Pekka Buttler, March–June 2021
While this article is part of a JAPB series of articles on the optical flaws of lenses (you can find the index of the series here), diffraction is NOT an optical flaw or aberration, because there is NO WAY you can design a lens that would not be affected by diffraction.
Instead, I’m mentioning diffraction as part of this series,
a) because some seem to think that diffraction is an issue similar to, say, spherical aberration, and would expect JAPB to mention it in a series of articles on optical flaws and aberrations; and
b) to explain what diffraction is and why it is not an optical flaw or aberration.
What is diffraction (optics)?
Diffraction affects all waveforms, including (among others) ocean waves, sound waves and – importantly – light. Diffraction is very much related to quantum mechanics, but here diffraction (regarding optics) will be explained in laymen’s terms (sacrificing some precision)
In optics, diffraction is an effect caused by light encountering an edge. While (when unobstructed) light would like to travel in a straight line, when light encounters the proximity of a hard edge (such as an opaque dust mote or the edge of a diaphragm blade), light is dispersed in all directions. With two qualifications.
- The phrase proximity (previous paragraph) is important. As light is a wave, light needs space (do not think of light as an infinitesimally small particle) which allows that wave to swing to and fro. In practical terms, light that hits an opaque obstruction is obstructed, and light which passes far enough past the obstruction is unaffected, but light that passes sufficiently close to the obstruction’s edge is affected.
- While light is dispersed in all directions (including at right angles, around the corner) , the majority of light continues onward with no or only a minor deviation to its original course.
How does diffraction affect images?
In practice, this means that as a photographer stops down their lens (making the aperture smaller) an increasing share of the light passing through the lens will be affected by the aperture’s edge. Therefore, light will increasingly end up in places on the film/sensor where it should not be, leading to a loss of contrast and sharpness. While the stopping down of a lens will first increase sharpness (as spherical aberration, longitudinal chromatic aberrations abate), from some point onward, all the improvement stopping down produces will be more than compensated by the increased effect of diffraction. This is why most sharpness tests show resolution dropping significantly from roughly f/8 onward.
Why is diffraction not a lens (design) flaw?
In the simplest term, diffraction is not a flaw, because there is simply nothing anyone can do about it. Importantly, diffraction is not something which occurs only when a lens is stopped down past f/11. There are a number of reasons why some measure of diffraction always affects lenses:
- Even a wide-open lens’ light path has edges, and while the share of light affected by those edges is minor, it is not entirely inconsequential,
- Optical glass is never entirely perfect (it is very good, but never perfect), leading to that even in seemingly transparent optical glass there are points which induce diffraction,
- Lenses typically gather dust, and while that dust may be impossible to see in the final image, dust creates diffraction ripples,
- Finally, every boundary between layers with different refractive indexes causes some diffraction.
Admittedly, the compound effect of these four systemic diffraction-inducing causes are not as visible as that of the diffraction caused by an aperture closed down all the way. Even so, diffraction is a central reason to why there can never be a truly optically perfect lens. While one may encounter a phrase such as “beyond f/16, the lens is diffraction-limited”, all lenses are – always – diffraction limited.
But wait, you might say, haven’t you reviewed lenses, even compared them, and said that some start softening after f/5.6 while most start softening after f/8 and some only after f/11. Doesn’t that mean that some lenses handle diffraction better than others? In short, the answer is: No.
Firstly, with legacy lenses there is always the possibility that (what the lens says is) f/11 is not really f/11. Secondly, and more importantly, going from any aperture stop to the next smaller aperture stop (e.g. f/8 to f/11) the picture will both improve (less spherical aberration, less longitudinal CA’s, less astigmatism) and deteriorate (more diffraction). The obvious reason for some lenses continuing to improve from f/8 to f/11 is not that diffraction would not yet kick in, but rather that the lens – on balance – still improves more than the negative effect of diffraction deteriorates. If you look at the kinds of lenses reviewed that show an improvement in sharpness from f/8 to f/11, they are typically lenses that have been somewhat underwhelming at wider apertures, whereas none of the lenses that performed well at wider apertures (and where the indicated aperture stops seem to be accurate) ever manage to improve going from f/8 to f/11. In fact, if you look at sharpness tests done on modern high-performance (sharpness) lenses, you will notice them reaching their peak sharpness way before f/8.
P.S. Should you feel that you could handle something beyond layman’s terms, look at the Wikipedia articles for diffraction and ‘airy disk‘
P.P.S. The bulk of this article was earlier published on JAPB as a glossary description.