A hundred and five years ago, in 1913, an asthmatic photographer who worked as an engineer for the Leitz optical factory in Wetzlar, Germany, seeking a less burdensome camera to carry, invented a handy little camera that was about as small as he could make it. It used 35mm movie film, so-called because its overall width, including the sprocket-holed edges, is 35mm. The film had the sprocket holes along both edges for registration in movie cameras and projectors, and, because it traveled a vertical path in both, each discrete image was 18x24mm, a 3:4 ratio. Barnack, who fitted his creation with a 40mm lens*, was constrained to use the same image area width of 24mm—but, because he used the film traveling horizontally in his camera, he was able to double the other dimension, resulting in the comparatively lavish larger image area of 24x36mm, a 2:3 ratio.
The first commercial Leica was produced eleven years later, in 1924, and for a number of years, "135" or "35mm," as it was called, was also known as "miniature" format. Most professionals and serious amateurs, who scorned it, at least used what we now call medium-format film, mostly 120 and 620, paper-backed film rolled around an open spool, and they often used sheet film. I worked for an older photographer in Maryland many years ago whose own older mentor had photographed assignments using a 4x5-inch Speed Graphic, sometimes taking only six sheets of film with him to cover an event. The skill of knowing which six exposures to make, and of making every shot tell, was considered an essential part of being a professional. With 35mm cassettes, not only could you load up your camera with up to 36 exposures at once, but you could also take along an extra cassette or two. Profligate and unnecessary!
During what later came to be known as the Korean War, a photographer named David Douglas Duncan discovered some remarkable screwmount lenses made in Japan by a lens manufacturing company called Nippon Kogaku, or Japan Optical. News of these lenses, which were impressively sharper (and cheaper) than German lenses, spread like wildfire among the war photographers in Korea and from there to the consumer culture in the West, and it put Nippon Kogaku—officially known since 1988 as Nikon—on the map.
As photographers became adept at composing within the overly long rectangle of 2:3, they used certain tricks, such as using a strong central vertical to break the space into two halves. Photo by Marc Riboud.
The 24x36mm image size, which by the 1950s was becoming accepted as a standard, was also widely known as a rather awkward rectangle. It suited landscape photos, but little else—it looked too tall in verticals, and made composition difficult, especially for photographers used to the square shape common in 120 cameras. Photographers learned various tricks to compensate—for example, using a strong center vertical element to break the shape up into two halves. But many people, especially committed amateurs, remained uncomfortable with the shape of the rectangle (called the "aspect ratio").
Nikon's first rangefinder camera, first called "The Nikon" then called the Nikon 1, of 1948 (now highly collectible with less than a thousand originally produced), had attempted to "correct" the shape by making cameras that made 24x32mm negatives and transparencies. It was a prettier shape—3:4 again—but it wasn't exported, and was quickly replaced, because slide mounts had already been standardized for 24x36.
Score one for 24x36mm.
The next possibility for moving away from the standard size happened in the 1960s, when "half frame" cameras enjoyed a brief vogue. Fitting twice as many exposures on a standard-length cassette of film, it appealed to pennypinchers. It was also the first time (to my knowledge anyway) that the word "frame," implying that there was a whole frame to be divided into half, was implicitly applied to the standard 35mm size.
Over the years that followed there were many attempts made by both camera and film manufacturers to leave 135 cassettes with 24x36mm image area behind—mainly led by Kodak as the leader of the consumer film industry. These included the highly successful 126 cassette for Instamatics (the result of research that revealed that the biggest problem consumers had with 35mm was loading and unloading their cameras properly—it was also, in my view, the product that convinced Kodak it could dictate the market, a big factor in the company's eventual almost-demise). It was introduced in 1963. The miniaturized version of 126, 110, came along in 1972; then the ill-fated Disc Film of 1982; and finally the grand finale, the big multi-company consortium push to APS in 1996. But every attempt to "smallen" 35mm failed. Mom and Dad Consumer were well satisfied for years with their point-and-shoots, 35mm color negative film, one-hour photo kiosks, and double 3R or 4R prints. Didn't need no steenking Discs, that's for sure.
In my view, the big problem with all these attempts to move away from 135 was that they only offered consumer products as alternatives—that is, they went downmarket from 35mm. When APS was in development, I was surveyed for my opinion and made the case to Kodak that if it wanted to establish the new film as a standard, it would also have to provide professionals and serious amateurs with an alternative that was better than 35mm—that is, balance the downmarket APS with an upmarket alternative that was better than 35mm. Rollfilm, which dated from way back in 1899, was kludgy and inconvenient and was initially designed for indoor studio use—handling it in bright sunlight could result in light leaks, although precision manufacture in later years minimized that problem. It was still far from ideal. A variety of experts proposed cassette-loaded, horizontally-traveling film closer in size to 645, with fewer "sprocket" holes (used for registration only) along only one side. But it never came to pass.
