First published on his filmmaker-centric site NoFilmSchool.com in 2010 and updated consistently since then, Ryan Koo's digital cinematography guide has been downloaded as a PDF over 800,000 times and is available to read with modified content as web pages on the No Film School site.
Culled from information he gained through online forums for the community fo DSLR users, Koo's guide is a user-friendly introduction and resource guide to the ins and outs of DSLR filmmaking.
As we continue to see DSLR's become the predominant medium for shooting documentaries as well as an incredibly popular choice for indie narrative filmmakers, we wanted to make sure Indiewire's growing Filmmaker Toolkit pointed our readers to one of the most basic and helpful DSLR filmmaking guides.
We're reprinting the 10 basic concepts essential to understanding DSLR filmmaking, taken from the PDF version of the guide. If you're hooked or want to read the more detailed content, check out the regularly updated web version or download the portable PDF version on the No Film School website here.
BASIC CINEMATOGRAPHY CONCEPTS: What should I know about (digital) cinematography?
From reader emails, I realized that a basic introduction to some of the concepts referenced later in the guide might be helpful. Many of you are already familiar with these terms, so feel free to skip this chapter! However, if your background is in still photography or if you’re new to digital imaging in general, this bonus chapter should help clarify some basic cinematography concepts that we’ll be working with going forward. By no means is this an exhaustive glossary, but it is a good starting point. I’m going to explain things from a practical, crash-course standpoint rather than a scientific, 100% semantically correct perspective, because I think it’s handier to know how something works in practice than it is to know all of the details of why it works — if you’re looking for knowledge of the latter, there are of course thousands of good resources on the internet to bolster your knowledge. In alphabetical order, then, here are ten basic concepts you should be familiar with:
1. Aspect Ratios & Anamorphic Lenses
Aspect ratio used to be a more prominent issue for digital cinematographers than it is today: before the advent of high-definition cameras, the standard 4:3 aspect ratio of standard-definition TV was generally seen as undesirable for anyone looking for a “cinematic” look, because 4:3 (or 1.33:1) content was associated with broadcast TV, while widescreen compositions were what people expected to see in the theater. When we say “4:3,” we mean the image is four units wide and three units high. When we “1.33:1,” we mean… well, you get it — the same thing. Many times the “:1” is removed because it is implied – shooters will simply say “1.85” instead of “1.85:1.”
HDTV today is widescreen by default, with a 16:9 aspect ratio that works out to be 1.78:1 — very similar to the traditional 1.85:1 aspect ratio of many feature films. Other than these two virtually indistinguishable aspect ratios, the
most common widescreen aspect is the CinemaScope ratio of 2.35:1, which appears most often in the multiplex in big-budget films.
2.35:1 films are typically shot with anamorphic cine lenses. Anamorphic lenses are not spherical in the sense that they squeeze an image to fill the negative or sensor, with an additional step necessary during projection to re-stretch the image to the intended size. The odd-looking image here of a lens with an oval aperture demonstrates the non-spherical nature of an anamorphic lens (the aperture is perfectly round, but the lens is distorting our view of it). While it is possible to attach an anamorphic lens to a DSLR, most of us will simply shoot at the native widescreen aspect ratio of 16:9.
Bokeh (pronounced like “bo” from “boat” and “ke” from “Kentucky”) is one of the chief reasons many shooters have switched to DSLRs. Bokeh is a term derived from the Japanese word “boke” which, roughly translated, means “blur quality.” Bokeh refers to the portions of an image that are defocused or blurry. In the filmmaker’s toolkit, bokeh is not only an aesthetically pleasing quality, but it also allows the filmmaker to focus the viewer’s eye on an object or area of interest in the frame. Bokeh is a function of shallow depth-of-field (see below).
3. Compression & Bit Rate
Compression refers to a method for reducing the amount of data a DSLR produces; in the case of video-shooting DSLRs, all cameras currently employ some method of compression. If you’re used to shooting photos in JPEG format, you’re used to capturing compressed images; while RAW can also employ compression, it is generally thought of as “uncompressed.” This is because, as far as shooters are concerned, when we’re talking about compression we’re talking about lossy compression — meaning, a codec (compression algorithm) that throws out data in order to reduce file size. As you can imagine, tossing portions of an image has negative side effects, and while many codecs deal with images perceptually in order to minimize their perceived impact, the difference is there. For example, if you upload a video to YouTube, the service recompresses your video in order to optimize it for internet delivery; you might not notice this compression, but check out this video that’s been recompressed a thousand times (original here) and you can see that every compression step throws out data along the way. On the positive side, however, lossy codecs are also the reason we can record hours of footage to inexpensive flash memory devices like CF and SD cards.
The most common compression formats in DSLRs are h.264 and MJPEG, and while both are lossy, h.264 is generally much more efficient (it introduces less artifacts at the same bit rate as MPJEG). Bit rate is the amount of data per time that a given codec adheres to; higher bit rates are almost always better because they use less compression. At press time there are no DSLRs that shoot uncompressed video.
4. Depth of Field
The amount to which objects in the foreground, mid-ground and background are all in focus at once is a function of depth of field. A shallow depth of field would mean that only one plane was in focus; a wide (or deep) depth of field would mean that all planes are in focus at once. Depth of field is determined by the focal distance and aperture size (see below for more on Aperture). DSLRs exploded in popularity almost singlehandedly because of their ability to render images with a shallow depth of field. This is chiefly due to their massive sensor sizes (see the next chapter, “Choosing a DSLR,” for an examination of sensor sizes), which are exponentially larger than previous video cameras. On a basic level, shallow depth of field (DOF) allows filmmakers to blur out areas of the image they deem to be unimportant or undesired.
