Palettist, a new color palette every day!


Palettist Manual

Palettist will automatically generate a new color palette every day, but you can also use custom settings to generate your own color palette.

Main screen

Settings. Press this to go to the Settings screen.
About. Press this to display some information about this app.
Share. Press this to save or share the current color palette.
Custom palette. Press this to display/compute your custom palette.
Today's palette. Press this to go to display/compute today's palette.

Settings screen

Red levels. Colors are represented as Red-Green-Blue (RGB) triplets. This is the number of subdivisions of the red component. The total RGB volume is this number, multiplied by the green and blue levels, R×G×B.
Green levels. The number of subdivisions of the green component.
Blue levels. The number of subdivisions of the blue component.
Single red value. When there's only one red level, you can select directly its value, a number between 0 (black) and 1 (red).
Single green value. When there's only one green level, you can select directly its value, a number between 0 (black) and 1 (green).
Single blue value. When there's only one blue level, you can select directly its value, a number between 0 (black) and 1 (blue).
Color space. Select sRGB if you want to use this palette in an old display, and Display P3 if you target an iPhone 8, an iPhoneX, a recent iMac Pro, or any display with P3 support. If you select displayP3 - sRGB, then only colors that are inside the Display P3 gamut, but outside the sRGB gamut, will be used, useful for evaluating Display P3 screens.
Gamma space. Select linear if you want to sample the RGB space before applying the gamma correction. The final color will be displayed with the gamma applied, but the total number of samples will be biased towards the bright end of the spectrum. If you select gamma, the samples will be taken at equal distances after having applied the gamma, and thus, more colors in the dark end of the spectrum will be represented.
Sample noise. Random additive noise applied to every sample in every color channel.
Color category. Filter samples by color category. Select "all" if you don't want to apply any filter.
Image size. The size of the output image containing the color palette. The maximum dimension is 2048. You can use "0" or "?" as wild cards, so the correct aspect ratio is applied, assuming each color bin is square. For instance, if you set this to "1024×?", and the color palette has 6 (2×3) colors, then the final image will be 1024×1536.
Sorting. How to sort the colors of the output color palette, assuming the output is discrete (i.e., it has 256 color bins or less). Select R→G→B to simply order colors by their sampling order. Select RGB distance to sort colors in a 2D grid by their Euclidean distance in Display P3 color space. Select L*a*b* distance to sort colors in the 2D grid by their distance in L*a*b* color space. The Euclidean distance in L*a*b* is more consistent with human color perception.
sRGB comparison shape. Given a discrete color palette (i.e., with 256 color bins or less), render a shape inside each color bin where the color is clamped to its equivalent color in sRGB color space. This is great to evaluate different displays and Display P3 capabilities. The available shapes are circles in 3 different sizes, and triangles. Read this blog post for details: Display P3 vs sRGB in Color Palettes
SOM quality. If the color palette contains more than 256 output colors, a continuous/smooth palette will be computed. This palette is computed using a type of neural network called Self-Organizing Map (SOM). It will take one minute or so to compute, depending on its quality, which can be tweaked with this parameter. The smaller this value, the faster it will get computed, but the noisier the output will be. If the output resolution of your palette is big, you may want to increase the quality. Read this blog post to find out more about SOMs: Exploring the display-P3 color space