AI image generation for projects: why a pixel-art image is not yet a sprite
A convincing pixel-art knight can still be unusable as a game asset: uneven pixel sizes, extra colours and a background that is almost—but not quite—solid.
Daniel Ball’s illustrated guide works through image-generation experiments for sprites, posters and game backgrounds. Its useful dividing line is practical: let the model propose the artwork, then use ordinary image-processing code for exact sizes, palettes, text and layout.
It also explains why cost per usable result matters more than the price of a single generation. A useful read for anyone turning generated pictures into something that actually ships.