Why computers needed a display card
Early personal computers had to put something on a screen, and the circuitry that does this is the display adapter, or graphics card. The adapter reads a region of memory, converts what it finds into signals and sends them to the monitor. Many early machines like the Apple II had this built into the main board. The IBM PC of 1981 instead used separate plug-in cards, which is why a market for graphics boards grew up around the PC standard.
Text-only beginnings
The IBM Monochrome Display Adapter, or MDA, launched with the PC in 1981. It showed crisp characters, 80 columns by 25 lines, on a single-color monitor, but it could not draw pictures. For business work such as spreadsheets and word processing, that was enough, and the text was easy to read.
A rival card from Hercules in 1982 kept the sharp monochrome text but added a graphics mode at a higher resolution, which made it popular with people who wanted charts without buying a color monitor. Compatibility with an existing standard, and software that supported it, repeatedly decided which cards succeeded.
First colors
Also in 1981, IBM released the Color Graphics Adapter, or CGA. It offered a handful of colors in low-resolution graphics modes, typically four colors at one resolution, and a text mode with 16 colors. The results were crude, and the color palettes looked garish to many users, but CGA brought color games and simple drawing to the PC.
In 1984 IBM introduced the Enhanced Graphics Adapter, or EGA, with sharper images and a 16-color palette picked from a larger set. EGA required its own type of monitor, and it also marked the point when graphical software on the PC began to look more convincing.
VGA becomes the baseline
The next major step was Video Graphics Array, or VGA, which IBM released in 1987 with its PS/2 line. It supported 640 by 480 pixels with 16 colors and a 320 by 200 mode with 256 colors, using analog signals that allowed many more shades than earlier digital connections. VGA proved durable. Clone makers copied it, and for years the VGA mode was the fallback that every PC could display. The standard connector from that era remained common on monitors for decades.
Third-party makers then built Super VGA cards that offered higher resolutions and more colors, though without a single official standard at first. A body of manufacturers later agreed on common interfaces to make software easier to write.
Windows and 2D acceleration
When graphical operating systems such as Windows became mainstream in the early 1990s, the processor spent a great deal of time redrawing windows, filling rectangles and moving blocks of pixels. Cards with 2D acceleration took on those jobs. Chips from companies such as S3 and ATI handled operations like drawing lines and copying areas of the screen directly, so the screen felt faster. These cards also increased video memory so that more colors and resolutions could be used at once.
The arrival of 3D cards
In the mid-1990s, games began using three-dimensional worlds, which demand heavy calculation: transforming shapes, applying textures and deciding what is visible. Dedicated 3D boards appeared. The 3dfx Voodoo, from 1996, was a widely noticed example. It was an add-on card that worked alongside an existing 2D card and handled texturing and shading for supported games. Soon after, cards combined 2D and 3D functions on a single board.
In 1999 Nvidia marketed its GeForce 256 as a graphics processing unit, or GPU, because it also took over some geometry calculations that had previously fallen to the main processor. The term “GPU” caught on, though other companies had similar ideas, and what counts as the first GPU depends on the definition used. Specialized graphics accelerators had existed for years in expensive professional workstations, so the PC cards were adapting older ideas to a lower price.
Connections and standards
The physical connection also changed. Cards moved from the older ISA slot to PCI, then to a dedicated Accelerated Graphics Port, and later to PCI Express. Each change widened the path between the card and the rest of the computer. Display outputs moved from analog VGA to digital interfaces such as DVI, HDMI and DisplayPort, which suit flat panel monitors. Features such as 4K displays require fast links and plenty of video memory.
Impact
Graphics cards shaped the culture of PC gaming and professional design. They also drove demand for better memory and cooling, and later became useful for other kinds of computation. As processors began to include basic graphics on the same chip, many users stopped needing a separate card at all, while enthusiasts and professionals continued to buy powerful ones.