Skip to content

Glossary

Terms as used in this handbook. Entries explain the distinction a filmmaker or post-production team needs in practice. Follow the chapter links for examples, specifications, and source references.

A

AAF: Advanced Authoring Format. A binary timeline interchange used mainly by Media Composer for picture turnover and by editing systems for Pro Tools audio turnover. It can carry multiple layers and speed-ramp keyframes. See AAF.

ACES: Academy Color Encoding System. A standardized color-management framework with one scene-referred interchange encoding (ACES2065-1), working spaces derived from it, defined transforms at each end of the pipeline, and a container. See ACES.

ACES2065-1: The scene-linear ACES interchange encoding, using AP0 primaries. It is distinct from the AP1 working encodings used for grading and compositing. See ACES encodings.

ACEScc / ACEScct: Two scene-referred AP1 grading encodings. Both use a logarithmic curve, but ACEScct adds a linear toe near black, changing how shadow adjustments behave. They are not interchangeable input labels. See ACES encodings.

ACEScg: The scene-linear AP1 working space intended for CGI rendering and compositing. Floating-point ACEScg can retain negative components and values above 1.0. Converting AP0 to AP1 does not itself require clipping. See AP1 does not inherently clip AP0.

ADR: Automated Dialogue Replacement, also called additional dialogue recording. Dialogue recorded in post to replace or supplement production sound, commonly matched to the filmed performance. See Post-production planning.

ADX: Academy Density Exchange Encoding. ADX10 and ADX16 encode Dmin-subtracted Academy Printing Density as 10-bit or 16-bit unsigned integer values. ADX is intended for motion picture color negative and internegative film scans. A generic Cineon-style scan is not automatically ADX. See ADX10 and film scans.

Alpha channel: A channel describing an element's opacity or coverage for compositing. It is separate from RGB color and is not a viewing transform. See OpenEXR.

AMF: ACES Metadata File. An XML sidecar that records which ACES transforms a shot or show uses: Input Transform, look transforms, Output Transform, so a pipeline's color configuration travels with the material instead of living in someone's notes.

AOV: Arbitrary Output Variable. A supplementary render output, such as a normal map, depth, or lighting component. AOVs can be stored as additional OpenEXR channels or separate files. See OpenEXR.

AP0 / AP1: The two ACES RGB primary sets. AP0 is used for ACES2065-1 interchange. AP1 is used for ACEScg, ACEScc, and ACEScct working encodings. A primary set alone does not specify whether the values are linear or log. See Color primary sets.

APD: Academy Printing Density. A standardized printing-density metric defined by spectral responsivities, measurement conditions, and a reference measurement device. APD is the density basis for ADX, not an image-file format. See ADX10 and film scans.

ARRICORE: ARRI's compressed, in-camera-demosaiced RGB recording format. ARRICORE is not Camera RAW, but the ARRI Image SDK retains selected exposure, white-balance, and tint controls. See ARRIRAW, ARRIRAW HDE, and ARRICORE.

ARRIRAW: ARRI's Camera RAW format, carrying sensor samples and metadata for image construction in post. ARRIRAW, losslessly encoded ARRIRAW HDE, and demosaiced ARRICORE are different recording representations. See ARRIRAW, HDE, and ARRICORE.

ASC CDL: American Society of Cinematographers Color Decision List. A deliberately restrictive color correction interchange format limited to slope, offset, power, and saturation. The ten values transfer reliably only when sender and receiver agree on the working space and application order. The CDL defines no color space of its own, so the same numbers applied to differently encoded images give different results. See Plate Pre-Grading.

ASC MHL: Media Hash List. An ASC-published format for recording checksums of copied media so that a transfer can be verified rather than assumed. The practical defence against silent corruption during offloads and archival.

Asset: A physical or digital object used to create or represent a creative work. Picture, audio, timed text, project files, graphics, metadata, and documentation can all be assets. An asset ID identifies the logical object. A production asset ID is assigned and controlled by the production. A filename is only one label or file detail. See Asset Identity, Naming, and Versioning.

Asset group: Assets managed as one unit, such as a VFX turnover, delivery package, or archive set. Group membership does not replace the identity of each asset.

ASSETMAP: The document that maps asset identifiers to file paths in a DCP or IMF delivery. It tells the receiving system where files are, not what order to play them in. See DCP package documents.

Asset register: The production-controlled record that connects stable asset IDs to descriptions, revisions, states, responsible roles, external IDs, and file locations. A small production can use a shared spreadsheet. See Keep One Simple Production-Owned Post Log.

B

Bayer pattern / CFA: A color filter array (CFA) places different filters over sensor photosites. The common Bayer pattern has two green, one red, and one blue sample in each repeating 2×2 block. Other arrays exist. Bayer describes a sampling pattern, not a RAW codec. See How sensor data becomes a color image.

Bit depth: The number of bits used to store a sample or component. An unsigned 10-bit integer has 1,024 possible code values. Floating-point storage uses its bits differently. Bit depth alone does not identify dynamic range, chroma sampling, or compression quality. See Working formats.

BRAW: Blackmagic RAW. The camera performs part of the demosaic before encoding, reducing decoding work while retaining adjustable RAW controls.1

Burn-in: Text or graphics rendered into the visible picture, such as shot name, revision, source timecode, or frame number. A burn-in remains visible without reading the file's metadata. See Offline reference.

C

Camera RAW: A recording of a digital camera sensor's corrected readout that defers part of the image construction to post-production. It is scene-referred and preserves deferred decode controls. Production archives the camera originals, but Camera RAW is not a standard VFX plate format or a self-sufficient preservation master. See Camera RAW.

