If a 3D artist would play a memory game and he would uncover a card representing a normal map and another one showing a high poly model he would state that this is the perfect pair. It’s just totally common that normal maps get created out of a 3D mesh (or sometimes the NVidia Photoshop Plugin). Except from some exceptions it’s not the best idea to manually paint in a normal map since every pixel is a tiny light vector and it’s hard to imagine how the resulting shading will look like.

But sometimes people don’t think too much about issues or problems and just do it. And sometimes the result is just stunning. In my eyes, this tool-project is a great example:

Source: Sprite Lamp

As you might know, a normal map stores one light vector per pixel and uses the R, B and G channel to store the X, Y and Z values of the vector. Sprite Lamp enables you, to paint the channels of the normal map separately and (more important) see the results quickly.

Source: Sprite Lamp

Imagine you would have to model a highpoly-retro-character like that to generate the normal map out of 3D data… If you’re interested in this tool, feel free to checkout its Kickstarter page.

Thanks to Anna for bringing us another example for normal mapped pixel art on environments:

Source: Chroma

By the way, this whole thing reminds me of a very interesting tutorial of how to create a normal map from the real world with a camera. This is pretty much the same idea of creating the different channels by hand (but using a camera instead of painting them):

Source: Tutorial by Ryan Clark

These examples showed me once more: Sometimes you need to think out of the box, stop worrying and just try it out.