5. Exposure & Aperture
Exposure refers to the amount of light allowed to enter the DSLR sensor (or any imaging surface). When shooting stills, DSLRs use a mechanical shutter to regulate exposure by opening for the desired amount of time (1/60th or 1/1000th of a second, for example) and then closing. DSLRs are generally rated to last for hundreds of thousands of shutter cycles, but at 24 frames per second, couldn’t your DSLR reach that limit very quickly? No, because in video mode, DSLRs use an electronic shutter — the sensor basically turns on and off to regulate exposure, instead of relying on a physical barrier (i.e., the mechanical shutter) to regulate light. Aperture refers to the adjustable opening near the rear of the lens that lets light through — the amount of light it transmits is generally referred to as the F-stop (T-stop is very similar, except it’s measured instead of calculated). We’ll go more into depth on aperture in the “Lenses” section of the guide, but keep in mind that the size of the aperture does not only affect the amount of light, but also the angle of light rays hitting the sensor — a narrow aperture creates an image with a wide depth of field, whereas a large aperture creates an image with a shallower depth of field.
6. Focal Length
Technically, focal length refers to the distance over which collimated rays are brought into focus. An easier way to think of it: focal length refers to image magnification. A longer focal length, e.g. 100mm, makes distant objects appear larger, whereas those same objects will appear smaller with a shorter focal length, e.g. 35mm. Focal length also refers to angle of view; longer focal lengths have a narrower angle of view, whereas shorter focal lengths have a broader angle of view. When it comes to focal length, a picture is worth a thousand words, so here are images taken with the camera in the same place, but with lenses of different focal lengths attached:
7. Frame Rate
Frame rate is the frequency with which your DSLR captures consecutive images. This typically corresponds to the number right before a “P” in the case of progressive images, so that 24p is 24 frames per second, 30p is 30 frames per second, and 60p is 6,000,000 frames per second. Just kidding. Different frame rates have very different motion rendering characteristics, which, combined with different shutter speeds, produce images that behave very differently. Motion pictures have had a standard frame rate of 24 frames per second since the 1920s, and audiences have come to associate this frame rate with cinematic content, so being able to shoot in 24p is essential if you’re planning on shooting narrative material. However, you don’t always have to shoot at the same frame rate at which you’re planning on distributing your material. For example, if your DSLR can shoot 60p, this is a very effective way of acquiring slow-motion footage — anything shot at 60p can be played back at 40% speed in a 24p timeline for a flawless slow-motion effect, and can generally be slowed down further in your editing system.
8. ISO & Noise
ISO is actually the International Organization for Standardization, which is why you see it used in lots of places beyond photography — many businesses are certified ISO:9001, for example. As cinematographers we’re concerned with just one “standardization,” however — the one that pertains to measurement of noise in photography. ISO as it relates to digital photography is based on analog standards of film speed — while we won’t be shooting a frame of actual film with our DSLRs, our cameras are calibrated so that an ISO of 400 on our camera is somewhat equivalent to a film SLR’s ISO 400. ISO is a logarithmic measurement, so ISO 400 is twice as sensitive to light as ISO 200, ISO 200 is twice as sensitive as ISO 100, and so on and so forth.
The relationships between sensitivity and noise is basically linear, however, so the higher the ISO, the brighter the image — and the more noise contained in the image. However, thanks to sophisticated noise reduction and other processing tricks, DSLRs have managed to dramatically reduce noise at higher ISOs, and can often blow film stock out of the water (this depends on which camera you’re shooting with, which we’ll cover in the next chapter).
9. Progressive vs. Interlaced
Interlacing was a workaround invented for oldertech CRT monitors in the 1930s that has lived far too long. In the early days, video bandwidth was more limited than today, and so engineers found a way to divide a frame into two images and display it using alternating fields. As you can see in this image of a tire wheel, interlacing can cause motion artifacts (as well as a host of other problems). We’re lucky to live in a predominantly progressive society today — in the imaging sense if not the political. Progressive scanning is a method that captures and displays the lines of an image in sequence, which is akin to motion picture film with regards to motion rendering. Compared to interlaced images, progressive images have a higher vertical resolution, lower incidence of artifacts, and scale better (both spatially and temporally). Friends don’t let friends shoot interlaced! Luckily, while there are plenty of video cameras that shoot interlaced footage, every DSLR I can think of shoots progressive footage.
10. Shutter Speed
Shutter speed refers to the length of time an image is exposed. For film SLRs, this would be measured by the amount of time the camera’s mechanical shutter is open, but for shooting video on DSLRs, this is simulated electronically. Shutter speed affects the amount of light that reaches the camera and also affects the motion rendering of the moving image. Lower shutter speeds yield a brighter and smoother image (up to and including water and light blurring tricks), whereas higher shutter speeds result in a darker and more stroboscopic image. Motion picture film cameras typically shoot with a 180-degree shutter, which means that the shutter is open 50% of the time (180 out of 360 degrees). This means the amount of time your shutter is open is half of the shooting frame rate; thus, at 24 frames per second, a 180-degree shutter is best emulated on a DSLR by choosing a shutter speed of 1/48.
This may not be possible depending on your DSLR, so the closest reading will do — 1/50 or 1/60, for example. This gives the most “filmic” rendering of motion, but can be varied greatly depending on your intention. Higher shutter speeds create “jerkier” images, as most famously seen in action films like Saving Private Ryan and Gladiator. Conversely, lower shutter speeds create “smoother” images due to increased motion blur. There is no hard and fast rule when it comes to shutter speed, but if you’re not sure of what shutter speed to select, go with the setting that’s closest to half that of your current frame rate.