USB set out to tidy the back of the PC and replaced nearly every port
In the mid-1990s, the back of a personal computer bristled with different sockets: serial, parallel, game and more, each needing its own fiddly setup. Seven companies set out to replace them with one plug. Today its descendant, USB-C, also carries video and charges laptops at up to 240 watts.
Development began in 1995 among Compaq, DEC, IBM, Intel, Microsoft, NEC and Nortel, and the standard was introduced in 1996. At Intel, Ajay Bhatt and his team worked on it, and Intel produced the first supporting chips. The idea borrowed from earlier shared buses, notably the one on Atari's 8-bit computers; its designer, Joseph Decuir, later worked on USB for Microsoft, and that old Atari design was cited as prior art when USB was defended against a patent troll.
The attraction for users was simplicity. USB configures itself, so nobody has to set interrupts or memory channels by hand. Devices can be plugged in and removed while the computer runs, and small gadgets draw power straight from the port instead of needing a separate adapter. Because the logo is only allowed after compliance testing, a certified device should simply work. For manufacturers it removed the need to invent a proprietary connector for every new peripheral, though selling a product requires paying the USB Implementers Forum for an ID.
There are trade-offs. Cables are short because the standard assumed devices on the same desk, and peripherals cannot talk directly to one another; everything passes through the host in a strict tree. Physical plugs multiplied too, with full-size, Mini and Micro versions of the older Type-A and Type-B connectors.
USB-C, which is reversible and required for USB4, now replaces all of them. Speeds have climbed from 12 megabits per second in the earliest version to 80 gigabits in the newest, and USB charging has cut the need for proprietary phone chargers.
Source: USB
