One tile is a different shade. Find it before the clock runs out.
One tile is slightly different. Tap it. Each level the grid grows and the difference shrinks. You have 30 seconds. Wrong taps cost 2 seconds.
Reaction Time TestHow fast can you click when it turns green
Draw a Perfect CircleOne freehand loop, scored to the percent
Typing Speed Test60 seconds. What's your WPM?
2048Slide and merge tiles to reach the magic number
Color MatchMix colours, bring it to life
I'm Not a RobotProve your humanity in hilarious ways Every round shows a grid of tiles in a single colour, except one that is a very slightly different shade. You tap the odd one out. Each level the grid grows and the difference shrinks, until you are staring at squares that look genuinely identical and guessing.
What this measures is colour discrimination, meaning how small a difference in hue or lightness your eyes and brain can separate. That is a different thing from colour blindness. A colour blind person confuses particular colour pairs regardless of how far apart they are, and this test is not designed to detect that. Plenty of people with entirely typical colour vision still stall around level twelve, because discrimination varies enormously between individuals.
Two things are working against you at once. The grid grows, so there is more to scan, and the colour gap narrows, so each comparison is harder. Human vision is also much better at judging two colours that touch than two separated by other tiles, which is why the odd tile is far easier to find when it happens to sit next to the tile you are looking at.
Your screen is a real limit here too. An older or cheaper panel simply cannot display the difference between two shades a few values apart, so beyond a certain level you are testing your monitor rather than your eyes. Brightness matters as well: at low screen brightness the darkest steps compress together and become genuinely indistinguishable.
Most people stop somewhere between level ten and fifteen. Past twenty is genuinely strong, and anything higher usually means both good discrimination and a good screen. Women are somewhat more likely to reach the higher levels, since red-green colour deficiency is far more common in men, affecting roughly one in twelve compared with about one in two hundred.
If you suspect an actual colour vision deficiency, this is not the right tool. Ishihara plates and proper clinical testing exist for that, and an optician can do it properly in a few minutes. Your best level here is saved in your browser and no account is needed.
No. It measures colour discrimination, which is how small a difference in shade you can detect. Colour blindness is confusion between particular colour pairs regardless of how far apart they are. If you suspect a deficiency, an optician can test properly with Ishihara plates.
Most people stop between level ten and fifteen. Past twenty is genuinely strong. Beyond that you are usually testing your screen as much as your eyes.
Partly your eyes and partly your display. Older or cheaper panels cannot render two shades that are only a few values apart, and low screen brightness compresses the darkest steps together.
A great deal. Turning brightness up is the single most effective thing you can do to reach a higher level.
Red-green colour deficiency is far more common in men, affecting roughly one in twelve, compared with about one in two hundred women. That shifts the upper end of the distribution.
Your best level is stored in your own browser. There is no account and nothing is sent to a server.
Yes. The daily gives everyone the same board, generated from the date, and it resets at midnight UTC. Your result is recorded and the strip above the game shows your last seven days and your current streak. Nothing is sent to a server, so the streak lives on your own device.