DSLR cameras by design have some inherent flaws and limitations. Part of it has to do with the fact that SLR cameras were initially developed for film. When digital evolved, it was treated just like film and was housed in the same mechanical body. Aside from the circuitry required for a digital sensor and other electronics, new digital film media and the back LCD, the rest of the SLR components did not change. Same mechanical mirror, same pentaprism / optical viewfinder, same phase detection system for autofocus operation. While new technological advances eventually led to extending of features of these cameras (In-camera editing, HDR, GPS, WiFi, etc), DSLRs continued to stay bulky for a couple of reasons. First, the mirror inside DSLR cameras had to be the same in size as the digital sensor, taking up plenty of space. Second, the pentaprism that converts vertical rays to horizontal in the viewfinder also had to match the size of the mirror, making the top portion of DSLRs bulky.
Lastly, manufacturers wanted to keep existing lenses compatible with digital cameras, so that the transition from film to digital was not too costly or too limiting for the consumer. This meant that manufacturers also had to keep the “flange distance” (which is the distance between the camera mount and the film / sensor plane) the same between the two formats. Although smaller APS-C / DX sensors and lenses seemed like a great way to reduce the bulk of DSLR systems, the flange distance / compatibility concerns left them fairly large and heavy physically. 35mm eventually came back with modern full-frame digital sensors, so the mirror and pentaprism sizes again went back to what they were in film days. On one hand, keeping the flange distance the same allowed for maximum compatibility when mounting lenses between film, APS-C and full-frame DSLRs, without the need to re-design and re-market lenses for each format. On the other hand, DSLRs simply cannot go beyond their minimum size requirements and the presence of the mirror is what continues to make them so much more complex to build and support.
DSLR Camera Limitations
Due to the mirror dependency of DSLRs for “through the lens” viewing, they have the following limitations:
- Size and Bulk: the reflex system needs space for both the mirror and the prism, which means that DSLRs will always have a wider camera body and a protruding top. It also means that the viewfinder must be fixed in the same spot on every DSLR, in-line with the optical axis and digital sensor – basically, there is no other place to put it. As a result, most DSLRs have somewhat similar exterior looks.
- Weight: large size and bulk also translates to more weight. While most entry-level DSLRs have plastic bodies and internal components to make them lighter, the minimum height and depth issue to house the mirror + pentaprism/pentamirror means lots of wasted space that needs to be covered. In addition, it would be unwise to cover such a large area with a very thin layer of plastic just to make it appear smaller / lighter – the underlying idea of a DSLR is ruggedness, even on a basic body. On top of that, DSLR lenses are typically large and heavy (especially those with a full image circle that were created for 35mm film / full-frame), so a super light camera body would result in balance issues. In essence, it is the larger physical size of DSLR systems that directly affects the weight.
- Complex Mirror and Shutter Design: every actuation requires the mirror to move up and down to let the light pass through directly onto the sensor. This alone creates a number of issues:
- Mirror Slap: the most amount of noise you hear on SLR cameras comes from the mirror slapping up and down (the shutter is much quieter in comparison). This mirror slap results in loud noise and camera shake. Although manufacturers have been coming up with creative ways to reduce noise by slowing down the mirror movement (Nikon’s “Quiet” mode for example), it is still quite loud. Camera shake can also become an issue when shooting at long focal lengths and slow shutter speeds. Once again, DSLR manufacturers had to come up with features like “Mirror Lock-Up” and “Exposure Delay” to allow mirror to be lifted, then exposure taken after a set delay – all to reduce vibrations.
- Movement of Air: as the mirror flips up and down, it moves plenty of air inside the camera chamber. And with air, it also moves dust and other debris around, which eventually ends up on the camera sensor. Some people argue that their DSLR cameras are better suited for changing lenses than mirrorless cameras, because there is a mirror between the sensor and the mount. There might be some truth to that. However, what happens with that dust after the mirror moves inside the chamber? All that dust will obviously circulate inside the chamber. In my experience shooting with a number of different mirrorless cameras, I found them to be actually less prone to dust than any of my DSLRs.
