When looking for digital cameras that are a step up from your basic point-and-shoots, the question of mirrorless vs. DSLR pops up fairly quickly, since the vast majority of digital cameras of this calibre are one of these two kinds.
Digital imaging is a fast-moving technology—and the trend shows no sign of abating. As the technology continues to develop, the problem of selecting just the right camera for your needs is something of a moving target.
In this article, I’ll attempt to chart a course through the maze of jargon and technical terminology so that you can make an informed decision on which type of camera to choose.
As cameras have moved from film to digital, some familiar features of their operation have been modified and improved but many of the essential parts have remained. However, recent advances in sensor technology are now poised to upset the applecart and force a re-think in camera design.
The dominance of the established DSLR is being challenged by newer ‘mirrorless’ cameras in a race to keep the public buying dedicated devices rather than slipping into smartphone photography.
The leading camera designs in this race are distinguished not so much by their size but by the presence or absence of an important camera component—the flipping mirror.
In order to understand how having or not having a mirror in your camera is so important, it’s necessary to explain how each one works.
The Sony RX100 Mk III mirrorless camera and the Canon EOS 50D APS-C sensor DSLR
What is a DSLR Camera?
DSLR stands for ‘Digital Single Lens Reflex’. Before the digital revolution, this design of camera would simply have been called an ‘SLR’ and it’s a design that neatly solved some of the problems experienced by early pioneers.
One of the problems with early film cameras was the difficulty in seeing beforehand exactly what you were about to record on film because only the film actually saw the image created by the lens.
Early Viewfinding Systems
The photographer had to rely on a viewfinder which showed a similar but not identical view via a small window on the front of the camera slightly to one side of the camera’s lens. Viewfinders helped frame the subject but they introduced parallax errors and couldn’t show if the subject was in focus.
One innovative solution to this problem was the addition of an extra identical lens—one through which the photographer peered, and one to form the image on the film. Both lenses were geared together so that focusing one would also focus its twin.
The twin lens camera showed what would be in focus but it was hardly a practical design for lens swapping and it still didn’t show you what the main lens was seeing, which was far from ideal when using filters.
The Defining SLR Feature: The Mirror
A quantum leap in the design was made when a certain Thomas Sutton patented a novel solution in 1861 that made it possible for photographers to at last compose their shots through the same lens that was about to take the picture.
Over a hundred years later, the basic idea is still incorporated into modern digital cameras almost unchanged. The most important components of the DSLR are shown in the simplified diagram below.
Light entering the lens and passing through the aperture control gets only as far as a mirror (shown in green) before being diverted at right angles and made to come to a focus on a ground glass focusing screen.
In early SLR designs, photographers would simply peer down onto this screen to view a laterally inverted image. These days, a pentaprism is positioned just above the screen and with the help of some eyepiece optics, the photographer can see the subject without any lateral inversion at what appears to be a comfortable viewing distance.
The film has long since been replaced by an electronic sensor (shown in red) that sits behind a mechanical focal plane shutter (not shown for simplicity) and, in this state, even if the shutter were to open, the sensor would still see only the back of the mirror. When the shutter button is pressed, the optical path is momentarily rearranged to look like this:
An electro-mechanical system flips the mirror up and out of the way of the light rays, simultaneously blocking the view through the eyepiece.
The focusing has already been done and since the distance from the mirror to the focusing screen was engineered to be the same as the distance from the mirror to the sensor for any given light ray, the image that finally falls on the sensor will have exactly the same focus as seen when composing the shot.
Once the shutter has exposed the sensor, it closes again ready for the next shot and a spring returns the mirror to its resting position.
As you can see from the diagrams, a DSLR needs a relatively cavernous void in which to swing the mirror and if it’s to take a series of exposures in rapid succession, it has to be very carefully designed to minimise noise and vibration and ensure a good level of reliability.
In addition to reading data from the sensor only after a photo had been taken, manufacturers realised that the DSLR could be made to stow the mirror away while continually reading the sensor.
This continuous stream of image data could then be used to make a video or just to show the view through the lens when composing a photograph. Canon called this mode ‘Live View’ and it’s now a common feature on many DSLRs—making the distinction between the two camera types somewhat less obvious at first.
What is a Mirrorless Camera?
Technically speaking, any camera without a built-in mirror and all the associated mechanical engineering could be classed as mirrorless and they generally tend to be much smaller and simpler than the the DSLR design.
The extra bulk of a DSLR and their clunky mechanical operation has long provided a niche for smaller, more discreet cameras that could be easily carried in a coat pocket.
A World War II ‘mirrorless’ spy camera
Today, the term ‘mirrorless camera’ evokes fewer memories of Box Brownies and more of recent smartphones innovations and the boundaries between Mirrorless and DSLR are not as clear-cut as they once were.
