One of the biggest complaints we hear about from photographers today is lack of innovation by DSLR manufacturers. Given how far mirrorless cameras have gotten in the last few years with the electronic viewfinder (EVF) technology, it is a given that DSLRs are looking archaic in comparison, particularly when it comes to intelligent information overlays, manual focusing, focus peaking, EVF image playback and other important advancements that make mirrorless cameras not just joyful to use, but also very helpful in reducing focus issues. When using classic lenses such as the Noct 58mm f/1.2 on a DSLR, I personally find it quite frustrating that I have to switch to live view to try to nail focus with the camera at my arm’s length. Not only does that result in potential instability and undesired camera shake, but it takes me away from the optical viewfinder (OVF) and slows down the whole process. But what if there was a solution to the problem? What if DSLR manufacturers came up with a way to integrate an EVF into DSLRs and make both OVF and EVF possible? Sort of a “transitional DSLR” with both OVF and EVF capabilities. How cool would it be, if you could switch from an OVF to an EVF with just a single button? I have been thinking about this concept for a while and I think there is a way to implement this, if camera manufacturers are willing to be flexible and put some R&D resources towards such a project. It would certainly reduce the potential of mirrorless cameras taking a huge market share away from DSLR sales, which have only been declining in the past few years.
When it comes to mirrorless vs DSLR, we know that DSLRs have a physical limitation in terms of the flange distance (which is the distance between the mount and the sensor) – due to the presence of a mirror and the lenses which have been specifically designed for such flange distance. Therefore, DSLRs will always have the additional bulk, at least in terms of camera width, when compared to mirrorless cameras. Although at some point in the future the mechanical mirror and the pentaprism most likely will have to be abandoned completely (once EVF refresh rates and overall responsiveness get as good as OVF), providing both OVF and EVF capabilities at this point would be an ideal solution in my opinion.
DSLRs still reign supreme when it comes to fast phase detection autofocus, start up time, shutter response and the choice of native lenses available at the moment, particularly super telephoto lenses with exceptionally fast focus motors. If you don’t shoot sports and wildlife, you might have talked to friends who do, and you know they will laugh every time they hear the words “mirrorless” and “action” together – despite all the efforts to make mirrorless cameras fast in AF speed, we know that mirrorless has a long way to go not only in terms of improving AF speed and accuracy, but also in providing solid native-mount lens choices for professional needs. As of today, not a single mirrorless manufacturer offers anything professional-grade above 300mm without use of messy adapters.
So if DSLRs were modified to be able to switch between OVF and EVF, they would give the best of the two worlds in a single package. Sports, wildlife photographers and those who prefer OVF would continue enjoying the benefits of classic DSLRs, while everyone else would have the choice to switch to EVF if they desire. But how would it be possible to achieve such a task?
The first answer lies in the mirror mechanism. All current-generation DSLRs already have the capability to switch to live view mode, where the image is projected on the LCD directly from the image sensor. In this mode, the mirror goes up and stays locked up until the photographer leaves the live view mode. Leaving the mirror raised does not consume a lot of battery life – what drains the battery are the sensor and the LCD actively capturing the live data. So there is no concern with leaving the mirror locked up in this position – having been doing a lot of videography lately, I can say with confidence that practically every DSLR today can easily do this already!
Here is a diagram of a DSLR in Live View mode:
As you can see, light rays are passed right onto the sensor in this mode. Note that the pentaprism is blocked in this mode, as explained below.
