How Photography Works | Gamut Mapping and Clipping — What Happens When a Colour Cannot Survive the Destination

The numbers 255, 0, 0 do not describe one universal red.

They describe a red only after a colour space tells the system what those numbers mean. Change the colour space and the same triplet can refer to a different visible colour. Change the display and the device may not reproduce the intended colour exactly. Send the image to print and the available range changes again.

A colour space gives numerical colour values a defined visual meaning. A gamut describes the range of colours that a device or colour space can represent.

This article continues the canonical How Photography Works | Every Photograph Leaves Something Out knowledge map. Colour Is Reconstructed, Not Collected Whole owns camera-side colour formation. This article owns the next handoff: how rendered colour values become portable coordinates that must survive software, displays, printers and other output systems.

Quick Read

RGB numbers do not have complete colour meaning by themselves. A colour space defines primaries, a white point and a transfer relationship so those numbers can be connected to intended colours. Different spaces cover different gamuts. sRGB is widely used for general web delivery; wider-gamut working spaces can represent colours outside sRGB, but only if the rest of the workflow understands the profile. When a colour falls outside the gamut of an output device, software must clip it, compress it or otherwise map it into a reproducible range. Missing or misinterpreted profiles can therefore make the same file appear oversaturated, undersaturated or shifted across devices.

numeric values + colour-space definition + device profile → intended colour → device-limited reproduction

RGB Values Need Coordinates

Imagine someone gives you the coordinates 10, 20 without telling you whether the map uses metres, degrees or another coordinate system. The numbers are incomplete. RGB values work similarly.

The colour space supplies the reference system. It tells software which red, green and blue primaries the numbers refer to, which white is considered neutral and how code values correspond to luminance.

sRGB Became a Common Delivery Language

sRGB covers a practical range designed around common display behaviour and became a widely supported interchange space for consumer imaging and the web. Its strength is not that it contains every visible colour. Its strength is predictability across a large ecosystem.

For an ordinary webpage, a correctly tagged sRGB image is often the safest delivery choice because browsers, operating systems and devices are generally built to handle it.

Wider Gamut Means More Possible Colours, Not Automatically Better Colour

A wider-gamut colour space can encode more saturated colours than sRGB in some regions. This can be useful for high-quality editing, certain displays and print workflows.

But wider is not automatically better for every delivery path. If the receiving software ignores the profile or assumes sRGB, the numbers may be interpreted incorrectly and the image can look dramatically oversaturated or shifted.

The Profile Travels With the Meaning

An embedded ICC profile identifies how the file’s encoded values should be interpreted. Colour-managed software reads that profile, transforms the values through a device-independent reference and sends appropriate values to the display or printer profile.

Without that information, a receiver may guess. Guessing works only when the assumed space happens to match the actual one.

The Display Has Its Own Gamut

A file can encode a saturated green that one monitor reproduces while another cannot. The second display must approximate it with the nearest or mapped colour it can produce.

The file has not changed, but the visible output has. This is why colour management extends beyond photography software into display calibration and device characterisation.

A Printer Has a Different Colour Shape Again

Displays emit light. Prints reflect light from inks or pigments on paper. Their colour capabilities are shaped by different physics. A printer may reproduce some colours differently from a monitor and may reach colours in one region while falling short in another.

The transition from screen to paper is therefore a gamut-mapping problem, not simply a matter of making RGB numbers permanent with ink.

Out-of-Gamut Colours Need a Decision

Suppose a flower contains a saturated colour outside the destination printer’s gamut. The output system cannot reproduce that exact target. Something has to change.

One approach clips the colour to the nearest reproducible boundary. Another compresses a wider region so relationships among colours are preserved more smoothly. Different rendering intents and conversion systems make different trade-offs.

Clipping Can Flatten Distinct Colours Into One

If several different saturated source colours all lie beyond the same output boundary, simple clipping can map them to similar output values. Distinctions disappear.

This is the colour-space version of highlights clipping: several distinct inputs collapse into one reproducible result, though the mechanism and domain are different.

Gamut Mapping Can Preserve Relationships by Moving More Colours

A perceptual-style mapping may move even some in-gamut colours so the overall relationship to out-of-gamut colours remains plausible. The entire colour field changes slightly to avoid a harsh boundary.

The output may be less numerically literal but more visually coherent. Again, representation is a trade-off.

Soft Proofing Is a Preview of the Constraint

Colour-managed editing software can simulate how an image may reproduce through a particular printer-paper profile on a calibrated display. This is called soft proofing.

The preview is still limited by the monitor’s own gamut and viewing conditions, but it can reveal where saturation, shadow separation or paper white will change before ink is used.

White Point Changes the Reference Neutral

Colour spaces and devices define or operate around particular white references. A display’s white, a sheet of paper under warm gallery light and a daylight viewing booth are not physically identical.

This connects colour spaces to White Balance. Capture neutrality and output neutrality are different stages of the same larger adaptation problem.

Same RGB Numbers, Different Spaces, Different Colours

Take a numerical triplet such as R=200, G=50, B=50. In sRGB it refers to one defined colour. Interpret the same numbers in a wider space and the primaries are different, so the resulting colour is different.

This is why assigning a new profile and converting to a new profile are not the same operation. Assignment changes what the existing numbers mean. Conversion changes the numbers so the intended colour remains as similar as possible in the destination space.

Assigning the Wrong Profile Reinterprets the File

If a wide-gamut image is mistakenly assigned sRGB without conversion, the same numerical values are interpreted with different primaries. Colours can change sharply.

