diff options
author | Max Kellermann <max@musicpd.org> | 2020-01-17 16:37:09 +0100 |
---|---|---|
committer | Max Kellermann <max@musicpd.org> | 2020-01-17 16:37:09 +0100 |
commit | c84bae739a29ec27790c5534b7fc777d6f40a02f (patch) | |
tree | c31b837eb57298d807cf3f2803017f354ad78ad2 /doc | |
parent | 925b5954c362c8283f321724823da83e65485a3f (diff) |
pcm/Convert: add option to enable the integer-only dsd2pcm implementation
Diffstat (limited to 'doc')
-rw-r--r-- | doc/user.rst | 20 |
1 files changed, 20 insertions, 0 deletions
diff --git a/doc/user.rst b/doc/user.rst index d75975d21..73559f122 100644 --- a/doc/user.rst +++ b/doc/user.rst @@ -568,6 +568,26 @@ Sometimes, music needs to be resampled before it can be played; for example, CDs Check the :ref:`resampler_plugins` reference for a list of resamplers and how to configure them. +DSD +^^^ + +There are three ways to play back DSD (Direct Stream Digital): + +- natively if your output supports it (ALSA only and requires a sound + card with native DSD support) +- `DSD over PCM <http://dsd-guide.com/dop-open-standard>`_: wrapped + inside fake 24 bit PCM samples (ALSA and OSX only and requires an + external DAC with DoP support) +- for everything else, MPD automatically converts DSD to PCM + +For the last option, the setting ``dsd2pcm_converter`` selects a +conversion method. The default is ``float`` which converts to +floating point samples. For CPUs with a weak floating point unit +(e.g. some ARMs like the one in the Raspberry Pi), it may be faster to +use ``integer``, which is an integer-only implementation of the +conversion algorithm. + + Client Connections ------------------ |