The full frame image sensor format has fully hit the digital cinema world – now at multiple tiers of budget – so it’s probably a good time to clear some stuff up. There is a good bit of confusion and some slightly flawed “rules of thumb” that have been going around, but if you know what’s up and you’re willing to possibly do a little math, you will pick the right camera and lens combination for your next project. (If you’re looking for the TL;DR – scroll down to the section with the trendy clapping emojis.)
While full frame in high-end digital cinema cameras is a pretty recent development, it’s been around in one way or another for decades in digital stills cameras and over a century in film photography. In fact, if you’ve ever used a Canon 5D or a Sony A7, you have used a device with a “full frame” sensor.
The term Full Frame is a reference to the standard of the imaging size of 35mm film running horizontally parallel along the strip of film, as it does in standard 35mm stills cameras. Super 35 is based on a standard which uses the same exact film, but instead is rotated 90 degrees and the exposed frame area is perpendicular to the direction of the film and passed through the camera vertically. This results in a smaller area of exposure, as the height of the Full Frame (24mm) has become the width of Super 35, and the width of Full Frame (36mm) has become the height of approximately two frames of Super 35 (18mm high each).
These formats evolved as a standard because there are limitations to manufacturing and distributing a physical item (film stock) that are less of a concern with a digital sensor. It didn’t make sense to have dozens of companies making all sorts of different sized film for different cameras with proprietary mechanisms that only worked with a specific size/type of film. It’s much more economical to create a standard that is used widely.
Today, the sensor is specifically manufactured for the camera, so there’s no specific limitations or standards to the size and shape of the imager as there was with film. The recorded end product can be screened on any number of displays, so it opens up the playing field of formats and aspect ratios, all of which then makes it even more difficult to standardize. While it is a helpful frame of reference, you’d have a tough time finding a digital sensor that is the exact same size as the physical 35mm film standards
For the sake of simplicity going forward in this article, we will consider a “Full Frame” sensor to be 36mm (w) x 24mm (h), and a “Super 35” sensor to be 24mm (w) x 18mm (h). This is not accurate for all cameras, but using round numbers makes it easier to illustrate the concept. Of course each camera has multiple modes of recording that each use different parts of the sensor with different aspect ratios and resolutions, and some sensors are Full Frame width but shaped with a different aspect ratio (like the RED Monstro, or Sony FX9), but the core of understanding the compatibility of lenses and cameras with different sized sensors comes down to understanding a couple different physical relationships between them.
Full Frame Coverage
The first question we will address is “Does (x) lens cover (x) sensor?”. There are two main variables here. One is the actual size of the sensor (in mm, diagonally), and the other is the “image circle” of the lens. The image circle is a measurement that describes the diameter of usable focused light at the lens’ backfocal distance. For example, a PL mount lens is designed so that the mount sits precisely 52mm from the sensor. If a given PL mount lens’ image circle is 34mm, that means that light comes through the lens, and when it exits the rear of the lens focused, and hits the sensor, the physical diameter of the circle of light created will be 34mm. So for a lens to “cover” a sensor, the lens’ image circle has to be at least as large as the sensor is, diagonally.
In this instance, if we put this lens on a Super 35 camera, that camera has a sensor that is 30mm diagonally. So, this lens will cover the sensor just fine. Our Full Frame sensor, however, is about 43mm diagonally, so the corners will not be covered by the light coming out of the lens. This is what causes vignetting. This is illustrated in the images below. In this example, green is the image circle, light green is the part of the image circle that covers the sensor, and grey is the part of the sensor that falls outside of the image circle. You can see that the same image circle falls short of the diagonal size of the Full Frame sensor, but illuminates the Super 35 sensor fully.
Most of the time, however, it is not a hard line that cuts off the light. Some lenses will have a soft roll-off into a vignette that is acceptable or in some cases, even pleasant and appropriate for the subject matter. Also, there are some times when a lens will illuminate the full sensor but was designed with a smaller sensor in mind. Lens design is a series of compromises, and usually the area of light that falls outside of the intended imager area will have aberrations that remain uncorrected, even though they do form an image. This is simply because correcting them for a larger imager would have been costly and unnecessary for the intended design. Again, like vignetting, for certain purposes these aberrations and imperfections can lend character to the image and open up plenty of new creative potential.
Conversely, something like the Cooke S7 series, which is designed for Full Frame imagers, has some softness in the edges built into the optical design as part of the “Cooke Look”. If you put this lens on a smaller sensor than was intended, you will be using just the sharpest and most corrected part of the image circle. This can be good or bad, but is a creative choice that needs to be considered.
