satellite setup</title>
<para>
<application>MPD</application> runs well on weak machines such
as the <ulink url="http://www.raspberrypi.org/">Raspberry
Pi</ulink>. However, such hardware tends to not have storage
big enough to hold a music collection. Mounting music from a
file server can be very slow, especially when updating the
database.
</para>
<para>
One approach for optimization is running
<application>MPD</application> on the file server, which not
only exports raw files, but also provides access to a readily
scanned database. Example configuration:
</para>
<programlisting>music_directory "nfs://fileserver.local/srv/mp3"
#music_directory "smb://fileserver.local/mp3"
database {
plugin "proxy"
host "fileserver.local"
}
</programlisting>
<para>
The <link
linkend="config_music_directory"><varname>music_directory</varname></link>
setting tells <application>MPD</application> to read files
from the given NFS server. It does this by connecting to the
server from userspace. This does not actually mount the file
server into the kernel's virtual file system, and thus
requires no kernel cooperation and no special privileges. It
does not even require a kernel with NFS support, only the
<link linkend="nfs_storage"><filename>nfs</filename></link>
storage plugin (using the <filename>libnfs</filename>
userspace library). The same can be done with SMB/CIFS using
the <link
linkend="smbclient_storage"><filename>smbclient</filename></link>
storage plugin (using <filename>libsmbclient</filename>).
</para>
<para>
The <link
linkend="config_database_plugins"><varname>database</varname></link>
setting tells <application>MPD</application> to pass all
database queries on to the <application>MPD</application>
instance running on the file server (using the <link
linkend="proxy_database"><filename>proxy</filename></link>
plugin).
</para>
</section>
</chapter>
<chapter id="use">
<title>Using <application>MPD</application></title>
<section id="client">
<title>The client</title>
<para>
After you have installed, configured and started
<application>MPD</application>, you choose a client to control
the playback.
</para>
<para>
The most basic client is <filename>mpc</filename>, which
provides a command line interface. It is useful in shell
scripts. Many people bind specific <filename>mpc</filename>
commands to hotkeys.
</para>
<para>
The <ulink
url="http://www.musicpd.org/clients/"><application>MPD</application>
Wiki</ulink> contains an extensive list of clients to choose
from.
</para>
</section>
<section id="music_directory_and_database">
<title>The music directory and the database</title>
<para>
The "music directory" is where you store your music files.
<application>MPD</application> stores all relevant meta
information about all songs in its "database". Whenever you
add, modify or remove songs in the music directory, you have
to update the database, for example with
<filename>mpc</filename>:
</para>
<programlisting>mpc update</programlisting>
<para>
Depending on the size of your music collection and the speed
of the storage, this can take a while.
</para>
<para>
To exclude a file from the update, create a file called
<filename>.mpdignore</filename> in its parent directory. Each
line of that file may contain a list of shell wildcards.
Matching files in the current directory and all subdirectories
are excluded.
</para>
</section>
<section id="queue">
<title>The queue</title>
<para>
The queue (sometimes called "current playlist") is a list of
songs to be played by <application>MPD</application>. To play
a song, add it to the queue and start playback. Most clients
offer an interface to edit the queue.
</para>
</section>
</chapter>
<chapter id="advanced_usage">
<title>Advanced usage</title>
<section id="bit_perfect">
<title>Bit-perfect playback</title>
<para>
"Bit-perfect playback" is a phrase used by audiophiles to
describe a setup that plays back digital music as-is, without
applying any modifications such as resampling, format
conversion or software volume. Naturally, this implies a
lossless codec.
</para>
<para>
By default, <application>MPD</application> attempts to do
bit-perfect playback, unless you tell it not to. Precondition
is a sound chip that supports the audio format of your music
files. If the audio format is not supported,
<application>MPD</application> attempts to fall back to the
nearest supported audio format, trying to lose as little
quality as possible.
</para>
<para>
To verify if <application>MPD</application> converts the audio
format, enable verbose logging, and watch for these lines:
</para>
<programlisting>decoder: audio_format=44100:24:2, seekable=true
output: opened plugin=alsa name="An ALSA output" audio_format=44100:16:2
output: converting from 44100:24:2</programlisting>
<para>
This example shows that a 24 bit file is being played, but the
sond chip cannot play 24 bit. It falls back to 16 bit,
discarding 8 bit.
</para>
<para>
However, this does not yet prove bit-perfect playback;
<application>ALSA</application> may be fooling
<application>MPD</application> that the audio format is
supported. To verify the format really being sent to the
physical sound chip, try:
</para>
<programlisting>cat /proc/asound/card*/pcm*p/sub*/hw_params
access: RW_INTERLEAVED
format: S16_LE
subformat: STD
channels: 2
rate: 44100 (44100/1)
period_size: 4096
buffer_size: 16384</programlisting>
<para>
Obey the "format" row, which indicates that the current
playback format is 16 bit (signed 16 bit integer, little
endian).
</para>
<para>
Check list for bit-perfect playback:
</para>
<itemizedlist>
<listitem>
<para>
Use the <link linkend="alsa_output">ALSA</link> output
plugin.
</para>
</listitem>
<listitem>
<para>
Disable sound processing inside
<application>ALSA</application> by configuring a
"hardware" device (<parameter>hw:0,0</parameter> or
similar).
</para>
</listitem>
<listitem>
<para>
Don't use software volume (setting <link
linkend="config_audio_outputs"><varname>mixer_type</varname></link>).
