A 6x7cm negative measures 56×70mm — more than four times the surface area of a 35mm frame. That physical size difference changes how depth of field falls off, how grain reads at print size, and why medium format bodies from the 1970s sounded like heavy machinery on every shutter press. The S 67 profile in RfCamera rebuilds all three effects from the optics up.
The Physics of a Bigger Negative
Depth of field depends on three coupled variables rather than f-stop alone: the absolute focal length of the lens, the aperture's physical diameter, and the circle of confusion your final print or scan can tolerate before blur reads as blur. A 6x7 body needs a longer lens to match the field of view of a 35mm camera at the same shooting distance — Pentax shipped the 6x7 with a 105mm f/2.4 as its standard lens, roughly double the throw of a 50mm normal lens on 35mm film. Because depth of field narrows as focal length increases for a fixed field of view and aperture, a medium format frame shot at f/2.8 renders noticeably shallower separation between subject and background than a 35mm frame at the same f-stop and framing. The effect compounds with the larger circle of confusion permissible on the larger negative.
Asahi's Studio Workhorse: A Short History of the 6x7 Format
Asahi Optical Co. released the original Pentax 6x7 in 1969, building a medium format body that handled like an oversized 35mm SLR instead of a waist-level twin-lens reflex. That was the point: studio and fashion photographers used to working fast with an SLR could pick up a 6x7 and keep their existing shooting habits — eye-level viewfinder, interchangeable lenses, focal-plane shutter — while gaining the negative size that made large prints and detailed retouching possible. The mirror needed to clear a 56×70mm frame is large and heavy, and the shutter had to travel across the same distance. Both slap hard enough that Pentax and third parties sold mirror lock-up accessories, and studio photographers nicknamed the camera "the rice cooker" for its blocky pentaprism housing. Asahi revised the line in 1989 as the Pentax 67, adding mirror lock-up as a built-in feature and refining the shutter damping, then again in 1998 as the 67 II with a redesigned metering prism and faster flash sync. A roll of 120 film loaded into a 6x7 back yields ten exposures — a hard limit that shaped how photographers worked with the camera. You didn't burst through frames hoping one would land; you composed, checked the light, and released the shutter once you were sure. That discipline, as much as the negative size, is part of what people are chasing when they reach for a 6x7 simulation today. RfCamera's S 67 profile carries the optical signature forward; it can't force you to slow down, but the format's history is worth remembering while you shoot.
Grain, Tone, and Why Medium Format Looks Smoother
Photographers loaded 6x7 backs with the same stocks available in 35mm — Kodak Portra 400, Kodak Tri-X 400, Ilford HP5 Plus, Ilford FP4 Plus — because grain structure is a property of the film emulsion, not the format. What changes is how much that grain gets magnified before you see it. Printing a 35mm negative to 8x10 requires roughly 8x linear enlargement; the same print size from a 6x7 negative needs only about 3.6x. Half the enlargement factor means the same physical grain clumps occupy proportionally less of the final image, so a medium format print from Tri-X 400 looks noticeably finer-grained than a 35mm print from the same stock and same developer, even though the film itself is identical. Tonal transitions benefit the same way. A gradient across a sky or a face is recorded by the same density curve regardless of format, but with less enlargement there is less amplification of any banding or grain-driven texture within that gradient, so it reads as a smoother, more continuous slope from shadow to highlight. This is why medium format work is associated with "creamy" skies and skin tones rather than any special property of the lenses or film chemistry. RfCamera's S 67 profile widens the tonal falloff curve in its colour matrix to mimic that reduced-magnification smoothness and layers a lower-opacity grain plate — scanned from real film, not procedurally generated noise — so midtones stay soft without going flat.
Building the S 67 Profile: Shader, Colour Matrix, and On-Device Bake
Every RfCamera profile, including S 67, runs through one pipeline twice. In the live viewfinder, widgets/film_view.dart applies a colour matrix for the base tonal response, then a GLSL fragment shader (shaders/film.frag) handles the optical side — barrel distortion and chromatic aberration modeled on the lens family the profile references — before grain, light leak, dust, scanline, vignette, and date-stamp layers composite on top in real time on the GPU. For S 67, the barrel distortion coefficient is set lower than on the 35mm rangefinder profiles, matching the relatively flat field of the format's standard lenses, and the grain plate opacity is reduced to reflect the lower-magnification effect described above. When you press the shutter, none of that gets thrown away and recomputed differently: core/bake.dart replays the identical pipeline — same colour matrix, same shader parameters, same grain and vignette layers — inside a compute() isolate, so the JPEG that lands in your photo library matches what you framed in the viewfinder rather than approximating it after the fact. The isolate keeps that processing off the UI thread so the interface stays responsive while a full-resolution frame bakes. Everything happens on-device; the app requests no INTERNET permission and never uploads a frame anywhere. The panel below documents the exact parameters the S 67 profile applies before you touch a single slider — adjusting exposure or grain intensity in-app shifts these values proportionally rather than replacing them outright.
Shooting the S 67 Profile: Framing and Pace
The 6x7 aspect ratio sits between a square and the familiar 3:2 of 35mm — roughly 0.86:1, close enough to square that centered, symmetrical compositions read naturally, but with just enough rectangle to still support a clear foreground-to-background arrangement. Framing for it works best when you stop thinking in terms of horizontal or vertical orientation and instead think about what sits at the center third versus what fills the edges, because the near-square canvas doesn't reward the rule-of-thirds crop the way a 3:2 frame does. RfCamera's S 67 profile locks the capture ratio to match, so what you see in the viewfinder is what bakes into the file — no post-crop guessing. The ten-exposure roll that originally shaped this format is worth borrowing as a constraint even on a phone: shoot the S 67 profile in short, deliberate bursts rather than continuous frames, checking the light and the edges of your composition before each shutter press. Because depth of field runs shallower at matched framing, focus confirmation matters more than it does on wider-angle 35mm simulations — give the app's focus indicator a beat to settle before you shoot a subject closer than a few meters away. Portraits and still-life subjects show the format's character most clearly; wide landscapes work too, but the payoff of the format — that smooth, low-magnification tonal falloff — is most visible wherever your frame has a genuine near-to-far falloff for the depth of field to act on.
The Mirror Slap: Why We Modeled the Sound
Medium format SLRs earned a reputation for their shutter sound because the mechanism is proportionally larger than anything in a 35mm body. Clearing a 56×70mm frame means swinging a mirror with roughly 4.5 times the area of a 35mm mirror, and driving a focal-plane shutter curtain across the same wider gap. Both events happen in quick succession and both are loud: a low, resonant clack rather than the crisp click of a rangefinder or the softer clap of a compact 35mm SLR. Photographers who used these cameras in quiet rooms — churches, courtrooms, small studios — learned to time their shots around it, because the sound telegraphed the exposure to everyone nearby. RfCamera's S 67 profile includes that sound as part of the simulation, synthesized from the acoustic signature of the mechanism rather than borrowed from a generic shutter sample shared across profiles. It plays through the same on-device audio pipeline as every other camera profile in the app, with no network round-trip and no dependency on a sound file being downloaded after install — it ships in the app bundle. The sound isn't just a novelty layer on top of the image processing. It's part of what made shooting a 6x7 body a physical, felt experience rather than a purely visual one, and reproducing it honestly is part of reproducing the format rather than just applying a tonal filter over a modern sensor's output.
