Video Aspect RatiosAugust 07, 2013
A video’s aspect ratio is an important creative choice that can affect the feel of footage, but it’s also a key technical consideration that affects how and where this footage can be displayed. This article gives an overview of the various capture settings and applications.
The origin and evolution toward the current cinematic aspect ratios has been driven primarily by prior formats and pioneering engineers. Standards have also had to adapt to new or competing technologies, such as audio and home television.
Despite this variation, virtually all cinematic imagery has been wider than it is tall, in part because this is well-suited to the angle of view of human stereo vision. The convention is therefore to describe an aspect ratio using width first, then height. The ratio can be expressed using either two integers, such as 3:2, or as a normalized width relative to a unit height, such as 1.5:1.
In the early days, 1.33:1 (4:3) was the most commonly used aspect ratio for 35 mm silent films, which in turn was chosen because 70 mm film was already being mass-produced for still photography. This then had to narrow slightly to 1.17:1 due to the unanticipated rise of “talkies” on PhonoFilm, which required space for the audio track. By 1932, the Academy of Motion Pictures agreed to revert back to a wider standard of 1.375:1.
The “Academy” standard persisted until the development of television drove film studios to distinguish themselves using even wider aspect ratios in the 1950’s, such as Cinerama (2.66:1) and VistaVision (up to 2:1). Aspect ratios remained diverse for several decades, in part because of various new audio technologies, but also because theaters had to get equipped with new masks and projectors. Throughout this timeframe, cinema generally remained wider than television.
Eventually cinema converged on two leading standards: a normal 1.85:1 widescreen and an anamorphic 2.39:1 widescreen. With television, the formats became 4:3 with standard definition and later 16:9 with high definition, which at 1.78:1 was a close match to 1.85:1 widescreen cinema.
Note: many say 2.39:1, 2.40:1 (“two-four-oh”) or 2.35:1, but nowadays these typically all refer to the same 2.39:1 aspect ratio. 2.35:1 was an earlier SMTPE widescreen standard prior to 1970, and two-four-oh is just an incorrect rounding from 2.39:1 to 2.40:1 instead of to 2.4:1.
Although having fewer aspect ratios has helped simplify distribution, complications can still arise when there is a mismatch between the capture and display. This was especially problematic when using standard definition 4:3 television to display widescreen DVD’s, which are often cropped dynamically using a “pan and scan” technique that alters the composition. Otherwise the displayed image has to be padded, which reduces the image area and perceived detail:
Padding footage with black strips at the top and bottom is also referred to as “letterboxing,” whereas a similar (but less common) padding on the left and right is also referred to as “pillarboxing.” These still happen frequently when anamorphic widescreen footage or older standard definition content are shown on modern 16:9 HD or 4K displays:
Cinematographers therefore typically choose an aspect ratio that minimizes the need for either scenario, or at least compose their scenes with the potential for cropping in mind. For example, many will shoot at 1.85:1 widescreen for theaters, but will also pad their composition with side cropping to 1.78:1 television in mind.
CHOOSING AN ASPECT RATIO WITH RED
The native full resolution aspect ratio of RED is 1.90:1 with the MYSTERIUM-X® sensor, and slightly wider with the RED DRAGON™ sensor. Since these are a middle-ground between the most common output formats, frame cropping is minimized on average:
When intending output in a specific format, each resolution setting can be saved using one of several preset aspect ratios: widescreen or WS (2.4:1), 2:1 and HD (16:9). The full resolution setting also has an anamorphic or ANA preset, which is designed for typical 2X anamorphic lenses. Using these presets typically makes more efficient use of available storage and potentially enables higher frame rates. Common workflows include:
However, if a different aspect ratio is needed afterwards, the only options are cropping or padding. Furthermore, the composition cannot be changed in post-production without also compromising resolution. Many cinematographers therefore capture the full frame all the time, but compose for their intended output aspect ratio. Other times, using a conservative aspect ratio (such as 2:1 even when outputting for 2.39:1) provides a good compromise between conserving storage space and being able to recompose. Capturing the full frame also means that a crop can always be expanded later if a different aspect ratio is needed—thereby enhancing as opposed to reducing clarity and sharpness.
USING FRAME GUIDES WITH RED
Frame guides are a helpful way to compose for a specific aspect ratio without necessarily recording at that aspect ratio. The region outside the frame guide (the “look-around area”) can be monitored during capture to ensure that boom microphones, camera operators and similar objects remain outside the full frame.
Frame guides can be specified under “Settings > Display > Guides > Frame Guide” within RED’s camera menus. Available modes include “Full,” 4:3, 16:9, 2.4:1, 1.85:1 and 1.9:1, but custom ratios are also possible.
Digital Cinema Packages (DCP). The chosen resolution setting and aspect ratio may also need to conform with a distribution standard. With cinema, this is typically a DCP, which currently accepts 4096×1716 (“scope” or 2.39:1), 3996×2160 (“flat” or 1.85:1) and 4096×2160 (“full container”).
Home Video. Many televisions have the ability to magnify 2.39:1 and other widescreen formats in order to utilize more area on displays which aren’t as wide. Cinematographers may therefore want to avoid placing critical subjects too near the horizontal edges of the frame.
Terminology. With digital, several types of aspect ratios exist: the display aspect ratio (DAR), the pixel aspect ratio (PAR) and the storage or sample aspect ratio (SAR). DAR refers to the viewable image dimensions, PAR refers to the physical dimensions of each pixel, and SAR refers to the number of horizontal versus vertical pixels. Unless specified otherwise, the term “aspect ratio” in this article refers to DAR.
Most modern displays utilize square pixels, in which case DAR and SAR are identical, and PAR equals one. PAR is usually only important with anamorphic lenses, in which case PAR may be set to 1.3 or 2.0 in order to stretch the image horizontally. With REDCINE-X®, this can be controlled automatically using the “Framing” right-panel.