C This is something I've never really understood. I know from C experience that some dust or scratches on the primary C don't degrade the image significantly. However, a quarter C wave problem on the primary will significantly degrade an C image. Since dust particles are definitely larger than a C wavelength of light why don't they cause significant errors C in the image? My guess is that a figuring error in the C primary is a systematic error that will degrade the entire C image while dust and scratches are random errors that will C tend to average out on the focal plane (but possibly reduce C contrast?) Can anyone help on this?
Now we finally boil down to RMS versus P-V ! Of course that no mirror can FULLY claim to be "1/8 wave peak to valey" - it is the question HOW MUCH energy gets scattered outside the diffraction disc. A spec of dust or scratch on the mirror will depart HUNDREDS (if not thousands) of waves, but it is insignificant in comparison with a majority of energy (probably few billion times greater in magnitude) which rest of the surface delivers. It will just increase background noise by truly undetectable levels.
Now while on subject of scratches, I am sure that all of us will be in for a big surprise if some of our pro colleagues give us some numbers on what is acceptable as "astronomical quality" optics, in a numerical form. (I was shocked to see what Maksutov considered as "insignificant defect" - he allowed that 150mm mirror could have scratches and pits up to 150 square mm (!!!), as long as they are randomly distributed)
Bratislav