How To Get the Best Performance from Your Camera Lens, Parts III, IV, and V

[Parts I and II are here—also, don't neglect to read Moose's excellent "Featured Comment" under that post, as it usefully updates these now 6-year-old articles.

Copyright 2008, 2014 by Michael C. Johnston. All Rights Reserved.]

-

III. Lenses in Use

It seems to me that there are two ways to use lenses intelligently, with conscious and informed knowledge of their imaging properties. The first is to learn the specific strengths and weaknesses of specific lenses, and beware of pictorial situations that won't match well with your lens's strengths or that will expose its weaknesses too unfavorably. The example I always give of what I mean is that many lenses have a lot of falloff wide open. But in many cases when you use a lens wide open, it's in "available darkness" situations with a central subject and nothing but darkness in the corners of the frame. No harm, no foul. If you know a lens is weak wide open, avoid using it that way! You can't always do so, but often you can. If you know the imaging properties of a lens, you can almost always "play to its strengths" and minimize its weaknesses. After a while this becomes habit.

Zander in Werner's garage. The lens and aperture used in this picture might result in falloff (vignetting) in the corners. But who could tell? No harm, no foul. In many way it's possible to mask or disguise the evidence of the performance shortcomings of even fairly poor lenses—but only if you know where those shortcomings are likely to show up.

The other strategy would be to pick a good lens, satisfy yourself that it behaves reasonably well in most situations, and then don't worry about it—shoot freely and take what it gives you.

To split the difference, in effect, a decent seat-of-the-pants working principle might be to "seek the middle" or avoid extremes where possible. That is, if you return to my original ten points for how to optimize a lens's performance, in Part I, you'll notice that most of the recommendations involved selecting a middle value and that most of the ways to "stress" a lens involve pushing it to one sort of extreme or the other. If you just keep this basic principle in mind—middle apertures, middle distances, middle of the zoom range on zoom lenses (at least, back off just a tad from the extremes), moderate contrast subjects, straight-on angles to your subject matter, middle ISO values on your digital camera, etc.—you'll have a decent shot at getting better optical performance out of your lenses.

-

IV. State of the Art

As I said at the outset, almost all lenses have a "sweet spot" where they will give their optimum results, and those optimum results are almost always excellent. The worse the lens, the more performance suffers in various ways as you move away from that sweet spot. What modern lensmakers work toward, and what you pay for in a really good modern lens, is performance that stays consistent and robust as you move increasingly away from that optimum. A really good lens is one that almost has no sweet spot because it's so consistently excellent under all conditions. It relieves you of the obligation of learning all its foibles because it basically doesn't have any. It will never let you down.

Take as examples the Nikon Nikkor AF-S Nikkor 24–70mm ƒ/2.8G ED or the Canon EF 24–70mm ƒ/2.8L II USM. Among the world's most beautiful camera lenses, they are best suited for use on full frame 24x36mm DSLRs. Neither lens imparts magical qualities to your pictures—they will not cut your eyeballs with zippety-zappety contrast or nuke your subjects with the X-ray vision of super-high resolution in just the axial zone; rather, these lenses are "merely" magnificently, supremely difficult to stress. Within physical limits (by which I mean, putting heads in the corner of the frame at 24mm will still result in perspective distortion, diffraction is still diffraction, and so forth), you may do almost anything either lens is capable of doing without negative optical consequences. Both foreground and background bokeh are subtle, gentle, and unobtrusive, especially for zooms, and is especially impressive with the lenses wide open, where even some good lenses are let down by some ugliness in the out-of-focus rendering. You may use any aperture and any focus distance with virtual impunity. And, assuming you can meter properly, you may put light sources in pictures virtually without fear: flare resistance is excellent and the flare that occurs is reasonably picturegenic (if that's a word; if it's not, it should be). The lenses' imaging properties are remarkably stable across the frame and throughout the zoom range. Both lenses are beautifully built and most likely can be depended upon to keep delivering their marvelous results imperturbably over time and in the face of hard use.

The more you understand about how lenses typically behave, the more you are likely to be impressed by lenses of the quality of these, especially as time wears on and you become more and more familiar with the one you own. Simply magnificent achievements of the lens designer's and maker's craft, richly deserving of the term "state of the art."   

