Revolutions are tricky things. Sometimes they spring up overnight, transforming a landscape in unpredictable ways. Other times, though, they creep up on people, providing plenty of advanced notice but not always a lot of forethought.
The digital photography revolution happened suddenly, and it brought huge changes to the photographic landscape. Companies rose, companies fell (some of them to rise again), and in what felt like an instant, the long-established world of photography was changed forever.
The mirrorless revolution hasn’t been as swift, but there’s a new epoch coming nonetheless. In the face of Sony’s rising market share numbers, Nikon and Canon have announced (officially/unofficially) that their own mirrorless systems are en route. With the question changing from “will all the major manufacturers offer mirrorless cameras?” to “when will all the major manufacturers offer mirrorless cameras?” (with an answer, perhaps, by the time you read this), photography is about to enter a new era.
Will SLR-style cameras continue to be produced in the new era or will their production dwindle as did film cameras? Will photographers be able to move their legacy gear over to the new platforms or will everyone have to start from scratch? And what will the balance of power be like in an era where the difference between the three dominant players isn’t dependent on which technology is at the core of their systems?
Mirrorless On The March
When I started to write about mirrorless cameras, they were an interesting novelty. Compact size was the dominant benefit of shooting mirrorless, and many adopted early cameras for their portability, despite their relatively poor performance compared to DSLRs.
For a few years, journalists like myself had to explain the benefits of mirrorless cameras, or more accurately, their potential benefits, as camera systems were far from being on par with the performance of DSLRs.
In the last few years, the industry has watched as the various mirrorless systems—be it Micro Four Thirds or APS-C or full-frame—have steadily encroached upon and then overtaken that of most DSLRs. All of the players in the mirrorless market have pressed ahead with technological advances that not only push the envelope of photographic possibility but also redefine that envelope.
Want 20-frames-per-second capture? Easy-peasy. Want to capture 4K video with full AF? No problem. How about viewfinders with no blackout, continuous eye detect AF and continuous exposure, silent shooting, in-body image-stabilization…? Mirrorless has you covered.
This isn’t to say that DSLRs aren’t eminently capable cameras. Some of today’s DSLRs are easily the best cameras ever offered. The mechanical nature of their operation, however, limits their potential—there are physical barriers that are simply hard for DSLRs to overcome. DSLRs are incredibly honed, incredibly evolved machines, but at their core they’re still analog to some degree. While the images produced by DSLRs are digital, and while the sensors, memory, processors and various components are digital as well, the physical mirror in a DSLR, which flips up and down to make image capture possible, also bakes in specific laws of physics.
At this point, there isn’t much cause to debate whether or not mirrorless cameras will dominate the photographic market—Nikon and Canon hint-announcing that they’ll be offering mirrorless systems puts an end to that speculation game and instead sets a deadline.
After The War
At some point soon, Nikon and Canon will release professional-level mirrorless cameras, and the balance will change. Both companies have dipped their toes in the water with compact mirrorless systems, but these have more in common with the earliest mirrorless cameras than they do with their pro offerings.
The rest of the photographic world, with a few unique use cases, has already gone full mirrorless. Olympus, which started the mirrorless revolution, and partner Panasonic, have the all-mirrorless Micro Four Thirds market wrapped up. Fujifilm dominates in the professional APS-C arena and has launched a medium-format mirrorless system. Sony leads the full-frame mirrorless charge and has a strong presence in APS-C. Leica has its full-frame mirrorless body, and its whole range of rangefinders are, by definition, mirrorless. Hasselblad has even moved into the mirrorless space with its newer bodies, although it and rival PhaseOne both offer medium-format systems built around reflex capture.
When looked at this way, it makes the position of Nikon and Canon seem even more extreme—they alone develop and release DSLR cameras.
So let’s fast forward to a time when Nikon and Canon release mirrorless cameras (assuming that this hasn’t happened by the time you read this); what’s this world look like?
Both Nikon and Canon have a full lineup of DSLR bodies, from entry-level solutions through professional models. There’s virtually no chance that the entire line of existing camera models would be replaced with mirrorless models at the same time. It’s more likely that mirrorless systems will be released to supplement what’s already on the market, with new models arriving as DSLR models reach the end of their life.