- Frame Speed Limitation: while the modern mirror and shutter mechanisms are very impressive, they are limited by the physical speed at which the mirror flips up and down. When the Nikon D4 fires at 11 frames per second, the mirror literally goes up and down 11 times within each second, with the shutter opening and closing in between! It has to be a perfect synchronization of both the mirror and the shutter in order for it all to work. Take a look at the below video that shows this in slow motion (skip to 0:39):
- Now imagine this process at 15-20 times per second – that’s probably physically impossible.
- Expensive to Build and Support: the mirror mechanism is very complex and consists of dozens of different parts. Because of that, it is expensive to build and provide technical support if anything goes wrong. Disassembling a DSLR and replacing internal components can be very time consuming.
- No Live Preview: when looking through an optical viewfinder, it is impossible to see what the image is actually going to look like. You have to look at the camera meter (which can be fooled in some situations) and adjust the exposure accordingly.
- Secondary Mirror and Phase Detection Accuracy: you might already know that all DSLR cameras with phase detection autofocus system (more on this below) require a secondary mirror. I wrote about this in detail in my “how phase detection AF works” article. In short, part of the light that reaches the mirror ends up on the smaller secondary mirror that sits at a different angle than the primary mirror. The purpose of the secondary mirror is to pass the incoming light to phase detection sensors that are located on the bottom of the chamber. The problem with the secondary mirror, is that it has to be positioned at a perfect angle and distance for phase detection to work accurately. If there is even a slight deviation, it will result in missed focus. And even worse, the phase detection sensors and the secondary mirror have to stay perfectly parallel to each other. If they don’t, some autofocus points might be accurate, while others will constantly miss focus.
- Phase Detection and Lens Calibration Issues: the problem with the traditional DSLR phase detection system not only lies with the secondary mirror alignment issues, but also requires lenses to be properly calibrated. It becomes a two way game – precise focus requires perfect angle and distance of the secondary mirror to the phase detection sensors (as explained above), and requires a properly calibrated lens to the body. If you had autofocus accuracy problems with your lenses in the past, you might have had experience sending your gear to the manufacturer. Very often, support techs will ask the lens in question to be sent together with the camera body. If you wondered why before, now you have the answer – there are basically two places where things could potentially go wrong. If the technician adjusts your lens to their standard camera environment and your camera is slightly off, your issues might get even worse after such tuning. That’s why it is best to calibrate both the camera and the lens to resolve those discrepancies.
- Price: although manufacturers have gotten much more efficient over the years in terms of DSLR production, assembling the mirror mechanism is no easy task. Lots of moving components mean high precision assembly systems, the need for lubrication in areas where metal components rub against each other, etc. In turn, this all results in increased manufacturing costs. And it does not stop there – if anything goes wrong with the mirror mechanism, the manufacturer must repair or even potentially replace it, which is a very labor-intensive task.
Mirrorless to the Rescue?
With the rise of cameras without a mirror (hence the name “mirrorless”), most manufacturers have already realized that traditional DSLR systems are not going to be the driving force of camera sales in the future. It makes sense from the cost standpoint alone, but if we really look at the current innovation, where are we at with DSLRs? With each iteration of DSLRs, it seems like we are getting closer and closer to hit the wall of innovation. Autofocus performance and accuracy have already pretty much hit the wall. Processors are fast enough to crank HD videos at 60p. Just to keep the word out and sales going, camera manufacturers have been resorting to just re-branding the same camera under a new model name. What else IS there to add? GPS? WiFi? Instant Photo Sharing? More in-camera editing? Those are all great bells and whistles, but are they innovations that will truly drive future sales? I don’t think so.
