Olympus PEN-F and Canon 6D
Note: Here’s a followup to this post, “Relentless Mirrorless Advances”.
I’ve been an early adopter of mirrorless cameras, ever since I bought my Sony NEX 5, back in 2010. While the NEX started out as my smaller, secondary camera, that changed as I realized its benefits. Nowadays, while I still own a few DSLRs, it’s my mirrorless Olympus cameras that are my primaries. I’ve been bullish on mirrorless for a while, and while it took longer that I anticipated for mirrorless to hit critical mass, most people now admit the tide has definitely shifted.
Nikon’s DSLR sales are in decline. Canon is holding on, but not growing. The interest these days are with Sony, Olympus, Fujifilm and Panasonic — all of the mirrorless players. What’s telling is, at Precision Camera, the biggest camera store in Texas, they’ve moved the mirrorless camera displays up front, relegating the DSLRs to the back of the store. They too know that future is mirrorless.
Why did I know mirrorless will dominate? Several reasons. One, the smaller cameras are easier and lighter to carry. Two, mirrorless is more interactive, with EVFs giving real-time feedback for color, exposure and composition. Three, mirrorless is better for video. Finally, and most importantly, mirrorless improves at the speed of Moore’s Law. This last point needs clarification as it’s the crux of my argument.
What’s Moore’s Law and what does it have to do with mirrorless? Everything. Moore’s Law is why all computers, smartphones, smartwatches and everything else on the internet is getting faster and cheaper at an astonishing rate. The gist of the law says the number transistors on chips double every two years. Which means that the performance of devices using these chips are improving really quickly.
Mirrorless cameras are mostly computers. Other than the mechanical shutter (some even have options for an electronic shutter), everything else is run by chips. Therefore, they increase in capability and performance at Moore’s Law speeds. DSLRs, because of their mechanical mirrors and phase detect focusing, are more dictated by physical constraints. They don’t improve as quickly.
So that slow mirrorless focusing, that people complained about in 2010, speeds up every year. Now they are faster than DSLRs. Their frames per second also increase, now besting DSLRs. Those grainy and laggy EVFs, which people hated, become higher and higher definition with faster refresh. As of 2017, most everything people originally disliked about mirrorless has been fixed and is often superior to DSLRs. And the relentless speed of improvement continues. This means the gap between mirrorless and DSLRs will only continue to increase.
Today, the only advantage DSLRs may still have, is their focus tracking speed for fast action, like sports. And even this is arguable with the latest generation mirrorless cameras such as Olympus’ OM-D E-M1 Mark II. In a few years, tracking speed should probably tip to the mirrorless camp. But the news gets even better for mirrorless, because of Moore’s Law.
The big trend in the computer business is the use of AI (Artificial Intelligence). All of the big tech companies are using it to understand natural language, to analyzing images and to predicting relevance in social media posts, among other things. Smart applications can even turn photographs into “paintings” in your favorite artist’s style. This is all due to AI powered by Moore’s Law.
So imagine this. What if you tell your future camera what kind of images you prefer. You can train it by “Liking” images you took or by giving it examples of your favorite images. Your camera’s processor can then intelligently scan the scene, optimize and make it easier to take the kind of photos you like. In fact, it may even take it for you automatically. Yesterday’s face detection and smile detection is merely a primitive beginning of scene analysis.
So that last bastion of DSLR superiority is sports. But what if I can train my future mirrorless camera for the type of sports scenes I want. First, I can have it memorize the colors of the team’s uniforms and tell it to focus on my preferred team. And what if I can pre-program the best poses for a particular sport. The camera can start shooting 8K images at 60 frames per second and the camera will auto filter the best poses by sport, exposure, color and composition. This is all possible because of AI and Moore’s Law.
Well, you say, DSLRs can do that too, right? Not quite. Remember that legacy mirror that’s used to bounce the image to your optical view finder? That mirror gets in the way of the on board computer analyzing the scene. The DSLR is basically blind until you press the shutter. With mirrorless, the camera can constantly scan and analyze the incoming image data. Basically, the future mirrorless camera will be smarter and faster than most photographers via their AI. All you need to do is point the camera in the general direction.
DSLR users may say, if I use LiveView, wouldn’t that work the same? Perhaps, but at that point, you’re really not using the mirror are you? You have a legacy SLR design, with the baggage of a mirror, that gets in the way. Why not just get rid of the mirror?
I’m sure this future picture-taking technology will scare the traditional photographer. It’ll be like the car enthusiasts complaining about self driving cars, when those get perfected. But the good news is, you don’t need to use this photo technology, if you don’t want to. You can opt to use your traditional DSLR and do it the hard way. That’s basically what traditional Leica users are doing with their old-fashioned range finder cameras. Heck, you can continue to shoot film too. But I contend that for pro and amateur sports photographers, the future AI enabled mirrorless will run circles around the dumb DSLR. And for everything else, mirrorless is already superior.
The DSLR’s days are numbered. Long Live the DSLR.
Please support this blog by clicking on my Amazon Link before buying anything.