By Ked · June 2026
June 2026
The 28mm focal length has an unusual history in the Leica system. It was available in screwmount before the M camera existed, briefly contemplated for M-bayonet in the mid-1950s and then shelved, redeveloped as a fast M-mount lens before any M body had a frameline to show it, and shot by some of the most important photographers of the postwar era without dedicated viewfinder support. Carl Zeiss made their first 28mm lens in 1933 and then effectively abandoned the focal length for half a century. Canon beat everyone to a fast 28mm by more than a decade. The M-system frameline arrived in 1980, fifteen years after the fast lens, and twenty-five years after the slow one.
A focal length this central to contemporary photography took a surprisingly convoluted path to institutional recognition.
My interest in this history is personal. I currently shoot five 28mm lenses on M bodies: the Canon f/2.8 LTM from 1957 via adapter, the Elmarit-M ASPH in its current form, both versions of the Summicron-M f/2 ASPH, the original with its chintzy clip-on plastic shade and the Version II in Safari finish with the integrated metal hood, and the modern Summaron-M f/5.6, Leica's 2016 reissue of the 1955 optical formula in native M-bayonet mount. That range spans sixty-six years of 28mm design, from a lens that predates the M camera to one that deliberately reached back to the screwmount era. Writing this history was partly a way of understanding what I had accumulated and why.
Leica's involvement with 28mm predates the M camera by two decades. The Hektor 28mm f/6.3 was first publicly demonstrated at the Spring Fair of 1935 (the year of Leica's 10th anniversary) and produced in LTM (M39 screw mount) through 1952 in three variants: nickel finish (pre-1937), chrome (1936–1942), and a coated version (1950–1952). Approximately 46,404 units were made across the full run.
f/6.3 was not a market positioning decision: it was the aperture a first-generation 28mm wide-angle design could achieve with mid-1930s glass types and manufacturing. General-purpose 35mm film of the period ran at roughly ASA 25–40 by modern equivalents. At f/6.3 on ASA 25 film, a tripod was functionally needed for anything short of direct sun. Handheld available-light use was not easy. This was a consequence of where the optical technology stood, not evidence of intent to target a specific market. The Hektor was not a photojournalist's lens in the modern sense of run and gun, but only because no 28mm lens of the era could have been.
What it did establish is that Leica recognized the 28mm field of view as legitimate territory, and it arrived several years before any significant competing design. The production numbers are worth noting: the Hektor ran to approximately 46,404 units over its seventeen-year life, while the faster and better-corrected Summaron that replaced it managed only 6,228. The Hektor's longer run, slower aperture, and higher volume suggest a market that was buying into the focal length itself, whatever light conditions it could be used in, rather than demanding speed.
Carl Zeiss has a longer and stranger relationship with 28mm than most photographers realize.
In 1933, the year after the Contax I debuted, Zeiss introduced the Tessar 2.8cm f/8 for the Contax rangefinder bayonet mount. It is historically notable as the first 28mm lens developed for a 35mm camera, and its four-element, three-group Tessar formula was well-understood at the time. Approximately 6,450 units were produced for the Contax bayonet across the prewar period. There is a footnote: during the wartime years, Carl Zeiss Jena produced approximately 200 units of the Tessar 28mm in M39 Leica screwmount as part of trade arrangements, Germany exchanging camera equipment for strategic materials with Sweden and other neutral states. These are rare wartime oddities, not a product line.
There is, however, a critical limitation to the Contax Tessar 28mm that the production numbers tend to obscure: it was not rangefinder-coupled. Despite being sold for a rangefinder camera, focusing was by scale estimation only. f/8 plus uncoupled focus made it a bright-light or tripod instrument, 2.4 stops slower than Canon's 1951 f/3.5, and 3 stops slower than the Elmarit-M that would arrive in 1965. Canon's marketing compared their Serenar favorably against it, but they were comparing against a lens that had been out of production for years and had never been properly coupled in the first place.