</para>
</listitem>
<listitem>
<para>
Don't force <application>MPD</application> to use a
specific audio format (settings <link
linkend="config_audio_outputs"><varname>format</varname></link>,
<link
linkend="config_global_audio_format"><varname>audio_output_format</varname></link>).
</para>
</listitem>
<listitem>
<para>
Verify that you are really doing bit-perfect playback
using <application>MPD</application>'s verbose log and
<filename>/proc/asound/card*/pcm*p/sub*/hw_params</filename>.
Some DACs can also indicate the audio format.
</para>
</listitem>
</itemizedlist>
</section>
<section id="dsd">
<title>Direct Stream Digital (DSD)</title>
<para>
DSD (<ulink
url="https://en.wikipedia.org/wiki/Direct_Stream_Digital">Direct
Stream Digital</ulink>) is a digital format that stores audio
as a sequence of single-bit values at a very high sampling
rate.
</para>
<para>
<application>MPD</application> understands the file formats
<link linkend="dsdiff_decoder"><filename>dff</filename></link>
and <link
linkend="dsf_decoder"><filename>dsf</filename></link>. There
are three ways to play back DSD:
</para>
<itemizedlist>
<listitem>
<para>
Native DSD playback. Requires
<application>ALSA</application> 1.0.27.1 or later, a sound
driver/chip that supports DSD and of course a DAC that
supports DSD.
</para>
</listitem>
<listitem>
<para>
DoP (DSD over PCM) playback. This wraps DSD inside fake
24 bit PCM according to the <ulink
url="http://dsd-guide.com/dop-open-standard">DoP
standard</ulink>. Requires a DAC that supports DSD. No
support from ALSA and the sound chip required (except for
<link linkend="bit_perfect">bit-perfect</link> 24 bit PCM
support).
</para>
</listitem>
<listitem>
<para>
Convert DSD to PCM on-the-fly.
</para>
</listitem>
</itemizedlist>
<para>
Native DSD playback is used automatically if available. DoP
is only used if enabled explicitly using the <link
linkend="alsa_output"><varname>dop</varname></link> option,
because there is no way for <application>MPD</application> to
find out whether the DAC supports it. DSD to PCM conversion
is the fallback if DSD cannot be used directly.
</para>
</section>
</chapter>
<chapter id="plugin_reference">
<title>Plugin reference</title>
<section id="database_plugins">
<title>Database plugins</title>
<section>
<title><varname>simple</varname></title>
<para>
The default plugin. Stores a copy of the database in
memory. A file is used for permanent storage.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>path</varname>
</entry>
<entry>
The path of the database file.
</entry>
</row>
<row>
<entry>
<varname>cache_directory</varname>
</entry>
<entry>
The path of the cache directory for additional
storages mounted at runtime. This setting is
necessary for the <command>mount</command> protocol
command.
</entry>
</row>
<row>
<entry>
<varname>compress</varname>
<parameter>yes|no</parameter>
</entry>
<entry>
Compress the database file using
<filename>gzip</filename>? Enabled by default (if
built with <filename>zlib</filename>).
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section id="proxy_database">
<title><varname>proxy</varname></title>
<para>
Provides access to the database of another
<application>MPD</application> instance using
<filename>libmpdclient</filename>. This is useful when you
run mount the music directory via NFS/SMB, and the file
server already runs a <application>MPD</application>
instance. Only the file server needs to update the
database.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>host</varname>
</entry>
<entry>
The host name of the "master"
<application>MPD</application> instance.
</entry>
</row>
<row>
<entry>
<varname>port</varname>
</entry>
<entry>
The port number of the "master"
<application>MPD</application> instance.
</entry>
</row>
<row>
<entry>
<varname>keepalive</varname>
<parameter>yes|no</parameter>
</entry>
<entry>
Send TCP keepalive packets to the "master"
<application>MPD</application> instance? This option can
help avoid certain firewalls dropping inactive
connections, at the expensive of a very small amount of
additional network traffic. Disabled by default.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section>
<title><varname>upnp</varname></title>
<para>
Provides access to UPnP media servers.
</para>
</section>
</section>
<section id="storage_plugins">
<title>Storage plugins</title>
<section>
<title><varname>local</varname></title>
<para>
The default plugin which gives
<application>MPD</application> access to local files. It is
used when <varname>music_directory</varname> refers to a
local directory.
</para>
</section>
<section id="smbclient_storage">
<title><varname>smbclient</varname></title>
<para>
Load music files from a SMB/CIFS server. It used used when
<varname>music_directory</varname> contains a
<parameter>smb://</parameter> URI, for example
"<parameter>smb://myfileserver/Music</parameter>".
</para>
</section>
<section id="nfs_storage">
<title><varname>nfs</varname></title>
<para>
Load music files from a NFS server. It used used when
<varname>music_directory</varname> contains a
<parameter>nfs://</parameter> URI according to <ulink
url="http://tools.ietf.org/html/rfc2224">RFC2224</ulink>,
for example "<parameter>nfs://servername/path</parameter>".
</para>
<para>
This plugin uses <ulink
url="https://github.com/sahlberg/libnfs"><filename>libnfs</filename></ulink>,
which supports only NFS version 3. Since
<application>MPD</application> is not allowed to bind to
"privileged ports", the NFS server needs to enable the
"insecure" setting; example
<filename>/etc/exports</filename>:
</para>
<programlisting>/srv/mp3 192.168.1.55(ro,insecure)</programlisting>
<para>
Don't fear: "insecure" does not mean that your NFS server is
insecure. A few decades ago, people thought the concept of
"privileged ports" would make network services "secure",
which was a fallacy. The absence of this obsolete
"security" measure means little.