Captions: Timed text conveying dialogue and relevant sounds for accessibility. Closed captions can be enabled separately. Open captions are visible to everyone watching that presentation. The term describes the presentation, not one file format. See Captions and subtitles.

Capture frame rate: The number of images a camera records per second. It normally matches the project timebase for sync sound, but deliberate overcranking or undercranking makes it differ for slow or fast motion. See Frame rate and timebase.

CBB: Could Be Better. A visual effects shot the filmmakers or supervisors can accept but would improve with another revision if schedule permits. See Digital Intermediate.

CGI / CG: Computer-generated imagery, including rendered characters, environments, effects, and other image elements. CGI may be part of a VFX shot, but VFX also includes photographed and painted elements. See VFX production.

Checksum / hash: A value calculated from a file's bytes to check whether they have changed. Matching values calculated with the same algorithm support transfer and preservation checks. They do not prove that a render looks correct or contains the intended revision. See Delivery checksums and preservation hashes.

Chroma subsampling: Storing color-difference components at a lower sampling resolution than luma. In common progressive formats, 4:2:2 halves horizontal chroma resolution and 4:2:0 halves both horizontal and vertical chroma resolution. 4:4:4 retains full component sampling. See Chroma subsampling.

Chromaticity: A description of a color's proportions independent of its luminance, commonly plotted as CIE x,y coordinates. A chromaticity diagram shows primary and white-point relationships, not the brightness range of a display. See Color spaces.

Clean plate: A photographed or constructed view without the foreground subject or object being removed, used to supply background detail for a composite. It is not simply an ungraded plate. See On-set VFX references.

CLF: Common LUT Format. An XML interchange that can carry a sequence of LUTs and mathematical color operations, including ASC CDL. It is a transform container, not a color space. See LUTs vs. transforms.

Clipping / clamping: Clipping loses distinctions when values reach a system's limit. Clamping explicitly restricts values to a chosen range. Clamping scene-linear data to 0–1 can discard legitimate highlights and negative color components. See AP0/AP1 round trips.

Codec / container: A codec defines how media samples are encoded and decoded. A container, or wrapper, organizes encoded picture, sound, and metadata into a file. ProRes is a codec family. QuickTime and MXF are containers. An extension alone does not specify image quality. See Frames, wrappers, and codecs.

Color gamut: The range of colors a device can capture or reproduce, or an encoding can represent under stated limits. An RGB primary triangle is not a complete description of a display's color volume across luminance levels. See Color spaces.

Colorist: The artist who performs the final color grade, working with the director and cinematographer to realize their visual goals and maintain continuity across the finished picture. See Responsibilities of the Colorist.

Color space: A defined system for interpreting color coordinates. For an RGB handoff, identify the primaries, white point, and transfer function, not just a label such as “log.” See Color spaces.

Comp: Composite. An image or shot assembled from photographed, rendered, or painted elements. A comp can be a work in progress or a final. Also the verb for assembling it.

Composition: Instructions that assemble assets into a presentation. An IMF CPL is a composition because it selects track files, timing, and order. The composition is not the rendered or packaged essence it references.

Concatenation: Combining compatible geometric transforms into one net operation so the image is resampled once. Intervening processing can prevent concatenation, depending on the application and operation order. See Concatenated Image Transforms.

Conform: Reconstructing the show's timeline at full quality from camera original and VFX elements, using metadata from the offline edit. See Editorial Turnover for DI.

Conform Check (Confidence Check): A review movie sent by the DI to picture editorial to verify that the conformed timeline matches the intended edit. The editor or assistant editor compares it with the matching offline reference to check cuts, timing, framing, opticals, titles, and VFX versions. It verifies edit integrity, not final color approval. See Conform checks: DI to editorial.

Count sheet: A shot-by-shot record of a VFX turnover, including plate names, frame ranges, handles, work descriptions, references, and relevant technical details. It describes what the vendor receives and is expected to return. See Count sheet.

CPL: Composition Playlist. A document specifying which picture, sound, and timed-text assets make up one DCP or IMF composition, including their order and timing. It references media rather than containing the images or sound. See IMF structure.

Creative transform: A repeatable creative treatment that establishes contrast, color relationships, film emulation, or another show look. It may be a LUT, DCTL, grade, or ACES Look Transform. It does not have to produce a display-ready image. See Translate the Color-Management Terms.

CTL: Color Transformation Language. A programming language used to express reference color operations mathematically, including the ACES reference transforms. Artists usually access their implementation through application color management or an OCIO config. See How ACES is delivered.

D

Dailies: Footage processed and delivered for review shortly after it is shot, usually with a viewing LUT applied, for the director, cinematographer and editorial. See Digital Dailies.

Data window / display window: In OpenEXR, the stored pixel region and the visible frame region. The regions can differ, but DI systems do not interpret the difference consistently. Resolve truncates pixels outside the display window. Test unequal windows in the target system. See OpenEXR.

DCDM: Digital Cinema Distribution Master. The graded, unencrypted image master used to author DCP picture essence. Conventional SDR DCDMs use DCI X′Y′Z′ encoding, commonly stored in 16-bit TIFF sequences. A pDCDM carries the defined 12-bit image values in losslessly compressed JPEG 2000 .pdc.j2c frames for exchange. A DCDM may also be delivered and archived as a mastering element. See Digital Cinema Mastering for image precision and file mapping.

DCI: Digital Cinema Initiatives, the organization that publishes the Digital Cinema System Specification. “DCI” identifies cinema requirements, not one universal RGB color space or image file format. See Digital Cinema Packages.

DCP: Digital Cinema Package. The picture, sound, timed text, and packaging documents prepared for cinema playback. A DCP may be encrypted or unencrypted. See DCP.

