I also don't understad the random criticism of pulse audio. is it bad, should I get rid of it>? If so, what replaces it?
Thanks
Alright, unrelated to JRiver but here are my thoughts.
I believe a lot of criticism on Pulseaudio come from the days where it wasn't very good or even stable. I think today, Pulse works fine and, for all intends and purposes, makes Linux work more like Windows with regards to audio and outputting sound from multiple programs at once.
By default, Pulse is configured for 16-bit audio and 48 kHz sampling rate using an efficient and fast resampler (good enough for normal use). This is safe, works on most machines/soundcards, doesn't use a lot of CPU cycles, but can be tweaked/tuned to your personal preferences. You could increase the resampler to a higher quality (using more CPU) or you can choose another resampler. I have it set to 32-bit float and the resampler uses SoX-vhq set to 96 kHz. When I use MC in shared audio mode, I have MC configured to resample to 96 kHz so Pulse will simply accept MC's input 'as-is' and will only remix if another application outputs something.
Using my reasonably good PC speakers or headphones, I cannot hear something wrong with audio quality, even in shared audio mode (MC using Pulse).
I haven't recently tested it (no longer see the need as Pulse works fine for me), but you could create a direct HW device in ALSA and output to the hardware directly with MC. Pulse (or any other application) won't be able to output anything this way, but if you're concerned about getting the best possible audio output then this is the way to go.
I think the above is valid for any recent Linux distro. I don't think one distro is better for playing audio than another. The only exception here is if you're into music production/editing, you might want to switch to an real time kernel (I think Ubuntu studio uses that?). I have no more info on that as I never used it and don't need it.