.jpg)
Every rangefinder camera made since 1954 has been, in one way or another, a Leica M: the same bayonet, or a copy of it, or the screw thread the M replaced. Except one. In September 1994 a Kyoto ceramics company that had bought a camera maker a decade earlier put out a titanium-shelled camera with a new bayonet, electrical contacts in the throat, a motor that wound the film and drove the lens, and a finder that looked like a rangefinder’s and worked like nothing anyone had sold before. It focused itself. It came with three lenses that carried the name Carl Zeiss and were built in Japan, and a fourth, a Hologon, that was built in Germany because nobody in Japan could. Joerg-Peter Rau’s line at Macfilos is the right epigraph: “When everyone thought the rangefinder had no future, the new Contax G System entered the stage.”
Eleven years later the company announced it was leaving cameras entirely, and the G went with it. This is the story of the last new rangefinder system, why it was built, why it was good, and why it could not have lasted.
How a ceramics company came to own Contax
The name is older than the camera. Zeiss Ikon of Dresden put it on its answer to the Leica in 1932 and made Contaxes until 1961; the Contax I, the Sonnars and the Kievs that continued the line in the Soviet Union are in this catalog under Contax rangefinder. When Zeiss Ikon gave up cameras the name sat idle until 1973, when Carl Zeiss licensed it to Yashica, a Japanese maker with electronic experience and an ambition to sell professional SLRs. Their first joint camera, the RTS of 1975, with ergonomics by Porsche Design, gave Yashica a prestige line; the arrangement was that Yashica made the bodies and Zeiss designed the lenses, most of them built in Japan under Zeiss’s supervision and all of them wearing the T* coating and the Zeiss name. In 1983 Yashica was absorbed by Kyocera, the Kyoto ceramics and electronics group, and the Contax licence went with it: “1983年ヤシカは京セラに吸収合併され、コンタックスブランドも同社へ引き継がれた.”
Kyocera was not a camera company in its bones, and that is part of the story’s end. But through the 1980s and early 1990s it did the licence proud. The RTS III of 1990 was the most elaborate manual-focus SLR anyone had made. The Contax T of 1984 and the T2 of 1991 put a Zeiss Sonnar in a titanium pocket camera and invented the luxury compact. And in 1993, Konica — not Kyocera — put out the Hexar AF, a fixed-lens autofocus camera in a rangefinder’s shape with a 35mm f/2 lens, which showed there were buyers for a camera that looked like an M and thought for itself.
What Kyocera built
The G1 was announced for Photokina 1994 and on sale that September at ¥143,000 — $2,210 with the 45mm in the United States the next spring. It was the first camera with interchangeable lenses and a rangefinder-style finder to be launched since the Leica M3 forty years earlier, and, as the Japanese trade put it, “レンズ交換式レンジファインダー機としては世界初となるAFカメラ” — the first AF camera among interchangeable-lens rangefinders. The shell was titanium in a champagne finish. The finder was a real-image optical finder that changed its magnification to match whatever lens was fitted, from 28 to 90mm, and corrected its own parallax as the lens focused, with an LCD below the frame for the distance and the shutter speed. Focusing was by phase detection through two windows in the front plate, an electronic version of the twin-window rangefinder — accurate in good light, slow and unsure in poor light and close up. A wheel on the top plate let you focus by hand, by wire, without ever seeing a rangefinder patch.
The lenses were the point. The launch trio was a Planar 45mm f/2 at ¥37,000, a Biogon 28mm f/2.8 at ¥57,000 and a Sonnar 90mm f/2.8 at ¥47,000, all T* coated, all designed by Zeiss in Oberkochen and built by Kyocera in Japan; the 45 would be measured by the Swedish laboratory Photodo as the sharpest lens it had ever tested for 35mm. The fourth, announced with the system and delivered later, was the Hologon 16mm f/8 at ¥280,000 — a five-element symmetrical ultra-wide at a fixed aperture, scale-focused, with a bayonet centre filter, and made in Germany, the one G lens that was. Four more followed: the Biogon 21mm f/2.8 (¥120,000) and Planar 35mm f/2 (¥47,000) in 1996, and, in 1999 at ¥125,000, the Vario-Sonnar 35–70mm f/3.5–5.6, the first zoom lens ever made for a rangefinder camera. The G1 needed a firmware update, marked by a green sticker in its film chamber, to drive the 21 and the 35, and could never drive the zoom: its mount had five contacts and the zoom needed seven.
The G2 of October 1996, at ¥163,000, was the camera the G1 should have been, launched for Zeiss’s 150th year. It added active infrared triangulation to the passive phase-detection so that the focus locked fast and in the dark, gave the finder a wider base, offered single and continuous focus, wound at four frames a second, raised the top speed to 1/6000 on aperture priority and sync to 1/200, moved the focus wheel to the front plate where a Dresden Contax’s had been, and carried the seven contacts the zoom would need. It was heavier, 560 grams, and ten millimetres deeper. On 10 September 1997 Kyocera sold a thousand in black, later called the Millennium Edition; a black-chrome version followed the black paint. The GD-2 data back that fitted it wrote exposure data between the frames. In America the G2 was $2,250 with the 45 in 1999.
