Autofocus Turns 50 in 2027, and the Camera That Started It Was a Konica

Japan observes a Camera Day on November 30, and the date comes from 1977, when Konishiroku Photo Industry put the Konica C35 AF on sale. It was the first mass-produced camera that focused itself, and the part doing the focusing was built by a Minneapolis company whose main businesses were thermostats, industrial controls, and defense electronics. Autofocus turns 50 in November 2027, and the anniversary belongs to a brand most photographers have quietly filed under "used bin."

Konica Shipped It First, and Japan Still Marks the Date

The C35 AF did not start as a technology showcase. Konica ran a market survey on why ordinary snapshots came back ruined, and according to Konica Minolta's corporate history, 36% of failed photographs failed on focus. Not exposure, not camera shake, not flash. More than a third of the pictures people paid to develop were soft because the photographer had guessed the distance wrong or done a poor job reading their viewfinder. Solve that one problem and you sell cameras to everybody who had given up on cameras.

The machine they built was modest. A Hexanon 38mm f/2.8 lens with four elements in three groups, a programmed leaf shutter with three speeds, a built-in flash, an automatic exposure system, and a focusing range that ran from 1.1 m to infinity. You wound the lever, pressed the button, and the camera worked out the distance on its own. There was no focus ring to turn and no rangefinder patch to line up.

In Japan it was nicknamed the Jaspin Konica. The slang is a compression of "just" and pinto, the Japanese word for focus, itself borrowed from the Dutch brandpunt, and it describes focus landing exactly where you aimed it. The camera was popular enough that the word outlived the product and settled into Japanese photographic vocabulary as a general term for a sharply focused frame. Konica sold roughly a million units in two years, and November 30 became a date on the Japanese calendar.

Konica had pulled this trick once already. The C35 EF of 1975, nicknamed the Pikkari Konica, was the first 35mm compact with an electronic flash built into the body, and it sold to the same audience for the same reason. The company's specialty was not building the most advanced camera in the room. It was removing the specific decision that kept ordinary people from getting a usable picture.

The Focusing Happened Above the Lens, Not Through It

Honeywell developed the autofocus module in the middle of the 1970s, called it Visitronic, and took out four patents on it. The company was not a stranger to photography. It had bought Heiland Research in 1954, sold Strobonar electronic flashes under its own name for years, and imported Asahi Optical's SLRs into the United States as Honeywell Pentax from around 1959 until Pentax set up its own American distribution in the mid-1970s. What Honeywell did not have was a camera of its own to put the module in, so it licensed the technology out.

Visitronic was passive, which means it emitted nothing and simply looked. Two small windows sat on the top plate, one on either side of the nameplate, viewing the same scene from slightly separated positions. The module compared the two views, slid one against the other until the patterns aligned, and converted the offset into a distance. Modern phase detection compares two displaced views the same way, but it splits light coming through the taking lens and measures how far the lens is from focus. Visitronic worked out how far away the subject was, entirely outside the camera's optical path. Cocking the shutter armed a spring-driven mechanism, and pressing the release let it walk the lens until the module was satisfied, at which point the shutter fired.

That measurement was happening on the roof of the camera, a couple of inches away from the lens, reading a scene from a different angle and through different glass than the film would ever see. Screw a filter onto the front and Visitronic had no idea it was there. Frame your subject off to one side and the module cheerfully measured the wall behind them. The camera was not focusing on your picture. It was focusing on a decent approximation of your picture, taken from nearby.

Rollei used the same module in the Rolleimat AF in 1980. Canon went the other direction in 1979, building an active infrared system that fired a beam at the subject and measured what bounced back, an approach most compact makers adopted over the following decade. Both approaches shared the same limitation: the sensor and the film were looking at the world from different seats.

Credit for inventing autofocus is messier than credit for shipping it. Leitz patented a run of autofocus and sensor designs between 1960 and 1973 and demonstrated a system called Correfot at Photokina in 1976, built into a Leicaflex SL2 body fitted with a servo motor on a 50mm lens and a pair of viewfinder LEDs to signal focus. Leitz never sold one. Leitz and Minolta also ran a technical cooperation agreement from 1972 to 1997, and whether the German patents were sold to Minolta, licensed, or simply shared is not something the public record settles cleanly. What is documented is that Konica put a working autofocus camera in stores and nobody else did.

Sonar, and Then a Detour Through the SLR

Polaroid attacked the same problem with sound. The SX-70 Sonar OneStep of 1978 mounted an ultrasonic transducer on the front of the folding body, above the lens, and did exactly what a submarine does. It sent out a pulse you could not hear, timed the echo, and converted the delay into a distance before driving the lens. That made it the first autofocus SLR anybody could buy, and it focused in total darkness, which no passive system could manage.

The failure modes tell you where the measurement lived. Shoot through a window and the pulse bounced off the glass. Shoot across a canyon and nothing came back at all. Polaroid eventually sold the ranging hardware as a standalone component, and its ultrasonic modules became fixtures in university robotics labs through the 1980s and 1990s, which is an unusual second career for a snapshot camera part.

Pentax moved the measurement behind the lens. The ME F of 1981 was the first SLR on the market with through-the-lens autofocus, its focus sensor sitting in the camera body and reading light that had already passed through the taking lens. The parallax problem was solved. The drive problem was not. The SMC Pentax AF 35-70mm f/2.8 built for it carried its own motor and its own set of four AAA batteries, which made it a heavy, expensive oddity, and it was the only lens on the mount that could autofocus. Pentax built the ME F until 1984.