The pixels did not become corrupted. Their coordinate system was changed underneath them.

Converting Tries to Preserve Appearance

A colour-space conversion calculates new destination numbers intended to reproduce the source colour as closely as the destination gamut permits.

This is a transformation of representation with the goal of preserving appearance rather than preserving numerical values.

Working Space and Delivery Space Can Be Different

A photographer may edit in a wide-gamut working space to avoid clipping saturated colours during major adjustments, then convert a final derivative to sRGB for general web use.

The master and delivery version therefore can inhabit different colour spaces by design. Versions owns the identity distinction; colour spaces explain one technical reason those versions may differ.

A Wider Space With Too Few Code Values Can Create Its Own Problem

Spread a fixed number of numerical steps across a wider gamut and the distance between adjacent encoded values can become larger. At low bit depth and under strong editing, this can increase the risk of visible banding or quantisation error.

Article 36, Bit Depth and Banding, follows that precision problem.

Web Platforms Can Convert or Strip Colour Information

A platform may resize, recompress or convert uploads. Modern browsers are generally colour-aware, but a downstream derivative can still differ from the master because the platform changed pixel values, profile data or both.

When accurate colour matters, verify the actual delivered file rather than assuming the upload pipeline preserved every property.

Screenshots Add the Display Into the Colour Chain

A screenshot captures rendered pixel values generated by software for a display environment. Depending on operating system and format, colour-profile handling can differ from the source image.

The screenshot is therefore not merely another copy. It is a new rendering event in the colour pipeline.

Colour Gamut Is Not Human Vision

No common RGB working space contains every colour humans can perceive. Every practical colour space is a selected region and coordinate system built for a job.

Even very wide spaces are representations, not the totality of visible colour.

Camera Gamut Is Not a Simple Triangle Either

Camera sensors have overlapping spectral sensitivities that do not correspond directly to standard display primaries. RAW conversion uses matrices, profiles or more complex transforms to map sensor responses into a working colour representation.

This is why camera colour cannot be understood as “the sensor records Adobe RGB” or “the sensor records sRGB” in a literal raw-measurement sense. Those are later encoding choices.

Scientific Colour Requires More Than a Pretty Profile

If colour is being used as a measurement, controlled illumination, calibrated targets, spectral characteristics and traceable processing matter. A pleasing colour-managed print is not automatically a scientific colour measurement.

Scientific Photography Is Measurement With a Camera owns that higher evidence contract.

Colour Spaces Change What “Out of Range” Means

A colour can be inside one working space, outside another display’s gamut and still visible to a human observer. “Out of gamut” therefore always needs an object: out of which gamut?

The phrase is relational, not absolute.

The Colour-Space Audit

1. Source profile: which space gives the file numbers meaning?
2. Working space: where is the image being edited?
3. Assignment or conversion: are numbers being reinterpreted or transformed?
4. Destination: browser, phone, wide-gamut display, printer or archive?
5. Device profile: is the output characterised?
6. Out-of-gamut colours: which colours cannot be reproduced exactly?
7. Rendering: clipping, relative mapping or perceptual compression?
8. White point: what neutral reference is assumed?
9. Bit depth: is numerical precision adequate for the space and editing?
10. Purpose: robust delivery, expressive editing, print matching or measurement?

Photography Laboratory 1: Assign Versus Convert

On a copy of an image in colour-managed software, compare assigning a different profile with converting to that profile. Observe how assignment changes appearance immediately while conversion tries to preserve it by changing numbers.

Photography Laboratory 2: sRGB Delivery Copy

Make a web-sized sRGB derivative from a wide-gamut master. Compare saturated colours in colour-managed software and inspect which values moved during conversion.

Photography Laboratory 3: Soft Proof

If you have a valid printer-paper profile and calibrated workflow, soft-proof a colourful image. Mark which colours change and how different rendering intents manage them.

For Primary Readers

Think of a colour space as a box of crayons. One box may contain colours another box does not. If the picture asks for a crayon that is missing, you must choose the closest one.

For Secondary Readers

Connect RGB values to coordinate systems. Explain why numbers need a defined set of primaries and white point before they identify a colour.

For Advanced Readers

Analyse RGB spaces through chromaticity primaries, white points and transfer functions, then connect ICC transforms to profile connection spaces, device characterisation, rendering intents and gamut-boundary mapping. Distinguish sensor spectral response from output colour-space coordinates.

Common Misconceptions

• “RGB numbers define colour by themselves.” They require a colour-space definition.
• “Wider gamut is always better.” It helps only when the workflow and destination support it appropriately.
• “Assigning a profile and converting are the same.” Assignment changes interpretation; conversion changes values to preserve appearance.
• “A monitor can show every colour in the file.” The display has its own finite gamut.
• “Out of gamut means invisible to humans.” It means outside a particular colour space or device capability.

Frequently Asked Questions

Why is sRGB commonly used online?

Because it is widely supported and provides a predictable general-purpose delivery space across browsers and consumer devices.

What happens to an out-of-gamut colour?

The conversion system must map it into a reproducible destination colour, often by clipping or compressing colour relationships.

Why does an untagged image sometimes look wrong?

Without a profile, software may assume the wrong colour space and interpret the same numerical values with the wrong primaries or transfer relationship.

Final Thought: Colour Needs a Map

A photograph does not carry colour as self-explanatory numbers. It carries coordinates whose meaning depends on a defined system, and those coordinates still have to survive the limits of the device that finally shows them.

Colour management begins when we stop asking whether the numbers stayed the same and start asking whether the intended colour survived the journey.