-

V. The Big Caveat

The big elephant-in-the-room qualifier: good lenses don't guarantee good pictures. Good optical properties are scientifically measurable, but good pictorial qualities are not! In fact, given visual sensitivity and sophistication and some luck, photographers can and often do use the bad or flawed properties of poor lenses to good pictorial effect—as Holga, Diana, Lomo and toy camera aficionados do regularly.

Conversely, the finest lens in the world is "nothing special." That is, using it for your pictures is no guarantee that your pictures will be any good at all. In fact, using the world's finest lens is no guarantee that your pictures will even be optically good, since equipment is only half the battle—the other half being your technique. A great lens + poor technique = a technically poor picture. And even a technically excellent picture is not necessarily an excellent picture.

I've found it most amenable to think of lenses and their properties as a hobby somewhat apart from, and ancillary to, my practice of photography. I get pleasure out of understanding how my lenses are responding to and recording the visual world, but that pleasure isn't always an integral part of the visual content and (especially) the artistic effect of pictures. Understanding how lenses work, however, can help you solve specific problems, apply specific solutions, and generally avoid the pitfalls of unpleasant optical surprises and disappointing performance. And the best way to understand how a lens works? Stress it, and see how it does. Owning a fine lens can be a source of pride and possibility, but only if you learn it well.

Mike

Copyright 2008, 2014 by Michael C. Johnston. All Rights Reserved.

Original contents copyright 2014 by Michael C. Johnston and/or the bylined author. All Rights Reserved. Links in this post may be to our affiliates; sales through affiliate links may benefit this site.

TOP's links!

(To see all the comments, click on the "Comments" link below.)
Featured Comments from:

Terry Letton: "Thank you for Part V. The whole discussion was beginning to feel like a formula for boring images, and photography is all about images as far as I am concerned. I long ago learned that the vision displayed in the production of an image, purposeful or not, is the most important component in the production of that image, not the gear. That said gear an be fun to discuss, test, and play with but that is not photography."

A. Costa: "You have just sank the 'you have to buy an excellent lens to make excellent pictures' ship. Camera makers will deeply hate you."

Ming Thein: "A great post, Mike. As for lenses, it's very much personal preference, but I personally think the very best lenses are those that are completely transparent: they impose no rendering or interpretation of their own into a scene, which of course means that it's all on the photographer—good and bad. You won't be saved by swirly bokeh but you also won't be distracted by it, either. They're not easy to use for precisely this reason. But they have an enormously wide shooting envelope and will always deliver. Right now there's only a very small handful of lenses like this; foremost in my mind are the Zeiss Otuses...."

Tim Auger: "I like the concluding paragraph particularly. Old lenses (old Pentax lenses in my case) are fun in themselves. Learning about and enjoying their quirks, and valuing them as precision objects, is a pleasure in itself. Taking good photos is another, more difficult thing, and the two things are not the same. But one does not exclude the other. That's my answer to accusations of gearheadism."

JK: "I guess I have to agree with Ming about the very best lenses being transparent—I certainly felt that about an Apo-Sironar I used to use on my view camera—but over the years I've come down on the side of character in my lenses. That is, optical imperfections that please me and that suit the way I shoot. Nothing too obvious, just a way of seeing the world that works for me.

"One side benefit of this is that when Leica or Zeiss announce some new uber-lens worth more than my car, I think well, that's cool but not for me. Set in my ways or just secure in what I'm using? A bit of both, probably, but I'm sure many other long-time photographers have come around to a similar point."

Comments

[Nicholas Condon]

I find that I want lenses that are "invisible," like those two huge zooms you mentioned. (Come to think of it, I want the same thing from a camera.) The shorter the list of things I have to worry about, the more mental effort I can put into nontechnical things like framing and timing.

[Steve Jacob]

Better gear only means fewer excuses ;-)

I wonder these days how much lenses really matter when LR makes adjustment so easy. I tend to look at a digital image as data - which I want to be as complete and uncorrupted as possible - which I will then use to create an image. I am not sure if some subtle lens qualities survive the transformation.

[sebastian]

should not "picturesque" be the word?

[Marco Sabatini]

Neither camera imparts magical qualities to your pictures—they will not cut your eyeballs with zippety-zappety contrast or nuke your subjects with the X-ray vision of super-high resolution; rather, these cameras are "merely" magnificently, supremely difficult to stress. Within physical limits (by which I mean, shooting in a dark room will still result in imageless pictures, image noise is still image noise, and so forth), you may do almost anything either camera is capable of doing without negative imaging consequences.

