← ShadeMatch

🔬 What we are learning

ShadeMatch doubles as a colour-vision study at Semmelweis University. The question behind the game: how precisely do people perceive and judge colour differences on their own, everyday screens? Every round you play is one data point. This page shares the picture so far — it refreshes as the dataset grows.

140participants
7 125rounds played
56%perfect matches

Where “no difference” begins

When players give up on a colour, they judge the remaining difference. Pairing that judgment with the measured colour error (ΔE) maps where perceived sameness begins — the heart of the study.

🟢 “No difference” · 768× median ΔE 0.72
🟡 “Small difference — acceptable” · 990× median ΔE 1.31
🔴 “Difference — not acceptable” · 269× median ΔE 3.74

The hardest colours so far

Ranked by how rarely attempts end in a perfect match (colours with at least 10 attempts).

Bay of Many 0% · 14×
Twine 0% · 10×
Burnt Sienna 0% · 11×
Bud 0% · 10×
Trendy Green 0% · 14×

Our pledge to participants

You are identified only by a random ID; your email is never linked to published results. Findings will be published openly, and this page will always show what your contributions are teaching us. You can withdraw your data at any time.

These are descriptive statistics from gameplay on uncalibrated screens — interesting signals, not medical measurements.