| Sheen | Roughness | What you see | Where it earns its keep |
|---|---|---|---|
| Matt | 0.95 | No visible reflection | Ceilings and imperfect walls — hides flaws best |
| Eggshell | 0.60 | A faint sideways glow near the window | Living rooms and bedrooms |
| Satin | 0.38 | A soft wide highlight | Hallways and kitchens — wipeable |
| Semi-gloss | 0.22 | A clear soft-edged reflection of the window | Woodwork and bathrooms |
| Gloss | 0.08 | A sharp mirror image | Skirting and frames — shows every imperfection |
Your monitor emits light; a wall reflects it. Two people looking at this page see different colours, because their screens differ and neither is calibrated. Compare choices against each other here, then buy a tester pot before you commit.
A saturated wall throws its own colour onto the ceiling and the floor. That is why a bold colour often feels heavier once four walls are done than the tester patch suggested. This preview models it; a flat swatch cannot.
Opposite walls are closer together, so light makes more short trips between them and the colour compounds. Try the same colour at each of the three sizes — the small room reads noticeably more saturated.
Skirting, coving and frames are usually the brightest, glossiest surfaces in the room. They frame every wall, and dropping them from brilliant white to a soft off-white changes how the wall colour reads against them.
Rays leave the camera, bounce around the room up to four times, and stop only when they find their way out of the window or reach a light. Nothing is painted on afterwards. That matters because the single biggest reason a colour surprises people is colour bleeding: a saturated wall throws its own colour onto the ceiling, the floor and the wall opposite, so four walls of a colour read stronger and warmer than the tester patch you judged it by. A flat swatch cannot show you that. This can.
Put a strong terracotta on the walls and a plain white on the ceiling. The ceiling does not stay white — it picks up a warm cast, strongest where it meets the walls. Switch the walls to sage and the same ceiling goes faintly green. Neither is a trick of the rendering; it is what the room will do.
Because it is lit at a glancing angle while the wall opposite catches the light square-on. Same tin, same coat, sometimes a startling difference. It is worth dragging the view round before judging a colour — and worth remembering when someone says the paint "looks different on that wall".
It changes what you see, which is the same thing in practice. Gloss reflects the light source itself — usually whiter and brighter than the paint — so a glossy wall reads lighter and less saturated where that reflection lands, and shows every roller mark and filled dent. Matt scatters almost everything and shows the pigment more honestly.
It is averaging random light paths, so it starts noisy and resolves as samples accumulate. More samples means a cleaner picture and a longer wait; the Quality panel lets you choose. Nothing about the colour changes as it converges — only the noise.
Opposite walls are closer together, so light makes more short trips between them and the colour compounds with each bounce. It is the same effect that makes a bold colour feel heavier in a box room than on a tester patch.
No, and you should distrust anything that claims otherwise on a screen. The colours here come from the mixing model next door, not from any manufacturer's product, and your monitor is not calibrated to anyone's. Use this to compare options against each other, then buy the tester pot.
A rectangular room with one window, four bounces of light, and a single sky model — enough to show how a colour behaves, not a lighting simulation of your house. Surfaces are perfectly flat and uniformly coated, so it cannot show brush texture, patchy coverage, or the sheen difference between a first and second coat. Skirting, coving and window frames are drawn as flat bands rather than mouldings, so they carry the right colour and sheen but not a profile. Furniture is blocked in for scale and for the light it bounces, not modelled. Sheen is mapped to roughness on an illustrative scale rather than measured gloss units, and paint colours come from the mixing model rather than from any manufacturer's product. Screen colour is emitted light rather than reflected pigment, so treat this as a way to compare options rather than as a match — test on the actual wall before buying.
No changes to this tool’s own behaviour since the earliest archived release (v2.2). The full history for the site is in the changelog.