DCTL: DaVinci Color Transform Language. Code used in Resolve to define image operations mathematically. A DCTL can also embed a LUT, so the file type alone does not mean every operation is an analytical formula. See LUTs vs. transforms.

Debayer / demosaic: Estimating the missing color components from a sensor's color-filter-array samples to construct an RGB image. Debayer refers specifically to a Bayer array. Processing may happen in camera, in post, or be divided between them. RAW still has a recorded sensor raster. See Hardware and software demosaic.

Delivery schedule: The distributor's list of materials, technical requirements, deadlines, and acceptance conditions needed to complete the sale. It covers more than delivery dates: picture, sound, captions, artwork, and supporting documentation may each have separate requirements. See Distribution Deliverables.

Digital intermediate (DI): The stage where a show is conformed, VFX are integrated, color grading happens, and every deliverable is mastered. Named for the photochemical intermediate elements it originally replaced. The DI is also shorthand for the company, facility, team, or individual handling that work. It includes the colorist and may include a DI producer, conform or finishing editors, optical or lite-VFX artists, pipeline specialists, and mastering technicians. See What “the DI” Means and Where the name comes from.

Display-referred: Image encoding defined in relation to display characteristics, with no numeric relationship to real-world exposure. The state of any image intended for exhibition. ISO 22028-1 calls this same state output-referred. The two terms are interchangeable and you will meet both. See Display-Referred Imagery.

DIT: Digital Imaging Technician. Supports the cinematographer with digital-camera workflow, monitoring, and on-set image management. Data offload and backup may be separate responsibilities assigned to other crew. See Digital Dailies.

DKDM: Distribution Key Delivery Message. A KDM addressed to an authorized mastering or distribution system so it can work with encrypted DCP material and issue playback KDMs. It is not a universal key that any theater can use. See DCP encryption.

DNx: Avid's codec family, rebranded in 2025 from the DNxHD/DNxHR names to five quality levels (444, HQX, HQ, SQ, LB). Resolution-independent, and all levels now support 8- to 16-bit encoding and an optional alpha channel. See Avid DNx.

Dolby Vision: An HDR mastering and delivery system that combines picture with dynamic metadata to adapt the image for different display capabilities. Creative trims can refine that mapping. It is not a synonym for PQ or for HDR generally. See HDR distribution formats.

Dolby Vision trim: A colorist-authored pass that adapts an HDR grade to lower-capability displays. The trim is carried as dynamic metadata instead of being baked into a separate master. See HDR Mastering.

DPO / DTO: Disney-specific conformance labels. DPO means Distribution Conformance, Program Only. DTO means Distribution Conformance, Textless Only. They are not general industry names for all texted or textless masters.

DPX: Digital Picture Exchange. An image-sequence format used for film scans and mastering. A DPX file's bit depth does not establish its density calibration or color encoding. See DPX.

Drop-frame timecode: A counting convention that skips selected timecode numbers so the count tracks elapsed time at rates such as 29.97. It does not discard image frames. 23.976 uses 24-frame non-drop counting. See Frame rate.

DRT: Display Rendering Transform. The scene-to-display operation that establishes tone scale, highlight behavior, and gamut mapping. DRT describes a function, not one universal standard. See Translate the Color-Management Terms.

DSM: Digital Source Master. The graded master a show's deliverables are derived from. DCI deliberately does not specify its format, color space or encoding. It is defined by what the production and its distributors agree on, and in practice one show may carry several (Rec.709, Rec.2020/HDR10, P3-DCI and Dolby Vision versions are all commonly requested separately). Sometimes skipped in favour of producing a DCDM directly. See DSM.

Dynamic range: The ratio between the highest and lowest usable levels in a system. Camera range concerns captured scene exposure. Display range concerns reproduced light. Neither is specified by recording resolution alone. See HDR Mastering.

E

EDL: Edit Decision List. A text-based timeline interchange. Traditional CMX-style EDLs mainly describe source clips, edit positions, and simple transitions. Enhanced variants add information that may not transfer to every system. See EDL.

EOTF / OETF / OOTF: Electro-Optical, Opto-Electronic, and Opto-Optical Transfer Function. The OETF maps scene light to code values (a camera's job). The EOTF maps code values to displayed light (a display's job). The OOTF is the net scene-to-display rendering, including the intended artistic rendering. Conflating the OETF and EOTF of a standard is a common and consequential error.

Essence: The picture, audio, or timed-text content carried in a media file, separate from its metadata and the documents that package it. In IMF and DCP, MXF track files carry the essence.

EXR: OpenEXR. The standard image format for scene-linear and VFX work, storing half-float or full-float pixels, arbitrary channels, and a data window that may differ from the display window. See OpenEXR.

F

FFOA: First Frame of Action. The reference point marking where the program picture begins, used to align head leaders and the 2-pop. Also written FFOP (First Frame of Picture). FFOA is the form used in most delivery specifications. See Leaders and Pops.

Fixity: Whether a preserved file's bytes remain unchanged over time. Periodic comparisons with recorded hashes check fixity. They do not establish whether the file can still be decoded or the project reconstructed. See Preservation hashes.

Flat / Scope: The standard digital-cinema presentation formats, approximately 1.85:1 and 2.39:1 respectively. At 4K their picture rasters are 3996×2160 and 4096×1716. A different creative aspect ratio can sit inside either, with unused picture area. See Digital cinema resolutions.

Floating point / half-float: Numeric storage that combines a significand and exponent to represent a wide range of values. OpenEXR commonly uses 16-bit half-float or 32-bit float, which can carry negative values and values above 1.0. Float storage does not itself make an image scene-linear. See OpenEXR.

Foot-lambert (fL): A non-SI unit of luminance still used for projected cinema screen brightness. One foot-lambert equals 3.426 cd/m². It measures light leaving the screen toward the viewer, not illumination arriving at the screen. See Nit.

