Ah yeah, to be fair it's not a _huge_ correction, but in what I do (single-mode confocal microscopy) it definitely becomes an issue around NA=0.3 when trying to

Ah yeah, to be fair it's not a _huge_ correction, but in what I do (single-mode confocal microscopy) it definitely becomes an issue around NA=0.3 when trying to characterize spot sizes and collection efficiency functions. Namely, the overlap integrals can change significantly if your source is also polarized. Also, I'm probably more sensitive to these things than most since I'm at the single-photon level usually.