Short answer: Predictability of rendering. Uniqueness of rendering. An outstanding trait.
While I get a lot of “what camera” questions, it’s really the “which lens” questions that predominate. These questions are also harder to answer.
Thing is, you pick lenses differently than cameras. So let’s start there.
What I’m finding with cameras is that everyone is still tending towards FOMO (fear of missing out). That immediately leads them to full frame, and then to maximum pixel count on the image sensor. Even if they don’t need full frame or maximum pixels. People are trying to bulletproof their camera body choice because they believe they’ll use it for an extremely long time. Plot twist: you’re likely to use your lenses for a longer time ;~).
Lens choices also tend towards FOMO, though in a different way. Zooms are picked over primes because, well, you never know what focal length you really need. In this line of thinking, a 24-200mm zoom gets regarded as a “better lens” than a 50mm prime.
These two things combined is leading to my having to deal with a lot of people in over their head with Z9 bodies and 24-200mm lenses.
Those of you who have been following along know that there’s a disconnect there: these folk are using Nikon’s best Z camera body with Nikon’s worst Z-mount zoom (pretty much throughout the focal length range, but particularly so at 200mm). These folk never seem to want to change a lens on the one Z System body that has a protective shield for when they're changing lenses. Hmm.
This isn’t to say the Nikon 24-200mm f/4-6.3 VR isn’t a decent lens. It’s one of the better superzooms made, but against other choices for specific needs, it’s always a compromise choice.
Which brings me to the one of the simplifications I mentioned at the onset: outstanding trait.
The outstanding trait of the 24-200mm is its wide focal length range. You can go from taking a wide angle of the town square to getting a tight image of the person that just tripped over the ledge of the fountain and fell in the water in a very fast flick of your wrist. 24-200mm is “more outstanding” in that respect than 24-120mm, which in turn is “more outstanding” than 24-70mm. In the situation just described, the 24-70mm might not have been the right lens for the job, as it gets one capture right but the other not so much.
Outstanding traits are the tricky part of lens choice. For instance, I get a lot of questions about astrophotography. These questions all invariably lead to “are stars still rendered as points in the corner of the frame” type queries. Well, on some lenses, yes, on others, not so much. The questioner here is looking for a particular outstanding trait because it has a critical impact on the type of photography they’re doing. (Aside: it’s rare that these same astrophotographers ask me about other aspects of the lens they’re considering. Vignetting, linear distortion, chromatic aberration all seem to fall off their list of things to ask about, even though those will also impact their results.)
So my first point is this: do you know which outstanding trait it is that you’re looking for?
For instance, in the Z-mount right now you can buy a 400mm f/4.5 VR S or order a 400mm f/2.8 TC VR S. Note the verbs. You can obtain the former today, but the latter only at some unpredictable time in the future. But I was writing about outstanding traits. With this duo of lenses we find conflicting outstanding traits. The 400mm f/4.5 VR S is easily the most hand holdable autofocus 400mm we’ve seen to date. It’s light, it’s smallish, it balances better on cameras, and doesn’t require a heavy duty gimbal (e.g. a Wimberely Sidekick works fine).
Meanwhile, the 400mm f/2.8 TC VR S lets in a stop+ more light and doesn’t need an external teleconverter to extend its range.
So which of these things is more important to you? You can’t say all three (lightest, biggest aperture, multiple focal lengths). And you pay a lot more (including a long wait) for the latter two over the first one, so you’d better really need the 400mm f/2.8 TC VR S's outstanding traits.
Canon, Nikon, Sigma, Sony, and Tamron all seem to be coming up with unique lenses these days. Probably because long-time customers may already own existing lenses that cover some of the same tasks when used on an adapter. Here’s just a few of the unique trait lenses at the moment:
- Canon RF — 28-70mm f/2, collapsing 70-200mm f/2.8, 600/800mm f/11 DO.
- Nikon Z — 400/600mm f/2.8/4 TC, 58mm f/0.95 NOCT, 26mm f/2.8 pancake.
- Sigma L/FE — 45mm f/2.8, 65mm f/2, 60-600mm f/4.5-6.3.
- Sony FE — 12-24mm f/2.8GM, 20-70mm f/4G, 14mm f/1.8G.
- Tamron — 20-40mm f/2.8, 35-150mm f/2-2.8.
I’m pretty sure you don’t have those lenses in EF, F, or A mounts that you could use on an adapter ;~).
Unique sometimes intersects with outstanding trait. It’s sometimes the outstanding trait that makes the lens unique. For instance, in the mid-range zoom world, we have a plethora of f/2.8 and f/4 24-70mm zooms. But Canon’s 28-70mm f/2, Sony’s 20-70mm f/4G, and Tamron’s 20-40mm f/2.8 lenses each extend a trait (faster aperture or focal range) into a new formula that creates a unique lens.
I’m seeing a lot of users these days with “unique envy.” Nikon’s 24-70mm f/2.8 S has to be one of the very best mid-range zooms ever made, if not the best. It’s hard to find a fault with it. Nevertheless, I keep hearing from readers who “wish” it were f/2, or went to 20mm, or picked up one of the unique lens traits that other companies have been introducing. Me, not so much. That Sony 20-70mm f/4G has an enormous amount of mustache distortion at the wide end, as well as other traits that make it a little tougher to use optimally. The Nikon? Just put it on the camera, dude, it just works.
Thus, I tend to not get excited about the unique trait unless it can be proven that I need (and will use) it. The built-in teleconverter thing? Yep, love it and use it. A wider focal range with more optical drawbacks? Not so much. Still, owning the unique lenses probably gives you bragging rights in the sauna at the local photo club (what, your photo club doesn’t have a sauna?).
And you might have noticed one other thing hidden (implied) in my description of the Nikon 24-70mm f/2.8: predictability. When this lens is on my camera, I know what the results are going to be. They’re going to be really, really good. I’m not going to fixing lens things in post.
We’re in an era now where the number of really good lenses has expanded greatly. I remember my mentor getting loaner lenses from Nikon after every launch, his looking at them and using them a bit, then sending them back to the mothership with the comment “not for me.” In private conversation he’d go further, saying that the sampled lens just didn’t push above any bars he had set for his work. If it didn't, then why did he need it?
So, if you’re looking for a great lens, make sure it delivers predictable results, has an outstanding trait, and maybe is unique (it probably won’t stay unique for long). Make sure it’s going to get over the performance bar you already have established for your work. Be thankful that you found any lens that succeeded.
That’s right, great lenses are much more rare than adequate lenses.
————
I didn’t really talk about the uniqueness of rendering in the main portion of this article, so because I don’t want you to go away disappointed, here’s a quick bit:
There was a time years ago when the best photo editors could pretty much tell you which lens someone had used in a photo, simply by the way it “rendered.” Two typical things that were giveaways (at least at the common focal lengths): (1) the acuity falloff from center to corner; and (2) the transition from focus plane to out of focus, and what that out of focus (bokeh) looked like.
One thing that’s an issue today is that image stabilization, lens design, and even things like electronic shutter all can have an impact on #2. The typical complaint these days is “busy bokeh.” But it sometimes isn’t even a predictable busy bokeh. If it was, many of us could adjust what’s behind our subject to minimize any busy-ness. Unfortunately, the stabilization-caused busy tends to change on some lenses based upon distance to the thing, how much stabilization was happening, and what was doing the stabilization. Not as predictable.
So when I write that a great lens is predictable in its rendering, I mean that I know ahead of time what I’m going to get from it, no matter how close/far I am to the subject, what the background is, or what settings I’m using.
