Color Blindness Simulator

• Deuteranomaly (~5% of men, 0.4% women) — green-weak. Reds and greens harder to distinguish; oranges look more like yellow. Most common form.
• Protanomaly (~1% men) — red-weak. Similar effect to deuteranomaly but with reduced red sensitivity; reds appear darker.
• Deuteranopia (~1% men) — green-blind. Cannot distinguish red and green; both shifted to a yellow-beige range.
• Protanopia (~1% men) — red-blind. Like deuteranopia but with reds darker.
• Tritanomaly (<0.01%) — blue-weak. Rare. Blues and yellows look similar; greens look more like blue.
• Tritanopia (<0.001%) — blue-blind. Vanishingly rare.
• Achromatopsia (~0.003%) — total color blindness. World in grayscale only; also typically accompanied by light sensitivity and reduced visual acuity.

The X-linked recessive inheritance pattern is why men are far more affected — they have only one X chromosome. Women need both X copies to carry the defect.

• Red error / green success badges — to a deuteranope, both look like beige/yellow. Pair color with shape (✗ vs ✓) or label.
• Red lines on a green map (or vice versa) — chart legends fail. Use shape, pattern, or contrasting luminance instead.
• "Red zone" warnings without text — passenger safety information that depends on color discrimination alone fails accessibility standards.
• Subway / transit maps with similar-luminance route colors — colorblind passengers cannot tell which line is which. Some networks (e.g., Berlin S-Bahn) deliberately vary luminance to remain readable.
• Heatmaps with red-green gradient — the most common heatmap color scale is the worst for colorblind viewers. Use viridis or other colorblind-safe palettes.
• Highlight selections with color only — selected state, hover state, drag-and-drop indicators. Add border, shadow, or background change.

• ColorBrewer 2 — http://colorbrewer2.org. Designed for cartography. Sequential, diverging, and qualitative palettes with colorblind-safe filters.
• Viridis (and inferno, magma, plasma) — perceptually-uniform, colorblind-safe scientific palettes. Default in matplotlib since 2017.
• Okabe-Ito — 8-color qualitative palette designed for color vision deficiencies. Used by many accessibility-conscious data viz projects.
• IBM Design Library — 5-9 color palettes with proven colorblind safety.

The common theme: rely on luminance differences alongside hue differences. Two colors with the same luminance but different hue become indistinguishable under one or more types of color blindness; different luminances stay readable.

• You are auditing a design system for accessibility and want to see how the palette holds up under each color vision deficiency.
• You inherited a chart or dashboard and want to verify it is interpretable without distinguishing red from green.
• You are explaining accessibility concerns to stakeholders and want a concrete visual demonstration of "what 8% of your users see".
• You are presenting research data and want to confirm that figures will be legible in print (grayscale) and to colorblind reviewers.

• It will not match every individual's experience. Two people with "deuteranopia" can have slightly different specific perception due to varying severity and individual differences. The simulation uses an average clinical model.
• It will not test other accessibility dimensions. Low vision, motor disabilities, cognitive disabilities each have their own simulation/test tools.
• It will not fix problems. The simulator shows what is broken; you redesign or recolor based on what you see.