What it was like
The G was, and is, a camera that divides people down a clean line, and the reviews on this site’s family pages show the line clearly. James Tocchio at Casual Photophile: “the G2 is what a modern rangefinder should be. Fast, accurate, incredibly capable” — point it and shoot. Michael Nguyen, writing for Japan Camera Hunter from Tokyo: “film rangefinder technology culminated in 1996 with the engineering marvel that is the Contax G2.” Ibraar Hussain kept one for thirteen years and called it the finest camera ever to see his mind’s eye — “lightning fast, locks instantly into focus and fires off like an AK47.” The Japanese shop columns say the G1 was a hit with landscape and snapshot photographers and that G1 owners traded up to the G2 in numbers.
And then the other line. Japanese Wikipedia’s article on the system carries the criticism in two clauses that could have been written in 1994: “オートフォーカスレンジファインダーと銘打たれたが、距離計式ではない” — marketed as an autofocus rangefinder, but not a rangefinder type — and, because the photographer never sees the two images meet, “焦点合わせの結果に不安がある,” there is unease about whether focus has been achieved. Every G user has the same story about the 90mm Sonnar: Drew Chambers’s G1 would not focus it reliably for portraits; Hussain’s G2 hit about half the time. Bob Janes loved his G1 and could not trust it — “the camera stops me — and for me that is an issue.” Matthew Thompson at Emulsive found the G2 “so close” and sold it, because when the electronics failed nobody would be able to fix them. The manual-focus wheel was, everyone agrees, a formality.
The professionals who had used Leicas for forty years were the ones Kyocera had hoped to win, and they were the ones who trusted it least. A Leica shows you focus; the G asks you to believe it. Rau’s judgement at Macfilos is that the system suffered from professional suspicion of autofocus through a rangefinder base, from fragile contacts, from high prices, and from the Leica M’s one unanswerable advantage: every lens made for it since 1954 still fits.
The ground moves
The G2 went on sale in 1996. The first consumer digital cameras were already in the shops; by 1999 Nikon had a professional digital SLR; by 2002 Kyocera’s own film cameras were, in Wikipedia’s dry phrase, “declining in sales.” Kyocera’s response was ambitious and late. The N system of 2000 gave Contax a new autofocus SLR mount; the Contax 645 of 1999 gave it medium format; and in 2002 the N Digital put a full-frame 24×36 mm CCD into a Contax body, one of the first cameras anywhere to do so — and it sold badly, ran slowly and ate batteries. The G series had no digital future at all: its lenses sat too close to the film for any sensor of the day, as anyone who has adapted a Biogon to a Sony has found.
By then the licence was the problem as much as the market. Zeiss had signed with Yashica for cameras; Kyocera was moving its resources into camera modules for mobile phones, and the Nikkei’s report of the withdrawal put its finger on it: the Contax contract “covered only camera-related products” and, it appeared, could not be applied to the modules Kyocera now wanted to make. On 10 March 2005 Kyocera announced it was leaving digital cameras; on 12 April it announced the end of Contax: production had already stopped in March, shipments of bodies, lenses and accessories would finish by the end of September, the 645 by December. The reason it gave was seven words long: “急激な市場の変化に対応できなかったため” — because it could not respond to the rapid changes in the market. Repairs were promised for up to ten years, and Japanese Wikipedia records the last date for the G, March 2015. A dealer’s retrospective on the exit has the epitaph: “時代が生み、時代に散った” — born of its era, scattered by it.
Carl Zeiss had already found its next partner. At Photokina 2004, seven months before Kyocera’s announcement, Zeiss showed the Zeiss Ikon rangefinder, built by Cosina on the bones of a Voigtländer Bessa, in the Leica M mount, with a new line of ZM lenses. The lesson of the G had been learned by the company that designed its lenses: if you want to sell Zeiss glass to rangefinder photographers, put it on the mount they already own.
What it left
No manufacturer has tried again. Jaron Schneider at PetaPixel, writing in August 2026 about why a G2 body now costs about $3,000: “it’s the only interchangeable lens, autofocus-equipped rangefinder system ever made by any brand.” The Sunrise Camera column says the same in Japanese — “オートフォーカスレンジファインダーカメラは、このコンタックスGシリーズの後が続かないジャンルとなりました” — a genre with no successor — and calls the G1 “perhaps the camera that most symbolises the 1990s.” Kyocera never published how many it made; nobody outside the company knows. Tens of thousands is the usual guess.
What survives is the glass, and the paradox. The Planar 45 and the Biogon 28 are among the best lenses of their focal lengths ever made for film, and they were bolted to the only mount in the world that cannot be repaired. So they are cut off their mounts: converted to Leica M by workshops in Japan and Germany, given helicoids, sold at twice their price to people who would never buy the camera they were made for. Rau’s counterfactual — “a digital G2 with those lenses would be a very attractive camera” — is the one every G owner has had, and Kyocera’s answer, in 2005, was to make camera modules for telephones. The camera that was supposed to be the future of the rangefinder was instead the last of them, and the M, which the G was built to replace, is still in production.