There is an enormous amount of information out there about so-called “full-frame” sensors for cinematography. Some of it might leave you with more questions than answers. The terminology is sometimes less than clear (“full” what, exactly?) and explanations about what it is that makes full-frame sensors different—why people prefer them for some applications and avoid them for others—is sometimes presented in fragmented or needlessly convoluted ways.
So let’s take an overview at this “full-frame” concept:
It generally refers to an imager that has the dimensions 36mm by 24mm. The term’s origins harken back to the early days of 35mm film at the start of the 20th century when the motion picture photography format was essentially standardized as 35mm film moving through a camera’s optical gate vertically with each frame being four perforations tall and having a width almost the size of the space between perfs. The actual picture real estate of 35mm motion picture frames got a bit smaller for various reasons from the late ’20s (sound-on-film came along and space was needed for the soundtrack and the image area subsequently shrunk as blank space was added between the frames to create more of a rectangular aspect ratio) but that was for all intents and purposes the standard.
Meanwhile, in the early part of the 20th century, still photography, previously the domain of larger film formats like 4×5 or 8×10, saw the beginning of an exciting new “miniature” style of camera—the 35mm still camera of the type that came to dominate news, sports, street, a lot of glamour and other many other types of work throughout the 20th century and still exists today.
This is a useful summary of the evolution of that format from the start to the roughly century-old standard. This format also uses 35mm film, but it’s oriented to pass behind the lens horizontally. A still frame in this format was eight perforations wide and essentially as tall as the space between the sprockets. This 36×24 mm frame size is, to this day, what people refer to as “full-frame.”
“Full-frame” Cinematography Comes to Feature Films
In 1954, Paramount Pictures, responding to widescreen processes, cropping and a combination of the two)—that were themselves the result of television’s encroachment into what had been movie theaters’ monopoly on motion picture entertainment—invented a motion picture format called VistaVision. For those interested in a deep dive into the process and the studio’s expectations, visit www.widescreenmuseum.com.
Read more: Camera Sensor Sizes Explained: What You Need to Know (StudioBinder)
For this format, regular 35mm film stock ran through a gate horizontally, the way it does in a still camera, and at the same 24 frames per second that had been the standard for motion pictures since 1927.
The resulting frames had more definition and could be blown up to larger screen sizes than equivalent 4-perf frames with less degradation. The term “8-perf” was also used to refer to 35mm film oriented in this way, as opposed to the standard “4-perf” used for all other motion picture photography. Some major filmmakers used VistaVision for movies such as Vertigo (Alfred Hitchcock), The Searchers (John Ford), and Richard III (Laurence Olivier), but the format, which naturally devoured twice as much film negative as 4-perf 35mm was essentially retired in America less than a decade after its auspicious debut.
However, VistaVision cameras continued to find uses among visual effects creators because of the format’s higher resolution. In the photochemical days, in which every dupe (copy) of an image resulted in some loss of quality, it was worth the extra cost in film to create visual effects elements that could withstand the degrading effects that optical printing and multiple duping impose. Many visual effects and animation companies of the day had one or more VistaVision (“full-frame”) cameras on hand to shoot what were then ground-breaking visual effects shots.
Still Photography Ushers in Affordable “Full-Frame” Video
When still photography quickly moved to digital shooting in the last years of the 1990s, a lot of the first generations of digital still cameras such as this and this, though designed to pull market share from photographers used to shooting film in the 36×24 mm frame, had sensors that were not quite as large as the 8-perf, or “full-frame”, 35mm film format, which would have been prohibitively difficult and expensive to manufacture for the intended buyers.
These smaller sensor sizes instead were roughly the size of the film format called APS (Advanced Photo Systems . 25.1×16.7 mm). There’s a nice history of that format’s development from film to digital cameras here.
With these new and smaller sensors being used by more and more still photographers, there was some already some confusing documentation that spoke of focal length multiplier and crop factor, which we’ll get into shortly.