But it’s also likely that DSLRs will continue to be produced for years to come. That’s because there are huge amounts of legacy gear in the DSLR world, and not everything will work well with mirrorless cameras.
While mirrorless cameras and DSLR cameras seem different only in the presence of a mirror, in reality there are some technological differences in the way lenses and other equipment functions in the two systems. One of the most significant differences is in the way that lenses focus, from a mechanical standpoint.
DSLR lenses have been around for a very long time, and during their lifespan, numerous different focus motor mechanisms have been employed as autofocusing has evolved and improved.
Mirrorless cameras, having come after many of these lens focusing systems were abandoned, have started off with (generally) more sophisticated focusing systems. (A good primer on this is available on the LensRentals site at lensrentals.com/blog/2016/04/a-look-at-electromagnetic-focusing.) The current mirrorless systems don’t have to worry about being able to efficiently drive older lens autofocus systems because manufacturers like Sony and Fujifilm and Olympus never made lenses with older systems.
This is the main reason that adapted lenses don’t perform as well as native lenses on mirrorless systems—the adapters can’t overcome the variety of non-native focusing motors, and the mirrorless cameras are optimized for their own native systems.
This is particularly important because when Canon and Nikon release mirrorless cameras, they have two choices in the way they can introduce them.
The first would be to simply remove the mirror in the camera and make a mirrorless version of their standard DSLR bodies, which can connect to their standard line of DSLR lenses. This would provide the largest compatibility with the decades of lenses available for the systems. It wouldn’t, however, provide for ideal focusing speed on older lenses; essentially, only the most modern of the company’s lenses, or ones designed for their mirrorless systems, would focus quickly, and the rest would focus with the same performance level of adapted lenses.
The second way to introduce new mirrorless bodies would be to release a brand-new line of lenses specifically designed for a new mirrorless camera. This would provide for the fastest focus, but it would put these companies in an odd position. With a full mirrorless-only lineup of lenses available on the various systems, would DSLR customers settle for just a handful of new lenses designed to maximize the performance of their systems?
Most likely, we’ll continue to see the development and release of DSLR bodies for the customer who wants to maintain use of their legacy glass and a new system gaining lens models running in parallel.
There are some other factors that could see DSLRs continue to be produced for some time, including the head start other companies have on their mirrorless development. Nikon and Canon had the best chance of catching their competitors before mirrorless AF and AE systems became so advanced, but companies—and particularly Sony—have poured a considerable sum of money into their sensor divisions, and the result has been some pretty advanced chip design. If Nikon and Canon’s initial mirrorless cameras aren’t on par with the competition’s systems, it’s likely that the DSLR lineups would continue to be developed to maintain competitive advantages over those competing systems.
In any case, the post-DSLR era is unlikely to be decided by the camera manufacturers and their plans as much as it’s to be decided by consumers. The tricky thing about technological change is that it alienates some customers by design. (See such minor changes as the iPhone dropping the headphone jack for an example of customer uproar.)
It’s likely that the switch to a largely mirrorless camera world will be tumultuous and the outcome as unpredictable—or more unpredictable—as the early days of the mirrorless revolution.
While mirrorless (or EVIL – Electronic Viewfinder Interchangeable Lans – a term I like better) certainly is the future, it seems to me DSLRs are still ahead in terms of the viewfinder. As much as I love and use my X-T2 and X-T3, there are times when the viewfinder lag is still, well, a drag. And given the fact that the camera is in essence a mini TV studio, where the light hitting the imaging chip is sent as a raw data stream to the computer in the camera to be demosaic’d and converted into a viewable image that is then sent to the viewfinder, it will likely be a while before that lag is eliminated. Photons bouncing off mirrored surfaces area as instantaneous as it can get; computations to generate images necessarily take longer. Just go to an electronics store and stand in front of a video camera connected to a TV and wave – the lag is quite noticeable – it’s much shorter in an EVF yet nonetheless there.