The second answer lies in the pentaprism, which is what flips and mirrors what you see through the lens into the viewfinder. The moment the camera mirror is raised, the pentaprism goes dead, as the light is completely blocked by the mirror (as illustrated above). This is why OVF is only available for us before or after the capture – you cannot see a thing when the camera is in the process of capturing an image, or when the camera is used in live view mode. The below diagram shows the normal operation of a DSLR, where the mirror optically projects the image into the pentaprism:
Now the biggest issue lies in the implementation of an EVF with the pentaprism in place. In mirrorless cameras, there is no pentaprism, since there is nothing to optically project into a viewfinder. Instead, there is a small digital screen (similar to an LCD screen on the back of the camera) that sits inside the viewfinder. So how can a DSLR integrate a digital screen inside the same area where there is an optical viewfinder? This is where the biggest challenge lies – engineers will have to think of a way where to best put the digital screen. Personally, I would propose to put the screen on the opposite side of the pentaprism, as illustrated below:
When the pentaprism is blocked by the mirror, the OVF goes completely black. Putting an EVF on the opposite side of the pentaprism would project the image just like it is projected normally, except the screen would only turn on in live view mode, when the mirror is raised. In normal mode with the mirror down, the screen would go black, essentially serving as the back side of the mirror. Considering how thin and tiny OLED screens are today, adding one right behind that area should not be an issue and should not result in a bulkier top of DSLRs. If putting an OLED screen where I indicated has space or other constraints, it could also be placed on the top of the pentaprism. The image would have to be turned upside down, but it would technically work.
I thought about other ideas on where to put the EVF, but they are going to be much more difficult and technically challenging to implement. For example, another idea would be to place the OLED screen right in the focus screen under the pentaprism, but since the OLED screen cannot be made semi-translucent, it would have to be mechanically inserted in live view mode and taken out in normal mode. Not a good idea at all, since it would only increase the complexity and result in another mechanical component that might fail at some point.
The Results: No More Focusing Issues!
The idea of a “transitional” DSLR sounds really good to me personally. It would give us both OVF and EVF in a single package and provide innovative features that DSLR users are craving when they are looking at mirrorless cameras. It would allow us to use our native-mount lenses that we have acquired over the years. It would make manual focusing a breeze and not keep us away from other technological advancements we see today. In the meantime, DSLR manufacturers could take this concept a step further – add phase detection pixels on sensors for faster focusing in EVF mode (this technology is already available) and further work on increasing AF speed in such mode. In addition, it would be wonderful if DSLR cameras had the capability to self-calibrate the phase detection sensor – now that would be a game changer! If phase detection sensors are on the sensor, the camera could probe for focus in live view / EVF mode and when a subject is in focus, calibrate the phase detection sensor so that it focuses exactly the same way. This would address any focus issues that we presently see on DSLR cameras. And we could take this a step further in terms of lens calibration too – have the phase detection sensor automatically adjust focus on each lens too! Since the camera would know exactly where focus needs to be, as it is already projected on the phase detection sensors on the image sensor, all focus issues could be completely addressed with such a system.
Does this sound too good to be true? Would love to hear the thoughts of our readers!
I had the same idea, but had not thought it through as you have. It seems so obvious. Why haven’t Nikon and Canon done it?
Let the mirror goes downward in stead of upward
I think that Nikon has a patent out for an EVF assisted OVF. And Sony is on the right track here trying to integrate more and more into/onto the sensor. Eventually electronics on the sensor will make them more energyefficient in live-feed mode. Also patches of pixels could be controlled for an electronic shutter and on sensor HDR.
I personally liked the idea of hybrid vievfinders.
EVFs are used at professional and consumer video cameras for years. We see people using them for long hours of video shooting. They use both EVF and displays for different shooting conditions. EVF is not new to this industry. They evolved a lot and likely to replace OVFs.
I’m an engineer working on sales and marketing side for nearly 20 years and an enthusiastic photographer. What we should not forget about new technologies and products is that they are sales and profit driven rather than technical and science. Theoretically right products does not always sell or generate profit.
I personnaly think the products that gives the good feeling of shooting photos and video will win. The product may be weak on paper and specs but should have a good user experience. It’s like cars: The best seller is not always the one with the higher technology but the one with best driving experience. Electric cars are good but most people like to hear the engine roar.
To me, the right camera is the one that gives me the best shooting experience. Today, it’s a Nikon DSLR camera, I’ll see what it will be for the years to come.
I try to stay away from being gear based. There are very different options you can buy to shoot but stay focused on the photography itself rather than the gear :)
I think hybrids are a bad idea. (I think they’re kind of a bad idea for cars, too.) You get additional complexity, more weight, etc. for the dubious benefit of having an extra viewfinder system you probably don’t use. I think the optimal solution for Nikon is to release a mirrorless F-mount with a flange distance of 26mm (Leica’s M-mount) or less if that makes sense, but give it first rate support for legacy F-mount (i.e. a better version of the FT-1). People want two viewfinders? Buy two bodies.
