Astronomy
My 10 inch Mirror with a 1/48 P-V Wavefront Error @ 550 nm The university was planning to contact us to ask me for permission to send my mirror out to a commercial optics lab with better test facilities.
Further testing at the second lab revealed that my mirror's P-V Wavefront Error was also better than their "fine referrence" which was known to have a Peak to Valley Wavefront Error of 1/24 @ 550 nanometers.  Thus my mirror's Peak to Valley Wavefront Error was measured as less than 1/24 @ 550 nanometers, which is 1/48 Wavefront Error at the mirror's surface at 550 nanometers.  The minimum P-V Wavefront Error attainable by theory is 1/50 because the Focault Test cannot reveal surface errors smaller than 1/100 of a wavefront at 550 nanometers.  This puts my mirror's quality within the same order of magnitude as theoretical perfection, - some 6 orders of magnitude better than Difraction Limited Optics.
Using the popular Root Mean Square statistical equivalent, - this is a Wavefront Error of 1 / 78 RMS @ 633 nanometers or 1 / 156 RMS @ 633 nm at the mirror's surface.  Such a fine surface needed protection from "Sagging." In those days amature telescopes were limited in size due to this phenomena. There was a good solution for small apature long focal length telescope mirrors in making the thickness of the mirror much greater than the mirror's "Sagita".  For wide aperture, Short Focal Length mirrors with larger Sagitas, - this requires such an increase in thickness that their shear mass becomes prohibitive.