Forced narrative: Timed text translating only essential foreign-language dialogue or on-screen information. Keep it separate from the only picture master so each territory can replace it.

Frame padding: Deliberately recording image area outside the intended frame lines to preserve latitude for stabilization and reframing. See Frame Padding.

Frame rate: The number of image frames captured, processed, or played per second. Specify which rate is meant when capture and playback differ. See Capture rate and timebase.

Framing chart: A photographed chart or generated pixel-accurate reticle communicating intended framing unambiguously. See Framing Charts.

G

GAM: Graded Archival Master, a term used in Netflix archival naming guidance. Identify its image state, color encoding, text state, raster, and representation rather than relying on the abbreviation outside that context.

Gamma: The exponent in a power-law relationship between signal and light, also used loosely for a curve or grading control. Not every transfer function is a simple gamma curve. PQ and camera log encodings need their specific definitions. See Transfer functions.

Gaussian splatting: A method of representing and rendering a scene using many spatial elements with color and opacity. Captured splat scenes can provide VFX appearance and spatial reference, but are not automatically accurate measured surveys. See VFX data capture.

GB / GiB: A gigabyte (GB) is 1,000,000,000 bytes. A gibibyte (GiB) is 1,073,741,824 bytes. Likewise, TB and TiB are different units. A smaller binary-unit readout does not mean the drive lost that amount of storage.2 The calculator uses decimal GB/TB and adds headroom separately. See Storage estimates.

GOP: Group of pictures. An encoded group that may use information from neighboring frames to reconstruct an image. Long-GOP compression can reduce data rates but makes decoding and frame access more dependent on other frames. See H.264 and H.265.

H

Handles: Frames beyond the edit at the head or tail. Plate handles are source frames supplied to VFX, including any needed for tracking or motion estimation. Comp handles are additional finished frames the vendor is contracted to return. The two ranges can differ or be zero. See Frame Numbering and Handles.

Harding test: A photosensitive epilepsy (PSE) test performed with HardingFPA software. It checks flashing images and spatial patterns against the selected guidelines. Other accepted tools can perform PSE testing. See Harding testing.

HDE: High Density Encoding. CODEX's lossless encoding of ARRIRAW sensor samples. It reduces storage and reconstructs the original ARRIRAW values exactly. HDE is encoded during offload or later with ARRI's HDE tool. It is not recorded in camera. See ARRIRAW, ARRIRAW HDE, and ARRICORE.

HDR: High dynamic range. In mastering and exhibition, a greater usable range from dark image areas to bright highlights than SDR. HDR also requires a specified transfer function and viewing target. A camera recording wide scene range is not automatically a finished HDR master. See HDR Mastering.

HDR10: An HDR delivery format using 10-bit PQ picture in a BT.2020 color container with static metadata. The name does not mean a 10,000-nit mastering display or ten stops of range. See HDR distribution formats.

HDRI: High-dynamic-range image. In on-set VFX work, usually an environment reference capturing lighting across a wide exposure range, often as a panorama for CGI lighting and reflections. It is not an HDR television master. See On-set VFX data.

HD / UHD: High definition and ultra-high definition. This handbook's common video masters use 1920×1080 for HD and 3840×2160 for UHD. UHD is not the 4096-pixel-wide digital-cinema 4K raster, and resolution alone does not imply HDR. See Video delivery formats.

HLG: Hybrid Log-Gamma. A relative HDR transfer function designed for broadcast, where the display maps the signal to its own peak. Contrast with PQ, which is absolute. Both are defined in ITU-R BT.2100.

HTJ2K: High-Throughput JPEG 2000. A faster block-coding method in the JPEG 2000 family. Its use as OpenEXR compression is separate from its use for IMF picture essence. Naming HTJ2K alone does not establish the container, profile, or receiver's support. See OpenEXR compression and IMF Application #2E.

I

IAB: Immersive Audio Bitstream. A format for carrying immersive sound in digital cinema and specified IMF workflows. It is not interchangeable with a home Dolby Atmos delivery file. See Dolby Atmos and IAB.

Identifier scope: The system or authority within which an asset ID is unique. A production database can guarantee that HBL_010_020 is unique for one show without making it unique across the industry. Record the ID and its scope when assets cross systems.

IDT / ODT / RRT: Input Device Transform, Output Device Transform, Reference Rendering Transform: the original ACES 1.0 names. Current ACES documentation says Input Transform and Output Transform. Since ACES 1.1, the RRT and ODT are concatenated into a single Output Transform rather than applied in sequence. The old names persist in software menus.

IMF: Interoperable Master Format. Component-based mastering in which reusable picture, audio, and timed-text assets can support multiple compositions. It is a package architecture, not a single movie codec. See IMF.

IMF-A / IMF-D: Disney-specific mastering classes, not IMF applications. Disney uses IMF-A for an archival-quality IMF master and IMF-D for a distribution mezzanine. State the actual IMF application, compression, and recipient specification.

IMF application: A constrained profile of the Interoperable Master Format specifying which essence encodings a package may use. App #2E is the common JPEG 2000 profile. Other applications support specific archival, ACES, and ProRes workflows. Use the application named in the distributor's specification. See IMF.

IMP: Interoperable Master Package. One IMF Packing List and every asset it references. A delivery can contain one or more IMPs.

Integer encoding: Storing samples as whole-number code values within a defined bit depth. The transfer function determines what those codes mean. Integer does not mean linear, and a 16-bit integer image is not equivalent to a 16-bit floating-point image. See Scene-linear and log encoding.

ISM: Image Sequence Master, a Disney-specific mastering class. Disney specifications can use TIFF, DPX, OpenEXR, or other approved representations according to the image state and workflow. The label is not a universal name for every image-sequence master.

