Farnsworth D-15 Test Online
Arrange 15 colour caps into a smooth hue run. Where you jump across the circle instead of stepping around it is what the test measures — the number of jumps suggests severity, their direction suggests type.
Educational screening toolMedical disclaimer
Arrange the caps by colour
Start from the fixed reference cap on the left. Drag each remaining cap so it sits next to the one it most closely resembles, working left to right — you are building a smooth colour run, not sorting light to dark. Keyboard: focus a cap, press space, move with the arrow keys, space again to drop.
Method
What the D-15 Actually Measures
It is not a test of whether you can name colours. It is a test of whether you can rank them.
The Farnsworth Dichotomous Test — the D-15 — was published by Dean Farnsworth in 1947 as a shorter companion to his 85-cap Farnsworth-Munsell 100 Hue Test. Where the FM-100 grades how finely someone discriminates hue, the D-15 asks a blunter question: is this person's colour vision broadly normal, or does it break along a particular axis?
Why arrangement reveals what naming cannot
Every colour on the cap circle has two neighbours. To someone with typical colour vision, the nearest cap is always one step away. To someone missing or with a shifted cone type, two caps on opposite sides of the circle look the same — so they place them next to each other. That jump is called a crossing, and it is the entire diagnostic signal.
This is why the test survives translation, illiteracy and young age in a way plate tests do not. There is no digit to read and no word to say. It also means the result is graded rather than pass/fail in nature, and spread around the whole hue circle rather than concentrated on red-green — which is why the D-15 can pick up blue-yellow loss that an Ishihara plate set is not built to test for.
How the score is read
A single crossing is treated as a slip — a moment of inattention or one cap dropped in the wrong slot. Two or more is the threshold at which the arrangement stops looking like carelessness and starts looking like a confusion axis. The direction of those crossings, not just their number, is what suggests which cone type is involved.
| Major crossings | Reading | What it suggests |
|---|---|---|
| 0 | Pass | Arrangement follows the hue circle. No axis to classify. |
| 1 | Pass | Treated as a slip rather than a deficiency — one cap out of place is common even with typical colour vision. |
| 2 or more | Fail | The pattern the test is designed to catch. The orientation of the crossings indicates the likely axis. |
Limits
Where a Browser Version Falls Short
Worth knowing precisely what this can and cannot tell you.
The clinical D-15 controls two things this cannot. The caps are Munsell papers whose reflectance is fixed and known, and they are viewed under a standardised illuminant — traditionally daylight around 6500K at a specified illuminance. Both exist so that the only variable left in the room is the person's vision.
On the web, neither holds. The colours reaching your eye depend on your panel, its calibration and colour profile, the ambient light, and any night-shift or adaptive-brightness feature your operating system is applying. A poorly calibrated screen can compress the very hue steps the test depends on, producing crossings that say more about the display than the viewer.
The cap colours here are also an approximation. Rather than claim to reproduce the licensed Munsell specifications, they are generated as equal steps around a perceptually uniform hue circle at constant lightness and chroma — which is how the physical set is conceived, and preserves the equal spacing the scoring relies on, but is not the same as the papers themselves.
FAQ
Farnsworth D-15 Questions
What people ask most about arrangement testing.
Is this the real Farnsworth D-15?
No. It follows the same principle — arrange caps into a hue run, score the crossings — but the real test uses physical Munsell caps under a standardised illuminant and is scored by an examiner. The colours here are generated approximations shown on your own display. Treat it as an exercise in what the test does, not as the test itself.
How is the D-15 different from the Ishihara test?
Plate tests are pass/fail by design: you either read the digit or you do not, which makes them good at flagging red-green deficiency and poor at describing it. An arrangement test grades the result and spreads it around the whole hue circle, so it can surface blue-yellow problems that a red-green plate set never probes. Our Ishihara test covers the plate format.
What is the difference between the D-15 and the FM-100?
Resolution. The FM-100 uses 85 caps across four trays and produces a fine-grained Total Error Score; the D-15 uses 15 and answers a coarser question — normal, or broken along an axis. The D-15 exists because an 85-cap sort is impractical in a screening context. Our colour perception test is built on the FM-100 principle.
Why does it say “red-green” instead of naming protan or deutan?
Because 15 caps cannot separate them reliably. The protan and deutan confusion axes sit close together on the hue circle, and telling them apart takes either more caps or a different instrument — typically an anomaloscope. Reporting a confident protan-versus-deutan call from this many samples would be overstating what the data supports.
I passed. Does that mean my colour vision is fine?
It means you did not produce the pattern this test looks for, on this screen, in this lighting. Mild anomalous trichromacy frequently passes a D-15 — the test is designed to catch moderate and severe deficiency, and a milder shift can order the caps correctly while still causing difficulty in real use. A pass is reassuring, not conclusive.
Can I use a phone?
Yes, the caps are draggable by touch. Be aware that phone displays commonly oversaturate by default and that features like Night Shift or True Tone shift the whole palette warm, both of which affect the hue steps. Turn those off first if you want a result worth comparing to a later attempt.