Short answer
ACES, the Academy Color Encoding System, is a published colour-management framework for film and television. It standardises the route, not the look: every camera's footage is converted by an input transform into one scene-referred encoding, work is done in defined working spaces, and a published output transform renders the result for each kind of display. Using it means cameras, visual-effects houses and grading suites agree on what the numbers mean. It does not decide how a film should look.
A production may shoot on film and on two makes of digital camera, send hundreds of shots to effects companies, and deliver for cinema, SDR and HDR. Each camera has its own log curve and primaries, and each destination its own standard. Without a shared reference each handover needs its own conversion, and a mistake in any one produces a shot that will not match. A cinematographer quoted in American Cinematographer put the appeal as control, repeatability and predictability. ACES answers with one encoding in the middle that everything is converted into and out of.
An input transform converts each camera's data into ACES2065-1, a linear, scene-referred encoding on very wide primaries called AP0, which is the form used for exchange and archive. Work is done in two related spaces on the smaller AP1 primaries: ACEScg, linear, for rendering and compositing, and ACEScct, quasi-logarithmic, for grading. A look transform can apply a creative look to the scene-referred data. An output transform renders it for a named display. The transforms used can be recorded in an ACES metadata file so that another facility can reproduce them.
Grading controls were designed for film-scan log, where equal steps of signal are roughly equal steps of exposure. Linear data behaves badly under them. ACEScct exists for that reason: its specification calls it a quasi-logarithmic encoding for use within colour grading systems. Above a small break point it is a pure logarithm; below it the curve becomes a straight line, a toe, so that lifting the shadows behaves more as it does with traditional film log. The specification is explicit that it is a working encoding, not one for interchange or archive.
In ACES 1 the rendering was described as two stages, a reference rendering transform and then an output device transform for each display. ACES 2 treats it as a single output transform and redesigns it. The Academy's summary lists a tone scale with lower mid-tone contrast and a gentler highlight roll-off, gamut mapping intended to reduce clipping, and simpler parameters for unusual displays. The rendering now works in a lightness, colourfulness and hue space taken from a colour appearance model, applying the tone scale, then chroma compression, then gamut compression. A project must hold one version throughout.
ACES does not supply a look, and its output transform is one rendering among several a production could choose; some work in a camera maker's pipeline or a facility's own. It does not guarantee that an input transform exists for every camera mode, or that two implementations match to the last digit. It is free and open, and adoption is a choice. The Vault, a small feature shot on two digital cameras, is recorded as finishing in ACES with advice from Academy staff, and its cinematographer noted that a production can adopt it in post as long as the camera's input transform is known.
| Component | Publisher | Edition | What it is | Limit |
|---|---|---|---|---|
| ACES2065-1 | Academy of Motion Picture Arts and Sciences | ACES documentation dated September 2025 | The ACES interchange and archive encoding: scene-referred, linear, with the very wide AP0 primaries. Input transforms convert each camera's footage into it. | An exchange encoding; grading and rendering are done in the two encodings below. |
| ACEScg | Academy of Motion Picture Arts and Sciences | ACES documentation dated September 2025 | Scene-referred and linear like ACES2065-1 but with the smaller AP1 primaries, intended for computer-graphics rendering and compositing. | A working space, not a delivery or archive format. |
| ACEScct | Academy of Motion Picture Arts and Sciences | ACES documentation dated September 2025 | A quasi-logarithmic encoding of ACES data on the AP1 primaries, for use inside grading systems. It is logarithmic above a small break point and linear below it, a toe that makes shadow controls behave more like those of film-scan log. | Transient by design: the specification says it is not for interchange or archive. |
| ACES 2 output transform | Academy of Motion Picture Arts and Sciences | ACES 2; documentation dated 10 September 2025 | The rendering from scene-referred ACES to a named display. ACES 1 described it as a reference rendering transform followed by an output device transform; ACES 2 treats it as one transform that applies a tone scale, chroma compression and gamut compression in a lightness, colourfulness and hue space. | An ACES 1 and an ACES 2 rendering of the same grade differ; a project has to hold one version throughout. |
| Common LUT Format (CLF) | Academy of Motion Picture Arts and Sciences | Version 3.0 | An XML format for exchanging colour transforms as a list of nodes: one-dimensional and three-dimensional lookup tables, matrices and an ASC CDL node among them. | Many lookup tables in circulation are in simpler vendor formats that record less. |
Each statement is labelled by kind — established fact, a standard’s requirement, observed market data, a convention, or Colourwise’s own interpretation or analysis — with the strength of the evidence behind it.
FactStrong evidence
The ACES documentation defines ACES2065-1 as the linear, scene-referred interchange encoding on AP0 primaries, with ACEScg (linear) and ACEScct (quasi-logarithmic) as working encodings on the smaller AP1 primaries.
Source: ACES Documentation: system overview, encodings and output transforms; ACEScct: a quasi-logarithmic encoding of ACES data for use within colour grading systems
StandardStrong evidence
The ACEScct specification defines an encoding that is purely logarithmic above a small break point and linear below it, identical to ACEScc above the break, and states that it is intended as a working encoding within grading systems and not for interchange or archive.
Source: ACEScct: a quasi-logarithmic encoding of ACES data for use within colour grading systems
FactStrong evidence
ACES 2 replaces the separately named reference rendering transform and output device transform of ACES 1 with a single output transform whose rendering applies a tone scale, chroma compression and gamut compression in a lightness, colourfulness and hue space.
Caveat: The documentation pages are dated September 2025 and do not give the release date of ACES 2.
Source: Output Transforms (ACES Documentation); About ACES 2 (ACES Documentation)
FactModerate evidence
The Vault, a low-budget feature shot on two Canon C300 Mark II cameras, was finished in ACES by a post-production house that took guidance from Academy staff.
Source: ACES for Indies (Debra Kaufman, American Cinematographer, 7 May 2018)
Colourwise interpretationModerate evidence
ACES standardises interchange and rendering, not appearance: two ACES productions share what their numbers mean and can still look entirely unlike each other.
Based on: Follows from the documented structure, in which the creative look is a separate look transform and grade applied between standard input and output transforms.
Source: ACES Documentation: system overview, encodings and output transforms
Reviewed 6 October 2026. Colourwise summarises its sources in its own words and does not reproduce standards text or proprietary colour data. Spotted an error? Tell us.