J

JPEG 2000: An image-compression family used in digital cinema, IMF, and some camera formats. It supports lossless and lossy coding. The name alone does not establish a compression ratio, bit depth, or delivery profile. See Camera compression formats and DCP picture encoding.

K

KDM: Key Delivery Message. The encrypted key information authorizing a designated cinema playback system to play an encrypted composition during a specified time window. The correct server certificate and dates are essential. See DCP encryption.

L

Lens distortion grid: A photographed geometric pattern used to characterize a lens's spatial distortion for VFX. It is different from a framing chart, which communicates the intended crop. See Lens distortion grids.

LFOA: Last Frame of Action. The reference point marking where the program picture ends, used to align tail leaders and the tail pop. Also written LFOP (Last Frame of Picture). LFOA is the form used in most delivery specifications. See Leaders and Pops.

LiDAR: Light detection and ranging. Laser-based measurement used to record distances and spatial geometry, such as an on-set point cloud. The VFX team coordinates survey capture and its alignment to the photography. See VFX data capture.

LMT: Look Modification Transform. In ACES, a look applied in the scene-referred domain, before the Output Transform, as distinct from a display-referred grade. The right place for a show LUT-style creative look inside an ACES pipeline.

Log encoding: A nonlinear encoding that allocates code values to a wide scene-exposure range. Camera curves are commonly quasi-logarithmic, with manufacturer-specific behavior near black. “Log” does not identify the primaries, exact curve, or creative look. See Camera Log.

Longplay: The film assembled as a single continuous sequence rather than divided into reels. See Working in Reels.

Lossless / lossy compression: Lossless decoding reconstructs the encoded samples exactly. Lossy compression discards some information or precision to reduce size. “Visually lossless” is a quality judgment, not a promise of identical sample values. See Choose a compression setting.

Loudness (LKFS / LUFS): A weighted measure of audio level intended to relate more closely to perceived loudness than sample peaks do. LKFS and LUFS use equivalent scales. A delivery target must also identify the measurement method, such as whole-program or dialogue-gated loudness. See Loudness targets.

LTO: Linear Tape-Open. A data-tape technology used for backup and preservation. Tapes still need compatible drives, catalogs, integrity checks, and planned migration. An LTO copy is a storage medium, not a complete archive strategy. See Archive storage.

Luma: The weighted nonlinear Y′ signal used with color-difference channels in component video, including Y′CbCr. Luma carries most fine light-dark detail, which permits chroma subsampling, but it is not luminance or perceived brightness. The Y′ component in DCI X′Y′Z′ is encoded CIE Y, not luma. See Chroma Subsampling and Luma, Luminance, and PQ Signal Values.

Luminance: A photometric measure of light weighted for human vision. Display luminance is measured in cd/m². In CIE XYZ, Y represents luminance on an absolute scale or relative luminance when normalized. Do not use luminance as the name of the Y′ signal in component video. See Luma, Luminance, and PQ Signal Values.

LUT: Look Up Table. Sampled input-to-output values for a technical or creative operation. The table records the result, not the reasoning or original mathematics behind it. A 1D LUT maps channels independently. A 3D LUT maps RGB combinations and can mix channels. See Look Up Tables.

M

Manifest: An inventory of the files in a delivery or archive. A useful manifest records each relative path, byte size, checksum value and algorithm, stable asset ID, revision, and representation. See Asset Identity, Naming, and Versioning.

Master: An authoritative source used to make deliveries, versions, or preservation copies. The word alone does not identify the asset. Qualify it by purpose, image state, text state, target, and representation, such as display-referred Rec.709 texted ProRes master.

Matchmove: Reconstructing camera or object motion so digital elements follow the photographed scene. Lens data, distortion grids, measured geometry, and tracking references help constrain the solution. See Lens grids and camera metadata.

Matte: A channel isolating an element of a composite, delivered to the DI so the colorist can grade composite regions selectively without re-rotoscoping. See Working with Visual Effects Mattes.

MaxCLL / MaxFALL: Maximum Content Light Level and Maximum Frame-Average Light Level. Static HDR metadata describing the content, specifically the brightest pixel and brightest frame average. This contrasts with ST 2086, which describes the mastering display. See HDR Mastering.

M&E: Music and effects mix prepared for dubbing and localization. A fully-filled M&E reproduces the original music and effects without relying on the original-language dialogue stem. Simply summing the music and effects stems can omit sounds recorded with dialogue, requiring additional Foley, effects editing, and mixing to fill those gaps. Some unscripted specifications permit a partially-filled or augmented M&E. The recipient defines the required depth of fill.

Metadata: Information describing media, such as source clip name, timecode, camera settings, color encoding, or revision. It may be embedded in a file or stored in a sidecar or register. Metadata can describe an operation without baking it into the image. See Asset identity.

Mezzanine: A high-quality intermediate used to make downstream deliverables. The word describes a role, not a codec or image state. ProRes, DNx, and JPEG 2000 in IMF can serve as mezzanines. The pre-sale picture mezzanine recommended here is graded, display-referred, and usually texted; it can also serve as the archived graded video master. See What a Mezzanine Is.

MXF: Material Exchange Format. A container for media essence and metadata, used for camera recording, editorial media, and DCP/IMF track files. MXF alone does not identify the codec or color encoding. See Frames, wrappers, and codecs.

N

NAM: Non-Graded Archival Master, Netflix terminology for a fully conformed, scene-referred picture archive without the grade or display transform. Use the term for a Netflix workflow or state the properties directly in a general production record.