Mirrorless cameras open up huge opportunities for innovation in the future and solve many of the problems of traditional DSLRs. Let’s go through each point above and discuss additional benefits of mirrorless cameras:
- Smaller Size / Bulk and Lighter Weight: removing the mirror and the pentaprism frees up a lot of space. This means that mirrorless cameras can be designed to be smaller, less bulky and lighter compared to DSLRs. With a shorter flange distance, the physical size of both the camera and the lens is reduced. This is especially true for APS-C size sensors (full-frame is tougher to address, as discussed further down in the article). No more wasted space, no need for extra ruggedness to give a feel of a bigger camera. Mirrorless cameras can be made much lighter than DSLRs.The rise of smartphones as compact cameras has taught us a very important lesson – convenience, small size and light weight can potentially overpower quality. The point and shoot sales are significantly down, because most people find their smartphones to be “good enough” for those snapshot moments. All smartphone manufacturers are currently pressing hard on camera features, because they want people to think that they are not just getting a phone, but also a great camera in a single compact package. And judging from the sales figures so far, it is clearly working – more and more people are embracing smartphones and leaving their point and shoot cameras behind. Simply put, smaller size and lighter weight in electronics win in today’s economy. We can observe the same trend in many other gadgets – thinner and lighter TVs, tablets instead of laptops, etc.
Hence, people will naturally go after lighter and more compact, especially if the quality is not compromised.
- No Mirror Mechanism: no more mirror flipping up and down means a lot of good things:
- Less Noise: no more mirror slap, just the click of the shutter is all you hear from the camera.
- Less Camera Shake: unlike the mirror in a DSLR, the shutter by itself does not produce a lot of vibrations, resulting in less camera shake.
- No Movement of Air: since there is nothing constantly moving inside the camera chamber, dust is less of an issue (at least in my experience).
- Easier to Clean: and if dust does end up on the sensor, cleaning mirrorless cameras is easier than DSLRs. You do not need a fully charged battery to lock up the mirror – the sensor is exposed once you dismount the lens. In addition, most mirrorless cameras do not have an opening under the mirror to house a phase detection sensor and other components, so there is very little chance for dust to circulate after the chamber + sensor are fully cleaned.
- Very Fast FPS Speed: having no mirror means that the capture rate (fps) does not have to be limited by the mirror speed. This means that mirrorless cameras could potentially capture images at much faster frame rates than 10-12 FPS we see today, with much less noise.
- Cheaper to Build and Support: less moving parts translate to lower cost of manufacturing and support for the manufacturer.
- Live Preview: with mirrorless, you can get a live preview of what you are about to capture – basically “what you see is what you get”. If you messed up White Balance, Saturation or Contrast, you will see it in live preview – whether in the EVF (see below) or the LCD.
- No Phase Detection / Secondary Mirror Alignment Issues: now that many of the modern mirrorless cameras are shipping with hybrid autofocus systems that utilize both phase and contrast detection autofocus, you do not have to worry about the alignment of phase detection and secondary mirror. On a number of new generation mirrorless cameras, the phase detection sensors are located on the actual sensor, which means that phase detection will never have to be calibrated for distance, since it sits on the same plane as the sensor that captures the image.
- Price: producing mirrorless cameras is much cheaper than producing DSLRs. As of today, most mirrorless camera manufacturers charge heavy premiums for their camera systems, because their overall costs are high. While the actual manufacturing costs are lower than DSLRs, companies have to spend plenty of R&D money on improving autofocus performance and other technologies like EVF. Plus, since mirrorless cameras are relatively new, companies have to increase their marketing budgets to educate people. Overtime, however, prices of mirrorless cameras will drop to lower levels than even entry-level DSLRs.
- Electronic Viewfider: now here comes the biggest strength of mirrorless cameras and the present + future innovation with it. Without a doubt, an EVF has huge advantages over OVF. While the current implementation of EVF might not be as robust and responsive as it should be (see below), it is just a matter of time before manufacturers fix that. Let’s go over some of the key benefits of EVF over OVF:
Information Overlay: with OVF, you never get to see more than some basic grids. There is some static information presented in the viewfinder, but it is always fixed and cannot be easily changed. With EVF, you can get any information you want displayed right inside the viewfinder – from live exposure data to histograms. Different warnings could be added, such as a warning for a potentially blurry shot.
