Audacious

There is nothing in the source code of Audacious' pipewire output plugin that confirms your assumption. Yet Pipewire itself may work differently than Pulseaudio or direct ALSA output due to handling per app volume in different ways. More trouble-shooting will be needed.

I suspect this is a valid defect.

ALSA volume levels are typically percentages (0-100%) that translate linearly to decibel levels. It's been that way as long as I can remember, though it beats me if that's documented anywhere. On my system, each 5% up/down translates to a 3 dB change. (Nowadays, alsamixer uses some different, incomprehensible scheme, but that's just at the UI level; the underlying API calls haven't changed.)

On the other hand, the PipeWire documentation for SPA_PROP_channelVolumes says "0.0 is silence, 1.0 is without attenuation". It's exceptionally vague wording, but that suggests to me that the 0-1 range is meant to translate linearly to some reference voltage range. If that's the case, we should probably do some math in the get_volume/set_volume() methods so that the volume level in Audacious still translates to a decibel level.

I'm not interested in PipeWire personally, but if someone else wants to take a stab at a patch, the general rule is: a +20 dB increase means the voltage multiplies by 10x. You can also find a sample calculation in the audio_amplify() function (libaudcore/audio.cc).

This is what MPV/audio/out/ao_pulse.c ( https://github.com/mpv-player/mpv/blob/master/audio/out/ao_pulse.c ) uses for PulseAudio:

#define VOL_PA2MP(v) ((v) * 100.0 / PA_VOLUME_NORM)
#define VOL_MP2PA(v) lrint((v) * PA_VOLUME_NORM / 100)

Not sure if it's relevant, sorry if it's not.

Though for audio/out/ao_pipewire.c ( https://github.com/mpv-player/mpv/blob/master/audio/out/ao_pulse.c ) it's quite different.

I have asked the PipeWire developer(s), their answer is that PipeWire uses a linear scale for the volume.

https://gitlab.freedesktop.org/pipewire/pipewire/-/issues/3623

@John: Do you have further questions for them? I'm not sure how to proceed now.

I think we have all the information we need. We just need to implement the decibel to linear translation now.

Please see/check my pull request:
https://github.com/audacious-media-player/audacious-plugins/pull/146