[pulseaudio-discuss] [pulseaudio-commits] src/pulsecore

[Arun Raghavan]

On Mon, 2011-09-05 at 22:32 +0300, Tanu Kaskinen wrote:
> On Fri, 2011-08-26 at 09:48 +0530, Arun Raghavan wrote:
> > On Fri, 2011-08-26 at 00:43 +0300, Tanu Kaskinen wrote:
> > > I think the volume should be set to PA_MIN(s->base_volume,
> > > PA_VOLUME_NORM). The base volume can be higher than NORM in some cases,
> > > which will cause software amplification if the code is left as it is
> > > now.
> > 
> > Ah, thanks -- I'll fix this up. Is base_volume > 0 dB a case that makes
> > sense, or is it just something that we might come across with broken dB
> > values?
> 
> By the way, what made you notice this problem in the first place? It
> sounds strange that base_volume would somehow be a better volume for a
> sink than PA_VOLUME_NORM when using passthrough. That's because if
> base_volume does not equal PA_VOLUME_NORM, that means that hardware
> volume with dB information is being used. I'd be surprised if on some
> hardware there was an alsa volume element that would actually do
> something useful in the passthrough mode.

This is all in-theory, to be honest. The hardware I have has base_volume
as 0dB, so doesn't really make a difference here.

-- Arun