In a pair of recent papers, I discuss the damping of a piston enclosing an ideal gas in a cylinder in the absence of any explicit friction, viscosity, or turbulence.
In the first paper, the damping is due to thermal transfer in and out of the cylinder:
Thus, unlike a mass on a Hookean spring, for which one can idealize away all dissipation so that the oscillations are undamped, it is never possible to do that for a mass bouncing on an ideal gas.
I claim no novelty to the ideas, as all the concepts already exist in the literature. My contribution has merely been to collate them and present them in a simple form. -Carl