NaN / Inf: Non-finite floating-point values. NaN means “not a number,” while Inf represents positive or negative infinity. Neither is an ordinary highlight value. They can propagate through color and spatial processing and must be detected and repaired before delivery. See NaN and Infinite Values.

Near-field mix: An audio mix prepared and checked for home or personal listening conditions, rather than a theatrical room. It may need different balances and dynamics, not just a lower playback level. See Theatrical vs. near-field.

Nit: Informal name for the candela per square metre (cd/m²), the SI unit of luminance. Contrast with the non-SI foot-lambert, still used for projected cinema. SDR mastering references 100 nits. HDR grading displays are commonly 1,000 or 4,000.

NLE: Nonlinear editing system. An application that builds an edit from referenced media without physically cutting or overwriting the source recordings. Its native timeline is not the same as an exported interchange file. See DI Editorial Turnover.

O

OCES: Output Color Encoding Specification. In the ACES 1 architecture, the reference output-referred image between the RRT and ODT. It is not the usual grading space or a requirement to render another intermediate file. See ACES transforms.

OCIO: OpenColorIO. An open-source color-management library and Academy Software Foundation project. It lets a facility apply one color configuration across different applications. See Color Management and OpenColorIO.

OCM: Original Camera Media, also called camera originals. The original files recorded by the camera, including their associated metadata and required folder structure. OCM may be Camera RAW or processed image recordings such as ProRes. See Digital Acquisition.

Offline reference: The editorial review movie supplied with a turnover to show the intended cut, framing, timing, effects, and sound. It travels from editorial to DI. A conform check travels back from DI to editorial. See Offline Reference and Conform Checks.

Optical flow: Motion estimation used to create intermediate frames or otherwise follow motion between images. It can help retiming but may produce artifacts around occlusions, difficult motion, or fine detail. See Opticals.

Opticals: Named for effects made by optical printing of film. In a modern DI, usually editorially driven image operations handled during conform or finishing, often without formal VFX shot numbers. Simple opticals include resizes, dissolves, stabilization, and straightforward timewarps. Complex opticals are small 2D composites such as paint fixes, removals, beauty work, or split screens. See Opticals List for scope and assignment guidance.

Optional: A separate audio track containing useful language-bearing or territory-specific material that may be retained, replaced, or omitted during dubbing. Examples include discernible walla, actor efforts, foreign dialogue, song vocals, broadcasts, and censorship effects. The recipient determines what belongs in the M&E, an optional, or a dialogue guide.

OTIO: OpenTimelineIO. An open-source, vendor-neutral timeline interchange format and API, originated at Pixar and now an Academy Software Foundation project. Current finishing and VFX applications can use it for timeline turnover. Pipelines also use its structured data to compare, translate, and validate edit lists. See OTIO.

OTIO bundle: An OTIO timeline packaged with its referenced media. An .otioz bundle is one ZIP file. An .otiod bundle is a directory. See OTIO bundles.

OTT: Over-the-top. Streaming distribution outside traditional broadcast and cable.

Output Transform: In ACES, the transform that converts graded ACES data to a specific display encoding. ACES 1 combined its RRT and ODT under this user-facing name. ACES 2 uses one redesigned Output Transform.

Overcranking / undercranking: Capturing faster or slower than the playback timebase to produce slow or fast motion. This changes the capture rate, not necessarily the project's timebase. See Frame rate.

Overcut: Replacing timeline material with corresponding plates, proxies, or VFX versions while preserving the intended cut. Overcutting with plate proxies makes the available pulled frame range visible to editorial. See Visual Effects Plate Pulls.

OV / VF: Original Version and Version File in DCP terminology. A VF supplies a new composition and changed assets while depending on unchanged assets from an OV. The receiving cinema needs those referenced assets too. See DCP package contents.

P

P3: An RGB primary set used in cinema and other display encodings. Specify the white point and transfer function as well. P3-DCI, P3-D65, and Display P3 are not interchangeable complete encodings. See Display Color Spaces.

Package: Assets, compositions, and metadata gathered for transfer, delivery, or preservation. A package name does not replace the identity of each asset inside it.

Peak luminance: The greatest luminance a display produces under specified test conditions, measured in cd/m² (nits). Small-window peak and sustained full-screen output can differ. A PQ signal's encoded target is not proof that a particular display reaches it. See Luma, Luminance, and PQ Signal Values.

Photogrammetry: Reconstructing spatial geometry from overlapping photographs. Measured references help establish scale and alignment for VFX. It differs from LiDAR's direct range measurement. See VFX data capture.

PKL: Packing List. The inventory of assets belonging to a DCP or IMF package, including integrity-check information such as hashes. It does not determine playback order. See DCP package documents.

Plate: Source imagery prepared as the basis of a VFX shot, normally pulled from the camera originals or film scans into the agreed working format. It is not necessarily the camera-original file and need not have a viewing look baked in. See Visual Effects Plate Pulls.

Plate pull: Preparing the selected source frames for VFX in the agreed format, color encoding, resolution, and frame range. It usually includes a controlled decode and transcode, not simply sending the camera-original clip. See VFX Plate Pulls and Turnover.

PQ: Perceptual Quantization, also called the Perceptual Quantizer. An absolute HDR transfer function. After signal-range and any color-difference decoding, its EOTF maps each nonlinear RGB component to an absolute linear-light display component. Pixel luminance is calculated from the linear components, not by applying PQ directly to luma. See PQ signal values.

Previs (previsualization, also previz): Planning a shot or sequence visually before execution, from storyboards and animatics to detailed 3D scenes. The required precision depends on the creative and production decisions it must support. See Previsualization.

Primaries: The reference red, green, and blue coordinates that define an RGB system's basis. Different primary sets assign different RGB numbers to the same color. Primaries do not specify the transfer function or white point. See Color spaces.