</para>
</section>
</section>
<section id="neighbor_plugins">
<title>Neighbor plugins</title>
<section id="smbclient_neighbor">
<title><varname>smbclient</varname></title>
<para>
Provides a list of SMB/CIFS servers on the local network.
</para>
</section>
<section id="upnp_neighbor">
<title><varname>upnp</varname></title>
<para>
Provides a list of UPnP servers on the local network.
</para>
</section>
</section>
<section id="input_plugins">
<title>Input plugins</title>
<section>
<title><varname>alsa</varname></title>
<para>
Allows <application>MPD</application> on Linux to play audio
directly from a soundcard using the scheme
<filename>alsa://</filename>. Audio is formatted as 44.1 kHz
16-bit stereo (CD format). Examples:
</para>
<para>
<filename>mpc add alsa://</filename> plays audio from device hw:0,0
</para>
<para>
<filename>mpc add alsa://hw:1,0</filename> plays audio from device
hw:1,0
</para>
</section>
<section>
<title><varname>cdio_paranoia</varname></title>
<para>
Plays audio CDs. The URI has the form:
"<filename>cdda://[DEVICE][/TRACK]</filename>". The
simplest form <filename>cdda://</filename> plays the whole
disc in the default drive.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>default_byte_order</varname>
<parameter>little_endian|big_endian</parameter>
</entry>
<entry>
If the CD drive does not specify a byte order,
<application>MPD</application> assumes it is the
CPU's native byte order. This setting allows
overriding this.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section>
<title><varname>curl</varname></title>
<para>
Opens remote files or streams over HTTP.
</para>
<para>
Note that unless overridden by the below settings (e.g. by
setting them to a blank value), general curl configuration
from environment variables such as
<varname>http_proxy</varname> or specified in
<filename>~/.curlrc</filename> will be in effect.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>proxy</varname>
</entry>
<entry>
Sets the address of the HTTP proxy server.
</entry>
</row>
<row>
<entry>
<varname>proxy_user</varname>,
<varname>proxy_password</varname>
</entry>
<entry>
Configures proxy authentication.
</entry>
</row>
<row>
<entry>
<varname>verify_peer</varname>
<parameter>yes|no</parameter>
</entry>
<entry>
Verify the peer's SSL certificate? <ulink
url="http://curl.haxx.se/libcurl/c/CURLOPT_SSL_VERIFYPEER.html">More
information</ulink>.
</entry>
</row>
<row>
<entry>
<varname>verify_host</varname>
<parameter>yes|no</parameter>
</entry>
<entry>
Verify the certificate's name against host? <ulink
url="http://curl.haxx.se/libcurl/c/CURLOPT_SSL_VERIFYHOST.html">More
information</ulink>.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section>
<title><varname>file</varname></title>
<para>
Opens local files.
</para>
</section>
<section>
<title><varname>mms</varname></title>
<para>
Plays streams with the MMS protocol.
</para>
</section>
<section>
<title><varname>nfs</varname></title>
<para>
Allows <application>MPD</application> to access files on
NFSv3 servers without actually mounting them (i.e. in
userspace, without help from the kernel's VFS layer). All
URIs with the <filename>nfs://</filename> scheme are used
according to <ulink
url="http://tools.ietf.org/html/rfc2224">RFC2224</ulink>.
Example:
</para>
<para>
<filename>mpc add nfs://servername/path/filename.ogg</filename>
</para>
<para>
Note that this usually requires enabling the "insecure" flag
in the server's <filename>/etc/exports</filename> file,
because <application>MPD</application> cannot bind to
so-called "privileged" ports. Don't fear: this will not
make your file server insecure; the flag was named in a time
long ago when privileged ports were thought to be meaningful
for security. By today's standards, NFSv3 is not secure at
all, and if you believe it is, you're already doomed.
</para>
</section>
<section>
<title><varname>smbclient</varname></title>
<para>
Allows <application>MPD</application> to access files on
SMB/CIFS servers (e.g. Samba or Microsoft Windows). All
URIs with the <filename>smb://</filename> scheme are used.
Example:
</para>
<para>
<filename>mpc add smb://servername/sharename/filename.ogg</filename>
</para>
</section>
</section>
<section id="decoder_plugins">
<title>Decoder plugins</title>
<section id="dsdiff_decoder">
<title><varname>dsdiff</varname></title>
<para>
Decodes DFF files containing DSDIFF data (e.g. SACD rips).
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>lsbitfirst</varname>
<parameter>yes|no</parameter>
</entry>
<entry>
Decode the least significant bit first. Default is
<parameter>no</parameter>.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section id="dsf_decoder">
<title><varname>dsf</varname></title>
<para>
Decodes DSF files containing DSDIFF data (e.g. SACD rips).
</para>
</section>
<section>
<title><varname>fluidsynth</varname></title>
<para>
MIDI decoder based on <ulink
url="http://www.fluidsynth.org/"><application>FluidSynth</application></ulink>.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>sample_rate</varname>
</entry>
<entry>
The sample rate that shall be synthesized by the
plugin. Defaults to 48000.