After the war, Zeiss made no successor at 28mm for either the West German Contax IIa/IIIa (1950–1961) or any other rangefinder system. The postwar wide-angle lineup for Contax from Zeiss Oberkochen covered the Biogon 21mm f/4.5 (1954, rangefinder-coupled) and a redesigned Biogon 35mm f/2.8; nothing between them. Carl Zeiss Jena in East Germany produced a 25mm Topogon and a 35mm Biogon, but again nothing at 28mm. The gap left by the uncoupled prewar Tessar was not filled by any Zeiss product for nearly five decades.
The Zeiss return to 28mm came in 1994 with the Biogon T* 2.8/28 for the Contax G system, a passive autofocus rangefinder-adjacent camera produced by Kyocera under Zeiss license. It was discontinued in 2005 when Kyocera shut down the Contax G system entirely. For Leica M shooters, the relevant Zeiss 28mm arrived around 2004: the Zeiss Biogon T* 28mm f/2.8 ZM, manufactured by Cosina in Japan under Zeiss contract, with a 46mm filter thread and approximately 230g weight. It remains in production today.
While Zeiss was absent from 28mm for the postwar decade, Canon moved decisively into the space.
In October 1951, Canon released the Serenar 28mm f/3.5 in LTM, a six-element, four-group Gauss-type design. It was the fastest 28mm lens in the world at the time. Leitz's contemporary 28mm offering was the Hektor f/6.3, 1.7 stops slower. The Summaron f/5.6 would not arrive for four more years, and even that would still be 2/3 of a stop behind the Serenar. Canon identified the gap and moved into it. 28mm was simply not available at a usable aperture from any European manufacturer, and a faster 28mm opened up the field of view to anyone who wanted it, for travel, street work, interiors, group shots, anything where a standard 50mm felt too tight and a 35mm still not wide enough.
The Serenar was not a mass-market product. Canon made only 670 units through April 1956, plus a black-chrome follow-on run of approximately 200 units, priced at 27,000 yen, about 30% more than Canon's own 35mm f/2.8 of the same period. It was also produced in Contax/Nikon RF mount, confirming that Canon saw demand across the international rangefinder market.
A revised Canon Serenar 28mm f/3.5 arrived in January 1957. It was replaced in June 1957 by the Canon 28mm f/2.8 LTM, again six elements in four groups, priced at 33,000 yen. This is the Canon LTM 28mm most commonly encountered on the used market today, where it trades for roughly $300–$500 on M bodies via standard LTM-to-M adapter. Compact, rangefinder-coupled, and optically in the mid-century register, moderate sharpness, some vignetting, well-controlled distortion, it remains a valid working lens for film shooters who want the vintage character at a fraction of current Leica prices.
Nikon's rangefinder system of the same era used a proprietary Nikon S-mount incompatible with LTM, and Nikon produced no 28mm lens for the rangefinder market in any form that maps to M-mount use. Canon is the relevant vintage Japanese reference for 28mm on M.
In 1955, Leica introduced the Summaron 28mm f/5.6 in LTM, designed as a direct replacement for the Hektor, with a six-element, four-group symmetric formula that was sharper and better corrected. Approximately 6,228 units were made through 1963. At f/5.6 it was 1 stop faster than the Hektor but still 1 stop behind Canon's 1951 f/3.5 and 2 stops behind Canon's 1957 f/2.8. It was a meaningful improvement over the Hektor for available light but not a low-light tool by any reasonable standard.
In the spring 1956 issue of Leica Photography Magazine, Leica announced a bayonet M-mount version of the Summaron 28mm. It was never produced. One prototype is documented to exist; the production M-bayonet Summaron was cancelled before it reached the market. Leica users who wanted 28mm in the M system had to use the LTM Summaron with an adapter, functional but not yet native M.
The screwmount Summaron was discontinued in 1963, leaving Leica without any 28mm lens in production for the next two years.
In 1964/1965, Leica introduced the first M-bayonet 28mm with general purpose working aperture: the Elmarit-M 28mm f/2.8, code 11801. Nine elements, mixed German and Canadian production, 280g, engravings in red or yellow. At f/2.8, it was the first Leica 28mm that could be used in ordinary available light, indoors, at dusk, in shade, without a tripod. Whatever the photographer was shooting, the focal length was no longer a bright-weather-only proposition. Canon had demonstrated this was optically achievable more than a decade earlier.