Printmaster: The final composite audio mix for a defined soundfield. Theatrical and near-field printmasters are separate deliverables because they use different monitoring conditions and dynamic-range targets.

ProRes: Apple's professional codec family, published as SMPTE RDD 36, with MXF carriage in RDD 44 and an IMF application in RDD 45. The 4444 variants sustain multiple generations without meaningful degradation. See Apple ProRes.

ProRes RAW: Apple's camera-acquisition RAW format (2018). Despite the shared name it is not a member of the ProRes intermediate family and is not applicable to transcoded VFX plate pulls.

Provenance: The record of who created an asset, from what source, when, and why. A useful record also names the source revision, change reason, state, approver, and approval date.

Proxy: A lower-resolution or lower-bandwidth representation used for editorial, review, or conform checking. A proxy can itself be a requested deliverable, but it does not replace the full-quality picture master. See Digital Dailies.

Proxy-C / Proxy-D: Disney-specific conformance proxies. The labels describe Disney workflows, not general proxy classes. Use the recipient's current definition when they appear in a schedule.

Q

QC: Quality control. Technical inspection of media and packages against the intended content and agreed specification. QC is distinct from creative approval and from automated file validation. See VFX QC and Delivery QC.

R

Raster: The image's rectangular pixel grid, stated as width × height. The stored raster can include padding or black outside the intended picture. It does not by itself identify the creative aspect ratio. See Resolutions and framing.

Rec.2020 / BT.2020: The UHD television recommendation that defines a wide-gamut RGB primary set with D65 white. Its primaries are also used in BT.2100 HDR. A BT.2020 color container does not mean every encoded color reaches its boundary or that a display reproduces the entire gamut. See BT.2020 as a color container.

Rec.709 / BT.709: The HDTV recommendation commonly used to identify SDR RGB primaries and D65 white. The same primaries can also be used for UHD SDR. Specify the viewing EOTF separately, normally BT.1886 / Gamma 2.4 for mastering. See Display Color Spaces.

Reel: A ~20-minute division of a feature, inherited from the physical limits of print stock and retained because it parallelizes editorial and grading labor. See Working in Reels.

Representation: The same underlying asset or revision in another technical form. A full-quality EXR sequence and its review QuickTime can be two representations of one approved VFX revision.

Resampling: Calculating new image samples during scaling, rotation, stabilization, or another geometric operation. The filter influences sharpness, aliasing, and ringing. Repeated resampling can change an image even when the final framing returns to its starting point. See Scaling Algorithms.

Revision: Changed work intended for the same context and use. A correction to an approved master creates a new revision. Another codec alone creates a new representation, not necessarily a new creative revision.

RGC: Reference Gamut Compression. An ACES transform that maps out-of-gamut camera values into a workable range. It prevents downstream artifacts around saturated highlights.

Ringing: Bright or dark ripples near a high-contrast edge introduced by some reconstruction filters. Undershoot can produce negative pixel values. Clamping those values does not remove the filter behavior that caused them. See Scaling Algorithms.

Rotoscoping: Creating and animating masks to isolate parts of an image over time, often by tracing photographed shapes. Roto mattes may be used in compositing or delivered for selective grading. See VFX mattes.

Roundtrip: Sending material out of a system and bringing it back, then verifying that nothing changed but the intended work. The standard proof that a pipeline is correctly configured. See zero net change.

S

Sample rate: The number of audio samples recorded or played per second, such as 48,000 samples per second (48 kHz). It is independent of picture frame rate. Shooting at 23.976 does not by itself require recording at 48.048 kHz. See Picture and sound rates.

Scene-linear: Encoding proportional to relative scene light. Doubling a value represents one stop more exposure. VFX commonly stores these values in floating point, with 18-percent gray at 0.18 and highlights allowed above 1.0. Linear describes the encoding, not the storage type. See Scene-Linear.

Scene-referred: Image values related to light or exposure in the photographed or synthesized scene, rather than final output on a target display. They may be linear or nonlinearly encoded. See Scene-Referred Imagery.

Scope: See Flat / Scope for the standardized 2.39:1 cinema format.

SDH: Subtitles for the Deaf and Hard of Hearing. SDH includes dialogue plus relevant speaker, music, and sound information. SDH describes the content. Closed describes whether the viewer can turn the presentation on or off. See Captions and subtitles.

SDR: Standard dynamic range. Display-referred mastering for conventional video or cinema targets rather than HDR. Traditional SDR transfer functions are relative, with actual light output established by display calibration. SDR does not mean low resolution. See SDR display transfer functions.

Shaper LUT: A one-dimensional mapping used to place a wide input range into the useful domain of another LUT, commonly scene-linear to log before a bounded 3D LUT. It is part of the transform chain, not an optional viewing decoration. See LUT types.

Show LUT: The production's approved viewing recipe. It may combine the creative look and display rendering in one LUT, or carry only the creative portion of a separately managed viewing chain. See The Show LUT.

Slap comp: A rough temporary composite made in editorial to convey a story beat before the real shot exists. See Creative Editorial.

Slate frame: An informational frame prepended to a delivered comp carrying shot, version, vendor, artist, and framing details. See Frame Numbering and Handles.

ST 2086: SMPTE's static HDR metadata standard describing the mastering display, including its primaries, white point, and minimum and maximum luminance. Frequently confused with MaxCLL/MaxFALL, which describe the content instead. See HDR Mastering.

State: A workflow decision recorded independently from identity and revision. The handbook uses WIP, in review, approved, rejected, final, delivered, and superseded. Final means creative and technical approval is complete. Delivered means that exact revision reached the named recipient.

Stem: A grouped audio component that combines with other stems to reconstruct a mix. Common stems include dialogue, music, effects, M&E, narration, and audience reactions.

