“…After repeated discussions and admonitions by Church [6,14], Church and Takacs [15][16], Stover [17], Germer and Asmail [18], Dittman [19][20] and others; most of the optical surface metrology community is aware that when we associate surface roughness with scattered light, we must specify the spatial frequency band-limits of the effective roughness that is relevant to the particular scattering application. In other words, we must replace the total or intrinsic rms roughness, , with the relevant (or effective) bandlimited rms roughness, rel , in Eqs.…”