APS ended up doomed by two things: first, it solved producers' problems mainly, and didn't offer consumers enough advantages over 35mm—or rather, its advantages weren't intuitively obvious to consumers in 1996—and second, of course, digital was on the horizon by then.
When digital began to be viable in the late 1990s, I assumed, along with many others, that we had finally met the thing that was going to end the reign of 35mm film. Which of course it did...
...But what it didn't do, obviously, was end the reign of 24x36mm as the standard size of recording substrate! Larger "chips" (sensors) were difficult and expensive to manufacture at first, and most early consumer digital cameras had smaller sensors. For more serious cameras, APS Type C sensors were initially the norm. These made more sense than 24x36mm for many reasons—cost, but also better depth-of-field (D-o-F) characteristics and the fact that smaller lenses were required to cover the format.
There was also a "focal-length multiplying effect," whereby lenses of shorter focal length could show the same field of view (FOV) of longer lenses for 35mm. When Four-Thirds came along in 2006 (or so—I don't remember the date, and the Internet is surprisingly deficient in answering the question "when was Four-Thirds introduced?"), the focal-length multiplier of 2X indicated to me that it would be hugely popular. Although I didn't use long lenses myself, I had been professionally involved with "Ad-Ams" (Kodak's term for advanced amateurs) for many years, and I was aware how seriously many photographers coveted longer and longer lenses. I assumed photographers would flock to Four Thirds and never look back, glorying in that focal-length multiplier and the newfound accessibility of long-lens angles of view.
But it turned out that the reason Ad-Ams loved big teles was for prestige, not picturetaking. They were big, expensive, and impressive, and that's why people coveted them so much. When "bridge cameras," with their absurdly long telephoto reach, removed the high status from extremely narrow angles of view, it turned out most people didn't care all that much about telephoto reach after all.
I also assumed that people would like smaller sensors because they made it easier to get better (more!) D-o-F. All through the history of photography photographers had worked to get more "pan sharpness," that is, everything in the image in focus from front to back; the manifestos of the then-radical Group ƒ/64 in the early 1930s called for sharpness from foreground to background, for example. For the first time, that got stood on its head from the early- to mid-2000s when digital cameras with 135-sized sensors started to become accessible to well-heeled amateurs and hobbyists. Cameras with 24x36mm sensors gave noticeably less D-o-F than images from smaller sensors, and this emerged as a virtue, not a liability. Of course, it didn't help that the U.S. magazine Photo Techniques had, in 1997, published a series of articles introducing to photographers in the West the idea that out-of-focus blur (something the magazine termed "bokeh," the first time that spelling had been used, significant in that it allowed the spread of the concept to be tracked) had aesthetic properties that could function indivisibly in certain pictures. Outrage and ridicule was widespread at first. But that was before people started looking around for reasons to prefer the larger sensors that they actually wanted.
The very terminology became biased. Lens focal lengths, to cope with the hodgepodge of different sensor sizes, were already being given in "35mm equivalents," to the extent that many digicam lenses were marked in equivalent focal lengths—the focal lengths in 35mm terms the FOV of which they mimicked—rather than their actual, real focal lengths. Photographers might have no idea what the FOV of a 4.3mm lens on a 1/2.3" sensor might be, but they understood the FOV of a 24mm lens on 24x36mm, so the description "24mm equivalent" was descriptive for that 4.3mm lens. "Focal length multiplier," which described an advantage, was replaced with "crop factor," which implies an adulteration from completeness—to have a crop factor, you have to be "cropping"—cutting down or limiting or curtailing—something. What? Why, 24x36mm, of course. Then, when the simple concept of "35mm size" was replaced by the brainless moniker "full frame" (brainless because all deliberate standard formats are full frame—is a 4x5-inch contact print from a negative made with a 4x5-inch camera not the full frame?), it was an unmistakable signal that bias was afoot and roving the landscape. If 24x36mm is "full" sized, then it implicitly stands to reason just from the terminology that anything bigger is too much and anything smaller falls short. Support was again gathering for the substrate size that refused to die!
Of course, this mainly had to do with the leading cameramakers' investment in SLR technology for the format—and, for consumers/photographers, two things. The first was the familiarity they had with the FOV of long-familiar 35mm lens focal lengths. Cameras that used 24x36mm sensors returned photographers' favorite lenses to the FOVs they were already long comfortable with. And, of course, the second was prestige. Status is a very strong motivator in the photography hobby. "Full-frame" sensor cameras were bigger and more expensive, their "image quality" at least detectably ahead of that of smaller sensors. Full-frame, as it was now called, picked up in DSLR sales to dedicated photographers of all stripes, and (even though Nikon tried hard to resist) after a while it became comfortably ensconced.
For some time it looked very much like the amateur alternative, the newer mirrorless types of cameras, would be where "FF" would not gain a foothold. Micro 4/3 was announced to great fanfare and continuing popularity (eventually killing off regular 4/3), and most other camermakers' mirrorless products were APS-C or smaller for a while. But that was not to last. When Sony began its "throwing spaghetti at the wall" marketing strategy, trying this, that, and the other thing, what was the stickiest and stuck? Full-frame mirrorless ILCs. The A7 and its proliferating heirs became one of the most successful among Sony's more serious photographic offerings.
