Cameras and Photography Explained

"The AF was so much better (and easier) than anything I have ever used that I am now seriously contemplating totally switching systems." --dpreview post

My guess is that this poster has his word priority backwards. What I'm finding again and again is that people are buying cameras due to not having to think about autofocus (the "easier" part of the post), and not about how good the resultant focus actually is (the "better" part of the post).

I've noted this before: a lot of the Z9 purchasers are buying it solely because of the perceived "I don't have to do anything, it'll just focus" marketing that Nikon has been doing. To a large degree, the same thing happened with the top Sony cameras, as well. 

Those buyers mostly get what they wanted: the new high-end camera performs totally automatic focus better than they ever achieved before. But I have to point out this: the Z9 buyer is paying US$3000 extra (today's pricing) in order not to study an autofocus system a bit and get it to work for them. On top of that, they get a camera that's far more complex, with far broader and deeper controls, and that seems to be tripping them up on other things ;~). Seriously. I've had people who don't know how to get a PDF file onto their computer/tablet (or even what a PDF file is) buy my 1200-page book on the Z9 in order to figure out their new "all automatic" camera. 

We seem to be in an era where we don't want to think (or learn) any more. Just let the machines do the work. My car wants to keep itself in lanes, but has a real problem when the lines start to widen at off ramps. It also prioritizes centering in the lane despite a vehicle encroaching on an adjacent lane. Autofocus systems aren't any different. They see general problems decently, but the marginal problems—of which more exist than you might realize—aren't the focus AI's forte, for sure. 

On every brand—and yes, I've used the latest and greatest from every brand—I find nuances that just get obliterated if you let the focus system do what it wants to. Sometimes that results in a slight misplacement of the focus plane, sometimes it's focusing on the entirely wrong object, sometimes the in-frame chaos just overwhelms its decision making. And, of course, the automated systems aren't thinking about depth of field or other factors you might want to consider. 

Great photographs don't happen simply because you point a box somewhere. As I've noted before, I've documented several hundred decisions that go into taking a photograph (I haven't gotten around to counting the decisions in presenting one ;~). The more of those you let some automated function decide, the less it's your photograph and what you actually saw or wanted to capture. 

The best practitioners, craftspeople, and artists in the photographic world tend to all do the same thing when it comes to automation: trust, but verify. We're—yes, I'd humbly include myself in that group—very quick to override something when the product(s) don't do what we need them to, but we also have learned that sometimes an automated control is fine if it lets us totally concentrate on something else. It used to be for focus that I did a two-finger dance (Back Button Focus). But today I do a four-button-and-two-dial dance, mostly because it allows me to take back control when I see the machine go not where I want it to.

I get it, though. the Z9/A1/R5 focus systems are better than the DSLR you were using when totally automating focus decisions. So, in essence, you're paying for something that moves your photography forward from where it was, at least if you were totally relying on autofocus to just do the right thing. But focus is just one sub-component of what makes a photograph great. I don't know of a camera that does it all, let alone correctly. Which means that if you want far better photographs the onus is on you, not the camera. 

"Maybe one day Nikon will offer a camera system and not some collection of products that  share a lens mount." email received

I don't think Nikon is the only "system camera" company that's not exactly making a system. Ironically, the newcomers, OMDS and Sony, probably come the closest, though I still see significant gaps in their systems. 

So what is a "system" these days?

1. Camera body 
2. Camera accessories (remotes, plates, batteries/power, cards, grips, straps, protection)
3. Lenses
4. Lens accessories (tripod collars, plates, straps, hoods, filters, protection)
5. Light (flash, strobe, triggers)
6. Support (tripod, monopod, gimbal)
7. Audio (microphone, mixer, etc.)
8. External record/storage (HDMI recorders, Computer/Mobile apps, etc.)
9. External Ecosystem (mobile apps and connectivity, desktop apps and connectivity, etc.)(not rated below, as was added after article was published)

Most Japanese camera companies spend most of their engineering money and time on #1, with #3 also getting a high share. #5 and #6 are mostly being ignored these days. #2 and #4 are afterthoughts if they're done at all. Some are doing #7 (via hot shoe connection), while most are working with third parties with #8, meaning that they don't want to spend a lot of money for this.