Nikon’s DSLRs have ludicrous battery life advantages. If battery life improves such that mirrorless battery life becomes halfway decent, then DSLRs will be like the old match needle SLRs that had batteries that lasted for some unknown number of years (my FM-10’s battery lasted for about a decade). I do think DSLRs need on-sensor phase detect (you may recall I hallucinated that Nikon had put it in the D750 :-) ). They can tack on an external EVF to such a camera and voila, hybrid!
I think your idea of putting the OLED display on the front of the pentaprism is pretty ingenious. I just don’t think it’s a good idea.
If Nikon and Canon were really smart maybe they’d pick a <26mm flange distance and make it an FX-mount competitor to M43. Then you'd be able to use Canon, Nikon, and pretty much any other lens (including Leica, legacy Nikon/Canon, Sony, etc. etc.). This would be good for them and bad for Sony, et al. In any event lock-in lens mounts suck and they're something that has to go.
I’m since a long time convinced precisely this concept will be the joker Nikon & Canon will put on the table when the MILC market growth would force them to revise their current strategies. Also hybrid – X100/X-Pro1-kind of solutions – can’t be too difficult – including manual focussing aids like focus peaking or a zoom view. Do I need it today? Not really. The CSC-story is propagating through a lot of hyping and blogging – in particular ridiculizing the DSLR market like Fuji did with the caveman. Despite that, CSC-manufacturers are still implementing the kind of solid performance, features and accessory-line that is about mainstream in the DSLR market.
Think it’s been done on the not so successful Sony A350.
It was patented a few years earlier and uses a mirror to flip the optical viewing path between the pentaprism and an EVF display.
P.S. There are actually at least two DSLR+EVF concepts out there. The Sony A350 where the pentaprism is replaced by two ordinary mirrors, where the forward mirror tilts slightly to view an EVF display. The other concept uses an ordinary pentaprism and slides a mirror in front of the eye to view an EVF display.
These are great ideas. Eventually, we could have a programmable drone that follows us around and constantly shoots 360-degree video that can be refocused later, by computer of course, at 36+ MP. We’d never even have to touch a camera again. Heck, we could send them out, remotely, while we watched the Kardashians at home.
It’s amazing to me that our collective knowledge is increasing exponentially while, individually, we’re becoming stupid and lazy!
If the magic drone helps me get better pictures – bring it! :)
I know (hope) you’re kidding but even now people think it takes no skill to get a good photo. Photographers often complain about this perception while at the same time asking for newer and better technology that would help justify it.
LOL…I love it, best comment I have read in a long time being on the blogs and forums.
As others pointed out the proposed solution is not possible. The most simple and natural is combining the focus screen with a transparent OLED screen. They exist, and some versions doesn’t seem to darken the display more than a traditional focus screen.
Considering a dual optical and electronic view finder solution is more expensive than only one kind such will not come first, if ever, to APS-C consumer cameras.
Introducing it in the coming D5 and later let it trickle down to the other full frame cameras would be great. It would be quite disappointing if the D5 didn’t have some new hot selling point except a new 24MP sensor and expected upgrades like faster processing and improved AF.
I expect to soon see an image sensor and processor fast enough to provide a fully electronic shutter plus continuous, parallel, live feed to the EVF display.
A colleague and I were talking about what we would want to see in a camera and we both agree that a square sensor would be great. According to my math (I am not the best in math) it would result in 100sqmm more sensor surface (961sqmm vs 861sqmm), that’s around 12% more pixels.
Also, it would negate the problem of going from landscape to portrait with the camera – hard on the wrist, flash re-position is needed constantly and as we know, metering is off when on portrait (at least with Nikon).
A button that by pushing alternates between landscape and portrait frame (shown in red) would be perfect.
Now add that to the option of having analog and digital viewfinder (for the reasons you mentioned above) and we have a much better tool to work with.