How to use the Color Blindness Simulator
- Load your artwork. Drop a file on the box, click it to browse, or paste a screenshot with Ctrl+V. Until you do, the simulator shows a built-in still life and sales chart packed with reds and greens. Everything is processed on your own device; the image is never uploaded.
- Pick a vision type. Deuteranopia is selected first because red-green confusion is by far the most common. For the three “-anomaly” types, set Anomaly severity: 10% is a barely shifted cone, 100% is the same as the full “-opia”.
- Compare. Split puts the original and the simulation in one frame — drag across the image or use the slider underneath. Side by side shows two copies, and All types lays out the original next to five simulations so you can scan them in one look.
- Export. Download PNG saves the current simulation at the image’s full resolution (up to 4096 px on the long side). Download comparison saves the six-panel sheet, which is handy for a client or an art director.
- Check a palette by numbers. Switch to the Palette check tab, paste your HEX codes and the tool lists every pair that falls under the ΔE you set in Flag pairs under ΔE. Copy report gives you the list as plain text.
Why red and green are the ones that collapse
Your retina sorts light with three kinds of cone. S cones peak around 420 nm (violet-blue), M cones around 530 nm (green) and L cones around 560 nm (yellow-green). M and L overlap heavily, and the brain gets “red versus green” from the small difference between their signals. Lose one of those two cone types, or shift its sensitivity toward the other, and that difference shrinks or vanishes. What is left is a blue-to-yellow axis plus lightness.
That is why red-green deficiency is so common. Roughly 8% of men and 0.5% of women of Northern European ancestry have some form of it, because the genes for the L and M pigments sit side by side on the X chromosome. The mildest and most common variant, deuteranomaly, accounts for about 5% of men on its own. Protanopia, deuteranopia and protanomaly each affect around 1% of men. The blue-yellow types, tritanopia and tritanomaly, are rare and affect men and women about equally because the gene is not on the X chromosome. Complete achromatopsia, seeing only light and dark, is rarer still, at roughly 1 in 30,000 people.
The simulations here use the model published by Machado, Oliveira and Fernandes in 2009, which estimates what each cone deficiency does to a color and turns that into a 3×3 matrix. The paper gives matrices for severities from 0.1 to 1.0; this tool interpolates between them, so 65% sits halfway between the 0.6 and 0.7 matrices. The math runs on linear RGB — light values before the sRGB gamma curve — because cone responses add up as light, not as the encoded numbers in a file. Applying the same matrices to raw 0–255 values darkens midtones and exaggerates the effect.
Protanopia also makes reds look darker, not just browner. Without L cones, long-wave red light barely registers, so pure #FF0000 comes out as a dark olive #6D5F00, while under deuteranopia it becomes a lighter mustard #A39000.
Measured: what happens to common color pairs
ΔE2000 is the standard measure of how different two colors look: around 1 is the smallest difference most people can see side by side, and under 10 two colors read as “the same color, slightly off” at a glance. These pairs were run through the same matrices the tool uses (severity 100%):
| Pair | Normal vision | Protanopia | Deuteranopia | Grayscale |
|---|---|---|---|---|
#D7263D red + #3E8E41 green |
67.9 | 19.1 | 3.7 | 5.6 |
#E4323F apple red + #A6D152 apple green |
66.8 | 36.5 | 18.0 | 23.1 |
#3E8E41 green + #F28C28 orange |
47.4 | 10.0 | 20.5 | 13.1 |
#D7263D red + #2A9D8F teal |
62.4 | 25.9 | 26.6 | 11.1 |
#D55E00 vermillion + #009E73 bluish green |
54.4 | 18.2 | 20.5 | 3.3 |
#E69F00 orange + #56B4E9 sky blue |
53.6 | 50.6 | 54.3 | 0.6 |
Three things stand out. The classic “stop/go” pair drops from 67.9 to 3.7 under deuteranopia: both turn into the same olive, #8A7D37 and #867B47. Swapping the green for a teal keeps the pair apart in every column. And the last two rows, from the Okabe–Ito palette designed for color-blind readers, survive every simulation yet have almost identical lightness. A palette can be color-blind safe and still fail when printed in black and white, which is why the All types view includes achromatopsia.
What a simulation can’t tell you
A simulation is an average. Two people with the same diagnosis can see the same swatch differently, and severity varies continuously, so treat the output as a strong hint rather than a photograph of someone’s experience. Your monitor matters too: an uncalibrated screen with a warm or cool cast shifts every result a little. And simulations show what is lost, not how people adapt. Many color-blind artists read value, saturation and context extremely well, so a pairing that “fails” here may still be workable when shapes and labels carry the meaning.
Tips from practice
Fix value first, then hue. If two areas of an illustration must read as different, give them different lightness. Run a value check in the Notan & Value Study Tool — a pair that separates in grayscale will almost always separate under every simulation too.
Push greens toward blue and reds toward orange. Blue-green against vermillion survives red-green deficiency far better than grass green against crimson, as the table shows. The shift is small enough that most viewers never notice the palette was adjusted.
Never let color be the only label. Map legends, sold dots on a price list, “available” badges in a shop listing: add a word, an icon or a pattern. To someone with deuteranopia, a red dot and a green dot are one olive dot, and milder red-green deficiency still blurs them for many of the roughly one in twelve men who have it.
Test thin lines and small text separately. Color fades out of fine detail before it fades out of large fields, so a colored outline or a tiny legend swatch fails sooner than the big shapes in the same image. Zoom the split view onto those areas, and check the text pairing in the Contrast Checker.
Ask a real person before you print 500 copies. If the piece is a poster, a game UI or a teaching chart, show it to someone with color vision deficiency. The simulator gets you most of the way; one honest viewer catches the rest.