Here's how I'd grade Nikon on these things:

1. Body = B. A little on the slow and late side, with gaps in the lineup, but generally they get bodies figured out well.
2. Accessories = C-. The only reason they get this good a grade is that they have been at the forefront of batteries and cards. Nikon got lithium batteries right very early and has kept them progressing with the tech without constantly outdating them, and XQD was a Nikon/Sony collaboration that turned into CFe. 
3. Lenses = A-. The Nikkor gang has been hitting it out of the park with almost everything. The small ding in their grade is that they're now late with three lenses and they still have delivery issues on significant lenses.
4. Lens accessories = D. Late or missing, inconsistent, totally dependent upon third parties.
5. Light = D. The only reason it isn't an F is that previous work still lingers ;~). It's supposed to be a secret, but much of the Speedlight tech was licensed and modified. It seems Nikon got tired of that, as 2016 is their last real effort at anything, and it's no longer fully compatible with #1. Oops. Press releases saying you're working with third parties are simply papering over a deficiency.
6. Support = F. Sure, VR solves everything.
7. Audio = F. Has Nikon engineering not seen what the others are doing? Hey, the Z9 is a great 8K video camera visually, but it mostly stinks as an audio source. And why doesn't a vlogging camera (Z30) have a headphone jack?
8. External record/storage = C-. Nikon's been working with Atomos and Blackmagic Design (apparently only Australian companies need apply to Tokyo ;~), so I'll give them credit for that. And they've hired some contract work to help with the connectivity software, though their frugality shows in what they end up offering.

So what is the overall system grade when your individual projects range between F and A-? 

And again, Nikon isn't alone in making systems cameras that aren't really fully defined systems. I could do this same report card for every other mount and I'd get different grades, but I'm not sure any of them don't have some F's in their submitted materials.

"While physical parameters such as flange distance and the throat diameter do make a difference to the performance of wide and super-wide angle lenses due to extreme angle of incidence, said parameters are almost irrelevant for long focal lengths, thanks to almost perfect telecentric nature of long focal length lenses." --Discus post about 400mm f/4.5

Quotes like this are the reason why I continue to say that Nikon has not adequately described how the Z-mount has given them new optical choices. The 400mm f/4.5 is actually a good example: rather than use the traditional optical design used in most previous telephoto lenses where the major corrections are done up front as the light enters, the 400mm f/4.5 reverses that: the primary optical corrections are done at the rear of the lens, and part of that cluster of elements has to create telecentricity very near the image sensor, unlike the traditional designs. That requires larger rear elements.

The net benefits of the 400mm's design is that it is (a) shorter than it would be using traditional design; and (b) the center of mass is much closer to the lens mount than forward (which isn't just a handling issue, it's also a stress on the mount issue). 

Now that we have several examples of what Nikon is doing in the Z-mount, it would be a very appropriate time for the Nikkor marketing team to step forward and point out what the larger, shorter mount change has allowed them to do that would be much more complicated in the restrictive F-mount. The chances of that happening, of course, are near nil, as (1) Nikon thinks they already described why they needed the new mount [they didn't]; and (2) Nikon doesn't recognize that this is still an issue with the millions of customers they built under the old, more restrictive mount, who don't see a difference because they've never been told what it is.

"70-200mm plus TC or 400mm? 100-400mm plus TC or 800mm?" --series of common posts lately

Let me restate these questions generically: "flexibility or image quality?" Do you know what the answer is now? ;~)

To some degree cost may come into play, too, but if cost is a primary decision factor, then you're going to get the least expensive option almost every time. That wouldn't be my recommendation, of course; I generally say that it's better to get the best over the cheapest. Even if that means you have to wait a bit to save up the money to get it.

Further to the original question(s): most asking this question are expressing a form of FOMO (FOFOMO! FOMO is fear of missing out). They fear that they'll regret their choice because someone alongside them might get a better image than they do with the other choice. Can't have that, now, can we?