Nikon's 1983 attempt was stranger still. The F3AF put the phase-detection module inside a detachable DX-1 viewfinder, so autofocus became an accessory you bolted to the top of a professional body, and Nikon built only two lenses for it, an 80mm f/2.8 and a 200mm f/3.5. There was one other way in. The TC-16 teleconverter, introduced alongside the camera, carried its own motor and focused by shifting its own glass rather than the lens in front of it, which handed autofocus to more than two dozen of Nikon's existing manual-focus primes. It stretched them to 1.6 times their focal length and cost a stop and a third of light. Chinon tried yet another route, hanging active infrared sensors on the lens itself. Everyone understood the destination. Nobody had worked out how to package it.

Why You Remember the Maxxum Instead

February 1985 is when the industry actually turned. Minolta shipped the Maxxum 7000, sold as the 7000 AF in Europe and the Alpha 7000 in Japan, and it was the first camera to put the autofocus sensor and the focusing motor inside the body and combine them with motorized film advance. Everything a photographer needed was in the machine. The lenses could be smaller, lighter, and cheaper, because the intelligence had moved out of them.

Minolta paid for that with a clean break. The Maxxum used a brand-new A-mount and abandoned the company's older manual-focus glass, which made a lot of Rokkor lenses instantly awkward. In exchange, Minolta got a mount designed from the beginning for autofocus rather than one patched to tolerate it, and a camera that behaved like a camera instead of an experiment with batteries hanging off it.

So the honest scoreboard reads like this. Konica built the first autofocus camera you could buy. Polaroid built the first autofocus SLR. Pentax built the first SLR that focused through its own lens. Minolta built the first one that made every other manufacturer rewrite its roadmap. Collective memory keeps the last name on the list, because being first matters far less than being the one everybody had to answer.

Honeywell Came to Collect

Honeywell filed suit in April 1987 against Minolta, arguing that the autofocus system it had built for itself read on Honeywell's patents. The case went to a jury in federal court in Newark, New Jersey. On February 7, 1992, the jury found infringement and awarded Honeywell $96.35 million. It was not a clean sweep. Jurors invalidated one of the patents at issue and declined to find the infringement willful, which capped what the award could have been.

The figure most people repeat is a different one. Weeks after the verdict, the two sides settled Honeywell v. Minolta for $127.5 million, a number that folded in the damages, pre-judgment interest, and a license letting Minolta keep building autofocus cameras. If you see $127.5 million described as a jury award, or $96 million described as what Minolta paid, both are wrong. One is the verdict and the other is the settlement, and they sit about a month and roughly $31 million apart.

Honeywell was not finished. That August the company announced a second round of settlements with seven more manufacturers, reported at a combined figure of roughly $124 million. Konica was on that list, which is a strange piece of symmetry: the company that had licensed Honeywell's module legitimately in 1977 ended up paying for what came after it.

Totals for the whole campaign are reported inconsistently and you should treat them as approximate. Camera historians generally put the industry-wide payout above $300 million. The law firm that tried the case for Honeywell puts it closer to $500 million. Honeywell's own filings from the period describe the settlements as producing gains in the hundreds of millions across 1992 and 1993. The precise ceiling is not settled, but the shape of it is: an American controls company that never sold a mass-market camera of its own extracted a fortune from the industry that did.

Fifty Years of Moving the Measurement Closer

Canon spent 1992 measuring something nobody else had thought to measure. The EOS 5, sold in North America as the A2E, watched where your eye was pointing inside the viewfinder and selected one of five focus points to match, which the Canon Camera Museum records as a first for a 35mm SLR. It only worked with the camera held horizontally, and it worked better for some eyes than others, but the ambition was right. The camera had started paying attention to the photographer.

Through the rest of the film era and all of the DSLR era, the viewfinder's focus sensor still lived somewhere other than the picture. Light hit the main mirror, bounced off a small secondary mirror behind it, and dropped into a dedicated autofocus module in the floor of the mirror box. That is through the lens, but it is not at the image. The module and the imaging plane could disagree, and calibrating that disagreement is exactly why front focus, back focus, and autofocus micro-adjustment menus became a normal part of owning a DSLR.

The fix was to stop using a separate sensor at all. Fujifilm got there first in a compact, announcing the FinePix F300EXR in July 2010 with phase-detection photosites built into the imaging sensor and a hybrid system that chose between phase and contrast detection. Nikon brought it to an interchangeable-lens camera in September 2011 with the Nikon 1 V1 and J1, which combined 73 phase-detection points with 135 contrast-detection points on a single chip. Canon pushed the idea to its conclusion in 2013 with Dual Pixel CMOS AF in the EOS 70D, splitting every imaging pixel into two photodiodes so the whole sensor was simultaneously the picture and the rangefinder.

Once the measurement lived inside the image, the interesting question stopped being where to focus and became what to focus on. Face detection came first, then eyes. Sony pushed real-time Eye AF to the a9 by firmware in 2019 and added animal eye detection to the same camera later that year. The current top bodies run trained subject-recognition models: the Sony a1 II, the Canon EOS R5 Mark II, and the Nikon Z9 will identify a person, a dog, a bird, or a car in the frame, lock the near eye, and hold it through a branch passing in front of the subject.

For portrait work, that quietly retired a technique. Focus and recompose existed because a center-weighted module forced it, and the whole ritual of locking on an eye, half-holding the button, and swinging the frame back into position was a workaround for measuring in the wrong place. Nailing the near eye wide open is now the camera's baseline job, not yours, which pushes the difficulty back into expression and lighting where it always belonged.

Fifty years separate two little windows on the roof of a Konica from a sensor that focuses with the same surface it exposes, and the entire route is one idea repeated with better parts each time: shorten the distance between where the camera measures and where the camera records. That distance is now zero.