Computer Atlas

Color Space

Also known as: colour space, color spaces, color model

core intermediate concept 3 min read · Updated 2026-06-07

A specific way of describing colours numerically — defining which RGB triplets (or other coordinates) mean which colours in the real world.

Primary domain
Graphics & Media
Sub-category
Photograph Manipulation & Image Compression

In simple terms

The same triplet (255, 0, 0) can mean a different shade of red depending on which color space it’s in. A color space defines what every numeric value actually corresponds to as light hitting your eye. Get this wrong and your reds look orange, your skin tones look ill, and your HDR video looks washed out.

More detail

Color spaces have two parts:

  • Color model — how coordinates are organised. RGB (red/green/blue), CMYK (printing), HSL/HSV (intuitive), Lab (perceptual).
  • Color gamut / primaries + transfer function — what those coordinates physically mean.

Common RGB color spaces in 2026:

  • sRGB — the long-standing web/desktop default. Limited gamut, gamma ≈ 2.2.
  • Display-P3 — wider gamut, default on modern Apple displays. ~25% larger than sRGB.
  • Adobe RGB — print/photo working space.
  • Rec. 709 — HDTV; essentially the same as sRGB.
  • Rec. 2020 / Rec. 2100 — UHD / HDR.
  • DCI-P3 — cinema projection.

Important details:

  • Transfer function (gamma / PQ / HLG) — converts between linear light and the encoded value. Real lighting math must be done in linear; the encoded value is non-linear.
  • White point — defines what “white” looks like (D65 for most displays, D50 for print).
  • Wide-gamut and HDR require not just more bits but knowing the color space at every step of the pipeline.

If your color space metadata is lost or wrong, software typically assumes sRGB — which is why HDR videos often look flat in old apps and modern photos look oversaturated in old browsers.

Why it matters

Color accuracy underpins design, photography, video, and any UI that needs to look the same across devices. Bad color management is a major source of “looks great on my Mac, terrible on the customer’s phone” bugs.

Real-world examples

  • Editing an HDR video in an sRGB tool yields washed-out clipped highlights.

  • A web design that looks great on a Display-P3 MacBook can look dull on an sRGB monitor — and vice versa.

  • ICC profiles are how operating systems tell apps the exact characteristics of a display.

  • Apple’s ProRes RAW and Sony’s V-Log capture in extremely wide color spaces (Rec. 2020, S-Gamut3) so colorists can grade afterwards into Rec. 709 for delivery without clipping.

Common misconceptions

  • “RGB is RGB.” Every RGB triplet is meaningless without a color space.
  • “More bits = better color.” Bit depth helps with banding; gamut and accuracy come from the color space, not the bit count.

Learn next

The pixel itself: pixel. How colors get stored in files: image format.

Relationships

Requires
Leads to

Neighborhood

A visual companion to the relationships above. Click any node to visit that topic.