The addition of ‘Live View’ on many DSLR cameras means that they can be used in much the same way as a mirrorless camera (if the extra bulk is not an issue) but at the moment, it’s generally not so easy to use a mirrorless camera for all the tasks to which the DSLR is suited.
Modern mirrorless cameras have no optical viewfinder so the only way to see through the lens when composing a photograph, is to activate the sensor in a video read mode to update it in real time (or nearly real time) without saving anything to the camera’s media card.
Seeing a constantly updated image on the camera’s screen is an intuitive way to compose a photo that’s familiar to any smartphone camera user but as many will tell you, taking photos can quickly drain the battery.
On either a DSLR or a mirrorless camera, viewing the image on the camera’s LCD screen causes the camera to draw significantly more current from its battery, and the bigger and brighter the screen illumination, the more current will be drawn and the faster the battery will become exhausted.
This effect is aggravated by the fact that many people buy a mirrorless camera because they want a small pocketable camera but this places constraints on the battery size and capacity.
Initially, the digital mirrorless cameras on offer provided a fixed rear screen as the only way to view your scene or photos. This can present a problem when trying to see your image in strong sunlight.
Allowing the viewing screen to swing out and away from the back of the camera (so called ‘articulated screens’) can be useful in this regard but it’s mostly intended to allow you to see yourself when taking photos or video. It’s also a useful feature if you want to see what you’re photographing while, for example, holding the camera above your head in a crowd.
Another problem that irks some older photographers using mirrorless cameras is that they can be difficult to see up close if your vision is not perfect.
Users of DSLRs are used to a diopter adjustment in the eyepiece which can correct for a range of myopia, whereas looking at a rear LCD screen offers no such correction. Clearly, some people would benefit from an optional electronic viewfinder.
Fortunately, the resolution of many rear LCD screens is approaching 300 pixels per inch, which means that fitting a good quality mini-screen into a viewfinder complete with dioptric adjustment is now possible. With this innovation, yet another DSLR-exclusive feature is lost to the mirrorless camp.
In this Canon G1X mirrorless camera, the viewfinder is an optional extra which fits into the hot shoe. It incorporates eye detection that switches off the main screen (thereby prolonging battery life) and activates the viewfinder screen when it detects you’re peering into it.
The viewfinder accessory displays all the shooting information you’d expect to see in an optical display and more but at around £200 it’s not exactly an impulse purchase—even if it does rotate through 90 degrees.
Sometimes it’s useful to be able to use the rear articulated screen to compose a shot and sometimes it’s easier to use a more conventional eyepiece viewfinder. Holding the camera to your face affords greater stability than when holding it at arm’s length and it allows you see the image your camera is displaying even in bright sunlight.
If you want a mirrorless camera but don’t want to be restricted to only using the rear screen, think about buying a camera with a built-in viewfinder.
The Sony MX100 Mk III is a popular mirrorless camera that packs lots of high-end features including an articulated screen and pop-up electronic viewfinder into a small rugged case that’s truly easy to slip into your pocket. It’s not designed to replace a DSLR. No single camera design can satisfy every requirement.
It’s therefore important when choosing a camera, to be clear about what kind of photography you’re likely to do. To that end, we need to consider some further properties of DSLR vs. mirrorless designs.
Most camera sensors are basically the same. They’re an example of Very Large Scale Integration (VLSI) silicon chips using Complementary Metal Oxide Silicon (CMOS) technology fabricated to contain an array of millions of photo-receptive sites (each of which equates to a pixel in your final image).
These chips release charge carriers in response to incident photons and accumulate them on a capacitor whose voltage is directly proportional to the product of the light intensity and duration of exposure for that specific site.
Assuming the sensors use similar technology, there are two statistics that matter—the total number of pixels and how much surface area a single pixel has. These two figures determine the physical size of the sensor’s imaging area.
Different digital cameras have different sized sensors just as different film cameras used different sized negatives. When digital cameras hit the consumer market for the first time, there were two predominant film sizes—the 35mm roll film and the relatively new APS film cartridge.
Paradoxically, the APS (which stands for Advanced Photo System) was aimed at less advanced photographers or at least those who couldn’t be bothered with threading 35mm roll film into their camera while seated in some shady spot. The APS film came in a sealed cartridge that was quick and easy to change even in direct sunlight and the film was smaller as well. The area of an APS film negative was about 38% that of a 35mm negative.
When cameras first went digital, the state of integrated circuit fabrication was such that making an image sensor chip the size of a 35mm negative was a very difficult task with poor yields and high costs.