The September Revolution
Now, with the "September Revolution" of 2018 in the history books, it's official: full-frame mirrorless (FFM as I call it—not sure if anyone else does) has been anointed with great portent and fanfare as the wave of the (near) future for mirrorless. The so-called "Bigs," Canon and Nikon, never before serious about mirrorless in any significant way (probably hoping to choke the upstart with lack of oxygen), both announced FFM systems. Panasonic, probably chafing even more at longtime rival Sony's successes, did them one better and formed a Micro-4/3-like consortium with Leica and Sigma. Finally, as if to put a ! on the proceedings, Zeiss (Zeiss?!) released a quirky, newthink fixed-lens FFM—24x36mm sensor, check—and put a coda on this remarkable month. As in the famous story of the size of the Space Shuttle Booster being based on the width of two Roman warhorses' asses, "specifications live forever."
Of course it remains to be seen if the market for mirrorless will sustain all those newcomers, and which of the new systems might thrive and which wither. But the biggest winner is a foregone conclusion: it's Oskar Barnack's 24x36 standard for substrate size. It's now far removed from antediluvian adapted movie film with unnecessary vestigial sprocket holes along both edges. But it's still that ancient, awkward 2:3 rectangle. The fateful doubling of the 3:4 movie film ratio that old Oskar (d. 1936) settled on (who knows how) way back in 1913 is alive and well—and stronger than ever.
[This article, excerpted from the first draft ms. of my future book for my Patreon subscribers, is copyrighted. You can link to it but please do not post it on forums or republish it anywhere. —MJ.]
*The Ur-Leica, now very likely the world's most valuable camera (although that proposition has not been tested—it's a treasured artifact owned by Leica Camera AG) indeed has a 40mm lens, not 50mm, according to Malcolm Taylor, the British Leica technician who was entrusted by Leica to clean and service it.
Original contents copyright 2018 by Michael C. Johnston and/or the bylined author. All Rights Reserved. Links in this post may be to our affiliates; sales through affiliate links may benefit this site.
(To see all the comments, click on the "Comments" link below.)
Featured Comments from:
Hans Muus (partial comment): "Great article; look forward to your book! I seem to have read a couple of times, in a now distant past, that Barnack's invention was initially intended as a light metering instrument for movie-making: using the same film and processing as used for the movie proper and developing the test exposures quickly on the set."
Mike replies: I've heard that too, and I don't quite buy it, and I'll tell you why. The reason is, why would he double the frame size if it were merely for testing? Setting it up as what we now know as "half frame"—images the same size as the movie negative, 18x24mm—would have allowed more test exposures and fewer films to develop per test, and been the same resolution as the movie. The idea that the Ur-Leica was intended as a test device sounds like a "back-formation" to me—an idea someone came up with later to justify the little camera as being acceptably serious and purposeful. Of course I'm just guessing.
Jakub: "Nice article and bodes well for your book. Like many others (I'm sure) I caught your little joke about Photo Techniques not helping. From Wikipedia:
The English spelling bokeh was popularized in 1997 in Photo Techniques magazine, when Mike Johnston, the editor at the time, commissioned three papers on the topic for the March/April 1997 issue; he altered the spelling to suggest the correct pronunciation to English speakers, saying "it is properly pronounced with bo as in bone and ke as in Kenneth, with equal stress on either syllable."
Mike replies: The common pronunciation today is "BO-kuh," not "bo-keh" as described above. But as we learned with "Nikon" recently, whatever conventional pronunciation develops culturally becomes proper, so I'm cool with it.
It's worth remembering that the concept—that the out-of-focus areas in a photograph sometime contribute to the aesthetic whole—was received with a good deal of indignation and hostility at first. I got long handwritten letters admonishing me to pay attention to the sharp parts and not look at what you're not supposed to look at!
Gordon Lewis: "What you implied but not say explicitly is that the persistence of 24x36 in the digital world was motivated as much or more by legacy film lenses. Pros did not want to have to buy a whole new set of lenses for 'cropped' formats and didn't like having their expensive wide-angle zooms transformed into normal range zooms. Manufacturers, in turn, did not want to have to build out two competing lens lines. What I find ironic is that so many photographers will bitch and moan about a lack of 'innovation,' but when it's actually introduced (mirrorless m43, for example) they will fiercely stick to their tried-and-true 24x36 DSLRs."
Paul Christensen: "What a great photo by Marc Riboud! For those interested and in the neighborhood, there will be a retrospective of his work at the Suermondt Ludwig Museum in Aachen which opens this Friday. More information (unfortunately only in German) here."
Jim Richardson: "In many ways owning a FF camera performs the same function as knowing the secret handshake of an exclusive club: it lets you into the brotherhood."