Stop: A factor of two in exposure or scene-linear light. One stop up doubles the value, and one stop down halves it. Equal stop changes are not equal increments in every encoded signal. See Scene-Linear.

Subtitles: Timed text presenting dialogue, commonly as translation. SDH also conveys relevant non-dialogue sounds and speaker information. Subtitles may be separate timed text or burned into picture. See Captions and subtitles.

Supplemental package: A package containing new compositions or replacement resources while referencing unchanged assets from an earlier package. IMF commonly uses supplemental IMPs to add languages, text configurations, corrections, or alternate editions without duplicating every track file.

T

Tape name: The unique source identifier that, with timecode, links a timeline clip back to camera original media. Without it, conform is guesswork. See Edit List Generation.

TD: Technical Director. A technical specialist supporting an aspect of VFX production. In this handbook's image-pipeline instructions, the relevant TD or pipeline lead defines and tests the color and image-processing workflow. See Visual Effects Quality Control.

Textless bed: The collection of complete clean replacement shots needed to rebuild localized graphics. It is not the same as a complete program-length textless master.

Through-edit: A cut between adjacent segments of one continuous source shot where the picture and its treatment do not change. It is sometimes referred to as a phantom edit or match-frame edit. Remove unnecessary through-edits before VFX or DI turnover. See Simplifying Timelines.

TIFF: Tagged Image File Format. An image format commonly used as a sequence for mastering and preservation. The extension does not establish bit depth, compression, or color encoding. See TIFF.

Timebase: The reference frame rate used to interpret and play a sequence. A camera can capture more frames per second than that timebase for slow motion without changing the production's playback rate. See Frame rate and timebase.

Timecode: A frame-addressing system using hours, minutes, seconds, and frame numbers. The number alone may not reveal the frame rate. 23.976 and 24.000 can show identical timecode labels while playing at different speeds. See Frame rate.

Timed text: Text with timing and presentation information, used for subtitles, captions, or forced narratives. It is distinct from text already rendered into picture. See Captions and subtitles.

Tone mapping: Mapping image levels into a target display's luminance range while managing contrast and highlight behavior. It can be part of scene-to-display rendering or adaptation of an existing display master. See HDR display adaptation.

Track file: In DCP and IMF, a media file carrying a component such as picture, audio, or timed text. A CPL selects its role, entry point, and duration in a composition. See IMF structure.

Transcode / rewrap: Transcoding decodes media and encodes a new representation. Rewrapping changes the container without re-encoding the compressed picture or audio essence. Neither term alone establishes the output color encoding or metadata. See Frames, wrappers, and codecs.

Transfer function: The mathematical relationship between encoded signal values and linear-light values. Camera log, SDR display curves, PQ, and HLG serve different purposes. A transfer function alone does not define a complete color space. See Transfer Functions.

Turnover: The delivery of edit lists, media, and references from editorial to the DI or to VFX. See Editorial Turnover for DI.

Two-pop: A single-frame audio and picture sync reference two seconds before first frame of picture. See Leaders and Pops.

U

USD: Universal Scene Description, also called OpenUSD. Pixar's open-source framework for describing, composing, and exchanging 3D scene data, including geometry, shading, and lighting. It describes a scene, not just a rendered image.3

V

Variant: Related work intended for another context, territory, language, audience, or presentation. A censored cut and a localized version are variants. A correction intended for the same use is a revision.

VFX: Visual effects. Image work that creates, combines, removes, or changes elements beyond what the principal photography alone supplies. It includes invisible repairs as well as elaborate CGI shots. See VFX Production Management.

VFX Reference Platform: An annual target set of software-library versions intended to reduce incompatibility across VFX tools. It does not replace a show specification or a tested color configuration. See VFX Reference Platform.

Viewing transform: A broad term for the operation or chain that makes a working image viewable on a selected display. It may mean only display rendering or a complete chain that also includes the creative look. The term alone is not a complete pipeline specification. See Translate the Color-Management Terms.

Virtual primaries: Mathematical RGB basis primaries outside the physically realizable range of human-visible color stimuli. They are not extra colors a camera can see. They provide a useful coordinate system for encoding and processing color. See Real and Virtual Primaries.

W

White balance: Camera or decode processing that adjusts color relationships so the selected illuminant is treated as neutral. It is not the same as calibrating a display's white point. Its adjustability after recording depends on the format and processing already applied. See Camera RAW or ProRes log.

White point: The chromaticity defined as neutral white for an encoding or calibrated display. It is separate from white luminance. A working-space white point need not match the display's white when the output pipeline includes the appropriate adaptation. See The ACES white point.

WIP: Work in progress. Material that has not reached final approval. A WIP VFX shot can be reviewed in editorial or inserted into the DI so work proceeds before the final arrives. See Digital Intermediate.

Working color space: The color encoding in which grading, compositing, or another image operation is performed. It can differ from capture and delivery encodings. Choosing it affects how tools behave, not just how the image is labeled. See Application-Native Color Management.

Z

Zero net change: The requirement that non-VFX regions return through compositing without an unintended color, encoding, or resampling change. Difference-test matching formats at a tolerance of zero. When formats differ, compare them in a common high-precision state against the numerical tolerance recorded in the VFX specification. Fix every unexplained difference before approval. See Zero Net Change.

  1. Blackmagic Design, “Blackmagic RAW,” named section “Advanced De-Mosaic.” ↩

  2. National Institute of Standards and Technology, “Prefixes for binary multiples,” tables “Prefixes for binary multiples” and “Examples and comparisons with SI prefixes.” ↩

  3. Pixar, “Introduction to USD,” named sections “What is USD?” and “What can USD do?” ↩