How to play

One tile is a slightly different shade from the rest, and it gets found. Tap to pause.

Scan the gridEvery tile is the same color except one that is slightly lighter.
Click the odd oneFind it and you move up a level. The hue changes every round.
Misses cost a lifeLose all three lives and the run ends. Your score is the last level you cleared.
It gets harderThe gap shrinks from about 13 points to 2, and the grid grows from 2×2 to 6×6.

What this test measures

Early levels test discrimination, the smallest lightness difference your eye can pick up. Vision researchers call it a just-noticeable difference.

Later the grid grows to 36 tiles, and the task changes. A difference you would spot at once between two squares side by side can hide in a crowd. From there, visual search starts to decide how far you get.

The grid grows every three levels while the gap keeps shrinking. The hue is also random each round, so you cannot get used to one color and coast.

What counts as a good score

In Normal mode, the badge on your result follows this table. Harder modes raise your score a little before the badge is picked. We based it on the shrinking lightness gap. At level 10 the gap is 4 points, about twice the difference people can barely notice (1 to 2 percent). From level 12 on, the gap stays at its 2-point floor.

Level cleared	Badge	What it means
18 or more	Superhuman	Six levels past the 2-point floor. Great eyes and a great screen.
14-17	Pro tier	A few levels at the floor, close to what the panel can show.
10-13	Skilled	Gaps of 4 points or less. A genuinely good eye for color.
7-9	Standard	Comfortable down to a gap of about 7 points. Average eyes.
6 or lower	Needs practice	Check your brightness and night mode before blaming your eyes.

Real scores recorded on this site

So far, 206 runs of this test have been recorded here. Computers and phones score quite differently, so each gets its own table.

Computer (mouse)

96 runs · median level 12

Range	Share
18 or more · Superhuman	5%
14-17 · Pro tier	21%
10-13 · Skilled	69%
7-9 · Standard	4%
6 or lower · Needs practice	1%

Phone (touch)

110 runs · median level 12

Range	Share
18 or more · Superhuman	13%
14-17 · Pro tier	15%
10-13 · Skilled	55%
7-9 · Standard	12%
6 or lower · Needs practice	5%

Every finished run counts, so someone who plays five times adds five runs. The numbers are recounted every day, and a device table shows up once it passes 30 runs. Last updated: 1 Nov 2026.

Why your score depends on your screen

At a 2-point gap, whether you can tell two shades apart depends on your screen. Its brightness, its color settings and the light in your room move the result more than your eyes do. Compare your score with someone else's and you are mostly comparing monitors.

Eye-clinic tests like the Ishihara plates avoid this by controlling what a web page cannot: printed colors under standard lighting. We think the comparison that means something is you against yourself, on the same device and in the same conditions.

Your screen is part of the test

The same person can score several levels apart on two devices in the same afternoon. These are the usual reasons.

Factor	What it does to your score
Panel bit depth	A good 8-bit panel shows a 2-point step as two different colors. A cheap 6-bit panel may round both tiles to the same color, and then no amount of staring helps.
Brightness	A dim screen squeezes shades together, especially dark ones. Raising brightness pulls them apart again.
Night mode	Blue-light filters pull everything toward orange and flatten differences, most noticeably on cool hues.
Forced dark mode	A normal dark theme is fine, but "smart invert" and forced dark modes repaint web pages and can distort the tiles.
Glare	A window behind you lays glare over every tile and eats subtle contrast.
Viewing angle	On cheap panels colors shift off-axis, so corner and centre tiles can look different even when they are not.

This is also why your phone and laptop rarely agree. They have different panels, and a phone's auto-brightness can change mid-game when a cloud passes the window.

Color vision, and who this test excludes

Around 1 in 12 men and 1 in 200 women have some form of color vision deficiency. Most tests on this site never use color as the only signal. Shape, position, motion or text always carries the message too.

This test and the Stroop Test are the two exceptions, because color itself is what they measure. If you are color-blind, some rounds here will be unfair to you, and we would rather say so plainly.

This test also cannot diagnose anything. A low score says nothing reliable about your eyes, because a bad screen gives the same result. If you think you might be color-blind, see an eye doctor. They can run the Ishihara plates or the Farnsworth-Munsell test under proper lighting.

How to see the difference for longer

Do not stare

Your eye adapts to a shade you stare at, and soon every tile looks the same. Sweep row by row and let the edge of your vision catch the odd one.

Work the borders

Two shades that look alike from a distance are easier to tell apart along the line where two tiles meet.

Set up first

Turn off night mode, raise brightness and face the screen straight on. Thirty seconds of setup is worth a couple of late levels.

Rest your eyes

If every tile starts to look the same, look at something far away for ten seconds. The differences come back.

Tilt the laptop

Raising a laptop screen a little reduces the brightness shift between the top and bottom rows.

Once the game reaches the 2-point floor, your display decides most of what happens next. We think technique buys you levels in the middle of a run, not at the end.

Color Match Test