[thomas hobbes]

definitely agree on thinking about lenses and their properties as a hobby separate from photography. a decent photographer can make due with any lens and get good results.

i could be perfectly happy shooting with what is technically my worst lens (probably a jupiter-3, though there are other contenders), yet i like to explore all kinds of lenses for the fun of it. mostly i'm not interested in getting the best performance out of them though (the differences in best performance between most lenses aren't that great or interesting in my opinion). i'm more interested in stressing them to see what kind of interesting looks that produces.

i have an rx1, which is technically amazing having almost no sweet spot and yet has a beautiful distinctive look to it. despite this i often choose to shoot instead with a leica m 35/1.4 lux pre-asph that has many flaws simply because i enjoy the way the optical flaws look too.

[Tim]

All of what you say may be true, but it omits the potential in post-processing - there's an assumption that I want to be using one preset (probably as close to no-op as possible) in ACR or similar and will just lap-up the lens's foibles.

We can correct for drop-off easily; not just by cranking some slider with a preconceived idea of what vignetting to correct, but precisely by taking a reference image of a dull wall and countering it.

Similarly we can correct for distortion easily; not just by cranking some slider for barrel or spherical controls "to taste", but with parameters precisely measured for the lens at various focal-lengths, apertures and subject-distances, thanks to the lensfun project.

Best of all, if the subject is conducive to a panorama treatment, we can throw naively converted files into Hugin (free and open-source), which can calculate and compensate for lens distortion precisely as seen in the source images; blend with enfuse biased toward exposure, and the vignetting drops out since corners that are dark in one frame are correctly exposed in the next.

TBQH I fail to see the fascination with sitting there plugging at one RAW file at a time in Lightroom - continually choosing what looks like the correct vignetting correction for each of 300 images then coming back and repeating it a month later after the rose-tinted goggles have cleared is akin to dust-spotting in terms of manual, repetitive, boring and menial labour.

And, as you imply, sometimes we point the camera in the right direction, too. Not much point correcting for every technicality if the ball's kicked out of frame.

[Kenneth Tanaka]

This has been a very nice, concise overview essay on practical camera lens issues, Mike. Thank you for "reprinting" it.

For readers who enjoy this type of material may I suggest Canon's EF Lens Work III. This was formerly a private-distribution book that Canon decided to make freely available as a series of PDFs. Even if you're not a Canon shooter there's plenty of excellent general material in there!

---

My own perspective is that today's lens optical qualities are so far down the chain of total image quality factors as to be nearly negligible. Yes, in the age of passive imaging media the lens generally had a powerful influence on the final image.

But lenses today are merely "Eliza Doolittle" players in Pygmalion-like performances. Enormous strides in computer-aided engineering design, materials sciences, and manufacturing technologies produce "kit" lenses that surpass yesterday's "premium" lenses. Beyond that, however, today's active electronic recording media and infinitely malleable digital post-capture chain can compensate for most lens design deficiencies, either in-camera or later. Light fall-off, chromatic aberrations, geometric distortions, purple fringing, even sharpness fall-off are all generally digitally fixable often automatically without your awareness. Quite amazing to me!

But that fluidity is where so much of image troubles really start. The same powerful tools that can nudge image quality upward can, and often are, used to tastelessly or excessively degrade and distort images. Spend an extra $1,000 on a premium lens that minimizes edge blur and fall-off only to later commit slidercide and make its images look worse than Henry Fox-Talbot's first tries.

Not only have electronics produced a golden age of cameras (as was speculated in an earlier TOP article) but it's also delivered a golden age of lenses. But I sometimes wonder if that's had any real influence on the general quality of photography. Maybe in some sectors (advertising). But it sure seems to me that the advancement of the lowly camera phone, coupled with "social media", has unarguably had the most powerful impact on photography in history, for "better" or "worse".

But I've digressed.

[Lubo]

Nice motto below the letter blocks on Zander's t-shirt.

[Rick in CO]

Mike, I like and subscribe to your message. I use the Zeiss ZM lenses. I have always liked their contrast and color rendition from my film camera days and was happy to be able to use them on an M8 and now M9. They have those middle of the road qualities; modest apertures allowing optimal optical correction, no focus shift (ex-the f1.5 'Classic' Sonnar), round apertures for more pleasing out of focus rendition, plus top quality manufacture even if built to a relatively expensive price point. More importantly, they are consistently good, within their limitations, which, with the ability to adjust in-camera ISO and use photoshop corrections, are a non-issue.