Why two numbered colours look identical

Number four and number seven are visibly the same colour in the palette strip, so you cannot tell which one you are placing. That is usually the palette working as designed rather than a fault.

You have number four selected. Number seven, two swatches along, looks like exactly the same colour. Neither one helps you find its regions on the canvas, and when a region does fill you cannot honestly tell which shade went in. It feels like the palette is broken, or like your eyes are.

Neighbouring numbers are close on purpose

A picture made of flat unrelated colours — one red, one blue, one yellow — is trivially easy to work and looks like a diagram. Anything with shading, depth or a gradient needs several steps of the same hue: a mid blue, a slightly darker blue for the shadow side, a slightly lighter one where the light falls. Those steps have to be small, because large jumps read as banding rather than as shading.

So a palette built for shaded artwork will contain pairs that differ by very little. That is the palette doing its job. The smoothness you like in the finished picture is bought with precisely the difficulty you are having in the palette strip.

There is a name for the size of that step. The just-noticeable difference is the smallest change in a stimulus a person can reliably detect, and shading palettes tend to sit near it deliberately. Two adjacent steps are meant to be separable when seen side by side as large areas — which is not the condition you are viewing them under.

Four things that make the gap look smaller than it is

A swatch carries less colour information than a fill

Colour discrimination improves with the size of the patch being judged. A palette swatch is a few millimetres of screen; the region it will fill might be a hundred times that area. Two colours that are obviously different once they are down on the canvas can be genuinely indistinguishable in the strip. If you have already placed one of them somewhere, look at the canvas rather than the palette — that is the size the difference was designed for.

Every patch is judged against its surroundings

Simultaneous contrast is the textbook effect where an identical grey looks lighter on a dark ground and darker on a light one. In a palette strip each swatch sits between two different neighbours, so each is being pushed in a different direction by whatever happens to be beside it. Swatch four flanked by two darks and swatch seven flanked by two lights will look closer together than they are. You usually cannot rearrange the strip, but knowing the effect exists stops you concluding that your eyes are failing.

The border biases the patch

Swatches are normally drawn with a thin dark outline, and the selected one often gets a heavier one, or is enlarged, or lifted. A dark ring around a small patch drags its apparent lightness downwards, and the smaller the patch the more of it is border. So the swatch you have selected is being rendered under different conditions from the one you are comparing it against, which is the worst possible arrangement for a fine judgement.

A dim screen compresses everything

At low brightness the whole range from black to white is squeezed into fewer perceptible steps, and the first differences to disappear are the smallest ones. Auto-brightness in a dark room, a battery-saver mode, or a warm night tint will each flatten a shading palette. A blue-heavy palette suffers most under a warm tint, because that is the axis being shifted.

Five things that actually separate them

1. Read the number, not the colour. It is the only unambiguous label on the screen, and the entire point of the format is that you never have to make a colour judgement at all.
2. Select each candidate in turn and watch the canvas rather than the palette. The highlighted regions will be in different parts of the picture; if the highlight jumps from the sky to a flower when you switch, you have your answer without ever comparing two swatches.
3. Zoom in. A larger patch is easier to judge and the printed number becomes legible again, which usually settles it in a second.
4. Turn the brightness up and switch off any warm-tint or night mode while colouring. This one is free and often decisive.
5. Fill one of the pair completely, then look again. Once four is down, everything still highlighted in that near-identical shade is seven. Elimination beats discrimination.

When it is not the palette

Two other causes are worth ruling out. Some form of red-green colour vision deficiency — deuteranomaly and protanomaly are the common types — is reasonably widespread among men, and for those readers pairs that most people separate easily will genuinely collapse. Blue-yellow deficiency exists too and is much rarer. If the numbers you cannot tell apart are always the reds and greens rather than a random assortment, this is the likely explanation, and working from the numbers instead of the colours is not a workaround but simply the correct way to use the format.

The other is a display filter you have forgotten about: a greyscale accessibility setting, a colour inversion mode, a tinted screen protector, or a strong blue-light filter. Each of these can be checked in under a minute, and it is worth ruling them out before deciding a picture is badly made.

Frequently asked questions

Related guides

• Why light colours are harder to place than dark ones
• Why colours look different when you zoom in
• Colour by number with colour blindness

All guides