</entry>
</row>
<row>
<entry>
<varname>soundfont</varname>
</entry>
<entry>
The absolute path of the soundfont file. Defaults
to
<filename>/usr/share/sounds/sf2/FluidR3_GM.sf2</filename>.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section>
<title><varname>gme</varname></title>
<para>
Video game music file emulator based on <ulink
url="https://bitbucket.org/mpyne/game-music-emu/wiki/Home"><application>game-music-emu</application></ulink>.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>accuracy</varname>
<parameter>yes|no</parameter>
</entry>
<entry>
Enable more accurate sound emulation.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section>
<title><varname>mikmod</varname></title>
<para>
Module player based on <ulink
url="http://mikmod.sourceforge.net/"><application>MikMod</application></ulink>.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>loop</varname>
<parameter>yes|no</parameter>
</entry>
<entry>
Allow backward loops in modules. Default is
<parameter>no</parameter>.
</entry>
</row>
<row>
<entry>
<varname>sample_rate</varname>
</entry>
<entry>
Sets the sample rate generated by
<filename>libmikmod</filename>. Default is 44100.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section>
<title><varname>modplug</varname></title>
<para>
Module player based on <application>MODPlug</application>.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>loop_count</varname>
</entry>
<entry>
Number of times to loop the module if it uses backward loops.
Default is <parameter>0</parameter> which prevents looping.
<parameter>-1</parameter> loops forever.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section id="sidplay_decoder">
<title><varname>sidplay</varname></title>
<para>
C64 SID decoder based on
<application>libsidplay</application>.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>songlength_database</varname>
<parameter>PATH</parameter>
</entry>
<entry>
Location of your songlengths file, as distributed
with the HVSC. The <varname>sidplay</varname>
plugin checks this for matching MD5 fingerprints.
See <ulink
url="http://www.c64.org/HVSC/DOCUMENTS/Songlengths.faq">http://www.c64.org/HVSC/DOCUMENTS/Songlengths.faq</ulink>.
</entry>
</row>
<row>
<entry>
<varname>default_songlength</varname>
<parameter>SECONDS</parameter>
</entry>
<entry>
This is the default playing time in seconds for
songs not in the songlength database, or in case
you're not using a database. A value of 0 means
play indefinitely.
</entry>
</row>
<row>
<entry>
<varname>filter</varname>
<parameter>yes|no</parameter>
</entry>
<entry>
Turns the SID filter emulation on or off.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section>
<title><varname>wildmidi</varname></title>
<para>
MIDI decoder based on <ulink
url="http://www.mindwerks.net/projects/wildmidi/"><application>libwildmidi</application></ulink>.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>config_file</varname>
</entry>
<entry>
The absolute path of the timidity config file. Defaults
to
<filename>/etc/timidity/timidity.cfg</filename>.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
</section>
<section id="encoder_plugins">
<title>Encoder plugins</title>
<section>
<title><varname>flac</varname></title>
<para>
Encodes into <ulink
url="https://xiph.org/flac/">FLAC</ulink> (lossless).
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>compression</varname>
</entry>
<entry>
Sets the <filename>libFLAC</filename> compression
level. The levels range from 0 (fastest, least
compression) to 8 (slowest, most compression).
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section>
<title><varname>lame</varname></title>
<para>
Encodes into MP3 using the <ulink
url="http://lame.sourceforge.net/"><application>LAME</application></ulink>
library.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>quality</varname>
</entry>
<entry>
Sets the quality for VBR. 0 is the highest quality,
9 is the lowest quality. Cannot be used with
<varname>bitrate</varname>.
</entry>
</row>
<row>
<entry>
<varname>bitrate</varname>
</entry>
<entry>
Sets the bit rate in kilobit per second. Cannot be
used with <varname>quality</varname>.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section>
<title><varname>null</varname></title>
<para>
Does not encode anything, passes the input PCM data as-is.
</para>
</section>
<section>
<title><varname>shine</varname></title>
<para>
Encodes into MP3 using the <ulink
url="https://github.com/savonet/shine"><application>Shine</application></ulink>
library.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>bitrate</varname>
</entry>
<entry>
Sets the bit rate in kilobit per second.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section>
<title><varname>twolame</varname></title>
<para>
Encodes into MP2 using the <ulink
url="http://www.twolame.org/"><application>TwoLAME</application></ulink>
library.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>quality</varname>
</entry>
<entry>
Sets the quality for VBR. 0 is the highest quality,
9 is the lowest quality. Cannot be used with
<varname>bitrate</varname>.
</entry>
</row>
<row>
<entry>
<varname>bitrate</varname>
</entry>
<entry>
Sets the bit rate in kilobit per second. Cannot be
used with <varname>quality</varname>.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section id="vorbis_encoder">
<title><varname>vorbis</varname></title>
<para>
Encodes into <ulink url="http://www.vorbis.com/">Ogg
Vorbis</ulink>.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>quality</varname>
</entry>
<entry>
Sets the quality for VBR. -1 is the lowest quality,
10 is the highest quality. Cannot be used with
<varname>bitrate</varname>.
</entry>
</row>
<row>
<entry>
<varname>bitrate</varname>
</entry>
<entry>
Sets the bit rate in kilobit per second. Cannot be
used with <varname>quality</varname>.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section>
<title><varname>wave</varname></title>
<para>
Encodes into WAV (lossless).
</para>
</section>
</section>
<section id="resampler_plugins">
<title>Resampler plugins</title>
<para>
The resampler can be configured in a block named
<varname>resampler</varname>, for example:
</para>
<programlisting>resampler {
plugin "soxr"
quality "very high"
}</programlisting>
<para>
The following table lists the <varname>resampler</varname>
options valid for all plugins:
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>
Name
</entry>
<entry>
Description
</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>plugin</varname>
</entry>
<entry>
The name of the plugin.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
<section id="internal_resampler">
<title><varname>internal</varname></title>
<para>
A resampler built into <application>MPD</application>. Its
quality is very poor, but its CPU usage is low. This is the
fallback if <application>MPD</application> was compiled
without an external resampler.