Live Preview: the same live preview on the LCD can be shown inside the EVF.
Image Review: another key feature that you will never get in an OVF is image review. How cool would it be to see the image that you have just captured right inside the viewfinder? With OVF, you are forced to look at the LCD screen, which is a big pain in daylight conditions. People end up buying a Hoodman Loupe just to be able to see their LCD screen in daylight! With EVF, you never have to worry about this, since you could use the viewfinder for reviewing images instead.
Focus Peaking: if you don’t know what focus peaking is, check out this video on Youtube
- Basically, you can nail focus when performing manual focus without having to rely on your eyes. The area that is in focus gets painted with an overlay color of your choice and you can stop exactly where you want it to be. You would never be able to do this with an OVF in a DSLR.
- No More Viewfinder Coverage: with OVF, you typically get something like 95% viewfinder coverage, especially on lower-end DSLR models. This basically means that what you see in the viewfinder is about 5% smaller than what the camera will capture. With EVF, you no longer have this problem, because it will always be 100% viewfinder coverage, since what you see in the EVF is what the sensor will capture.
- Much Brighter Display: if the light conditions are poor, you cannot really see much through an OVF. Focusing with OVF in low light is also difficult, because you cannot really tell if the subject is in focus until you take the picture. With EVF, brightness levels can be “normalized”, so that you can see everything as if it was daylight. Some noise might be present, but it is still way better than trying to guess when looking through an OVF.
- Digital Zoom: this one is by far my most favorite feature! If you have used a Live View mode on your DSLR before, you know how helpful zooming in can be. With most modern DSLRs, you can zoom in to 100% and really nail focus. Well, with mirrorless cameras, this feature can be built right into the viewfinder! So imagine manually focusing with a lens, then zooming in to 100% right inside the viewfinder before you take the picture. A number of mirrorless cameras are already capable of doing this. It goes without saying that an OVF would never be able to zoom like that.
- Face / Eye Tracking: now we are moving to the coolest part of the EVF technology. Because the EVF shows what actually happens on the sensor, additional technologies for data analysis can be utilized to do very cool things, like face and even eye tracking! I am sure you have seen face tracking on point and shoot cameras, but if you take it a step further, you could have the camera automatically focus on the nearest eye of the person that you are photographing. How cool is that? Sony is already doing this on their new Sony A7/A7R cameras!
- Potentially unlimited focus points: as you already know, most DSLR cameras have a limited number of focus points that are distributed mostly around the center of the frame. While it works out in most situations, what do you do if you need to move the focus point to an extreme border of the frame? The only option is to focus and recompose, but that might not be always desirable, since you are also shifting the plane of focus. In addition, anything away from the center focus point is typically inaccurate and could result in “focus hunting”, where the camera struggles with AF acquisition and goes back and forth continuously. With mirrorless cameras and phase detection sensors placed directly on the imaging sensor, this limitation can be lifted. Contrast-detection is already possible anywhere in the imaging sensor, while on-sensor phase detection will eventually get to the point where focus points will be distributed all over the sensor.
- Subject Tracking and other Future Data Analysis: if things like face and eye tracking are possible with mirrorless cameras, you can only imagine what camera manufacturers will be able to do in the future. Imaging having a complex tracking system that intelligently combines sensor data with autofocus and uses it to track a given object, or subject in the frame. Even the top of the line DSLR cameras today have challenges with full subject tracking. If you have tried photographing birds in flight with a DSLR, tracking can get challenging, especially when the bird moves out of the focus point area, or when the light conditions are less than ideal. If data is analyzed on a pixel level and there is no real autofocus area to concentrate on, subject tracking could potentially get super advanced with mirrorless cameras.
- Eye damage: when looking through a viewfinder, one has to be extremely careful about photographing the sun, especially with long focal length lenses. With EVF, the image is projected through the sensor and there is no harm to your eyes.