So, for event shooting (particularly indoors) where the fleeting expression on a child’s face is missed a fraction of a second later, the EVF may not be up to the job. Or when I shoot macro using a macro lens and ring flash in a botanical garden, the ability to react instantly when the subject flower blowing in the breeze is in the exact proper position means I revert to my DSLR. The Fuji 80MM macro lens is great (their 60 was a dog) when the subject is static, but the EVF lag becomes a major impediment as compared to the Nikon.
Hi there f8Lee
I’m David, the former Editor of DPP, and I wrote this article, which ran in print a few months ago.
A few replies to your comment: optical viewfinders certainly have advantages in terms of the ability to see a scene exactly as it looks, you’re correct on that. I’m not going to debate here the merits of optical viewfinders, as they’re great.
When it comes to the comment that “it will likely be a while before that lag is eliminated” that lag has already been eliminated. The Sony A9 shoots with no blackout and no lag. That’s because the EVF is getting data from the stacked CMOS sensor and being displayed in real time. Any lag inherent in this system is so low as to be immeasurable.
Technology in this regard is only getting better, so it’s not going to be “a while” before lag is eliminated in mirrorless cameras in general. It’s a matter of processing and location of memory.
Your description of a mirrorless camera being like a little TV studio isn’t quite accurate. In a TV studio, a signal is coming in, being sent to a control board, being manipulated, and being sent out via one of a number of links to a transmitter. Not only does live TV operate with a built-in delay (usually 8 seconds) but the act of moving the signal between the different parts of the studio add delay.
It’s actually more accurate, if you want to use the TV analogy to describe the TV control room, where the signal is optimized to be viewed by the director and producer in real time. If that signal were meaningfully delayed, cuts between scenes would not work properly. In those situations there are high-bandwidth channels designed to move the data quickly (if digital).
Likewise with a TV in a electronics store. You’re being captured on a device with a lag in the output, fed into a board that spits the signal to the TVs and fed back in via one of the inputs, and then being displayed by the TV. There’s an inherent lag there, because Best Buy isn’t concerned with reducing latency.
Another way to look at this would be a fast WiFi access point connected to a poor internet connection and trying to compete with someone online. You’d have part of the system optimized for speed, but it wouldn’t matter as the poor internet connection would have high latency (I.e., lag).
“Much shorter” is important here. There’s already an inherent lag in what we see, it’s about 80ms for a new image to register in the brain. If you introduce something that causes lag that’s longer than what the human brain can perceive, then you’ll notice lag. If it’s less than that, there is no perceivable lag.
Let’s got back to the Sony A9 and the lack of blackout. When you’re shooting with a DSLR, every time the mirror flips up, you’re losing visual perception of the item you’re photographing. That means you’re introducing a period of re-acquisition of a subject, at rates up to 12 times a second. There’s a much larger reduction in the ability to accurately capture a moment when that moment might be blacked out.
Now let’s look at your body. The typical reflex arc (the time it takes to cognitively recognize a stimuli, to react to it and the brain to update) is about .25.
If you’re shooting a flower, and you’re not catching “peak” movement when it’s blowing in the breeze, then there is something malfunctioning with your camera. While the X-T2 did not have a great refresh rate, and so would have been a poor choice for fast action, the X-T3 EVF lag time has been measured at 0.005-second and its refresh rate has been upgraded to 100fps.
Simply put, there’s no way that the X-T3’s EVF, with a .005 lag time is responsible for lack of capturing peak action in a flower in the breeze.
Of course, using an EVF opens up all sorts of other possibilities. You could capture that flower blowing in the breeze at 20fps, so you’d be sure to get peak action, and since there’s no mirror flip to break the AF cycle, there’s a greater chance of in-focus images.
You can also see the exposure as it will be captured in an EVF, so that when you’re photographing that flower, you know if you’re going to be over or under. (In either a DSLR or mirrorless, you won’t be able to evaluate the ring flash exposure.)
Having shot sports with an EVF and caught peak action quite often, I’m having some issues with the argument that an EVF will take some time to allow that to happen. Not every brand is at a lag-free EVF, but many, including Fujifilm with the X-T3 have lag down below the level of human perception.