This is where vintage full frame stills lenses will shine, as they have been designed to expose a full frame imager. There will still be some issues that would have been corrected had they been designed as cine lenses (such as chromatic aberration, focus breathing, and parfocality with zooms), but often a properly rehoused stills lens will produce a high-performing image with great aesthetic qualities at a lower price point than possible when using lenses that were designed from cinema use from the ground up.
One concept everyone needs to start ignoring entirely is referring to a lens as covering a certain resolution. There is no such thing as a lens that “covers” 4K, 6K, or good old 480. There is no direct correlation between image resolution, image circle, and sensor size. The resolution is a product of the size of the sensor in relation to the size and density of its photosites. If you have larger photosites on one sensor than another, but the dimensions of the sensor are the same, the one with larger photosites will have a lower resolution. This doesn’t mean lower image quality. A lens can only “cover” a physical area, and the physical area of a sensor is not directly related to the number of pixels it creates. The only two factors to truly consider are the image circle and the sensor size.
How is Shooting Full Frame Compared to Super 35?
The other big question is “how is shooting full frame different than shooting Super 35?”.
Simply put, a lens with a given focal length will have a wider field of view on a Full Frame (larger) sensor than on a Super 35 (smaller) sensor. Thereby, on a Full Frame sensor, you will need a longer focal length than you would on a super 35 sensor to accomplish the same framing. By using a longer focal length, even if the size of the subject in your frame is the same, you will have a more shallow depth of field. This is what all the 1st Assistant Camera technicians are upset about!
PSA: Plan to add rangefinders and high performance wireless follow focus systems to your camera orders if you are choosing Full Frame. Pulling focus is significantly more difficult with larger sensors and ACs deserve to have the tools necessary to do a good job with this more demanding format!
In the following example we have two images. The first is a 55mm lens shot with a Super35 sensor. The next image is also a 55mm lens, at the same exact distance from the subject, but on a Full Frame sensor. You can see that the second example has a significantly wider field of view. The only thing that has changed between the two is the physical size of the sensor that captured the image. The sensor, by nature of being larger, is simply capturing a larger portion of the lens’ image circle.
In this next example, the constant factor between the two images is the field of view. In order to maintain the field of view between a Full Frame sensor and a Super 35 sensor, you need to change the focal length between the two. Using the 35mm sensor and the 55mm lens from the previous example as our starting point, we have to change the focal length to 85mm to maintain the field of view. This will keep your relative framing the same, but will have a noticeable change when it comes to depth of field. The same frame will have a much more shallow depth of field on a full frame sensor as a result of using a longer focal length to achieve the frame. Add or subtract a couple millimeters depending on the specific size of the sensor.
The main thing to understand about this whole mess is that it really comes down to relationships between the lens and the sensor. Because there are so many variables – sensor size, aspect ratio, spherical vs. anamorphic, etc. – there’s no such thing as a “50mm equivalent”, for example. One person could be using a 50mm on a stills camera as their point of reference and one could be thinking about how it looked on their RED camera with a wacky sensor size. Because everything is so variable, it’s time to start getting specific in the way we think and talk about it. The majority of confusion when it comes to this topic stems from the fact that most terms we use refer in some way to an existing standard. When someone says that a lens is, for example, a “50mm equivalent”, what they maybe should say is more like: “this lens will produce a field of view on a Super-35 sized imager similar to the field of view created when using a 50mm lens on a full frame stills camera”. There’s a lot of information in that statement that is implied, and with the ever-increasing number of formats in the stills world as well as the digital cinema world, it’s important to specify your sensor size, especially when discussing lens “equivalency”.
👏 THAT 👏 BEING 👏SAID: If you are used to using spherical lenses on a Super 35 format (Alexa, FS7, Gemini, et cetera), and you want to use that as a frame of reference for a focal length’s field of view or “size” on a Full Frame camera, divide by 1.5 and you will be in the approximate area. For instance. If you have a 50mm lens and you put it on a Full Frame camera, the field of view you can expect will be approximately the same as what a 33mm lens gives you on a Super 35 camera (and then you’ll round it to 35mm because when’s the last time you got a lens set with a 33mm lens in it? This is why it’s an imperfect system!). The chart below is a quick reference guide using rough approximations that are close to common focal lengths.
I know this might have been a lot to take in – so if you ever need clarification or want to see the differences yourself, feel free to reach out to us here at Expressway and we’ll be more than happy to discuss or show you!