What size change counts as a JND?

It depends on the colors and the viewing conditions, but in perceptual-space terms, changes below a small fraction of the space's typical scale are generally invisible — the tool's reading lets you compare against that order of magnitude.

Why not just measure RGB distance?

Because the eye does not live in RGB: equal RGB steps are unequal to vision. The perceptual reading predicts what a person notices; the RGB reading predicts what a machine stored.

My pipeline shifted a color — the JND says invisible. Do I care?

Probably not visually — but consistency audits may still care that the value changed. The tool tells you users will not see it; the rest is policy.

Does the JND differ for color-blind viewers?

Yes — thresholds shift under CVD, which is why the simulators exist. This tool answers for typical trichromatic vision.

What does the just noticeable difference calculator not do?

In each case the delta is real in the machine but possibly imaginary to a human — and the color delta threshold read settles which. The color comparison tool frames the same measurement as a general comparison; this one asks the narrower, sharper question — the smallest visible color change boundary — and answers it fo

Just Noticeable Difference Calculator

Color A

Color B

Distance

RGB Euclidean

261

OKLCH ΔE (approx.)

0.349

You Might Also Need

Color Distance Calculator

Compute perceptual distance between two colors

Color Comparison Tool

Compare two colors side-by-side

Color Blend Tool

Blend two colors with adjustable ratio

Perceptual Lightness Calculator

Compute OKLCH perceptual lightness

The just noticeable difference calculator measures two colors' distance in perceptual terms — the color delta threshold read that says whether a change is visible or imaginary.

What is Just Noticeable Difference Calculator?

Somewhere between identical and obviously different lies the boundary the just noticeable difference calculator maps: the JND — the smallest change a typical observer can detect. The tool takes two colors, computes their separation in perceptual space, and reports the distance in terms that mean something to the eye — so the question is this change visible? stops being a matter of squinting and becomes a measurement.

The perceptual framing is what separates this from raw channel arithmetic. RGB distance measures machine coordinates; a JND measures how far apart the colors look, which requires a space built to match appearance — the same perceptual reasoning that powers the OKLCH tools here. The two-color comparison reports both readings, and the disagreement between them is the lesson: pairs that score far apart in RGB can sit within a JND perceptually (invisible change), and modest RGB shifts can cross the threshold where the eye is sensitive (obvious change). The number that predicts what a person notices is the perceptual one.

The tool's home turf is the changes-too-small-to-eyeball territory: did the export pipeline shift this brand color? did the compression move this gradient stop? is this token change a real change or a rounding artifact?
In each case the delta is real in the machine but possibly imaginary to a human — and the color delta threshold read settles which.
The color comparison tool frames the same measurement as a general comparison; this one asks the narrower, sharper question — the smallest visible color change boundary — and answers it for a specific pair.

Difference measurement exists because the eye is a poor judge of small changes. Two values viewed in sequence are easy to compare; the same two viewed minutes apart, on different surfaces, or by different people are not, and the error grows exactly in the range where the decision matters — the hover state that must be visible, the disabled state that must be distinguishable, the border that must not appear. A color delta threshold tool turns those questions into a number computed in a perceptually uniform space, so the answer is stable and repeatable and can be attached to a review comment instead of an opinion. The threshold is the useful part: below it the pair reads as one colour, above it as two, and where a pair sits matters more than how far apart the two values look in a picker's coordinates.

In practice the number is used as a floor rather than a target: pick the value you like, then check that its distance from its neighbour clears the threshold for the job. That habit removes most of the mystery from apparently identical states and makes the difference between a deliberate step and an accidental one visible.

What the tool will not do is certify invisibility universally: JNDs vary with age, display and adaptation. But as a jnd color difference reference for the changes that must be judged — visible or not — it replaces the squint with a number, and the number is computed in the space where the eye actually lives.

How to use Just Noticeable Difference Calculator

Set color A — the original or reference value.
Set color B — the changed value under test.
Read the perceptual distance and judge whether the change crosses the visible threshold.

When to use Just Noticeable Difference Calculator vs related tools

Use this for pipeline shifts, compression checks and token diffs where the change may be sub-visible. Use the color comparison tool for general two-color measurement, the pixel color picker to obtain the values, and the CVD simulators when the visibility question involves color vision differences. For the neighbouring steps, see the color wheel theory guide and the bezold effect visualizer.

Privacy & Security

This tool runs entirely in your browser — no data ever leaves your device. There is no server round-trip, no upload, no logging, and no account required. Your input is processed locally using client-side JavaScript and is never stored, transmitted, or accessible to anyone else. When you close the tab, everything disappears.