Why Don't We Shoot Square Format and Crop Later?
Most cameras shoot rectangular images. That allows them to be rotated ninety degrees to achieve a different result. But in this world of digital imaging when at least some decisions can be deferred to posts processing, why don't digital cameras shoot square images? That way, we could crop horizontal or vertical, or both, after the fact.
First off, not all cameras are created equal. Hasselblad and other manufacturers have long made square format cameras. More recently, Lytro introduced their revolutionary Light Field digital camera that allows you to refocus its images after the fact. Its images are also square. But various rectangular formats have long been prevalent.
In the days of 35mm film, the standard 24 x 36 millimeter frame was standard for no other reason than the fact that it was popular. And of course it was popular because it was the standard. Originally, there were numerous proprietary film formats. But around the beginning of the twentieth century, a series of events featuring players including Thomas Edison George Eastman (of Eastman Kodak fame) and others adapted 35mm movie film stock for use in still cameras. This eventually led to the development of the Leica (a contraction, or technically a portmanteau, of "Leitz" and "Camera") 35mm system. And the rest, as they say, was history. Film cameras needed a standard format to allow for film and cameras to interoperate. And 35mm was as good as anything else. Nobody actually sat down and came up with this as a standard, but that's the way it turned out.
That all happened quite a few years ago, but for as long as film cameras ruled the day, the need for a standard kept this format as the standard. Not really anything more than that. Yes, there were a few ill-fated attempts to introduce new standards, but none ever really caught on the way the 35mm format did. At one point for a trip abroad, I bought an APS camera, a format that in hindsight seems more than a bit ironic given that its name stood for "Advanced Photo System." As anyone who still remembers this would agree, it was anything but that. But I digress. As digital photography came on the scene, APS died with few regretting its passing. 35mm film lives on albeit in a diminished role for most photographers, but APS film is gone.
But once digital cameras did become economically feasible, there wasn't much standardization in frame sizes. But all of them were small since the bigger the sensor was, the more it cost. So if camera makers wanted people to buy their products, they had to use small sensors. My first digital camera was a Nikon Coolpix 990 with a sensor measuring about half an inch across. By comparison, my film cameras of the same vintage all shot the same standard 35mm film, about an inch and a half across.
In a sense, one could argue that nobody really cared how big the sensor was, so long as the images that it produced were good, but photographers transitioning from film had certain expectations. Not only do larger sensors lead to better images, everything else being equal, changing the size of the sensor changes the coverage one achieves with a given focal length. Digital SLR's that accepted standard film SLR lenses produced cropped results due to the smaller sensor sizes. Nikon, Canon and others retained the same 1:1.5 aspect ratio in their attempts to live up to the expectations of transitioning film photographers, but making the sensor "full frame" was, for a long time, cost prohibitive.
There are still arguably other advantages of Nikon DX and similar smaller formats in terms of weight and other factors, but now that full frame cameras are becoming more affordable, it seems worth asking why we still stick with the same rectangular format we've had for over a hundred years now. After all, if digital sensors were square, there would no longer be any need flop your tripod head over to shoot vertical format images. Not only this, but photographers would be able to defer the choice of frame orientation until post processing just as we now can with white balance and other factors. With a square format camera, photographers could simply shoot and worry about cropping later, or even keep the image square if that were their preference. Maybe they could tell everyone they shot that square image on a Hasselblad.
Regardless, the square frame has a lot to be said for it. So why are we still shooting rectangular? Personally, I generally prefer the rectangular shape for a final result to the square since square images can often look somewhat static. But it sure would be great if I could decide after the fact. Then I could just keep on shooting away when in the field, not needing to remount the camera back and forth between horizontal and vertical all the time.
Companies such as Really Right Stuff and Kirk Enterprise Solutions have made a lot of money selling camera "L-plate" brackets that make switching from horizontal to vertical easier, but I doubt there's an evil conspiracy afoot to retain the rectangular frame solely to sell more L-plate brackets. Surely there's some other reason for the continued dominance of the rectangular camera sensor.
In the end, I think it comes down to two factors, both of which have been part of the calculus for sensor size all along.
First, photographers have certain expectations, and let's face it, all of us are simply used to shooting cameras that produce images that shape. Whether you started with film and moved over to digital, or you started photographing more recently and have shot digital from the start, you've probably used cameras that shot rectangular images, most likely with the same 1:1.5 aspect ratio that Leica standardized so long ago. There are exceptions of course, but odds are at least most of you fit this pattern.
And second, and almost certainly most importantly, it comes down to the familiar problem of economics. The same need to make cameras affordable that initially gave us nothing but small sensors still keeps us limited to rectangular ones. Think about it for a minute. To create a sensor big enough to crop both ways means a bigger sensor — quite a bit bigger. Since the standard 35mm "full frame" format is half again as long as it is wide, a square sensor big enough to crop to that format would need have that long dimension along both axes. If the result would need to be 24mm x 36mm, the starting sensor size would have to be 36mm x 36mm. This would allow you to crop to either 24 x 36 or 36 x 24 as desired. But by doing so, you would be lopping off the unwanted 12 x 36, thereby wasting a full third of the pixels you started with. So even if everything else were the same, making a sensor fifty percent bigger to accommodate that crop should cost half again as much to make. The realities of manufacturing though mean the cost would go up even more, but even if it were only a fifty percent cost increase, that's enough to create a potential cost advantage for the competition. How much more would you be willing to spend for the luxury of shooting a square format?
For the same reason that the cost of digital photography continues to drop even as megapixel counts keep going up, the economies of larger sensor production will almost certainly result in big, square sensors down the road. But for now, we can only dream. That is, unless square just doesn't fit your expectations for an ideal dream format.
Updated 9/16/2013 — Reader TH writes to remind me that "an equally annoying factor is having to adjust the polarizing filter (when used) each time the camera's orientation is changed." Excellent point.
Also, reader CT writes to say that even if the sensor "was made square the lens is round so the corners would just disapear. What we need is round sensors and crop to suit!" True indeed. Thanks for bringing that up. Reminds me of an article I posted back in 2006 entitled "Factoring in the Digital Crop Factor" where I bring up this very point. Every image is cropped. It only depends on how much and what the final shape is.
Updated 9/21/2013 — Reader HM adds another valuable point: "Also the mirror and the shutter must increase in size and especially the mirror would be problematic as it needs to flip away." Reader SB echoes this and goes further, reasoning that if the mirror were taller, the camera would likely be both taller itself and therefore heavier.