AIColors

PALETTE × 8 CVD TYPES

Color Blindness Simulator

Test an entire palette against eight forms of color vision deficiency at once. See whether your "red error" and "green success" collapse into the same color for the 5% of users who can't tell them apart.

Typical vision

Prevalence: 92%

Deuteranomaly

Prevalence: 5.0% (men)

Deuteranopia

Prevalence: 1.1% (men)

Protanomaly

Prevalence: 1.3% (men)

Protanopia

Prevalence: 1.0% (men)

Tritanomaly

Prevalence: 0.01%

Tritanopia

Prevalence: 0.008%

Achromatomaly

Prevalence: <0.005%

Achromatopsia

Prevalence: 0.003%

Why palette-level

Most simulators show one color at a time. That misses the actual problem.

When designers test for color blindness, they usually pop one color at a time into a single-color simulator, see it changes slightly, and conclude "looks fine". The real failure is not "this red looks weird" — it is "this red and this green produce identical colors under deuteranopia". Single-color tools cannot show that.

We render your whole palette across all 8 CVD types simultaneously. Two cells in the same row that produce the same color are a fail — that pair of colors is informationally indistinguishable for that user, and any UI that relies on the difference (a green-good / red-bad button, a chart legend, a status badge) will break.

The fix is rarely a different hue — it is adding a non-color signal. Icons next to status badges. Patterns or position in charts. Text labels next to legend swatches. Color is one channel of information; design that depends on color alone is design that breaks for 1 in 12 men and 1 in 200 women.

The eight types, ranked by prevalence

Color vision deficiency is far more common than designers tend to assume. Roughly 8% of men and 0.5% of women have some form of CVD — about 300 million people worldwide.

Red-axis (Protan: protanopia + protanomaly)

Reduced or missing sensitivity to red. Reds appear dim and shift toward dark olive or brown. About 2.3% of men. Red text on dark backgrounds becomes very hard to read; "red bad / green good" semaphores fail.

Green-axis (Deutan: deuteranopia + deuteranomaly)

The most common form — about 6% of men. Reds and greens are confused. Anomaly (mild) at 5% is the single most prevalent CVD overall. Test your palette under deuteranomaly first if you only have time for one check.

Blue-axis (Tritan: tritanopia + tritanomaly)

Rare — less than 0.01% of the population. Blues and yellows are confused. Less common as a design concern but worth checking if you ship financial dashboards or aviation tools where blue/yellow distinctions carry meaning.

Total / partial achromatopsia

Total absence (or near-absence) of color vision — the world reads in grayscale. Vanishingly rare (~0.003%) but the simulator output is identical to what your design looks like to anyone viewing it on a monochrome display, projected through a blocked-out projector, or printed without color.

Designer's checklist

Use this list before you ship a color-coded interface.

Frequently asked questions

Is this color blindness simulator free?
Yes — fully free, no signup, runs entirely in your browser. We never send your palette or any other data to a server.
How accurate are these simulations?
They use the standard Brettel / Viénot / Mollon matrix approximations widely used in tools like Sim Daltonism and Color Oracle. They are great for catching obvious failures but cannot perfectly reproduce what a colorblind user sees — the experience differs from person to person. Treat the output as a strong indicator, not a proof.
Which type should I prioritize testing against?
Deuteranomaly — it is by far the most common form. If your palette works under deuteranomaly, it will work for around 90% of users with any form of CVD. Protanopia and tritanopia are worth checking too but deuteranomaly catches the most cases per minute of testing.
What is the difference between protanopia and protanomaly?
Protanopia is the complete absence of red cones; protanomaly is reduced red sensitivity. The same prefix logic applies to deutan- (green) and tritan- (blue). The "-pia" suffix means total dichromacy; "-omaly" means anomalous trichromacy (partial deficiency).
My palette looks fine in the simulator — am I done?
No. The simulator catches the "two colors collapse into one" failure, but accessibility for CVD users also requires non-color information channels (icons, patterns, text). Run through our checklist above before shipping.
Can I share my palette test with my team?
Yes — the share button copies the page URL with every color in your palette encoded in the hash. Anyone opening the link sees the exact same simulation matrix.

Pair with our other color tools