Lets say I have a cube in an isometric pixel art with the 'normal' color #ff0000 (red) and I want to shade remaining sides, is there a rule for selecting the top (brighter red) color and the darker color?
I guess I want to know if there is a formula for calculating the color values, and also what side is most used to be darker, left or right?
There is no rule for shading isometric pixel art. It's an aesthetic design decision which depends on the vibe you want your artwork to have.
Even though pixel art is often highly stylized and mostly doesn't involve realistic lighting, adding shading is an emulation of a light source. It can be helpful to be aware of some of the basic principles of light and shadow.
The shades of the front faces reveal the direction of light.
The light source is conceived to be opposite of the darkest side. If the left and right sides are similar it indicates that the light source is either directly in front of or behind the scene.
I wouldn't say that there is a real convention for the direction of light. In user interfaces (and graphic design with "artificial depth" in general) the light mostly comes from the top left (like the light from an ordinary desk lamp), so my initial guess would be that the same goes for isometric pixel art, but a google image search for isometric pixel art reveals that different light directions are common.
The contrast between the shades on the faces can tell something about the diffusion and reflection of light and the amount of ambient light. It's probably most common to use the same contrast for all colors, but it's also possible to use different contrasts for different materials.
Low contrast indicates lots of ambient light and no directional light and can be used for light emitting materials. Medium contrast indicates a mix of ambient and directional light and can be used for ordinary matte materials. High contrast indicates no ambient light and bright directional light and can be used for reflective materials.
The color of light can altered be by tinting the shades, highlights and the black and white contours.
Warm tones like orange highlights and purple shades can look like a candlelit room or a sunset. Neutral tones can look like bright daylight. Cool tones like cyan highlights and blue shades can give the impression of being indoors in electric light.
There are many different approaches to achieving consistent shading depending on your personal taste, which application you use and your workflow. Here are some general examples.
A simple method could be to mix the colors with fixed amounts of black and white. In this example the highlights are the base color mixed with 40% white and the shades are the base color mixed with 50% black.
Another method is to use the base color for the top face and tint both the front faces darker. In this example the front faces are the base color mixed with respectively 30% black and 60% black.
Here is an example where the highlights are the base color mixed with 50% #ffcd70, the shades are the base color mixed with 50% #641862, the white contours are #ffcd70 and the black contours are #641862.
Pixel art emerged in the 1980s and back then the computers had very limited fixed palettes of for example 4, 16 and later 256 colors. The artists couldn't choose carefully calculated shades, but had to make do with what they had at hand. This limitation actually served as a catalyst for creativity as artists were forced to choose more sophisticated color combinations.
Have a look at the default EGA 16-color palette used on PC from the mid 1980s.
If we were to only use these colors we would be forced to introduce inconsistent contrast and to allow hue shifts. This can, in my eyes, add to the artistic value of the coloring.
A way to achieve a wider array of perceived colors with a limited palette is to use dithering.
Here I've just used Photoshop's Indexed Color with Pattern dithering, but a true pixel artist would of course do the dithering "by hand" to have maximum control.
Lots of pixel art doesn't 'calculate' any shading or other lighting colours, for they use a fixed palette. Just find the most appropriate colour (or the least unappropriate, in some cases) in the palette you're using and go with that.
This palette limitation is to emphasise pixel art's simplicity and it harkens back to the old computer systems the medium was born on: those couldn't display more than a set amount of colours. As 'restrictions breed creativity' (— Mark Rosewater), lots of pixel artists like to limit themselves colourwise as well as resolution-wise.
Good examples of palettes are Fleja's Master Palette and the PICO-8 palette.
Because exact lighter or darker versions of colours are (most often) absent in these palettes, pixel artist usually apply hue shift in their lighting. That is, if a lighter or darker colour variant is needed, the colour is slightly changed around the colour wheel. For a lighter red, a pixel artist might use, in sequence, a dark orange, a light orange, and a yellow. This 'ramp' as it is called, pushes the lighter colours to the warmer side of the colour wheel. Darker red ramps can be made with purples, blues or even deep greens, pushing towards the cooler side of the spectrum.
Pixel art custom palettes are commonly built around these ramps. The Fleja Master Palette I mentioned earlier is a good example of this.
Here's an example of one of @Wolff's cubes with (rather extreme) hue shift for lighting, and some shading:
Well, there is a formula (or many formulations) but that depends on what material you have and what direction your light is shining from, the formula uses stuff like Sin and Cross product.
However, nobody uses this, because they can just choose colors by themselves. After all if you are manually drawing the isometrics its mostly about taste.
OTOH, if you are making a computer program to do this. Then i suggest you look up a tutorial to 3D graphics programing