Mirrorless Camera Limitations
We’ve gone over the many advantages of mirrorless cameras over DSLRs. Now let’s talk about some of their current limitations:
- EVF Lag: some of the current EVF implementations are not very responsive, resulting in considerable lag. While this is certainly a nuisance compared to OVF at the moment, it is a matter of time before that lag is eliminated. The latest EVFs are already much better than what they used to be before. But as EVF technologies evolve, the lag issue will be resolved completely.
- Continuous Autofocus / Subject Tracking: while contrast detect has already reached very impressive levels on mirrorless cameras, they are still very weak at continuous autofocus performance and subject tracking. This makes them pretty much unusable for wildlife and sports photography at the moment. However, with the rise of hybrid autofocus systems and their continuous development, we will soon start seeing mirrorless cameras with much better continuous autofocus capabilities. One of the reasons why mirrorless cameras have been slow in this department, is because most mirrorless systems are small and not well-suited to handle large telephoto lenses. So manufacturers have not been putting much of their R&D efforts into this specific area. Again, it is a matter of time until this is implemented on mirrorless cameras.
- Battery Life: another big disadvantage of mirrorless cameras at the moment. Providing power to LCD and EVF continuously takes a toll on the battery life, which is why most mirrorless cameras are rated at about 300 shots per battery charge. DSLRs are much more power efficient in comparison, typically in 800+ shot range per charge. While it is not a huge problem for typical camera use, it could be an issue for someone that travels and has very little access to power. Still, I believe that the battery issue is also something that will significantly improve in the future. Batteries will be more powerful and power-hungry LCD screens will be replaced with OLED and other much more efficient technologies.
- Red Dot Patterns: due to the very short flange distance, most mirrorless cameras suffer from a “red dot pattern” issue, which becomes clearly visible when shot with the sun in the frame at small apertures. Basically, light rays bounce back and forth between the sensor and the rear lens element, creating grid patterns of red (and sometimes other colors) in images. Unfortunately, there is no way around this limitation on all mirrorless cameras with a short flange distance, as discussed here.
- Strong EVF Contrast: most EVFs designed today have very strong, “boosted” contrast, similar to what we see on our TVs. As a result, you see a lot of blacks and whites, but very little gray shades (which help to understand how much dynamic range can be captured). While one could look at the histogram overlay in EVF, it is still a nuisance. Manufacturers will have to find ways to make EVFs display images more naturally.
As you can see, the list is rather short and I expect it to get even shorter within the next few years. I believe that all of the above issues are addressable and they will get better with each iteration of mirrorless cameras.
In summary, I would like to say that DSLRs simply have no way to compete with mirrorless in the future. I am not saying that everyone will be switching to smaller and lighter mirrorless cameras soon – no, we are still far from that point. However, it simply does not make sense for manufacturers like Nikon and Canon to continue investing into making DSLRs better, when the technology advantage is clearly with mirrorless. Below is what I believe what Nikon and Canon should do in the near future.
Nikon’s Mirrorless Future
Currently, Nikon has three different formats and two mounts:
- CX – Nikon CX mirrorless mount, cameras with 1″ sensors. Current camera line: Nikon 1 AW1, J3, S1, V2
- DX – Nikon F mount, APS-C sensors. Current camera line: Nikon D3200, D5300, D7100, D300s
- FX – Nikon F mount, 35mm full-frame sensors. Current camera line: Nikon D610, D800/D800E, D4
When everyone was going mirrorless, Nikon ended up creating a new mirrorless mount – CX with a small 1″ sensor. While the imaging and autofocus technology of Nikon 1 cameras is good and the overall system is fairly compact, the biggest issue is the small sensor size. With a 1″ sensor (which is much smaller than APS-C) as shown below, the Nikon 1 cameras simply cannot compete with APS-C in image quality, bokeh and dynamic range, just like APS-C cannot compete with full-frame, or full-frame cannot compete with medium format. Simply put, Nikon has a sensor size disadvantage with its CX / Nikon 1 system.