Smaller image sensors were more cost effective and their smaller size meant that the lenses used to form the image could be smaller, lighter and cheaper too. The market now offers a wide range of different cameras and the statistic that most people tend to compare as a yardstick to quality is the number of megapixels but this can be misleading.
Resolution and Pixel Area
Sony RX 100M3, Canon EOS 50D APS-C and 5D MkIII full-frame sensor sizes compared
The sensor used in a full-frame DSLR camera such as the Canon 5D Mark III crams 22 million pixels into the same surface area as a 35mm negative, which restricts each pixel to an area of just seven thousandths of a millimetre (seven microns) square. This sensor will produce a high resolution photo with 5760 pixels along the long edge—easily enough to make an excellent quality A3 print.
The Canon EOS 50D is another DSLR but one that uses the smaller APS-C sensor size of 22.3 x 14.9 mm. This fits in almost as many pixels—still enough to make a good A3 sized print but notice the pixel size. The area of a pixel on the full-frame sensor is over twice that of the APS-C camera’s sensor.
The larger the surface area of a pixel in the sensor, the more light it can capture and the bigger the signal produced for a given level of illumination. The bigger the signal your sensor produces in response to light, the less of a contribution the ever-present electrical noise makes to the end result.
Now take a look at the RX 100M3. This mirrorless camera is designed to be small and pocketable but this requires a smaller sensor. The resolution of this sensor is actually somewhere between the APS-C and the full frame DSLRs but in order to achieve this, the designers have had to reduce the size of the pixels to an area of just 2.6 square microns. An individual pixel on this camera will gather 18 times less light than a full frame DSLR.
Smaller sensors are not in themselves less sensitive than larger sensors of the same technology but they will not perform as well as larger sensors in low light situations. Smartphone sensors are even smaller than the RX 100M3 example above and they are notoriously bad at low light photography. Recent advances in sensor design have reduced the inherent noise in sensors but the greatest signal-to-noise gains are still to be found in having a large pixel area.
It’s generally true to say that mirrorless cameras have smaller sensors than DSLRs (and for this reason, are not so good in low light conditions). However, there is no law of physics to say this must be so – it has more to do with economics and marketing than science. The trend towards mirrorless cameras is gaining ground in the upper sectors of the market and over recent years, many retailers have noted a gradual decline in the sales of traditional DSLRs while smaller mirrorless camera sales are holding their ground in the face of a generation who think the camera on their phone is good enough.
Full frame mirrorless cameras such as the Sony A6300 have very good low light performance as far as the sensor is concerned but electronic viewfinders can struggle to produce a bright clean image in low light conditions. The conventional DSLR with its optical viewfinder still has a distinct advantage when shooting in low light as it does not have to wait to gather sufficient light to display on an LCD screen.
Autofocus – the Final Frontier
If your main interest is in studio portrait work or landscape photography, you’ll be used to manual focus. You have time to set up the shot and you can zoom in to magnify the image for pin-sharp focus.
Sports and wildlife photographers may well have a different perspective because fast action is difficult to predict and there is often no time to get a shallow depth-of-field and spot-on focus. For situations such as this, the autofocus system in the camera is of prime importance and although some striking developments have taken place in recent years, there are two still two main methods of autofocusing—contrast and phase detection.
Early mirrorless cameras or DSLRs using Live View had to autofocus by using a process called ‘contrast detection’. This works by analysing a portion of the image to see if there were any sharp transitions between light and dark.
An out-of-focus area will not exhibit high contrast–that is, the edges of objects will be blurred together, and there is less of a difference between light and dark tones.) But, as focus is attained, edge detail becomes visible until the contrast reaches a maximum. The contrast detection autofocus algorithm then seeks to move the lens to achieve the highest contrast possible in the area of interest.
The problem with this autofocus method is that the camera has no way of knowing which way to move the lens and so it may actually move the lens further out-of-focus before it brings it back into focus.
Furthermore, the lens has to be moved through the point of optimal focus and then back again in order to establish the maximum contrast point.
In practice, this causes the lens to ‘hunt’ for focus – making this method of auto-focus frustratingly slow – particularly if the lens is very defocused. In its favour, contrast detection is capable of achieving very accurate focusing.
If your mirrorless camera only has contrast detection autofocus and you want to shoot video, forget autofocus altogether and switch your camera to manual focus—you’ll get better results.
The other autofocus method you’ll find is called ‘phase detection’. This works by splitting rays of light from the object to be focused into two beams on opposite sides of the lens diameter.
The camera compares how the two beams line up on a special light sensitive strip – effectively creating a rangefinder system. In a DSLR, the optics and electronics for this system are housed at the bottom of the camera.
