"If it sounds good, it is good."
I'll definitely defend anyone's right to complicate things as much as they care to. But it still amuses me, I have to admit, when people complicate something that's essentially so simple.
If you want to get reliably good optical quality out of your lens, here's what you do: determine the best aperture for your camera and lens. Use that and the aperture on either side of it freely. (Those are known as "the middle apertures.") Depart from that when you have to, need to, or want to.
No further complication necessary. Unless, as I say, complication is fun for you.
For full frame format (a.k.a. 35mm or 24x36mm), for me, the best aperture is ƒ/8. Ninety percent of the time, for 90% of [single-focal-length or "prime"] lenses, ƒ/8 is the aperture that gives me the all-around best-looking pictures, optically speaking. As usual, I figured this out empirically, i.e., by taking pictures and looking at them carefully, not by feeling the elephant's leg or pulling his tail ("testing"). You look at the whole picture, not just one little stupid aspect of it*.
So, for full frame, I use ƒ/5.6, ƒ/8, or ƒ/11 when I can.
...Departing from those apertures when I have to, need to, or want to. (I have to add that again, for the very few people who will forget I just said it and hence will want to argue with me as if I never did say it.)
The middle aperture changes with format. For 6x7 (cm), my middle aperture is ƒ/11; for 4x5 (inches), it's ƒ/22. For APS-C, 4/3, and Micro 4/3, it's ƒ/5.6. Think I'm wrong? No problemo; take your own pick. I'm not standing next to you with a ruler threatening to rap your knuckles if you depart from the gospel of Mike.
So then, what is the middle aperture for fingernail sensors? Like the ones in, say, the Canon G11 or the Pentax Q? That I don't know.
(Thanks to Scott Baker)
* Someone, somewhere, is bound to howl that I have no idea what I'm talking about, because you "have to look at" this or that property of IQ or you "have to measure in order to know." Actually, I know exactly what I'm talking about: for every property that that person fixates on, he's missing others that are of equal or greater importance to pictures, whether it be microcontrast, shadow detail, edge effects, tonal gradation, flare suppression, color differentiation, etc., etc., etc. A lens image is always a balance of many properties. Always. Look at images; look at the whole image; look, in the image, at the objects you're trying to image. In photography you don't get extra credit for resolving infinitesimally greater detail under ideal conditions; you get points if your pictures look better. (And sometimes not even then. Heh heh.) So look at the pictures and use your holistic all-encompassing sense of aesthetic discrimination—you know what a picture is supposed to look like, because you look at a lot of pictures; if your picture looks right to you, it is right. Don't let anyone talk you out of that.
UPDATE: Several commenters so far have agreed about ƒ/4 being the general sweet spot for fingernail-sized sensors.
Note: Links in this post may be to our affiliates; sales through affiliate links may benefit this site. More...
Original contents copyright 2011 by Michael C. Johnston and/or the bylined author. All Rights Reserved.
Featured Comment by Jim Richardson: "Right on! 'Nuff said, as they say.
"And because I have some sort of perverse mean streak that comes out in the presence of camera brand orthodoxy, I have, over the years, engaged in a bit lens bashing. This usually occurs when someone insists (with reverent tones) that this or that particular lens is absolutely perfect, the greatest thing ever. Mostly they think this because it is a Nikon or Canon lens, sometimes because it is a Leica lens (more rare but also more virulent), and sometimes a Zeiss (rarer yet but the teutonic nasal overtones.)
"The gist is that the lens, in their view, is without faults. If the lens in question is a wide angle or a zoom of almost any focal length I know I've got them.
"So I then take them out do a very simple test (which works best in the autumn and winter, for reasons that will become apparent.) I point the camera up at a tree with bare branches agains the sky and take a picture. Then I reframe so that the same area of branches that was at the center of the frame is at the far corner of the frame. Then we go look at the pictures.
"Put side by side it is easy to see that the corner image sucks. The branches that had been nice and crisp in the center are now fuzzy with aberrations, fringed with color, weirdly stretched, or just otherwise crummy. Suddenly they can see that their prized perfect lens has feet of clay. (To be fair really good macros and long, expensive telephotos fare pretty well in this test.)
"Suddenly their world crumbles around them. And we can then start an honest conversation about working with real world lenses to get the best quality out of them that they have to offer.
"It's cruel, I know, but there it is. Sorry."
Featured Comment by John Camp: "I think I touched off Ctein's lecture on diffraction with a question I posed in an earlier blog entry. His answer was perfect: before, I had these vague worries about diffraction. Now I don't. Combined with today's column, I've dropped a lot of my photographic worries. To sum up: a guy like me, who doesn't really do art photography or super-high-res landscape photography, will be just fine if he sticks around ƒ/4, ƒ/5.6, or ƒ/8 with my Micro 4/3 gear, but if I need to go to ƒ/11 or ƒ/16 (or ƒ/2.8, for that matter), that's okay too, because I'm not the kind of person who'll be hurt by diffraction that can only be seen with a magnifying glass, or by a little corner distortion. Right? In other words, the fact that I hand-hold almost everything will cause me more (technical-photographic perfection) problems than diffraction or lack of edge sharpness ever will. Whew. Thanks."
Mike replies: Right. You got it.