It wasn’t long before companies started releasing cameras with sensors that were “full-frame”, or, more specifically, had an image area of 36×24 mm just like 8-perf 35mm cameras so many amateurs and professionals were used to. Several manufacturers came out with these in around 2000, but the two that changed still photography were Canon’s EOS 1Ds in 2002 and Nikon’s D3.
The cameras had live video view on the LCD screen on the back of the camera body but it wasn’t until Canon added as an afterthought, a way of recording a clean video feed.
Some lower-end consumer cameras had a video function but the industry did not expect prosumers and professionals to use a still camera for video. The 5D MK II in 2008 changed all that. When people started shooting video seriously with DSLRs in that “full-frame”, there was a huge growth in interest in doing this and videos started cropping up everywhere from TV to YouTube and Vimeo, projects emerged with the visual characteristics that naturally this new, larger, full-frame image area.
Among the earliest of the still photography pros to make a point of exploring these new video options in their still cameras was Vincent Laforet and Philip Bloom, who looks back on that breakthrough here. Sony also broke through with its mirrorless still/video hybrid cameras starting with the A7, that actually led the way in the direction these hybrid cameras have taken and since Nikon did something DSLR shooters never thought they would and introduced their mirrorless “full-frame” Z7 and Canon EOS R.
The “Full-Frame” Look
So all those music videos and indie films shot in “Full-frame” on the Canon 5D MK II and subsequent cameras from Nikon and many other manufacturers had that kind of shallow DOF look that distinguished much of the work from imagery shot for 35mm film and even more so than images from 2/3″- and smaller chip-sized video cameras.
“Reverie” is perfect example of that look from Laforet, who for a time was among those early adopters who almost became synonymous with this “full-frame” DSLR approach to cinematography. This was a time where a lot of shots of models with one eye in focus, or tiny l portion of a brick wall or a single finger in a shot of a hand were in clear focus with the rest decidedly not. This was, of course, an aesthetic choice, not a technical necessity of shooting with a “full-frame” camera, although in some ways, the use of the larger sensor did have some influence on this approach to depth of field and a major signifier that a piece of video was shot with a “full-frame” sensor.
The Equivalency Conundrum
No, it’s not a lost Robert Ludlum manuscript. It’s a way of speaking about terminology that has been used over the years, which came about to simplify a concept but which has often made that co more confusing.
The longer a lens’s focal length, the shallower the DOF (all else being equal). In fact, all lenses of the same focal length (at the same aperture) possess the same DOF characteristics. There are factors that might somewhat mask that fact (the out-of-focus areas captured by a lens that isn’t very sharp at any point will not seem as obviously soft as they do on a lens that can get tack sharp when the subject is as very sharp). But a 50mm lens will have the same DOF whether it’s mounted in front of a “full-frame” sensor or an APSC-C sensor or a giant 65mm sensor or the 1/4″ imager still used in a lot of camcorders.
Sensor size determines image area or crop factor. Imagine you’re shooting a model in a bedroom. A 50mm lens on a “full-frame” sensor will take in more of the room than on an APS-C sensor from the identical camera position. If you shoot to an APS-C sensor with a 100mm lens, you’re cropping in much closer to the model’s face than if you’re using a 100mm lens on a “full-frame” sensor.
This is physics. All else being equal (frame size, aperture, distance from camera to subject) images shot on “full-frame” cameras will have shallower DOF—less will be in focus behind and in front of the subject.
People freely speak of “equivalent” focal length optics for different image areas. (A 35mm on the smaller APS-C sensor is the “equivalent” of a 50mm for a “full-frame” sensor). This is true in one sense: The field of view or crop factor will be the same—but in another way it’s completely untrue—the 50mm lens and the 35mm lens retain the exact same DOF characteristics regardless of sensor size.
A 35mm lens is a 35mm lens and retains the same DOF characteristics, whether it’s covering an APS-C or APS or “full-frame” or a 1/4″ chip camcorder any other sized imager.
To capture an image at the same angle of view or crop factor with a larger sensor, you will need a longer lens and with a smaller sensor, a shorter lens. Which will mean that material shot on a larger sensor will cause you to use a longer lens and so the resulting image will tend to have a shallower DOF.