When the mirror is down in a DSLR, most of the light rays are diverted onto the focusing screen where you can see the image and manually focus it.
However, the autofocus system is housed in the lower regions of the camera. The centre of the main mirror is translucent in order to allow some light to pass through it. This light is reflected by two further mirrors onto a chip containing an arrangement of light sensors each corresponding to an autofocus point in your camera’s display.
When the main mirror is moved out of the way at the instant the photo is taken, the autofocus sensor can no longer see the scene. In situations where the the mirror is flipping up and down during continuous shooting mode, the camera’s computer attempts to predict where the subject will be for the next exposure.
The reason this extra complexity is justified is that unlike contrast detection, this method can determine which way to turn the focus ring on the lens to bring the intended part of the image into sharp focus. There is no hunting back and forth so autofocus is fast and responsive.
Careful design has enabled this combination of optics, mechanics and electronics to achieve autofocus that’s approaching the limits of its performance. For example, Canon’s EOS-1DX Mark II DSLR can autofocus while firing off sixteen exposures per second.
Impressive though it is, there comes a point where if you move things fast enough, they will break—rather like over-revving an engine—so reliability will be an issue if this technology is pushed much further. However, this is still the fastest method of autofocusing, which is why action photographers are still holding onto their top-end DSLRs for the time being.
Recent Mirrorless Developments
Mirrorless cameras have gained a reputation for slow autofocus over the years—particularly in low light conditions. However, their performance has been steadily improving. Initially limited to only contrast detection, advances in sensor design to incorporate phase detection autofocus have made them serious rivals for the DSLR.
Truly mirrorless cameras have to do everything using their image sensor so that means any autofocus detection of the phase detection variety has to be built into the image sensor. This is indeed the case with higher-end mirrorless cameras that now have a pattern of dedicated phase detection pixels built into their sensors.
This has enabled them to take advantage of the speed of phase detection autofocus and the fine tuning accuracy of contrast detection autofocus in a hybrid system that’s been delivering some impressive performance in cameras such as the Fuji X-T2 and the Olympus OM-D E-M1 Mark II.
Not Quite Mirrorless
The main advantage of a ‘mirrorless’ camera is not so much that it lacks a mirror but rather that it lacks a mechanical component that has to be swung up and down, often at high speed, between shots. Considerations of noise, vibration and reliability are all due to the movement of the mirror and not the mirror per se.
This realisation has led camera manufacturers to seek a third alternative instead of continuing to fuel the ‘mirrorless vs. DSLR’ debate. This, in turn, has led to attempts to engineer a camera which retains the advantages of the mirror and eliminates the drawbacks of its movement. Sony’s flagship full-frame Alpha A99 II is a good example of a camera that is built with this aim in mind.
This top-end camera is expensive, but the technology it incorporates clearly shows that the writing is on the wall for the DSLR. By using a fixed translucent mirror, Sony has eliminated the problems associated with a flapping mechanism while retaining the advantages of a separate dedicated autofocus system.
The Alpha A99 II’s ‘fixed pellicle mirror’ is a thin membrane that transmits around 70% of the light to the image sensor and 30% to the dedicated autofocus system. Since the optical path is not altered in the manner of the DSLR, the dedicated autofocus system can continue to accurately track focus during continuous burst shooting.
The main image sensor of this full-frame camera has 42 megapixels incorporating 399 phase detection autofocus points and a further 79 autofocus points on the dedicated autofocus sensor. This system enables the Alpha A99 to shoot at 12 fps.
Innovative solutions such as this one, however, are still very much the exception rather than the rule on the market and choosing between mirrorless and DSLR looks to be the norm for at least a little while longer.
As digital imaging technology continues its relentless march, making a camera purchase is like hitting a moving target. The market is in a state of flux. Not all camera manufacturers have embraced the concept of mirrorless cameras to the same extent and not all photographers are happy to embrace the changes either.
If you’ve already invested in a DSLR system and have a good selection of lenses, there’s no need to sell all your gear just yet but all the signs are that DSLRs are on their way out—it’s just a question of how fast they disappear.
Now that mirrorless cameras are on the verge of outperforming DSLRs in every area from landscape to sport, it would be wise to make plans to move away from the DSLR and look towards the range of zoom and prime lenses for this new generation of cameras is sufficiently extensive and the autofocus performance you need is within your price range.
A note from Josh, ExpertPhotography's Photographer-In-Chief:
Thank you for reading...
if you want to capture breathtaking images, without the frustration of a complicated camera.
It's my training video that will walk you how to use your camera's functions in just 10 minutes - for free!
I also offer video courses and ebooks covering the following subjects:
You could be just a few days away from finally understanding how to use your camera to take great photos!
Thanks again for reading our articles!