Pictures Made of Dots
A digital image on a screen is made of tiny dots called pixels. Each pixel has a color, and a picture is simply a grid of them. On a modern display the dots are so small that they are hard to see. On the screens of early home computers and game machines, however, the grid was coarse, and individual pixels were clearly visible. Artists who worked with those grids developed a style now called pixel art.
The word pixel is a contraction of “picture element.” It was used in computing and image research in the 1960s and later became part of ordinary language.
Why Memory Was the Limit
To show an image, a computer must remember the color of every pixel. That takes memory, and early machines had very little. A simple calculation shows the problem. Suppose a screen is 256 pixels wide and 192 pixels tall, and each pixel can be one of just two colors, so that one bit per pixel is enough. The number of pixels is 256 times 192, which is 49,152. Dividing by 8 bits per byte gives 6,144 bytes, or 6 kilobytes, for the whole picture.
If each pixel is allowed one of sixteen colors, it needs 4 bits, which is four times as much: 24,576 bytes, or 24 kilobytes. Many early computers had total memory of about this size or less, and some game machines had far less. So designers could not simply give every pixel a free choice of color. They invented compromises.
Compromises With Color
Different machines handled color in different ways, and the details vary, but several common ideas appeared:
- Small palettes. Only a few colors were available at once, chosen from a larger set.
- Color shared among neighboring pixels. Some computers assigned color to blocks of pixels rather than to each one, which saved memory but sometimes produced visible clashes at the edges of objects.
- Indexed color. Instead of storing a full color for each pixel, the picture stored a small number that pointed into a table of colors, so a change to the table could recolor many pixels at once.
Artists became skilled at working within these rules. Because they could not rely on smooth gradients, they used patterns of alternating dots, a technique called dithering, to suggest additional shades. Viewed from a distance on an old television, the dots blended and appeared as a different color.
The Sprite
Redrawing the entire screen many times a second is slow, especially for the modest processors of the time. A sprite is a way to avoid it. A sprite is a small image, such as a character or a ship, that can be moved around the screen as a unit. On some machines, special hardware handled this. The programmer told the hardware where the sprite’s image was stored and where to place it, and the hardware drew it over the background with little effort from the main processor.
Hardware sprites were typically limited in number and size. A machine might allow only a handful per screen line or restrict each to a certain width. Developers often combined several small sprites to make a larger figure, or flickered them to show more objects than the hardware officially allowed. When a machine had no sprite hardware, programmers wrote routines to move images in software, which was slower and used the processor more heavily.
Tiles and Backgrounds
Backgrounds also needed economy. Many game machines stored a small set of tiles, usually square blocks of eight by eight pixels, and built the background from a grid of references to those tiles. A wall, a patch of grass and a section of sky might each be a tile reused many times. Using this approach, a large scrolling world could be built from a small amount of art. Designers took care that tiles joined neatly, so that repetition was not obvious.
This technique is part of why the early cartridge era was able to deliver detailed worlds on small memory chips, as described in the article on game consoles.
From Constraint to Style
Some early arcade games had a very distinct look, with simple characters made of a few dozen pixels that were nevertheless easy to recognize and often memorable. Artists learned that a few well-placed dots could suggest a face or an expression. Animation was done by swapping between a small number of frames, and just two or three images could create a convincing walk or wave.
When later hardware offered more colors and higher resolution, the older style did not vanish. Many developers continued to use pixel art deliberately, either from affection for the earlier period or because it suits small teams. Making a handful of well-drawn sprites can be faster than building a full three-dimensional model, so the style remains practical as well as nostalgic. Today, pixel art is used in games, illustrations and animation, and has its own tools and communities.
Seeing the Pixels Today
Modern displays can complicate how old art looks. A game designed for a cathode ray tube television was often meant to be seen with some blurring between dots, which softened edges. Shown on a flat screen with sharp pixels, the same art may look harsher than the artist intended. Some players and emulation software add filters to reproduce the soft look of the original television, while others prefer crisp squares. Opinions differ, and neither is the single correct way.
The broader lesson is that picture quality is not only about the number of pixels. Skilled use of a small number can produce images that people remember for decades, as the history of computer graphics repeatedly shows.