On 9/23/22 08:17, stan via users wrote:
On Thu, 22 Sep 2022 15:35:05 -0700 Samuel Sieb samuel@sieb.net wrote:
On 9/22/22 12:35, stan via users wrote:
You can set up and try a new user, but be sure that you have logged out your current user before you try with the new user. The first user to log in gains total control over pipewire / pulseaudio unless it is being run as a server, which is not the default.
Actually, switching users also switches control of the audio output. I haven't tried leaving something playing in the first user to see what happens, but when switching to a second user with the first one still logged in, the second user has sound.
Well, your direct experience trumps my doc reading. But, how does that work? Unless the device has parallel paths, how can two people use the same device at the same time? And what about the speakers? I remember that it was a security issue when it was first implemented, to ensure that sound was not shared. At the time, if a sound played, everyone heard it. Perhaps, pulesaudio / pipewire have become more sophisticated and now tag sound and segregate it by user. Or maybe sound cards can process interleaved streams if they are tagged properly. Or, could it be tied to the screen visibility? Not questioning your experience, but wondering how it is implemented.
I believe it's something with the systemd session management. The display control gets transferred (no root permissions any more), so I don't think audio would be any harder.