</para>
</section>
<section id="libsamplerate_resampler">
<title><varname>libsamplerate</varname></title>
<para>
A resampler using <ulink
url="http://www.mega-nerd.com/SRC/"><application>libsamplerate</application></ulink>
a.k.a. Secret Rabbit Code (SRC).
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>
Name
</entry>
<entry>
Description
</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>type</varname>
</entry>
<entry>
The interpolator type. See below for a list of
known types.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
<para>
The following converter types are provided by
<application>libsamplerate</application>:
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>
Type
</entry>
<entry>
Description
</entry>
</row>
</thead>
<tbody>
<row>
<entry>
"<parameter>Best Sinc Interpolator</parameter>" or
"<parameter>0</parameter>"
</entry>
<entry>
Band limited sinc interpolation, best quality, 97dB
SNR, 96% BW.
</entry>
</row>
<row>
<entry>
"<parameter>Medium Sinc Interpolator</parameter>" or
"<parameter>1</parameter>"
</entry>
<entry>
Band limited sinc interpolation, medium quality,
97dB SNR, 90% BW.
</entry>
</row>
<row>
<entry>
"<parameter>Fastest Sinc Interpolator</parameter>" or
"<parameter>2</parameter>"
</entry>
<entry>
Band limited sinc interpolation, fastest, 97dB SNR,
80% BW.
</entry>
</row>
<row>
<entry>
"<parameter>ZOH Sinc Interpolator</parameter>" or
"<parameter>3</parameter>"
</entry>
<entry>
Zero order hold interpolator, very fast, very poor
quality with audible distortions.
</entry>
</row>
<row>
<entry>
"<parameter>Linear Interpolator</parameter>" or
"<parameter>4</parameter>"
</entry>
<entry>
Linear interpolator, very fast, poor quality.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section id="soxr_resampler">
<title><varname>soxr</varname></title>
<para>
A resampler using <ulink
url="http://sourceforge.net/projects/soxr/"><application>libsoxr</application></ulink>,
the SoX Resampler library
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>
Name
</entry>
<entry>
Description
</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>quality</varname>
</entry>
<entry>
The <application>libsoxr</application> quality
setting. Valid values are:
<itemizedlist>
<listitem>
<para>
"<parameter>very high</parameter>"
</para>
</listitem>
<listitem>
<para>
"<parameter>high</parameter>" (the default)
</para>
</listitem>
<listitem>
<para>
"<parameter>medium</parameter>"
</para>
</listitem>
<listitem>
<para>
"<parameter>low</parameter>"
</para>
</listitem>
<listitem>
<para>
"<parameter>quick</parameter>"
</para>
</listitem>
</itemizedlist>
</entry>
</row>
<row>
<entry>
<varname>threads</varname>
</entry>
<entry>
The number of <application>libsoxr</application>
threads. "0" means "automatic". The default is "1"
which disables multi-threading.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
</section>
<section id="output_plugins">
<title>Output plugins</title>
<section id="alsa_output">
<title><varname>alsa</varname></title>
<para>
The <ulink
url="http://www.alsa-project.org/"><application>Advanced
Linux Sound Architecture</application>
(<application>ALSA</application>)</ulink> plugin uses
<filename>libasound</filename>. It is recommended if you
are using Linux.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>device</varname>
<parameter>NAME</parameter>
</entry>
<entry>
Sets the device which should be used. This can be
any valid ALSA device name. The default value is
"default", which makes
<filename>libasound</filename> choose a device. It
is recommended to use a "hw" or "plughw" device,
because otherwise, <filename>libasound</filename>
automatically enables "dmix", which has major
disadvantages (fixed sample rate, poor resampler,
...).
</entry>
</row>
<row>
<entry>
<varname>use_mmap</varname>
<parameter>yes|no</parameter>
</entry>
<entry>
If set to <parameter>yes</parameter>, then
<filename>libasound</filename> will try to use
memory mapped I/O.
</entry>
</row>
<row>
<entry>
<varname>buffer_time</varname>
<parameter>US</parameter>
</entry>
<entry>
Sets the device's buffer time in microseconds.
Don't change unless you know what you're doing.
</entry>
</row>
<row>
<entry>
<varname>period_time</varname>
<parameter>US</parameter>
</entry>
<entry>
Sets the device's period time in microseconds.
Don't change unless you really know what you're
doing.
</entry>
</row>
<row>
<entry>
<varname>auto_resample</varname>
<parameter>yes|no</parameter>
</entry>
<entry>
If set to <parameter>no</parameter>, then
<filename>libasound</filename> will not attempt to
resample, handing the responsibility over to
<application>MPD</application>. It is recommended
to let <application>MPD</application> resample (with
<application>libsamplerate</application>), because
ALSA is quite poor at doing so.
</entry>
</row>
<row>
<entry>
<varname>auto_channels</varname>
<parameter>yes|no</parameter>
</entry>
<entry>
If set to <parameter>no</parameter>, then
<filename>libasound</filename> will not attempt to
convert between different channel numbers.
</entry>
</row>
<row>
<entry>
<varname>auto_format</varname>
<parameter>yes|no</parameter>
</entry>
<entry>
If set to <parameter>no</parameter>, then
<filename>libasound</filename> will not attempt to
convert between different sample formats (16 bit, 24
bit, floating point, ...).
