Color Volume vs Color Gamut Color Volume vs Color Gamut

What gamut and volume each measure

Color gamut, in its traditional sense, is the convex set of chromaticity coordinates a display can reproduce, conventionally drawn as a triangle on the CIE 1931 chromaticity diagram whose vertices are the display's red, green, and blue primaries. The diagram intentionally omits the luminance dimension. Color volume extends gamut into a third dimension by adding luminance — it describes which colors a display can reproduce at every brightness level, not just at one reference luminance, and is visualized as a 3D solid rather than a 2D triangle.

To clarify the distinction, standards organizations have shifted terminology — the triangular area on a CIE 1931 or 1976 diagram is now called chromaticity gamut, while color gamut is reserved for the three-dimensional shape that includes luminance. Color volume is essentially the modern, correct expansion of what 'color gamut' implied.

Why does the third dimension matter? Because a display's color performance is intrinsically three-dimensional: the colors it can render are a function of the luminance level of the picture. Analyzing only one luminance level — the standard SDR practice — hides how the display behaves across the wider luminance range that HDR exercises.

How each is measured

A 2D gamut measurement reads the chromaticity of the red, green, and blue primaries at a single reference luminance and reports the resulting triangle's coverage of a target space (Rec.709, DCI-P3, BT.2020). A color volume measurement, by contrast, sweeps those primaries — and intermediate stimuli — across multiple brightness levels and integrates the resulting 3D solid.

The perceptual color space chosen for that integration is consequential. ICtCp (also written ITP) is a color representation specified in ITU-R Recommendation BT.2100, the same recommendation that defines HDR television. It was developed by Dolby Laboratories from the IPT color space of Ebner and Fairchild and is the perceptual color space most often used to compute HDR color volume. In ICtCp the I component is a luma channel that represents brightness; C_T is a blue-yellow chroma component (named from tritanopia) and C_P is a red-green chroma component (named from protanopia). The encoded I correlates with true luminance at 0.998 versus 0.819 for Y'CbCr, which is why ICtCp is preferred for HDR color-difference and color-volume work. ICtCp's improved hue linearity versus Y'CbCr is one of the explicit reasons it is recommended for color volume mapping — hue stays consistent as luminance changes, so the integrated 3D shape is a more faithful perceptual representation than the same calculation done in Y'CbCr.

Dolby's Measuring Perceptual Color Volume methodology converts a display's measured response into either CIE L*a*b* or ICtCp coordinates and reports the result in Millions of Distinguishable Colors (MDC). The MDC scaling is calibrated so that one unit in ICtCp is roughly one just-noticeable difference, and ICtCp is recommended over CIELAB because it is more perceptually uniform for HDR. Note: no SDO-mandated stimulus set or fixed list of brightness steps for color-volume measurement was located.

Why volume matters more than gamut for HDR

Two displays measuring the same on a 2D chromaticity gamut chart can deliver very different HDR color experiences. A worked example: one WOLED monitor measures 101.6% DCI-P3 in a traditional 2D gamut measurement, but once the luminance range required for HDR is included its coverage falls to 67%. A QD-OLED panel maintains DCI-P3 across its full luminance range, reaching 116% color volume.

The architectural reason is specific. WOLED panels lose color saturation as overall luminance rises: the white sub-pixel that supplies WOLED's 'white boost' increases the proportional white content in the output, washing out saturated primaries. Color luminance therefore cannot scale with white luminance — TFTCentral measured colors on a QD-OLED panel reaching roughly 140% higher (2.4×) luminance than on a comparable WOLED. This is also why reducing a display's brightness performance to a single white-luminance number is incomplete and can mislead. Two panels with the same peak-white spec can have very different color brightness — the very thing color volume captures — because color luminance and white luminance scale separately on architectures that use a white sub-pixel or color filters.

A meaningful color-volume plot makes visible what a 2D gamut chart hides. Portrait Displays' worked example notes that 'Display 1 clearly is unable to render highly saturated colors in the upper luminance ranges of DCI-P3.' That failure mode does not appear on a chromaticity diagram because the diagram has no luminance axis. Note: no measurement-tier numerical comparison of mini-LED LCD vs OLED color volume in 1000+ nit highlights was retrieved.

Common gamut/volume mistakes

The first mistake is treating '100% DCI-P3' as an HDR claim. DCI-P3 was defined by the Digital Cinema Initiative for theatrical projection at a nominal white luminance of 48 cd/m² (14 ftL) under a 2.6 gamma. A '100% DCI-P3 color volume' claim only spans the cinema reference container, which tops out far below the 1,000–4,000 nit peaks of HDR mastering.

The second mistake is comparing percentages across target spaces. Quoting color volume against DCI-P3 is far easier than quoting it against BT.2020. DCI-P3 covers 53.6% of the CIE 1931 visible-color area; BT.2020 covers 75.8%. A '100% DCI-P3 color volume' headline is therefore not comparable to a percent-BT.2020 figure, and the former says little about how a display will fare with future BT.2020-mastered content.

The third mistake is comparing MDC numbers without checking the underlying coordinate system. Dolby's perceptual-color-volume methodology supports both CIE L*a*b* and ICtCp and recommends ICtCp as more perceptually uniform for HDR; Portrait Displays' write-up describes color volume exclusively in CIELAB (L*a*b*) terms. Absolute MDC or volume figures from the two coordinate systems are not directly comparable, and any cited number should declare which space underlies it. Note: paired MDC values for the same display computed in CIELAB vs ICtCp were not captured.