I'd also tend to say that if you're asking this type of question you haven't really figured out what it is you truly seek to do and what they really need. Again: do you need flexibility or quality? Many of the responses to posts of these questions start asking things that would actually get the original poster to identify what it is they really want to accomplish, fortunately.

I should disclose that my fellow workshop instructor and several pros I work with often have very similar discussions, though not often zoom+TC versus fixed focal length. But in our debates we're usually positing a specific need and simply trying to figure out the best option to fill that need. 

I'll give you an example: traditionally on safari the "right approach" has been two bodies, one with a 70-200mm and the other with a long telephoto, typically 500mm. Note that gap between 200-500mm (which is a lens you can buy), and that to go beyond 500mm requires a teleconverter, which you generally don't want to be taking on and off in dusty safari environs. This is why the 400mm f/2.8 works for me: it has that built-in 1.4x, so on safari I have 70-200mm, 400mm, and 560mm available. I'm perfectly happy with that. 

The 800mm PF lens appearance has a lot of us debating whether we actually want to add one to our kit or not. Note what it does to your flexibility if I just sub it into the traditional approach: 70-200mm, 800mm. That's a huge range in between that's not covered, and it's a range I like to be in due to perspective and backgrounds. So, no, I'm not opting for an 800mm. 

But wait, how about a 100-400mm and 800mm? Okay, that's better, but I'm now f/5.6 or worse through much of the focal lengths, versus f/2.8 with the approach I'm now using. Again, it comes down to perspective and backgrounds. 

As I noted in an article on zsystemuser.com, the introduction of the 400mm f/4.5 introduced a new fly in the ointment: smaller and lighter than the 400mm f/2.8 I'm using, so more manageable in the front seat of the Land Cruiser, and 400mm f/4.5 isn't terrible with backgrounds. So is the answer now 70-200mm f/2.8 and 400mm f/4.5? I'm thinking not. But I'm not sure until I test a couple of things, including just going 70-200mm and 100-400mm. 

Thing is, for lens questions like these, there is only one right answer for you. My answer isn't your answer. In order to answer these questions you have to carefully evaluate what it is you want to do, and what your priorities are. There's no shortcut. 

Thus, when you ask questions like this, expect to be asked questions back ;~).

"[Cropped sensor] lens value 24b yen, [Full frame] lens value 124b yen. And we expect [camera companies] to deliver more [crop sensor] lenses? With these numbers they won't." --Discus comment on CIPA numbers

Not quoted was this: cropped sensor lenses shipped was 1.5m units, full frame 1.8m units. The camera makers are certainly delivering crop sensor lenses in quantity. The problem traditionally has been that the camera makers do that only with kit lenses bundled with the equipment, and then don't supplement that with much in the way of additional lenses. Unfortunately, even the companies that do make a fair number of additional crop sensor lenses don't effectively market them (I'm looking at you Sony). 

Further: when the crop sensor and full frame mounts are the same, as they are now for Canon, Nikon, and Sony (e.g. RF and RF-S, DX and FX, or E and FE), you don't really need additional crop sensor lenses above about the normal focal length, as there's no clear size/weight advantage to do so. Kudos to Sony for recognizing that and concentrating mostly on wide angle to normal focal lengths in the E mount lately (see, I can praise them as well as criticize them ;~). Thus, even when the major camera makers do build out a crop sensor lens set the volume will be more modest than full frame. 

I'll stand by my long-held belief: by not building out crop-sensor lens sets and marketing those effectively, the camera companies ultimately do themselves a disservice. It's really easy to see that with the Nikon Z DX cameras, of which we now have three, but only three DX lenses, only one of which hits 24mm equivalent. Three new well chosen lenses in the 10-28mm range and a fast 16-40mm lens would sell more DX cameras, I'm absolutely certain. So, does Nikon want to sell those DX triplets, or not?

In the myths are persistent category, I notice that Nikon's Z9 firmware update brought about a ton of "where's the autofocus firmware update for the Z6 II/Z7 II" posts. Embedded in most of those are comments like this: "...the Z6 II/Z7 II are now known through the user world as having very poor autofocus compared to Canon and Sony." So...