</entry>
</row>
<row>
<entry>
<varname>dop</varname>
<parameter>yes|no</parameter>
</entry>
<entry>
If set to <parameter>yes</parameter>, then DSD over
PCM according to the <ulink
url="http://dsd-guide.com/dop-open-standard">DoP
standard</ulink> is enabled. This wraps DSD
samples in fake 24 bit PCM, and is understood by
some DSD capable products, but may be harmful to
other hardware. Therefore, the default is
<parameter>no</parameter> and you can enable the
option at your own risk.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
<para>
The according hardware mixer plugin understands the
following settings:
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>mixer_device</varname>
<parameter>DEVICE</parameter>
</entry>
<entry>
<para>
Sets the ALSA mixer device name, defaulting to
<parameter>default</parameter> which lets ALSA
pick a value.
</para>
</entry>
</row>
<row>
<entry>
<varname>mixer_control</varname>
<parameter>NAME</parameter>
</entry>
<entry>
<para>
Choose a mixer control, defaulting to
<parameter>PCM</parameter>. Type <command>amixer
scontrols</command> to get a list of available
mixer controls.
</para>
</entry>
</row>
<row>
<entry>
<varname>mixer_index</varname>
<parameter>NUMBER</parameter>
</entry>
<entry>
Choose a mixer control index. This is necessary if
there is more than one control with the same name.
Defaults to <parameter>0</parameter> (the first
one).
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section>
<title><varname>ao</varname></title>
<para>
The <varname>ao</varname> plugin uses the portable <ulink
url="https://www.xiph.org/ao/"><filename>libao</filename></ulink>
library. Use only if there is no native plugin for your
operating system.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>driver</varname>
<parameter>D</parameter>
</entry>
<entry>
The <filename>libao</filename> driver to use for
audio output. Possible values depend on what libao
drivers are available. See <ulink
url="http://www.xiph.org/ao/doc/drivers.html">http://www.xiph.org/ao/doc/drivers.html</ulink>
for information on some commonly used drivers.
Typical values for Linux include "oss" and "alsa09".
The default is "default", which causes libao to
select an appropriate plugin.
</entry>
</row>
<row>
<entry>
<varname>options</varname>
<parameter>O</parameter>
</entry>
<entry>
Options to pass to the selected
<filename>libao</filename> driver.
</entry>
</row>
<row>
<entry>
<varname>write_size</varname>
<parameter>O</parameter>
</entry>
<entry>
This specifies how many bytes to write to the audio
device at once. This parameter is to work around a
bug in older versions of libao on sound cards with
very small buffers. The default is 1024.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section>
<title><varname>fifo</varname></title>
<para>
The <varname>fifo</varname> plugin writes raw PCM data to a
FIFO (First In, First Out) file. The data can be read by
another program.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>path</varname>
<parameter>P</parameter>
</entry>
<entry>
This specifies the path of the FIFO to write to.
Must be an absolute path. If the path does not
exist, it will be created when
<application>MPD</application> is started, and
removed when <application>MPD</application> is
stopped. The FIFO will be created with the same
user and group as <application>MPD</application> is
running as. Default permissions can be modified by
using the builtin shell command
<filename>umask</filename>. If a FIFO already
exists at the specified path it will be reused, and
will not be removed when
<application>MPD</application> is stopped. You can
use the "mkfifo" command to create this, and then
you may modify the permissions to your liking.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section>
<title><varname>jack</varname></title>
<para>
The <varname>jack</varname> plugin connects to a <ulink
url="http://jackaudio.org/"><application>JACK</application></ulink>
server.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>client_name</varname>
<parameter>NAME</parameter>
</entry>
<entry>
The name of the <application>JACK</application>
client. Defaults to "Music Player Daemon".
</entry>
</row>
<row>
<entry>
<varname>server_name</varname>
<parameter>NAME</parameter>
</entry>
<entry>
Optional name of the <application>JACK</application>
server.
</entry>
</row>
<row>
<entry>
<varname>autostart</varname>
<parameter>yes|no</parameter>
</entry>
<entry>
If set to <parameter>yes</parameter>, then
<filename>libjack</filename> will automatically
launch the <application>JACK</application> daemon.
Disabled by default.
</entry>
</row>
<row>
<entry>
<varname>source_ports</varname>
<parameter>A,B</parameter>
</entry>
<entry>
The names of the <application>JACK</application>
source ports to be created. By default, the ports
"left" and "right" are created. To use more ports,
you have to tweak this option.
</entry>
</row>
<row>
<entry>
<varname>destination_ports</varname>
<parameter>A,B</parameter>
</entry>
<entry>
The names of the <application>JACK</application>
destination ports to connect to.
</entry>
</row>
<row>
<entry>
<varname>ringbuffer_size</varname>
<parameter>NBYTES</parameter>
</entry>
<entry>
Sets the size of the ring buffer for each channel.
Do not configure this value unless you know what
you're doing.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section id="httpd_output">
<title><varname>httpd</varname></title>
<para>
The <varname>httpd</varname> plugin creates a HTTP server,
similar to <ulink
url="http://www.shoutcast.com/"><application>ShoutCast</application></ulink>
/ <ulink
url="http://icecast.org/"><application>IceCast</application></ulink>.
HTTP streaming clients like
<application>mplayer</application> can connect to it.