One complication: the Version 1's rear element protruded far enough that it was not compatible with the Leica M5 (1971) or the Leica CL, both of which have internal metering mechanisms that the rear element would damage. When Leica redesigned the Elmarit-M in 1972, compatibility with the full M body range was part of the brief.
The lens also arrived with no dedicated 28mm framelines on any M body. The M4, introduced in 1967, activated the 90mm framelines when a 28mm lens was mounted. The M3 and M2 offered nothing at 28mm at all. The fast practical 28mm now existed in M-mount; the viewfinder infrastructure to support it would not arrive for fifteen more years.
Version 1 (code 11801, 1964/65–1972): 9 elements. The founding design, heavy and large. Not M5-compatible due to rear element protrusion. Collectors call this "the 9-element." Used examples trade at $1,800–$2,500 in good condition, the most valuable pre-ASPH Elmarit-M 28mm, made in modest numbers and prized for its rendering character.
Version 2 (code 11802, 1972–1979): 8 elements in 6 groups. Redesigned by Dr. Walter Mandler at Leica's Canadian facility in Midland, Ontario, the engineer responsible for most major M-lens redesigns of the 1970s. Redesigned specifically for M5 and CL compatibility, resolving the rear element clearance issue. Approximately 7,050 units produced.
Version 3 (code 11804, 1979–1993): 8 elements in 6 groups. Also a Mandler design, refining the V2 formula. This is the version that coincided with the M4-P's introduction and the arrival of 28mm framelines. It is the most commonly encountered pre-ASPH Elmarit-M on the used market, regularly available at $600–$900.
Version 4 (code 11809, 1993–2006). The final pre-aspherical generation.
ASPH (code 11606, October 2006). One aspherical element, actual focal length 28.5mm. Updated with metal hood as code 11677 in January 2016 with identical optical formula but refined mechanics, this is the version referred to when photographers discuss the "current Elmarit-M ASPH."
Between 1955 and 1980, no M-system body had a dedicated 28mm frameline. The focal length nevertheless built its reputation in this period through photographers who composed by feel, by the outer edges of the finder, or through shoe-mounted accessory viewfinders.
William Klein photographed New York between 1954 and 1956 using wide-angle lenses in the 28mm and 21mm range, working closer and more aggressively than any street photographer before him, using a LTM-mount Leica with no 28mm frameline. The resulting book, Life Is Good & Good for You in New York, published in 1956 and awarded the Prix Nadar in 1957, proved that the wide angle could distort and compress in ways that made the image more emotionally true, not less.
Garry Winogrand experimented with 21mm, 28mm, and 35mm in the early 1960s, rejected the 21mm for its distortion, and settled on 28mm as his primary focal length. He received his first Guggenheim Fellowship in 1964, the same year Leica introduced the Elmarit-M, and produced through the decade the work that would appear in The Animals (1969) and Public Relations (1977). He composed intuitively against a finder that showed no 28mm line.
Josef Koudelka documented the Soviet invasion of Czechoslovakia in August 1968 on an Exakta Varex with a 25mm Flektogon f/4, a wide-angle SLR setup, not a rangefinder, producing wide-angle frames of extraordinary compression from street level. Lee Friedlander, William Eggleston, and Bruce Gilden (28mm on a Leica M6 with a handheld flash, camera in one hand, Vivitar in the other) extended the aesthetic through the 1970s and 1980s. Daido Moriyama settled permanently on the focal length, ultimately through the fixed-28mm Ricoh GR.
The pattern is consistent: 28mm's legitimacy as a documentary and street focal length was established empirically, through published work, by photographers who used it without institutional support from the Leica viewfinder. When the frameline arrived in 1980, it was capitalizing on twenty-five years of proven practice.
The M4-P (1980) introduced 28mm framelines to the Leica M system for the first time, fifteen years after the Elmarit-M entered production and twenty-five years after the Summaron. The Minolta CLE, also introduced in 1980 with M-bayonet mount, carried 28mm framelines from the outset; the two arrived concurrently and independently. All previous Leica M bodies had omitted 28mm entirely: the M3 covered 50/90/135, the M2 covered 35/50/90, the M4 covered 35/50/90/135. The M4-P added 28mm and 75mm simultaneously, establishing the six-frameline set, paired as 28/90, 35/135, 50/75, that every subsequent M body has retained. Viewfinder magnification: 0.72x.
