This reminds me a lot of videotape or film. In video there are 30 frames per second. If you try to improve this image (using the original 30 fps) it can only be assumed that you MUST be adding (THAT which was not already THERE to begin with).
One could duplicate each frame an additional 1, 2, 3 or however many times, and play them back at a faster rate. This might impart a "richer" more real looking image, but it's still only made up of what was there to begin with.
One might compare two consecutive images and interpolate what an image between these might look like and insert this creation. THIS still doesn’t give you higher resolution.
To my way of thinking up sampling isn't real sound improvement, its sound smoothing. The very nature of interpolative up sampling seems to fly in the face of what many folks on this site claim to desire: The BEST… The most accurate reproduction of the original sound - TAS.
I suspect that THIS version of “up sampling” being interpreted as “hi fidelity” is why so many folks continue to tout their vinyl.
It only stands to reason that the only way to gain higher resolution is to sample more frequently (capture more information) and play this information back at the appropriate corresponding rate(s).
So... 24/96... 24/192... HELL YES!