Skip to content
CHRM-MID◆Midnight Editorial
AtmosphereSartorial Silence
Accent—
Surface—
001▲
MONO-07·VOL. II · NO. 1
◆Color Science & Chromatic Theory

The OKLCH Mathematical Manifesto

Useful coordinates, explicit gamut limits and a contrast calculation

AUTHOR
ARCHIVAL SERIESThe Chromologium · Curatorial Monographs
DATE PUBLISHED
READING TIME4 MIN ESTIMATE · 629 WORDS
CURATORIAL THESIS

Choose colors in a useful space, then evaluate the actual pair and presentation; no coordinate can stand in for that second task.

Read the abstract
ABSTRACT:

OKLCH separates a perceptual lightness coordinate, chroma and hue for practical color work. That coordinate system does not guarantee contrast or physical display coverage. A neutral calculation and the clipped hue study distinguish authored colors, represented gamut and the foreground/background pair the reader actually sees.

Editorial essay; not externally peer reviewed. Estimated at 200 words per minute for the abstract and main prose; references and interactive studies take additional time. Publication details revised 10 October 2026.

Revised 2026-10-11. Complete design essay with declared examples and study-specific limits. Unsupported universal effects, measurements and historical-art attribution withdrawn.

The coordinate is useful, not sovereign

Ottosson’s original 2020 post introduces Oklab for image-processing tasks, with aims including useful perceptual relationships and smoother color manipulation. OKLCH expresses the related coordinates as lightness, chroma and hue. It gives a designer a practical way to vary one coordinate while keeping the others fixed. That is a design advantage, not proof of perfect uniformity for every viewing condition.

The source title is A perceptual color space for image processing, not the earlier bibliography’s graphics variant. CSS Color Level 4 describes the web color representation and its conversions. A specification for coordinates does not turn a chosen palette into measured material reflectance or make every requested color reproducible on an sRGB display.

A neutral pair exposes the distinction

Consider opaque neutral OKLCH with L=0.65 and C=0. For the neutral case, the linear sRGB channels are L³=0.274625 before quantization. Against ideal black, the WCAG relative-luminance ratio is (0.274625+0.05)/0.05≈6.49:1. Against ideal white it is 1.05/(0.274625+0.05)≈3.23:1. The same color therefore supports different ratios depending on the background.

The calculation uses an ideal, unquantized neutral and opaque sRGB black and white. An 8-bit preview rounds channels and gives slightly different values. WCAG’s contrast guidance describes the ratio and relevant text criteria. Passing a ratio for one pair does not establish complete accessibility, and a lightness coordinate alone cannot select the right background.

The earlier specimen advertised a verified invariant 14.6:1 ratio without identifying a complete foreground/background pair. That number is withdrawn. So are fixed percentages presented as universal fractions of all visible colors in sRGB or Display P3, and a universal chroma threshold for exceeding sRGB. Gamut boundaries depend on the coordinates under consideration.

FIG. 01 / A WORKING ARGUMENT

Equal numbers, different light

Sweep the hue through paired HSL and OKLCH scales. The markers identify OKLCH samples that exceed sRGB before display mapping.

HSL / 65% S
30
40
50
60
70
80
90
OKLCH / .15 C
30 †
40 †
50 †
60 †
70 †
80 †
90

† Outside sRGB before clipping. Preview uses clipped sRGB channels. Numbers are each model’s lightness coordinate, not contrast ratios.

OBSERVE / COMPARE / RETURNPerceptual lightness is not a contrast score. These two scales use different coordinates; neither their numbers nor their gamut limits are interchangeable.

Read the preview’s transformation

The study pairs an HSL scale at 65% saturation with an OKLCH scale at chroma 0.15. Its slider changes hue. The labels are each model’s lightness coordinates; an HSL label of 65 and an OKLCH value of 0.65 do not mean that their samples have equal luminance or equal appearance.

The study converts OKLCH to linear sRGB and marks samples whose channels lie outside the supported interval before clipping. Its preview clamps channels into that interval and encodes an sRGB color. The dagger marks an out-of-gamut request. This is explicitly a clipped teaching preview, not a demonstration of a particular browser’s more elaborate gamut-mapping procedure.

Inspect the color the reader receives

A palette workflow should save the authored coordinate, intended output space and any mapping or clipping step. Then inspect the rendered foreground against its actual background, including opacity, overlays and states such as hover or disabled. A contrast computation for a solid pair cannot certify text placed over a moving image.

For a declared design task, use OKLCH to develop related ink candidates, convert them through the supported display path and choose among them using the actual reading pair. If a vivid candidate changes substantially after mapping, the design decision may need to change too. The color-space advantage is retained while the display result is allowed to interrupt it.

A manifesto with a checkable claim

I favor OKLCH because it makes some palette decisions easier to express and inspect. I do not favor calling its numbers an accessibility invariant. A useful color argument names a pair, its conversion and its criterion; a useful visual judgment also pays attention to the text and setting.

The study makes the conversion’s limitation visible, and the neutral example makes contrast’s dependency visible. Together they support a narrower, stronger claim than the old guarantee: coordinates help author the palette, while explicit evaluation helps decide whether a particular reader can use it.

FROM READING TO PRACTICE

Put the argument to work.

Build a palette in OKLCH, then check each actual foreground/background pair for contrast and inspect it on the displays you support.

A useful boundary. Perceptual lightness is not a contrast score. These two scales use different coordinates; neither their numbers nor their gamut limits are interchangeable.

Continue in the laboratory
REFERENCE REGISTER / 03 SOURCES

Sources & further reading.

The books and papers behind the argument, with curatorial notes and references formatted in APA 7th edition. The citation style identifies the source; it does not claim external peer review.

Ottosson, B. (2020). A perceptual color space for image processing. Björn Ottosson. https://bottosson.github.io/posts/oklab/

Why it belongs here Original post dated December 23, 2020; later introduction is separate. Describes Oklab and its design aims, not an invariant accessibility guarantee.

View source ↗

World Wide Web Consortium. (n.d.). CSS Color Module Level 4. World Wide Web Consortium. https://www.w3.org/TR/css-color-4/

Why it belongs here Current specification consulted October 11, 2026. Color coordinates, conversions and gamut behavior do not substitute for evaluating the rendered foreground/background pair.

View source ↗

World Wide Web Consortium Web Accessibility Initiative. (n.d.). Understanding Success Criterion 1.4.3: Contrast (minimum). World Wide Web Consortium. https://www.w3.org/WAI/WCAG22/Understanding/contrast-minimum.html

Why it belongs here Defines the relative-luminance contrast calculation and relevant text thresholds. The numerical example below uses opaque sRGB colors, not a complete conformance evaluation.

View source ↗

Archival Cross-References

◈AEO 2026 · RECURSIVE SEARCH TREE ARCHITECTURE

Archival Inquiries: The OKLCH Mathematical Manifesto

Bespoke recursive search tree answering core questions on The OKLCH Mathematical Manifesto.

Topic: Archival Inquiries: The OKLCH Mathematical Manifesto1 Core Inquiries
About the author

Evelyna Bellamy

Curator & Principal Systems Architect · Temporal Architecture

Evelyna’s work brings together perceptual colour, cognitive ergonomics and tactile interface design. Her background in nonfiction writing, computer science and communications informs the questions explored in this archive.

Contents 0% read