The M6 (1984) through the MP (2003), the complete film M lineup from the frameline's arrival to the digital era, all carry the same 28/90, 35/135, 50/75 frameline set at 0.72x. The 0.58x magnification option (introduced with the M6 TTL in 1998) shows the full 28mm frame comfortably for eyeglass wearers. The 0.85x finder, available across the M6, M6 TTL, M7, and MP, omits 28mm framelines entirely: the geometry cannot accommodate the wide frame at that magnification. Photographers shooting 28mm on any of these bodies need the 0.72x or 0.58x version.
The M8 (2006), the first digital M, introduced a complication: its APS-H sensor imposes a 1.33x crop, so a 28mm lens produces a field of view equivalent to approximately 37mm full-frame. Photographers wanting a true 28mm field of view on the M8 used a 21mm lens instead.
The M9 (2009) restored full-frame with its 36x24mm sensor, the first digital M on which a 28mm lens showed a 28mm field of view.
The M10 (2017) raised viewfinder magnification from 0.68x to 0.73x, the most significant finder change in the digital M era and a close match to the film-era 0.72x standard. The 28mm framelines, while sitting near the outer edges of the finder at any magnification, are noticeably more usable in the larger M10 finder than in the M9 or M240. The M11 (2022) carries the same 0.73x finder with no changes relevant to 28mm.
In 2000, Leica introduced the Summicron-M 28mm f/2 ASPH, the first fast compact 28mm for the M system, and the lens that established f/2 as an achievable aperture at this focal length in M-mount. The original version was capable but carried a clip-on plastic lens shade widely regarded as out of character with the rest of the M lens family.
Version II arrived in 2016 alongside the updated Elmarit-M ASPH, bringing a screw-on rectangular metal hood, reduced field curvature, and improved MTF across the frame (289g, 54mm length, 46mm filter thread).
Version III was announced October 26, 2023, available globally from November 30, 2023: 9 elements in 6 groups, one aspherical surface, 275g, 55mm length, 46mm filter diameter, minimum focus 0.4m in live view / 0.7m via rangefinder, at $5,295.
Leica produced the Summilux-M 28mm f/1.4 ASPH first as a limited run of 100 silver-finish examples in 2014 for the brand's centenary, before regular production began in 2015. Its optical design (10 elements with one aspherical surface in 7 groups) holds distortion to 1.1%, a figure that is genuinely unusual for a fast wide-angle design. Vignetting at f/1.4 is approximately 3.4 stops. Filter thread: E49 (49mm). It is the lens you choose for its unique rendering and maximum aperture. The cost and size alone dictate that it be a very deliberate choice.
In a decision that surprised much of the M-system community, Leica announced the modern Summaron-M 28mm f/5.6 on October 19, 2016 (code 11695, silver chrome), available to special order (full prepayment required, not kept in stock) at $2,495.
The optical design is completely unchanged from the 1955 LTM original: six elements in four groups, symmetric arrangement, no aspherical elements. Modern multi-coatings and 6-bit coding are applied; the mount is native M-bayonet. It is the most compact M-mount lens currently available, protruding less than 2cm when mounted, and the only current Leica M lens whose optical formula is taken directly from the screwmount era.
A matte black paint version (code 11928) followed in 2019 as a limited edition of 500 units worldwide, optically identical to the silver, with aperture engravings filled in red. Used black paint examples currently trade from $6,500 to over $10,000.
The modern Summaron-M completes an arc that spans nine decades: from the Hektor f/6.3 in the mid-1930s, through the LTM Summaron of 1955, the cancelled M-bayonet prototype of 1956, the four optical generations of the Elmarit-M, and the fast glass of the 2000s and 2010s. One of the most photographically productive focal lengths in 35mm history also happens to be the one that took Leica longest to make its own.