Of course, this aesthetic isn’t “better”, it’s different. People shooting more documentary style footage where careful follow-focus wasn’t practical, tended to not want to give up their smaller sensor cameras which went with shorter focal lengths and therefore deeper focus (more in front of and behind the subject of a shot remains sharp).
To the extent that “full-frame” shooting does ultimately push shooters into using a focal length with less DOF, that wasn’t aesthetically better or worse. If you like the look of a model where just one eye is in sharp focus and want a lot of those kinds of shots, you may want to shoot “full-frame”. But if you’re shooting a documentary and really prefer to have almost everything in focus because you don’t have the luxury of pre-planning focus marks (or a focus puller for that matter), you might be better off using a camera with a smaller sensor and therefore, all else being equal, you’ll use lenses with shorter focal lengths and therefore have deeper focus.
Which is why you see so many varieties of cameras still being manufactured, with 1/3″ and 1/4″ and a wide assortment of differently-sized sensors. Here, Canon’s Larry Thorpe answers a number of questions about the market’s ongoing need for multiple sensor sizes.
Professional Shooters Embrace “Full Frame” and Beyond
Makers of high-end cinematography cameras—companies such as ARRI, RED and Panavision—also wanted to fill an increasing demand for “full-frame” cameras too but doing so for the pro presented a number of issues that DSLR makers didn’t need to worry about.
In order to actually process, move and store each individual frame, these early entrants into the “full-frame” arena (the DSLRs) had to make use of various types of compression to deal with throughput and storage limitations. The movie files used MPEG or other codecs but even before that compression, the vast amount of sensor date was reduced through processes including pixel binning and line skipping.
High-end cinematographers or even videographers needed higher-end gear so the introduction of higher end “full-frame” cameras took a while longer. Today, we have quite a few cameras in a very wide range of prices that can offer “full-frame” shooting without those issues.
The term “full-frame” is applied in professional cinematography to other sizes too, not just the 36×24 mm described above but also the “full” super 35 frame, which got its name from a once popular film format of just over 24.89 ×18.66 mm.
As the manufacturers of the very high-end cinema cameras have found solutions to the throughput and storage problems that limited the amount of picture information that could be recorded from these larger sensors, they’ve started to make use of the large sensors and attendant optical factors that go with the use of longer lenses.
Those cameras include:
- Sony VENICE
- Canon EOS C700 FF
- ARRI Alexa LF (It is called “large format” although the sensor size comes in very close to “full frame” at 36.70 x 25.54 mm.) This camera has been used on such groundbreaking features as 1917. Here, internationally-renowned cinematographer Roger Deakins, ASC, BSC explains why.
- And taking more of a jump the 40.96 x 21.60 mm RED Monstro Sensor, included in RED’s Weapon.
- Panavision’s Millennium DXL2
- And taking an even larger jump in size is the ARRI Alexa “65mm” sensor, which is an astounding 54.12 x 25.52 mm, based on the 5-perf, 65mm film format used for shooting many epic features, including Lawrence of Arabia and 2001: A Space Odyssey.
Here is an interview with Oscar-winning cinematographer, Emmanuel “Chivo” Lubezki AMC ASC about his reasons for using the Alexa 65 and the qualities the large sensor size brought to the aesthetics of Oscar-winning feature The Revenant, directed by Alejandro Iñárritu. The issue for him in this visceral man-vs.-wilderness drama was not the longer focal length/shallower DOF that is often associated with a the larger image area of a bigger sensor but, rather, the effect of capturing a wider image area with a longer lens that looks more like it does to the naked eye than it would using the wider angle lens that would capture the same angle of view.
The format, in other words, enabled the makers of The Revenant to get extremely close to lead actor Leonardo DiCaprio’s face while still being able to show a significant amount of the forest he inhabits.
Today’s shooter has a greater choice than ever before of sensor size. The designation of “full-frame”, while it primarily comes from the 36×24 mm 8-perf 35mm format, is frequently used to describe imager dimensions that are actually of a number of different sizes.
Cinematographers have made excellent use of some of the characteristic looks associated with “full-frame” sensors but those don’t make the pictures better, just different.
Articles in This Series