Once again I have to repeat something: the focus systems even on the original Z6 and Z7 are quite good, and very usable both in sports and wildlife situations, as I've proven multiple times. However, it takes getting to know the new focus system and how it works in order to get good results. 

I would further point out that I have issues with every autofocus system that's been introduced by every camera maker. I can find idiosyncrasies and problems with every last one, and things that you need to study, learn, and control in order to get the best possible results. I was noticing just yesterday that the Canon R7—which supposedly has the R3 autofocus system—struggles to maintain animal detection in mottled, contrasty light. There is no perfect autofocus system, period. They all must be studied, mastered, and controlled to get top level results.

The persistence of the "Nikon doesn't focus well" myth comes from one thing, and one thing only. That's users who set the camera to Auto everything and expect better results than they've been getting with the camera they already use with Auto everything. In that case, I'd say yes, Sony does a better job of "set and forget" than Nikon. However, that "better job" is not perfect, either. On the Sony A7 Mark IV, Sony's latest camera and you'd assume the one with the most engineering history in its autofocus system, doesn't always nail focus in "set and forget" modes. I find that the A7 Mark IV often gets somewhat close to focus, but does not hit the exact focus point I expect. That's particularly true with Face/Eye detect. Of course, getting close apparently is better than the folk who keep promulgating the myth have done in the past ;~). 

Dead-on-nailed-it focus is hard, and not a single camera I've used does that without my understanding and help. In the present cameras I'd say that Canon, Nikon, and Sony all do a good, but different, job in what they get right about autofocus. In all three cases, in all models from each, I still have to perform a finger dance to get the best results possible. When I do, there's not a lot of difference between the results I get from the three vendors' cameras. 

Would I like a Z6 II/Z7 II firmware update that addressed some autofocus issues? Absolutely. For two reasons: (1) my finger dance could get easier; and (2) the percentage of "perfect" results from my finger dance might go up. Definitely good things to pursue. But does the lack of an update keep me from using the cameras? Not at all. Indeed, if Nikon would add only one thing (AF-ON+AF-area mode) I'd say that the Z6 II, for instance, is a better choice for autofocus than the Sony A7 Mark III and maybe even Mark IV. Hmm. What's "well known" again? ;~)

And a corollary: "[poor autofocus] is affecting the [Z6 II/Z7 II] sales." 

A tough one to prove or disprove. However, Z6 II and Z7 II sales seem to be pretty near Nikon's capacity to produce at the moment, so I'm pretty sure that Nikon isn't feeling sales pressure to improve the cameras. Moreover, the visible sites and discussions on the Internet—including my own—are really only the tip of the iceberg: they're visible to fewer folk than all those swimming below the surface. The general consumer doesn't see the disinformation and misinformation that pervades many sites, and thus wouldn't even have heard the "poor autofocus" false argument. 

Moreover, I'd point out that the Z5, which is not as refined in autofocus as the Z6 II and Z7 II, still sells remarkably well. 

Much of the discourse on the Internet is effectively like a bunch of fifth graders trying to assert bragging rights. Not for their photos, mind you, but for the gear they purchased. 

"Birds get scared off by a loud shutter." -- several different fora posts I've seen lately

The Canon R7 seems to have made this complaint louder ;~). Yes, the R7 has a fairly loud (and clunky) mechanical shutter sound. But I have two responses to this kind of complaint about BNIF (bird not in flight) photography.

First, in my experience, if a shutter sound were enough to trigger a bird to move, you're very close to the bird. It's much more likely that any motion, including just swinging the camera lens some, will set the bird off into flight. So if you're not in a blind using camouflage techniques, I just don't see shutter sound as a problem worth noting.

But beyond that, I tend to want a BNIF to turn into a BIF! Worst case—if these complaints are actually valid—is that I have a first image that's BNIF, and then a series of images that are the bird taking off, which would be much more interesting images, and often tend to reveal more of the bird's coloration. Yes, rollers sit on sticks and look pretty, but they look far prettier and you can actually see how they got their name when they transition from BNIF to BIF. 

Amusingly, I've sat next to rollers on sticks (BNIF) and clacked away and they wouldn't fly. Movement makes them fly, not sound.