The Canon 28mm f/2.8 LTM (1957) is the Canon 28mm most useful to M shooters via standard LTM-to-M-bayonet adapter. Rangefinder coupling functions through the adapter for M cameras. Current used-market prices: approximately $300–$500. The Canon 28mm f/3.5 (revised January 1957) is sometimes described as the optically cleaner variant for those who can accept the slower aperture.
The Minolta M-Rokkor 28mm f/2.8, announced February 1981, was designed specifically for the Minolta CLE and is one of the more consequential 28mm lenses in M-mount history precisely because it was not made by Leica. Seven elements in five groups, native M-bayonet mount, 135g, 40.5mm filter thread, 0.8m minimum focus distance. At 135g it is substantially lighter than any Leica 28mm of the period (the Elmarit-M V3 weighed approximately 230g) and it is widely regarded as optically matching or exceeding the Leica equivalent in sharpness and distortion control.
Its reputation carries one significant caveat: the lens is prone to developing "white dot" degradation (sometimes called schneiderites), appearing as white spots that reduce contrast and flare resistance by allowing stray light to bounce around inside the lens. The problem is well documented and affects a meaningful proportion of surviving examples. Professionally cleaned samples remain functional but the issue recurs; buyers should inspect carefully or budget for servicing. Reliable examples currently trade at $450 and up.
The M-Rokkor functions on all M-mount bodies. On the Leica CL it can be mounted, but the CL has no 28mm framelines: the CL's finder covers only 40mm, 50mm, and 90mm, so composition at 28mm requires estimation. On the CLE, it triggers the dedicated 28mm framelines. On any Leica M from the M4-P onward it activates the 28/90 frameline pair normally.
The Zeiss Biogon T* 28mm f/2.8 ZM, manufactured by Cosina Japan under Zeiss contract and introduced around 2004, is the current Zeiss option for M-mount shooters. The near-symmetrical Biogon design delivers excellent edge-to-edge sharpness on film and on the M9/M10/M11 full-frame digital bodies. Like all near-symmetrical wide-angles, the rear element protrudes close to the sensor and projects light at a steep angle. On digital M bodies, this can produce cyan color shift in the corners: a documented issue, not a myth, though Leica's in-camera lens correction profiles (triggered by 6-bit coding) address it substantially. On film the ZM Biogon is a straightforward alternative to Leica's own 28mm glass at a price below the Elmarit-M ASPH. It remains in current production.
Cosina/Voigtlander has produced a more comprehensive M-mount 28mm range than any other third-party manufacturer, built up over more than twenty years:
The cultural meaning of 28mm shifted decisively in the 2010s, and not through anything Leica or Zeiss or Voigtlander did.
The original iPhone (2007) used a primary camera at approximately 37mm equivalent. Over the following years, as smartphone manufacturers optimized for indoor, close-quarters, and social photography, primary camera focal lengths drifted shorter. By the mid-2010s, the primary camera on leading smartphones had reached the 26–29mm range, the iPhone 6S at approximately 29mm, the Galaxy S7 at approximately 26.6mm. This was not a decision any user made; billions of people became daily users of a roughly 28mm field of view without ever choosing a focal length. The slightly expanded space, the environmental context around subjects, the sense of closeness without telephoto compression: this became the default grammar of contemporary photography.
Recent iPhone generations have moved the primary camera wider still, toward 24mm equivalent, with the iPhone 15 Pro and iPhone 16 both defaulting to 24mm. But the decade of smartphone 28mm dominance is the period when the aesthetic that Klein and Winogrand had staked out for documentary photography became genuinely universal. Working deliberately at 28mm with a rangefinder, in available light, at close range, is now defined against that background.
The Leica Q (2015) and Q2 are fixed-lens cameras built around a 28mm Summilux f/1.7, a separate optical design from the interchangeable Summilux-M, optimized for the Q's autofocus and leaf shutter system. The Q is not an M camera. But Leica's choice of 28mm for its generalist documentary compact is a clear statement about the focal length's current centrality, and the Q2 body-plus-lens at current market prices represents roughly the same value as a standalone Summilux-M 28mm f/1.4 ASPH.
All Leica M 28mm lenses tracked at usedlenstracker.com, including the Elmarit-M 28mm and third-party M-mount options.