The next thing will be electron microscope scans of of the instruments the artists used.
I'm joking. I really want well digitized vinyl
PVC grains are typically 100s of um in size, but the monomers themselves are ~10nm. Note that with 120dB claimed dynamic range this would be 37nm of deflection (at 45deg or 25nm lateral) for full resolution. That is close (3x off) to molecular Atomic Force Microscopy.
Of course what is encoded is actually force/velocity which in turn is inductively coupled out, but it is no surprise that simple optical scanning gives you mediocre results. Even 10x lower dynamic range requires 250nm optical resolution. It's doable, but not easy.
Here's a good place to look for vinyl LP info: