News/Views

News and commentary about the Nikon DSLR world and photography in general. This page automatically updates with links for each new news/views story and is a good place to bookmark if you want to see the traditional bythom "front page" type of story. 

Latest News/Views stories (top is most recent):

Is it Time to Replace Your Camera?

I've had advice on when to upgrade your Nikon DSLR (and to what model) for some time now. I'm going to have to start revisiting that again soon with more advice on whether an upgrade should be a DSLR or a mirrorless camera. But I came across an article in another field the other day that had me having an Aha! moment of simplification. 

This is the straw man proposal resulting from my Aha! moment:

  • If your current camera is 12mp or less: You should upgrade your camera. Period. Many things changed dramatically since you bought your camera. Images from new gear will look better than yours. But here's the awkward caveat for the camera companies: if you don't need more than 12mp and aren't a lens junkie, then you should probably upgrade your camera to something like the iPhone 11 Pro. Yeah, I just wrote that. I'll have more to say about the iPhone 11 Pro soon, but basically for sharing images electronically, you don't need more than 12mp and the new Apple smartphone is a pretty phenomenal 13-50mm camera, even in low light. It's only going to be the crowd that needs more than 50mm or shoots action in low light that will find the iPhone 11 Pro wanting (true for the equivalent smartphone competitors, too). 
  • If your current camera is 16-24mp: Maybe you should upgrade your camera, maybe not. This is where my Nikon DSLR upgrade advice comes most into play. Some of those 16-24mp cameras would have better choices available today. I think, for example, about the Nikon D7000. That was a camera that just doesn't hold up against the current model (D7500), and for a lot of reasons, including focus performance. Meanwhile perhaps the Nikon Df was exactly what you wanted, plus there's no real upgrade available in Nikon's lineup, anyway, so you don't upgrade. This category of cameras can go either way. If you're in it, this is where you need to pay some attention. And yes, there's one other aspect that comes into play here: 16mp on a m4/3 sensor is a lot different than 24mp on today's full frame sensors. 
  • If your camera is 26mp or more: Nope, you don't need to upgrade. I'm pretty sure of that. All the Nikon DSLRs that fit this bill shoot pretty fine images if you've taken the time to learn your camera. That includes the D800, D810, D850, and the mirrorless Z7. So do virtually all of the other 26mp+ cameras I can think of from other camera makers. And if you haven't taken time to learn your camera, why the heck would you want to buy a new one that you'll need to take the time to learn? ;~) 

That's it. Anything outside what I just wrote means that you're upgrading solely because you want to. Nothing wrong with that. However, once something becomes purely "want", then rational choice flies out the window. It's like love: you either want it (them) or you don't, and only you can figure that out. 

Do We Need a 120-300mm f/2.8?

No sooner had Nikon pre-announced the new 120-300mm f/2.8 lens I started getting the questions about whether we need this lens or not. 

Of course, part of that is driven by the fact that we already have such a lens, from Sigma. But several of the questioners were seriously questioning the "tweener" focal range.

Lens focal length is not really a good way to judge lens differences. What you really want to look at is the angle of view and the capture rectangle at the distances you'd use the lens. So let's do that. First, the angle of view:

  • 70-200mm = 29 to 10°
  • 120-300mm = 17 to 7°
  • 180-400mm = 11 to 5°

Those degrees are all the horizontal FX frame coverage. 

More to the point, though, the question is about what you can capture when you are position constrained, which you often are in sports photography. Sports users being one of the primary buyers of the longer than 200mm lenses.

Let's consider that I'm in the end zone and the line of scrimmage is the 10 yard line. I'm 60'+ away from the quarterback, I'm 30' away from the successful play's finish (goal line). What do I get at the long end of each lens at each of those distances?

  • 200mm and 60': 7' high, 10.75' wide (rounded numbers)
  • 300mm and 60': 4.75' high, 7' wide 
  • 400mm and 60': 3.5' high, 5.5' wide
  • 70mm and 30': 10.25' high, 15.5' wide
  • 120mm and 30': 6' high, 9' wide
  • 180mm and 30': 4' high, 6' wide
bythom sports CU vs WSU Nov10 2018 Z726536

Above is what you get when a player is at the 10-yard line with the right lens. And the following is what happens immediately after that player scores and comes right in your face (which is why we have multiple cameras with multiple zooms hanging from our necks):

bythom sports CU vs WSU Nov10 2018 Z726553

Even if I turn the camera vertically as I did above, 180mm is too much lens on a score, while the 70-200mm and 120-300mm give me different ranges to consider (the 70-200mm is better if I'm concentrating on the score, the 120-300mm better if I'm concentrating on the start of the play).

Thing is, every shooting position for every sport has a different set of calculations. I'm all for having more flexibility in lens choice, because it opens up additional shooting possibilities. 

Indeed, there's at least one missing lens even with the new 120-300mm in the mix! We sports photographers probably all would like to see a 35-135mm f/2 (or f/2.8), as well. That's doubly true if you shoot with a D500 or other DX crop camera. Since you're probably wondering, here's the comparable stats on such a lens:

  • 35-135mm: 54 to 15°
  • 135mm and 60': 10.75' high by 16' wide
  • 35mm and 30': 20.5' high by 30.75' wide

Note how the narrow angle about doubles from the 180-400mm versus 70-200mm, and does the same thing for the 120-300mm versus 35-135mm. The wide side does more than double with each of those pairings. That's about right, in my book.

Personally, I'm glad to see Nikon doing "zoom overlap". Of course, now I want it in the Z mount, too ;~).

Do We Need a D6?

Now that Nikon has officially acknowledged that the D6 is coming, and likely well in time for the 2020 Olympics, the question is simple: do we need it?

D6 24 70VR front

My answer might surprise you: maybe. Probably not if it's filled with new tech. 

I really find nothing particularly problematic with my D5 for event, sports, and some wildlife use. Sure, it's a DSLR so its focus sensors don't extend all the way across the image area, but I don't think a D6 is going to change that (though note my comment below). 

To explain my answer, let's start at the sensor. The D1h, D2h, D3, D4, and D5 all had one thing in common: they produced fewer pixels than the more common cameras of their time. While this got a little out of hand with the D2h when it was only 4mp versus the Canon 1D's 8mp, good reasons exist as to why the pro PJ/sports camera have tended to be underpixeled: (1) they perform better in low light; and (2) photographers on assignment have low resolution bars coupled with very short deadlines. 

That first reason is going away a bit now that we have BSI sensors and the overall base level of sensor tech has stalled and sensors are evening out in ability. At most of the ISO levels I tend to use, the 24mp, 36mp, and 45mp sensors all look about the same these days when output at the same size newspapers, magazines, and Web sites are demanding. That said, when you push insanely high with the ISO, there's nothing better in Nikon/Sony realm than the D5. 

The second reason has everything to do with time management and image transfer. I've written this before, but most of us shooting sports and events have really tight deadlines. Really tight. If you're scrambling to get your picture used by the wire services, you'd better be the first to upload. That means putting at least some images out not long after a sporting event starts, and certainly getting solid, evocative images out by the end of half-time (or the first quarter/period interval). 

When I shot the NCAA Lacrosse Championship earlier this year, my images of the post game celebration literally went onto the winning school's site within moments of me uploading them, and it was all I could do to try to turn that around within minutes after I finished shooting the celebration. There were several other photographers still in the press room trying to finish their uploads as I was leaving. I think they were using higher megapixel cameras ;~). In fact, I know two of them were.

So when time and transfer come into play, megapixels are not your friend. The next day after the championships, I pushed another 1GB of JPEGs up to my client. Very few of those images got used for anything, because they simply weren't timely enough. (There is usefulness in being complete and having all your shot images in the client/agency archive, because you never know when an image you took previously will become useful for some reason. But that usefulness is not why they hired you in the first place. ;~)

Shooting JPEG helps with the size issues, but it's still a game of bytes. A JPEG Fine Large from a D5 is about 10.5MBs. From a D850 that expands to 22MBs, or basically double. If I shoot 1000 images in the first half of a football game, the difference between ingesting those D5 images and D850 images can be measured in minutes (and is worse if you use Lightroom ;~). Likewise, when you push your selected, annotated, and cropped images back out to your client, you don't want double the upload time, either. Even shooting a D5 it's rare that I make it out of half-time in the press box back to the field in time for the 3rd quarter kickoff.

So what's my expectation for the D6? I'd be happy with 24mp, actually. Give me BSI and all the other newer sensor technologies tuned the same way the previous D# cameras have been tuned, and I'd be more than happy. A few more pixels gives me a little more cropping flexibility without chewing up card space and ingest/upload time.

Unfortunately, the expectations—particularly among the Sony A9m2 prognosticators—is that we'll get 36mp in the next high-speed pro camera generation. The leaked Sony Semiconductor IMX435 sensor details seem to suggest that a useful high-speed camera with a 36mp sensor is possible today, and some rumor sites also point to the 48mp IMX311 as a possibility. But I'm not at all sure that I want all those pixels to deal with in my PJ/sports camera.

What I want more than pixels is speed. No, not higher frame rates. I stopped using the 20 fps frame rate on the Sony A9 for the same reason I don't want more pixels: if I shoot 2x the images I normally do, it takes more than 2x the time to ingest, select, and output my images. Why "more than 2x"? Because you'll spend more time looking at a sequence of images to figure out which one is the one you're going to send, or you'll send more images and let someone else deal with figuring out which one to use.

No, speed is now defined by something different: how fast can I crop, annotate, and upload an image?

Technically, I can do all those things from my D5. These issues make the speed of doing that problematic, though: (1) cropping means I create even more images in the DCIM folder, and also means I have to remember which one is which since I don't have the ability to usefully rename them, either. Adding a star rating helps a bit, but only a bit. Why can't I crop in camera, give the image a meaningful name, and SAVE it to a SELECTS folder? (2) annotating is done with the touch screen right now, and while the D5's is quite good for that compared to anything else that exists today, the implementation for doing that for an already shot image is awkward at best. What I wouldn't give for a Siri-like capability "Hey Nikon, add the caption "Max Borghi scores go-ahead touchdown from quarterback Anthony Gordon in first quarter of the WSU versus Northern Colorado game at Martin Stadium in Pullman, WA." (3) uploading directly from the camera doesn't generally work well unless I've got a wired Ethernet connection to my ftp server. Not all stadia, and certainly not all shooting positions, have such capabilities, though the D5 has the ability to make that connection if it's present.

So far what you're seeing from my list—24mp, speed features oriented to client upload—would only come from talking closely to the prospective users of said camera, and savvy users, at that. A lot of the D5 ranks are filled with photographers who just took what they got from Nikon, and then grumble under their breath that they don't understand something or vaguely want "more," but can't enumerate that.

One area that will come up with any camera in this range is focus performance. 

I'm on record as saying that I believe that the D5 has the most consistent and usable focus system of any PJ/sports camera currently (and yes, the Sony A9 is getting close, but I find it tends to drift a bit from absolute focus). That's not to say that the D5 is perfect or that it can't be improved. I wonder, for instance, if putting phase detect also on the viewfinder sensor* wouldn't allow the D5 to track a subject outside the dedicated focus area. The camera already has an uncanny ability to track a subject outside the dedicated focus area and back via the color information produced by that sensor: perhaps this can be improved to build an even more functional focus ability.

*Technically, it might have to be on the focus screen. But yes, there's a patent for that ;~).

What was most useful on the D5 (compared to the D4) when it came to focus though was the addition of AF-ON+AF Area combinations for various programmable buttons. While it was a bit difficult to get used to at first (just as using AF-ON for back button focus is in the first place) this became second nature with a bit of practice. Note that such a feature is just another example of doing something useful to solve a user problem, not adding tech or pixels or absolute speed. 

So that's sort of my answer to the question I pose in the headline. If Nikon focuses on the tech side in creating a D6, they missed the point and I don't need one. We don't need more pixels (at least not many), we don't need higher frame rates, we don't even really need any more dynamic range. What we need are things that improve the user experience and let us get our jobs done faster. If Nikon continues to give us that, then maybe the D6 is what we need.

Sony's 3.7 micron Sensors

bythom sony 60mp

Jim Kasson, someone I correspond with from time to time behind the scenes, was the first one I know of to post this, but it's something I had noted earlier and wondered whether or not it was worth speculating on: Sony Semiconductor now has four different sensor sizes with much the same pixel structure and technology.

That would be:

  • APS-C 28mp
  • Full Frame 61mp
  • Small Medium Format 100mp
  • Large Medium Format 150mp

These all appear to be the same basic BSI pixel design at about 3.7 microns, with Exmor dual gain capability, and using copper wiring to do the heavy bandwidth lifting. The next step is likely 3 micron pixels, with perhaps some additional new technologies added in. 

Nikon, for some reason, has been dabbling with sensors right around 4.3 microns (20mp APS-C and 45mp full frame), but not exactly the same pixel size. It's unclear why Nikon picked those odd sizes, but the basic photo diode can be easily scaled, so there must have been a benefit Nikon saw to going slightly off Sony Semi's standard releases.

But there's a bigger picture thing to see here: it used to be that we got distinctly different moves in pixel size and technology at different dedicated camera sensor sizes, but now the same basic structure has rolled into basically all the Sony Semiconductor sensors, so we're seeing more technology parity across the sizes. One aspect of this is that it tends to make the oft-stated rule of thumb about sensor size more accurate (e.g. full frame is about a stop better than APS-C, all else equal; well, now most everything else is equal).

Unfortunately, a not-so-much-discussed aspect is now also being revealed: when companies like Sony Semi basically release the same tech across their entire lineup in a very short time period, that is often the hallmark of nearing the end-of-the-line in an iterative-type technology progression. Much as Moore's Law has slowed in CPUs, whatever improvement line we were following with image pixels has also slowed. To the point where it is simply being replicated upward through the sensor sizes.

What we need next is a new disruptive image sensor technology. 

The one that everyone has tried to perfect is what I call rollover saturation: in other words, break the saturation limit on an individual photosite without breaking anything else. I'm aware of at least four different basic approaches that have been attempted, so far not particularly successfully. Something else tends to break, unfortunately. Still, there's a lot of potential here if you can get past the issues. The whole notion of limited dynamic range would completely disappear if you could get past the saturation point. (The other end is the noise floor, and we're now in a situation where we fairly accurately record the randomness of photons, so we're not likely to improve the floor much; it's the saturation ceiling we need to break through.)

What Sony Semi says they're actually working on next is something completely different, though: Artificial Intelligence (AI) on (or with) the sensor. Some logical candidates exist for how to focus that AI: noise reduction, object/edge recognition, and color integrity (or manipulation) come to mind as clear possibilities. Other ideas get slightly into the "crazy" realm, but I'm all for crazy, as that is often where true innovation is found. 

I'm still a little leery about AI, though, as it suggests that we're moving away from collecting more accurate real data to creating faux data that mimics reality.

The real problem for dedicated cameras isn't really the image sensor, though. We've got plenty of good data to work with from pretty much any modern camera with a decent lens out front. The real problem is the fact that the data from an image sensor is two dimensional. 

You're probably thinking I'm about to talk about curved image sensors. Nope. True, a curved image sensor solves one optical problem when done correctly in conjunction with lens design, but it's still basically a two-dimensional approach.

Where the smartphones have been going for some time is in building depth data. You can do that using multiple cameras or by using Time of Flight type devices (a sensor looking at projected laser light reflections). You can also do it using some other techniques, but the two I mention are the ones that have the most work behind them at this point. 

As we throw more AI and computational photography at the underlying data, having three dimensional data sets is much more useful than two dimensional data sets. To me, this is the thing that the dedicated camera makers are going to have the most issues with in keeping the smartphones from continuing to erode the bottom end of the camera market. Already there are things that you can do with some smartphones—such as remap lighting—that are tough to do with dedicated cameras. This problem is only going to get worse as long as we're stuck in the 2D world. 

Thing is, though, as long as the higher end of the camera enthusiasts keep reacting positively to "more pixels" and "more dynamic range" and the basic, simple, iterative things that are going on with large image sensors, the more the camera makers are going to find themselves going down the wrong rabbit hole. 

At this stage, I'm still welcoming "more" because we haven't quite hit the limits where small gains on pixels and dynamic range just aren't visually useful in any way. But we're getting close. Plus each iteration is generating another group of Last Camera Syndrome users, where they fail to see the gain they get for the price they pay, and thus just happily stay with what they own.

Simply put: we're still in the rough patch for dedicated camera makers. The pool is getting smaller, but we still have large sharks trying to eat. Basic sensor iteration isn't changing that. Indeed, it's aggravating it and making the pool (number of customers) smaller. 

The "More is Better" Problem

I've long been on record as writing (and saying) that more sampling is always better. Put in a camera context, more pixels are better, all else equal. 

Recently I've gotten some pushback on that from engineers who are at the forefront of semiconductors. I need to adjust my statement ever so slightly, I believe: we're currently still in the more sampling is always better phase

Let's talk about full frame for a moment. In recent times we've gone from 24 to 35 to 42/45, and now to 60mp. The constraint in the "more sampling" is that the actual linear increase in resolution is getting smaller and smaller with each step, while the recorded diffraction impacts get clearer and clearer and nibble at lens capability at lower apertures. 

In conjunction with a computational imaging talk I was putting together just after the turn of the century, I took the time to run the complex math that starts to get involved. I actually needed help from a couple of friends whose math skills are better than mine to complete the analysis. Sensors were Bayer APS-C at the time, so that's the type and size we worked with. 

The results were that past about 24mp, while there would be recorded data differences, you ran into almost a wall in terms of getting something useful out of those differences. If you keep an AA filter over the sensor, the antialiasing and diffraction impacts simply get in your way. If you take the AA filter off, the faux data that gets generated beyond the Nyquist frequency is chaotic and doesn't reliably net you anything real. Add in the Fourier transforms used by JPEGs, the Bayer demosaic artifacts, and more, and 24mp sure seemed like a logical stopping point for APS-C. That's what I wrote in 2003, and that's basically what I still believe today.

As you probably noted, that didn't stop Sony Semiconductor from going to 26mp, Samsung to 28mp, and now Canon to 32.5mp with APS-C sensors. At 60mp, the full frame sensors are now hitting that same level of pixel density and thus expose the same issues as >24mp APS-C.

Of course, during that same period just after the turn of the century I also remember a respected camera engineer telling me we'd never have 1 micron photosites ;~). Because they wouldn't render anything useful. Obviously wrong. But why? 

Basically, when engineers hit a ceiling, they start getting creative. 

The most common creative approach at the moment is to apply AI-type analysis to the data structure you have and create a "better" or "fuller" structure. The pixels are no longer "real," but they look just fine. Both Topaz Labs and Skylum seem to be deep into doing this post mortem with post processing software, but we're going to see more companies try to move this into the hardware (and Sony Semiconductor has basically announced this as a goal, too). 

What could possible be driving the need for more pixels in cameras (and smartphones)? After all, the number of folk that ever print larger than 8x10" or display photos on more than an HD device (2mp) is incredibly small. 

Well, it has been small, but it's almost certainly going to get larger. 

That's driven by displays. HD was 2mp, 4K is 8mp, 8K is going to be 32mp. At the point where your Living Room wall is no longer painted gypsum but rather a large flat display, we will be talking needing far higher pixel counts in order to put excellent images on them. It's clear that's where we're headed. The only question is when we'll get there.

One of the things I started experimenting with a few years ago is taking low pixel count images and using them on large displays. They look bad. I apparently wasn't the only one that thought of that and came to that conclusion. Just as we got line doublers back in the last days of the original NTSC, I'm now seeing plenty of data interpolation capabilities appearing everywhere, some of which use AI to try to identify the objects to be resized. The 4K television I bought had a rudimentary form of data interpolation—US cable TV was 1080P at best at the time—but what I'm seeing now are much more complex algorithmic, and yes, AI approaches. 

A couple of years ago I would have said you couldn't probably do reasonable upsizing on existing consumer hardware in anything approaching a reasonable time frame. Today, I acquiesce. I'm still not sure you can do it for real time video, but processing a lower pixel count still image to something that looks good—or maybe I should say looks better than it should—on a big screen is starting to take very little time on a high end workstation.

Pretty much any trend eventually ends (or at least plateaus). But with a 5K display on my desk and likely an 8K and larger screen coming to my Living Room some day, I don't think we're near the end of this. We simply are going to find we like images with more pixels better than ones with fewer pixels.

So do I want to explore the Sony A7Rm4 (60mp) and the Fujifilm GFX100 (100mp)? You bet. I like living on the front edge of tech. It's where I grew up (Silicon Valley; Cupertino to be exact), it's where I spent most of my business life, and I am still fascinated by the game of trying to extract "more" from what basically starts as sand. 

Here's the thing, though: the camera makers are forgetting the primary user problems while pursuing problems we don't really have yet. 100mp is not yet solving a user problem. Sharing images would be. While I'm happy that some engineers are chasing down the pixel rabbit hole, that's not the rabbit hole that will pay their current salary.

So I find myself on the same soap box today as I've been on: yes, I'll take more pixels, and please figure out to make cameras share images more easily.

Nikon Pre-announces the D6

bythom nikon d6

Someone's getting a little antsy in Tokyo. 

Announcing the development of the D6 by press release seems like another of those Nikon "we need to announce something" announcements. The marketing department has always seemed a bit paranoid that if they're not constantly making some sort of noise, they'll be forgotten. It doesn't help, of course, that Sony discovered the "announce every month" tactic as is using it aggressively, even when the actual product timing doesn't quite work out to match the announcement timing. 

Still, it seems that there's a bit of self esteem lacking in Nikon's Tokyo headquarters and that they worry that Nikon pro (and prosumer) faithful take any quiet time as an indication that London Bridge has fallen and it's time to think about a new destination (e.g. camera brand). Unfortunately, a short, five-paragraph press release in which there are no details doesn't exactly say much. 

Moreover, it appears that Nikon finally got around to recognizing that this was not only the 60th anniversary of the F-mount, but the 20th anniversary of the game-changing D1 (a fact that it seems only I had pointed out prior; Nikon missed the actual anniversary by two months).

Nikon's real problem isn't at the top end of their lineup. As I write this, the D5 is still the best action/sports camera you can get (the Sony A9 being a close second, though). As I write this, the D850 is still the best all-around camera that you can buy (the Z7 and Sony A7Rm3 being a close second; I haven't evaluated the A7Rm4 yet). The Z6 holds its own against the third generation of the Sony A7 and against Canon's two mirrorless full frame cameras. The D500 is still probably the best APS-C camera you can buy, despite being hobbled with an incomplete lens set. Heck, the D750 is still holding up well despite being five years old, something that's remarkable in the digital camera realm.

No, it isn't in the high end that Nikon needs more strength, though none of us high-end Nikon users are going to turn down better gear. Moreover, cameras like the D6 (e.g. D3, D4, D5, etc.) don't sell in large quantities. They might have a few months of sales in the 5K unit range before settling down into numbers that sometimes are measured in three digits. Nikon's bottom line won't get any real punch from the D6, though their reputation might.

Which brings us back to point.

Clearly, the Z9 isn't ready yet and Nikon is worried about the Sony A9 or A9 Mark II stealing away some of the long-time Nikon DSLR users. The D6 is basically a statement that Nikon won't stand by for that, though it's a lousy statement with absolutely no details of how that will happen. Whether or not the D6 will hold the fort until the Z9 is ready is a story for a different day, a day after we've had some real experience with a D6 in battle. A day which none of us knows when will happen.

Nikon's real problem is at the bottom of their lineup. Again. 

This is not the first time that Nikon has struggled to hold onto "consumer market" sales. We had at least three episodes of that in the film era, plus some minor lapses in the DSLR era (I'm looking at you D80). 

What Nikon doesn't have at present is a smaller sensor, affordable entry point product line (e.g. DX or APS-C, or something else). On top of that, dedicated prosumer compacts are gone (though Canon and Sony still sell such things), action cameras are gone, and all that's left at the very bottom of the camera line are some reheated leftovers. The Coolpix-with-the-long-lens trick has pretty much played out. I can't off the top of my head tell you the difference between a D3300, D3400, and D3500 (ditto the D5xxx line), which says something about how poor the product iteration in that line was. 

In my article on recent APS-C mirrorless cameras over on sansmirror.com I wrote APS-C is for the masses, who don't want to spend a lot, carry a lot, or set a lot. To that I'd add: and who want to share images easily instantly. That all pretty much defines the cameras that are missing from Nikon at the moment. Call them the Z30 and Z50 (please don't drop everything to all single digit numbers, Nikon!). Something that has all the positive attributes of the Z6/Z7, but doesn't cost as much, is yet smaller and lighter, and gets automation and sharing right without compromising anything else.

That's not a simple task. No one's mastered it yet, though Canon and Sony are nibbling at the edges in different ways. 

Thing is, without establishing a healthy entry base, there's probably not enough upper market (pro/prosumer) to maintain the number of camera brands we have. If the camera market collapses all the way back to the 5m units/year level it was pre-D1, there's simply not enough money left on the table for everyone to keep playing full hands. 

What Nikon's marketing department needs to be antsy about is not whether people know that a  D6 is coming, but whether Nikon has an answer for what happens beneath the Z6 point. 

Aside: Sony didn't really answer that question for themselves with the recent A6100 and A6600 announcements, in my opinion. Despite all the hype, the A6600 is really just a recombination of things already existent in the Sony APS-C line (IBIS from A6500, focus from A6400) coupled with an A7 battery and as high a price as anyone is likely to get from an APS-C camera now. The A6100 is a de-contented A6400, so not exactly exciting, either. The A6xxx lineup seems confused, gadgety, and unimaginative now. 

Meanwhile, the interesting part of Nikon's five-paragraph announcement was a lens. Specifically, that a 120-300mm f/2.8E FL SR VR lens is in development. We never got an updated version of the 300mm f/2.8. It looks like this lens will be what we get instead. If it's as good as the 180-400mm f/4E is, then we've got another winner, though I'd expect this lens to be priced higher than the current 300mm f/2.8. (And no, I don't know what the "SR" stands for, either.)

Frankly, I find "development announcements" to be non-useful. As I've described before they're a form of FUD marketing; an attempt to get you to stall buying decisions when considering alternatives because something "better" might be coming down the line. I've not seen any evidence that Nikon's previous development announcements changed sales any, but apparently the folk in Tokyo have nothing else to do with their time while they wait for new product to come out of the development side.

What Nikon should have done is this: in July on the 20th anniversary of the D1 they should have had a big compendium of how the D# series changed photographers' lives, pointing out all the things that Nikon pioneered in those cameras, and showing off the best photos that were obtained from them. As part of that, then you add "and coming later this year we'll do it again when we introduce the D6." 

MBA students that aren't getting A's in their marketing classes can write better press releases than Nikon did for the D6 announcement. Fortunately, the development team still has top practitioners in it, but they must be wondering why they're paying the salaries of all those non wordsmiths...

Oh, and bonus points for this: the nikonimaging.com Web site is no longer. All those links are now broken and need repointing. We now only have nikon.com (corporate/global) and the individual subsidiary sites, though we do still have nikonimglib.com for some reason. Looks like a make-work project to me, but it wasn't announced that I know of.

2019 Starts Badly

The CIPA numbers for the first half of the year are available, and they're not very positive. DSLRs are selling at only two-thirds the rate of 2018, and even mirrorless unit volume is down 14% (though dollars taken in for mirrorless is up slightly, basically due to all the activity with full frame). Lens sales are down, compact sales are down, everything's down. 

So much so that the pattern suggests that interchangeable lens camera unit volume may drop to somewhere around 8m units or less this year (last year was almost 10.8m units). 

Which brings us to a segue: Nikon's CEO was quoted by Nikkei as saying that Nikon's camera unit volume now is less than one-sixth what their peak volume was in 2012 (that includes compacts, of which Nikon had many more models then than they do now). 

Somehow, though, Nikon has managed to stay profitable throughout that long contraction. Which brings us to Nikon's first fiscal quarter results.

Nikon is making some quiet noise that indicates that they think they've begun to turn a corner. The first interesting point in their results was that they're making no changes to their fiscal year forecasts. In other words, they see nothing in the next nine months that would cause a further slump in their camera sales or lower their profit further from the 6.5% level they're currently predicting (note: that's down from 11.4% last year, though).

Indeed, now that we can see some specifics in their full forecast data, they're expecting the second half of their fiscal year—October through March—to be clearly better than the first half. They're predicting a 17% increase in sales and 200% increase in profits in the second half over the first half. That means that there is significant new product in the pipeline destined for that period, and that Nikon thinks it will perform decently.

Meanwhile, camera/lens inventory has come down a bit (5%) and is projected to come down further (16%). Cameras will still make up 47% of Nikon's sales for the year (same as last year) according to Nikon's forecasts. Market share is expected to be 19% in ILC (down from 20%).

Nikon has been managing to contraction for some time now. That started in the semiconductor equipment business, then expanded to the camera business. Even Nikon's "growth" initiative, healthcare, is not growing at the moment. In many businesses you'd think that this would result in the business being less than healthy. In Nikon's instance, they've done a good job of staying ahead of lots of problem areas that arise from getting smaller. Cash flow is still quite positive, liabilities have been in good control, and R&D expenses have been going down slightly (while increasing as a percentage of sales, an indication that they're controlling those expenses, but still trying to push forward with new product). 

I'm sure it's humbling for Nikon to go from a strong #2 in cameras to fighting for that position with Sony. Still, in terms of business health, they appear to have done all the right things short term. The question now shifts to longer term: are we really at the bottom of their contraction within the overall market, and how much more will the overall market itself contract? 

While Nikon's CEO Umatate-san says that they still have efficiencies they can gain through rebuilding their sales structures and overhauling procurement costs, I suspect we're nearing the limit in terms of what such changes can do for profitability as long as the camera market itself keeps getting smaller. 

Put another way, Nikon has some new product coming. How well that product is received will go a long way towards further stabilizing their situation and perhaps even improving it. 

Meanwhile, Canon's chairman Fujio Mitarai is out talking with the Japanese business press and saying that the number of compact cameras and the number of ILC cameras will each drop to between 5 and 6 million units in 2020 (that's down from 8 (compact) and 10.8m (ILC) units in 2018. Best case scenario according to Canon is that we go from 19.4m total digital cameras in 2018 to 12m in 2020, or another 38% drop. The drop from peak cameras is attributed to "market structure caused by advent of smartphones."

Uh, yeah. I'm going to get back up on the soap box I've been on since 2007 and say this: exactly what have the camera companies done to make dedicated cameras as convenient for image sharing as phones? Of course sales are down and continue to go down when you don't actually solve the biggest user problem. The camera makers haven't actually embraced the new era of how images are shared. Simple as that. 

So let's look at how things will play out if Canon is right about how far the ILC market will contract and market shares stay about the same. 2020 unit volume:

  • Canon 3m units
  • Nikon 1.2m units
  • Sony 1.2m units
  • Everyone else: 400k units

I think Nikon/Sony are fairly locked in with about 40% of the market between them. And staying above 1m units/year gives them enough ramp to stay profitable, though there will be great stress in doing so. The real question in my mind is Canon versus everyone else. Can Canon really hold 50% market share in ILC. Certainly not with their current products. Everyone else—that's Fujifilm, Olympus, Sigma, and Panasonic basically—actually needs to grab share from Canon, I think. I can't imagine building 100k units a year being a profitable ILC operation. 

Do You Need "That" To Be Better?

Today's announcement of the Sony RX100 Mark VII (I abbreviate this RX100m7) has me raising a question to camera makers that I've been seriously pondering for some time: do I need that to be better?

bythom sony rx100m7

"That" is some specific feature or performance metric, which is generally the big marketing punchline for the new product. For the Sony A7Rm4 "that" was 61mp. For the Sony RX100m7 "that" appears to be blackout free continuous shooting at 20 fps with tracking focus ala the A9.

No doubt these are technological improvements worth noting. But the relevant question in photography is: do you really need "that"? 

The question comes up in camera pairings, too. For instance the Z6 and Z7 are basically the same camera with different sensors (so 45mp becomes "that"), as are the Panasonic S1 and S1R and Sony A7m3 and A7Rm3 (basically the same "that" as with Nikon). 

One thing about technology: if you continue to hire and employ talented engineers, they continue to iterate it ;~). While some keep writing that what Sony is doing is innovation, I'd tend to say it for the most part "that" turns out to be expected iteration. Innovation comes when you don't make the expected iterations but you instead deviate from the course and try something different. The A7S, for instance, was an unexpected course change, thus I'd argue that it was innovative. Of course, we're still left with the question of "do you need 'that' to be better?" ;~)

Since I espouse optimal data as my photographic mantra, it's rare that I'd turn down "that", whatever it turns out to be, as long as "that" is an improvement that allows me to collect more optimal data. But to the vast majority of those practicing photography with dedicated cameras these days, I often wonder whether or not most people need whatever the latest "that" turns out to be. 

Of course, marketing is going to tell you that you need "that". Moreover, today it's nearly impossible to tell if you're being gamed in any messaging on the Internet, where you're more likely to see That! or THAT!!!. 

Here in the US we have regulations (from the FTC) that require that you identify "influence", but they're not followed by most and the FTC has been lax at enforcing them in all but the most extreme cases. (Disclosure: I try to follow the spirit and letter of regulations as best I can; all ads on my sites are identified as such, and anything I receive from a company I write about is fully disclosed (mostly loaner equipment, but there have been a couple of other different "perks" over the years, as well).)

Now it might seem like I'm picking on Sony today, but they just happened to make an announcement on the day I was contemplating writing this article. Here's what I think: the addition of the microphone jack and the other improvements on the video side were more "that" for me (disclosure: I bought, own, and use a Sony RX100m6). 

But the details of the RX100m7 still left me a little cold on the video side. 

Because it's iteration, not innovation. 

The first question you have to ask yourself is this: does the Vlogging world need a camera that fits a shirt-pocket? If the answer to that question is "yes," then the RX100m7 isn't really the answer. (Oh dear, what happened here?) 

Sony emphasizes vlogging in their press release and marketing materials. Here's how they want you to do that: add an external microphone, add a shooting grip, and buy a lot of batteries. Suddenly it isn't a shirt-pocket vlogging machine anymore (and wait, didn't Sony try this trick already with the RX0m2?). Not unless you have substantially bigger shirt pockets than I do.

Here's how innovation works to solve the problem: fold-out shotgun (or variable angle) mic, fold-out grip. Fits in a shirt-pocket. Here's how iteration solves the problem: add someone's existing microphone with a cable using an age-old connector, add a variation on an add-on grip you already make. Doesn't fit in a shirt-pocket, so why are we starting with a shirt-pocket camera?

So the question for today (and every day for that matter) is this: do you need "that"? If not, keep shooting with what you've got and put the money you didn't spend into savings. 

I'm sure the Sony RX-100 Mark VII is an excellent camera. After all, the Mark VI it derives from was, and the new version is now I better ;~). 

What is Tack Sharp?

You hear the term all the time, what does it mean?

Unfortunately, different things to different people. Consider this sample from an image:

bythom sharp1


I've run a strong deconvolution sharpening on it without regard to noise propagation so that we can better see if the animal's fur is "in focus." (More on that term, in focus, in a bit.)

No, that sample is not tack sharp. We can kind of make out the individual hairs, but there's not strong edge acuity to them. You should see a small bit of antialiasing blur on them, despite my attempts to "pull the edges in."

How about this part of the same image?

Nope, still not tack sharp. While very close, there's still a bit of small blur to the fine detail, just like before. We're closer here to where we want to be, but not quite there.

How about this one?

bythom sharp2


Hmm, that looks a little better, doesn't it? The individual hairs have more clear delineation, and you can see exactly how fine and distinct they are at the upper right edge. Actual focus is probably just a bit to the upper left of center on this sample. And that's just a bit forward of the eye position on the animal.

You may have guessed by this point that these samples all come off the same image:

bythom sharp4


The first sample was on the back of the front shoulder, the second obviously the face, and the third in the middle of the raised leg.

Overall, this 45mp D850 image renders just fine when processed decently (not as much exaggeration as I've used here). Focus is very close to where I wanted it (the eye), but not quite there. Focus is clearly on the front of the animal, and there's very clear depth cues because the back of the animal is not in focus. That depth cue actually helps people say this image is in focus. 

Which brings me to "in focus." 

I keep encountering other photographers, other Web sites, and even photo editors that will say that something is "in focus" when I can assure you that the focus plane is actually not where it is supposed to be. Often they try to emphasize that by saying those examples are tack sharp (and that's after applying some form of sharpening, as I did ;~).

In the heat of the moment shooting sports and wildlife in particular, you will have a difficult time putting focus dead square where it should be. If you rely upon the automatic tracking focus modes, the camera will also have a hard time putting the focus plane on precisely the same spot of the subject over and over again (and yes, that's despite face and eye detection). I keep seeing sequences posted by others where they say their camera was precisely tracking a subject, but when I look closely at the sequence at 100% view, I can see clear front/back focus shifts on the subject through the sequence. 

Hint: when someone shows you sports images, take the time to look at the grass/field and see if you can pinpoint the focus plane and then follow that through the sequence. This is often how I can see the "waver" of the actual focus plane. You can sometimes do this with wildlife shots, too, though the fields animals are in generally are not as consistent and flat as they are in sports, so it can be more difficult to ascertain where the focus actually was.

What we have in most pronouncements of "in focus" is actually another form of "good enough." Between the camera's tracking mechanism, the depth of field, and the tendency on clear action for our eyes to take in the action without regard to focus, some wobble to the focus plane in action sequences is generally tolerable. But to call the results tack sharp would be wrong. 

And yes, different cameras produce slightly different results in automatic tracking. I'd say the two best of the bunch right now are the Nikon D5 and the Sony A9. Picking hairs (pardon the pun) between them I'd say the Nikon is a bit more consistent in placement of the focus plane being in the exact right place than the Sony, but I can accept the results from either.

In the Nikon-only world, I'm going to shock you. Here's the order in which I think the automatic focus tracking does the right thing: D5 is best, Z6/Z7 is second, D850/D500 is third, D7500 is fourth, D750 is fifth. But be forewarned, there's another variable here: the D5 is best at finding the focus plane in the first place on moving subjects (one reason the Z6/Z7 come in second). If you can't adapt to EVF lag or get the focus sensors positioned right initially, the D850/D500 would be second, and the Z6/Z7 might slip further than third. The DX cameras do well because they put their focus sensors fully across the screen, and thus compensate for the camera not staying perfectly framed on the subject somewhat better (another reason why the Z6/Z7 are higher than you probably expected). 

Canon users are probably shocked that I'm putting both Nikon and Sony ahead of them. The Canon 1DXm2 isn't really much behind the D5/A9. But it's now behind. Indeed, I tend to find that at every level when comparing Canon/Nikon products until we get down to some of the recent consumer cameras, where Canon has deployed dual-pixel focus and Nikon is still using old technology, like the 11-sensor focus system on the D3500. 

Now for the kicker (I really feel like I'm a mirrorless promoter today ;~): for static subjects, pretty much all the mirrorless cameras are more precise and consistent in setting their focus plane than the DSLRs. That gets even more pronounced with fast prime lenses, but it applies to landscape work, too. I'd say that if you're shooting static subjects with a mirrorless camera and missing focus, you're doing something wrong. Very wrong.

To me, tack sharp requires three things: (1) the focus plane be where it's supposed to be; (2) a lens of high MTF capability and little or no astigmatism/coma; and (3) a high resolution sensor capable of clearly resolving the detail produced by the first two. Anything less starts to fall into the "good enough" bin.

Thing is, the camera makers are all going up-scale in their offerings, and we've got better gear today than we've ever had. Gear that's capable of making better images than we used to be able to achieve. 

That means that we also have to go up-scale with our assessments, too. My definition of tack sharp today is clearly different than it was a decade ago. If you go back and look at the three ingredients I said were needed you'll see why: (1) focus systems have gotten better; (2) lenses have gotten better; and (3) we've gotten higher resolution cameras.

Next time someone uses a term like "in focus" or "tack sharp," ask them to define their terms. If they stumble at doing that or equivocate at all, then I'd say you should be looking for other experts to listen to.

This article is also getting posted to Lenses/Lens Articles.

The 20 Year Anniversary

June 15th marked the twenty-year anniversary of Nikon's announcement of the D1, the camera that most feel kicked off the DSLR era. The camera actually didn't ship to customers until early 2000, but a number of us got a chance to use it briefly in 1999.

Upon initially handling the D1—despite its many modal UI flaws (all fixed in the D1h)—I knew that the serious photography world was about to change.

Yes, I'm well aware of and even used some of the Kodak SLR conversions and the Nikon/Fujifilm E2 experiment in the 90's. I don't dismiss those products, but it was clear with the D1 that something different was about to happen: a new era of cameras you would find in every camera store and which carried on the mantel from the SLR bodies. 

It's illustrative to look at the specifications of that original D1 to see just how far we've come.

  • 2000 x 1312 megapixel images (3mp). 
  • A top ISO of 1600 (base of 200). 
  • An APS-C sized CCD image sensor producing 12 bits per photosite. 
  • A 2" LCD with 130k dots. 
  • A claimed top frame rate of 4.5 fps. 
  • The F5 autofocus system. 
  • All for US$5500. 

That frame rate turned out to be not quite true with the then current CompactFlash cards, particularly with the old Microdrive cards. Plus the buffer really worked out to be less than a dozen raw images (less than two dozen JPEGs). 

Still, as limiting as those specifications might seem today, for photojournalists in particular there was clear promise here, to the point that the D1 really changed serious photography as we knew it very quickly. 

Canon shot low, with the consumer D30 DSLR as their first effort in 2000, and feeling a bit rushed to market at that (sound familiar here in the mirrorless transition?). Nikon shot high, with the D1 being followed by the very solid D1h, D1x in early 2001 and the also serious D100 in early 2002. Canon responded with the 1D—featuring a stitched sensor from a non-Canon source, again a sign of rush—in 2001 and the 1Ds in 2002. Those six cameras pretty much were the kick-in-the-butt that blasted the DSLR era into high growth and killed the SLR.

I note that dpreview gave the D1 a Highly Recommended rating in its review. I didn't. I declared the D1 as being too modal and likely to trigger you to miss shots. It wasn't until the D1h came along that I could recommend to others that the DSLR era had truly arrived with a highly usable camera.

So twenty years on, what have we gotten from the DSLR iteration highway?

  • A transition from CCD to CMOS. That didn't come without complaint. Both semiconductor approaches have their pluses and minuses. That said, anyone in the silicon business knew that CMOS was going to be the winner if you could address its (then) image quality  shortcomings. That's because of the ability in CMOS to address cells individual and to add additional electronics into the image sensor itself, things we have in spades in today's cameras, and which make them better.
  • More and better pixels. A lot of people don't know that the original D1/D1h were actually 10.4mp sensors. Say what? Nikon took some Sony Semiconductor pixel technology and had them bin it! An individual pixel in the D1/D1h was actually four sub pixels binned together. The D1x saw a different binning approach, with only two horizontal pixels binned together. When people today talk about the camera makers being beat to the punch in computational photography by the smartphones, that's not exactly true. That D1x in 2001 used computational methods to build JPEG images that were bereft of short axis pixel information yet still looked quite good. That's because of computational work done in the imaging ASIC chip of the D1x.
  • Bigger and better LCDs. One of the primary benefits of the digital camera was its ability to let you immediately review what you just shot and evaluate if you need to change a setting or shoot again. As I've written before, this was one of the things that, once discovered by consumers, triggered the rapid change from SLR to DSLR. It's kind of amazing that this was clear even with a 2" display that only had 160k dots (today's screens are typically a minimum of 3" and a minimum of 1m dots, so larger and much more detailed).
  • More images faster and for longer periods of time. The D1 was effectively a 4 fps camera with a 2.5 second buffer (and remember, this is for a top-end professional camera). Today the D5—despite the extra megapixels and bit depth—is a fully functional 12 fps with a 17 second buffer. I remember clearly the first time I was on the football sidelines with a new deep buffer Nikon body standing next to a bunch of Canon shooters and I just decided to hold down the shutter release (as you all well know, I don't shoot long bursts to "save my butt," but am much more selective about timing and bursts). Every Canon user's head turned my direction in disbelief. That's basically where we are today with current cameras: anyone using an old one is going to wonder how it is that you're still shooting, should you care to fill you card up.

Looking at a D1 today, you see that the D5 isn't all that far from it, other than the things I just mentioned. Sure, Nikon added some additional controls (e.g. the thumb stick) and buttons, and changed card formats. But the bones and muscular structure are all there and are still being inherited today. Indeed, the D5 is really just another continuation of the all-electronic design Nikon put forth with the F5 in the 1990's. The things that weren't broken weren't fixed. Things that were missing were added. A few things have been fleshed out and made better.

You'll note that most of the changes and benefits we've gotten in the 20 years post D1 are internal. Indeed, that's something to consider in these days of transition from DSLR to mirrorless. 

Anyone who has been involved with the technology industry knows that the primary thrust in true tech product change comes from the silicon inside (and by inference, the software). That's because there are huge benefits to reaping the results of Moore's Law to combine, group, and simplify into fewer and fewer components that can be mass produced and assembled using automation. 

We got all the major changes in the DSLR I outlined above because of semiconductor advancement (yes, even the LCD addition). The transition that's going on right now in ILC is mostly a logical continuation of that. Mirrorless cameras remove most of the mechanical parts from DSLRs and put more emphasis on parts that can be mass produced on automated machines. The good news is that the constant push in semiconductor technology continues and will result in even more happening inside our cameras (see my other article today on where cameras are headed).

I haven't run into anyone lately still using a D1 era camera (e.g. the D1h). I have encountered one recently who was still shooting a D2Hs. And I still see quite a few dragging D3's around. D4's are almost as common as D5's when I look around the events I shoot at. 

Which tells another story about this anniversary: while we've had 20 years where Nikon (and Canon) has put out new and better iterations every four years, our DSLRs have proven to be long-lived even as the potential technology inside started to pass them by. I fully expect to see folk still shooting D500, D850, and D5 cameras (and similar Canons) four years from now. 

So happy birthday, D1. It's been a long and wonderful journey you started us on.


text and images © 2019 Thom Hogan
portions Copyright 1999-2018 Thom Hogan-- All Rights Reserved
Follow us on Twitter: @bythom, hashtags #bythom, #dslrbodies