Re: [fluid-dev] alsa audio driver only accepting huge period size

[Marcus Weseloh]

Am Sa., 8. Mai 2021 um 18:04 Uhr schrieb la Fleur <lafleur@boum.org>:

> I really wonder how pulseaudio manages to have a very low latency with the 
> very same hardware. Any clue ?

No really, but I must admit I know very little about PulseAudio. I
assumed that it simply uses ALSA as a sink, so with the same audio
device, it should be affected by the same limitations.
Have you actually measured the latency difference between PulseAudio
and ALSA? And are PulseAudio and ALSA really using the same audio
device?

One thing that comes to mind is the multi-channel feature of the LPass
driver. It can use up to 8 channels, but period_bytes_min/max is fixed
and not dependent on the number of channels. So if PulseAudio is
configured for the full 8 channels, then maybe the buffer fills up 4
times as fast, leading to a quarter of the perceived latency compared
to stereo mode. But that is just a wild guess. Looking at the
PulseAudio config might clear things up.

Cheers,
Marcus