He told me that these objectives
are polished to 1/20th wave after the 3rd polish.

Well

Given that a REFRACTING surface effects the wavefront roughly 1/4 as strongly
as a mirror surface.....this means that any given refracting surface that hase
1/20 th wavefront error is roughly that of a 1/5 wavefront error mirror in
terms of absolute surface accuracy requirements.....and given moderate,
spherical surfaces.....the claims may be slightly exaggerated but the are not
neccessarily "astronomical"

Blll