Appendix: Reference Source

A.1 Video-Data Sizes

The following table lists common video formats, their specifications, and standard, and the parameters of their equivalent digital counterparts.

A.2 Film-Data Sizes

The table that follows lists the physical dimensions and equivalent digital sizes of different film formats.

A.3 Relative Format Quality

The following table is a rule-of-thumb guide to the relative quality of different formats. Some debate is still questioning the quality of HD video compared to 16mm film., Certainly other factors such as compression play a role in the resulting quality of any medium.

A.4 Relative Picture Size

The following diagram shows the relative size of different digital formats, by resolution.

A.5 Data Storage

The following table lists the typical capacity of different storage media. Note that factors such as transfer speed, reliability, and cost should also be considered when choosing a data storage device. In addition, certain formats may require the use of separate readers/writers.

A.6 Characteristics of Still Image Digital File Formats

This table lists the specifications for different file formats.

A.7 Digital Image Operations

Several common digital image operations are available to most pipelines. The operations may be reversible even after they’re applied, and they may or may not degrade the image. Many have an optical equivalent; some don’t.

A.8 Common Color Temperatures

The following table lists the correlated color temperature of some common light sources.

A.9 Cmx 3600 EDL Format

The following illustration is of a typical CMX 3600:

Although it’s optional, the first line of the EDL usually contains a title. The second line, and all the subsequent lines, consist of the following components:

1. The event number. Differentiates between different lines in the EDL. Problems arise, however, when the lines are incorrectly ordered or when duplicate event numbers exist.

2. The reel number (or reel name). Identifies individual source reels and can usually consist of both numbers and letters. However, it’s generally safer to use numbers only, if possible, and to avoid using names longer than six characters.

3. The track. Video only (V), in most cases; however, numerous audio channels may be included (e.g., A12).

4. The transition type. Usually a cut (C), but dissolves (D followed by the duration), wipes (W followed by the wipe SMPTE number and the duration), and motion effects (M2 followed by the duration) may also be present.

5. The timecodes. A pair of source timecodes (i.e., in and out points) are mapped onto record timecodes (also in and out) for every event.

Note that the frame rate isn’t intrinsically specified in the EDL, which may lead to problems.

Conversion between EDLs of different frame rates involves converting all timecodes to frame numbers, typically using a sync reference frame—a timecode at which both EDLs line up (usually 00:00:00:00)—and then converting the frame numbers to the new timecode.

A.10 Key Numbers

A typical key number reads as follows:

IS 00 9123 1234•

The first two digits contain the manufacturer code and film type, the following 10 numbers are the serial number for the particular reel, repeated every foot with the frame zeroed at the dot. This is repeated every 16 frames (i.e., 1 foot) for 35mm film, and so individual frames are counted as the zero frame key number + number frames. In this instance, 5 frames from the zero frame is key number IS 00 9123 1234+05, and so on. It’s also possible to do this as negative numbers, so that 4 frames before the zero frame, in this case, is IS 00 9123 1234-04 (or IS 00 9123 1233+12). The same is true of 16mm film, except that it has 20 frames per foot.

A.11 Widescreen Safe Areas

A.12 Cie Chromacity Diagram

CIE chromacity diagrams, such as the one provided later in this Appendix, map all colors in the entire visible gamut (enclosed by the curve on the graph) as points on a graph. For every visible color, an x and y coordinate can be found on the graph (luminosity is treated separately and normally denoted as of a z value). Furthermore, you can use the diagram to determine the boundaries of different color spaces (e.g., sRGB or NTSC video) by plotting the maximum value for each component (i.e., the maximum values for red, green, and blue output). Where color spaces overlap, the enclosed colors may be accurately reproduced by each color space. Otherwise, the colors are out of gamut and must be altered before they can be displayed.

A.13 Digital Cinema Master Specifications

The following guidelines are based on version 5.1 of the DCI’s specifications, as pertaining to the image content of a Digital Cinema Distribution Master (DCDM).

• Images may be encoded in one or more reels, which are later digitally spliced together.

• Images should be 12 bits per pixel JPEG2000 format images, encoded in an X’Y’Z’ color space (i.e., gamma-corrected CIE color space coordinates).

• Spatial resolution should be 2048x1080 (2k) or 4096 × 2160 (4k) pixels, or an image area letterboxed to fit these areas, depending on the aspect ratio. For example, a 4:3 image at 2k would have an image area of 1440 × 1080 pixels, and a 16:9 image at 4k would have an image area of 3840 × 2160 pixels.

• 4k images should be accompanied by 2k counterparts.

• Frame rate must be 24 or 48 Hz for 2k images, 24 Hz for 4k images.

• The data must then be compressed, encrypted (using AES 128 encryption), packaged in an MXF container file, and finally combined with audio and other elements to form a Digital Cinema Package (DCP), the size of which should not exceed 500GB.

• Images must have a pixel aspect ratio of 1.0.

For more information on these specifications, visit the DCI’s website at www.dcimovies.com.