</para>
<para>
It is highly recommended to configure a fixed
<varname>format</varname>, because a stream cannot switch
its audio format on-the-fly when the song changes.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>port</varname>
<parameter>P</parameter>
</entry>
<entry>
Binds the HTTP server to the specified port.
</entry>
</row>
<row>
<entry>
<varname>bind_to_address</varname>
<parameter>ADDR</parameter>
</entry>
<entry>
Binds the HTTP server to the specified address (IPv4 or
IPv6). Multiple addresses in parallel are not supported.
</entry>
</row>
<row>
<entry>
<varname>encoder</varname>
<parameter>NAME</parameter>
</entry>
<entry>
Chooses an encoder plugin. A list of encoder
plugins can be found in the <link
linkend="encoder_plugins">encoder plugin
reference</link>.
</entry>
</row>
<row>
<entry>
<varname>max_clients</varname>
<parameter>MC</parameter>
</entry>
<entry>
Sets a limit, number of concurrent clients. When set
to 0 no limit will apply.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section>
<title><varname>null</varname></title>
<para>
The <varname>null</varname> plugin does nothing. It
discards everything sent to it.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>sync</varname>
<parameter>yes|no</parameter>
</entry>
<entry>
If set to <parameter>no</parameter>, then the timer
is disabled - the device will accept PCM chunks at
arbitrary rate (useful for benchmarking). The
default behaviour is to play in real time.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section id="oss_output">
<title><varname>oss</varname></title>
<para>
The "Open Sound System" plugin is supported on most Unix
platforms.
</para>
<para>
On Linux, <application>OSS</application> has been superseded
by <application>ALSA</application>. Use the <link
linkend="alsa_output"><application>ALSA</application> output
plugin</link> instead of this one on Linux.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>device</varname>
<parameter>PATH</parameter>
</entry>
<entry>
Sets the path of the PCM device. If not specified,
then <application>MPD</application> will attempt to
open <filename>/dev/sound/dsp</filename> and
<filename>/dev/dsp</filename>.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
<para>
The according hardware mixer plugin understands the
following settings:
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>mixer_device</varname>
<parameter>DEVICE</parameter>
</entry>
<entry>
<para>
Sets the OSS mixer device path, defaulting to
<filename>/dev/mixer</filename>.
</para>
</entry>
</row>
<row>
<entry>
<varname>mixer_control</varname>
<parameter>NAME</parameter>
</entry>
<entry>
<para>
Choose a mixer control, defaulting to
<parameter>PCM</parameter>.
</para>
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section>
<title><varname>openal</varname></title>
<para>
The "OpenAL" plugin uses <ulink
url="http://kcat.strangesoft.net/openal.html"><filename>libopenal</filename></ulink>.
It is supported on many platforms. Use only if there is no
native plugin for your operating system.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>device</varname>
<parameter>NAME</parameter>
</entry>
<entry>
Sets the device which should be used. This can be
any valid OpenAL device name. If not specified, then
<filename>libopenal</filename> will choose a default device.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section>
<title><varname>osx</varname></title>
<para>
The "Mac OS X" plugin uses Apple's CoreAudio API.
</para>
</section>
<section>
<title><varname>pipe</varname></title>
<para>
The <varname>pipe</varname> plugin starts a program and
writes raw PCM data into its standard input.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>command</varname>
<parameter>CMD</parameter>
</entry>
<entry>
This command is invoked with the shell.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section id="pulse_output">
<title><varname>pulse</varname></title>
<para>
The <varname>pulse</varname> plugin connects to a <ulink
url="http://www.freedesktop.org/wiki/Software/PulseAudio/"><application>PulseAudio</application></ulink>
server.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>server</varname>
<parameter>HOSTNAME</parameter>
</entry>
<entry>
Sets the host name of the
<application>PulseAudio</application> server. By
default, <application>MPD</application> connects to
the local <application>PulseAudio</application>
server.
</entry>
</row>
<row>
<entry>
<varname>sink</varname>
<parameter>NAME</parameter>
</entry>
<entry>
Specifies the name of the
<application>PulseAudio</application> sink
<application>MPD</application> should play on.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section>
<title><varname>roar</varname></title>
<para>
The <varname>roar</varname> plugin connects to a <ulink
url="http://roaraudio.keep-cool.org/">RoarAudio</ulink>
server.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>server</varname>
<parameter>HOSTNAME</parameter>
</entry>
<entry>
The host name of the RoarAudio server. If not
specified, then <application>MPD</application> will
connect to the default locations.
</entry>
</row>
<row>
<entry>
<varname>role</varname>
<parameter>ROLE</parameter>
</entry>
<entry>
The "role" that <application>MPD</application>
registers itself as in the RoarAudio server. The
default is "music".
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section>
<title><varname>recorder</varname></title>
<para>
The <varname>recorder</varname> plugin writes the audio
played by <application>MPD</application> to a file. This
may be useful for recording radio streams.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>path</varname>
<parameter>P</parameter>
</entry>
<entry>
Write to this file.
</entry>
</row>
<row>
<entry>
<varname>format_path</varname>
<parameter>P</parameter>
</entry>
<entry>
<para>
An alternative to <varname>path</varname> which
provides a format string referring to tag values.
The special tag <varname>iso8601</varname> emits
the current date and time in <ulink
url="https://en.wikipedia.org/wiki/ISO_8601">ISO8601</ulink>
format (UTC).
Every time a new song starts or a new tag gets
received from a radio station, a new file is
opened. If the format does not render a file
name, nothing is recorded.
