diff options
| author | David Flater <dave@flaterco.com> | 2012-02-01 03:45:20 +0100 |
|---|---|---|
| committer | David Flater <dave@flaterco.com> | 2012-02-01 03:45:20 +0100 |
| commit | 15029f2439781c5da7ffcdf18e39ea8ea92b17e9 (patch) | |
| tree | a14600714e4df64b2d60e33d7f8fb2bef700367f /src | |
| parent | 01062a87c0acfe71f6f5be14dbeb9e0ab66ca7f0 (diff) | |
| download | xine-lib-15029f2439781c5da7ffcdf18e39ea8ea92b17e9.tar.gz xine-lib-15029f2439781c5da7ffcdf18e39ea8ea92b17e9.tar.bz2 | |
Remove ALSA resampler bypass
In 2005, xine-lib 1.1.1 added the following change:
xine-lib (1.1.1) 2005-11-15
* Improve sound quality when using alsa 1.0.9 or above.
When playing a 44.1khz stream on a 48khz only capable sound card.
It bypasses alsa-lib resampler and uses xine's
That made sense when Alsa's resampling was the worst thing ever. But now,
if one installs alsa-plugins-1.0.25 and puts this in ~/.asoundrc:
defaults.pcm.rate_converter "samplerate_best"
Alsa then does markedly better resampling (using libsamplerate) than
xine-lib (using its own algorithm). The bypass added in 2005 is therefore
preventing the use of a higher-quality algorithm and blocking the ability to
tune that algorithm in ~/.asoundrc, and no longer seems to have any
redeeming value.
--HG--
extra : transplant_source : %AE%96zte%B0%ED%B6C%BC%89%02%1F%05%21%BD%ED%D6%8C6
Diffstat (limited to 'src')
| -rw-r--r-- | src/audio_out/audio_alsa_out.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/src/audio_out/audio_alsa_out.c b/src/audio_out/audio_alsa_out.c index af8b4c351..b8cc1a8ca 100644 --- a/src/audio_out/audio_alsa_out.c +++ b/src/audio_out/audio_alsa_out.c @@ -458,7 +458,7 @@ static int ao_alsa_open(ao_driver_t *this_gen, uint32_t bits, uint32_t rate, int this->num_channels, err, snd_strerror(err)); goto close; } -#if SND_LIB_VERSION >= 0x010009 +#if 0 /* Restrict a configuration space to contain only real hardware rates */ err = snd_pcm_hw_params_set_rate_resample(this->audio_fd, params, 0); #endif |