</para>
<para>
A tag name enclosed in percent signs ('%') is
replaced with the tag value. Example:
<parameter>~/.mpd/recorder/%artist% -
%title%.ogg</parameter>
</para>
<para>
Square brackets can be used to group a substring.
If none of the tags referred in the group can be
found, the whole group is omitted. Example:
<parameter>[~/.mpd/recorder/[%artist% -
]%title%.ogg]</parameter> (this omits the dash
when no artist tag exists; if title also doesn't
exist, no file is written)
</para>
<para>
The operators "|" (logical "or") and "&"
(logical "and") can be used to select portions of
the format string depending on the existing tag
values. Example:
<parameter>~/.mpd/recorder/[%title|%name%].ogg</parameter>
(use the "name" tag if no title exists)
</para>
</entry>
</row>
<row>
<entry>
<varname>encoder</varname>
<parameter>NAME</parameter>
</entry>
<entry>
Chooses an encoder plugin. A list of encoder
plugins can be found in the <link
linkend="encoder_plugins">encoder plugin
reference</link>.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section id="shout_output">
<title><varname>shout</varname></title>
<para>
The <varname>shout</varname> plugin connects to a <ulink
url="http://www.shoutcast.com/"><application>ShoutCast</application></ulink>
or <ulink
url="http://icecast.org/"><application>IceCast</application></ulink>
server. It forwards tags to this server.
</para>
<para>
You must set a <varname>format</varname>.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>host</varname>
<parameter>HOSTNAME</parameter>
</entry>
<entry>
Sets the host name of the <ulink
url="http://www.shoutcast.com/"><application>ShoutCast</application></ulink>
/ <ulink
url="http://icecast.org/"><application>IceCast</application></ulink>
server.
</entry>
</row>
<row>
<entry>
<varname>port</varname>
<parameter>PORTNUMBER</parameter>
</entry>
<entry>
Connect to this port number on the specified host.
</entry>
</row>
<row>
<entry>
<varname>timeout</varname>
<parameter>SECONDS</parameter>
</entry>
<entry>
Set the timeout for the shout connection in seconds.
Defaults to 2 seconds.
</entry>
</row>
<row>
<entry>
<varname>protocol</varname>
<parameter>icecast2|icecast1|shoutcast</parameter>
</entry>
<entry>
Specifies the protocol that wil be used to connect
to the server. The default is
"<parameter>icecast2</parameter>".
</entry>
</row>
<row>
<entry>
<varname>mount</varname>
<parameter>URI</parameter>
</entry>
<entry>
Mounts the <application>MPD</application> stream in
the specified URI.
</entry>
</row>
<row>
<entry>
<varname>user</varname>
<parameter>USERNAME</parameter>
</entry>
<entry>
Sets the user name for submitting the stream to the
server. Default is "source".
</entry>
</row>
<row>
<entry>
<varname>password</varname>
<parameter>PWD</parameter>
</entry>
<entry>
Sets the password for submitting the stream to the
server.
</entry>
</row>
<row>
<entry>
<varname>name</varname>
<parameter>NAME</parameter>
</entry>
<entry>
Sets the name of the stream.
</entry>
</row>
<row>
<entry>
<varname>genre</varname>
<parameter>GENRE</parameter>
</entry>
<entry>
Sets the genre of the stream (optional).
</entry>
</row>
<row>
<entry>
<varname>description</varname>
<parameter>DESCRIPTION</parameter>
</entry>
<entry>
Sets a short description of the stream (optional).
</entry>
</row>
<row>
<entry>
<varname>url</varname>
<parameter>URL</parameter>
</entry>
<entry>
Sets a URL associated with the stream (optional).
</entry>
</row>
<row>
<entry>
<varname>public</varname>
<parameter>yes|no</parameter>
</entry>
<entry>
Specifies whether the stream should be "public".
Default is <parameter>no</parameter>.
</entry>
</row>
<row>
<entry>
<varname>encoder</varname>
<parameter>PLUGIN</parameter>
</entry>
<entry>
Chooses an encoder plugin. Default is <link
linkend="vorbis_encoder"><parameter>vorbis</parameter></link>.
A list of encoder plugins can be found in the <link
linkend="encoder_plugins">encoder plugin
reference</link>.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
<section>
<title><varname>solaris</varname></title>
<para>
The "Solaris" plugin runs only on SUN Solaris, and plays via
<filename>/dev/audio</filename>.
</para>
<informaltable>
<tgroup cols="2">
<thead>
<row>
<entry>Setting</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry>
<varname>device</varname>
<parameter>PATH</parameter>
</entry>
<entry>
Sets the path of the audio device, defaults to
<filename>/dev/audio</filename>.
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</section>
</section>
<section id="playlist_plugins">
<title>Playlist plugins</title>
<section>
<title><varname>embcue</varname></title>
<para>
Reads CUE sheets from the "CUESHEET" tag of song files.
</para>
</section>
<section>
<title><varname>m3u</varname></title>
<para>
Reads <filename>.m3u</filename> playlist files.
</para>
</section>
<section>
<title><varname>extm3u</varname></title>
<para>
Reads extended <filename>.m3u</filename> playlist files.
</para>
</section>
<section>
<title><varname>pls</varname></title>
<para>
Reads <filename>.pls</filename> playlist files.
</para>
</section>
<section>
<title><varname>xspf</varname></title>
<para>
Reads <ulink url="http://www.xspf.org/">XSPF</ulink>
playlist files.
</para>
</section>
</section>
</chapter>
</book>