History of Computers/Early Innovations
The Vacuum Tube
[edit | edit source]The vacuum tube, also known as a valve because it controls electron flow the way a faucet valve controls the amount of water flow, was originally developed to amplify and demodulate ("detect") radio signals. Later on, it also amplified audio for sound reinforcement. In World War 2, radar (and sonar) used tubes for more-sophisticated processes, many of which were described at the time as "pulse circuits". Once the War ended, digital computers using tubes were built, however the sheer number of logic functions meant that very large numbers of tubes were needed. Heater power, on the order of a few watts per tube, had to be provided, and the heat removed. Large cooling systems were needed to carry away the heat. While tubes designed for simpler systems such as radios, as well as stereo and musical instrument amplifiers could last for decades, for digital computers, tubes were not adequately reliable. Later, tubes specially designed for computers, such as the RCA "Special Red" series, type numbers 5961, 5962, and 5963, were better suited, although their heaters still consumed significant power.
[Nevertheless, the SAGE (Semi-Automatic Ground Environment) air-defense computers in the USA, although using only tubes, were extremely reliable because the computers were installed in essentially-identical pairs, only one being in service. The other was run periodically at reduced power, which made weak tubes fail, so they could be replaced. Periodically, the standby computer was placed into service, relieving the other computer for maintenance. However, for civilian work, duplicate computers were impractical, and tube reliability was not really satisfactory.]
The Transistor
[edit | edit source]Shockley, Bardeen and Brattain's discovery of the point-contact transistor, followed by the more-practical junction transistor resolved the vacuum tube limitations and in the process started the shrinking of computers.
The transistor is a solid-state amplifier that, when driven to fully-on or fully-off states, serves as a logic gate switch: it does not need to heat a filament nor open or close physical contacts to cause a change in electron flow. The functional part is made of a semiconductor, a solid material that can be one of two types ( p or n) of conductor (or resistor) depending on concentration and choice of trace amounts of chemical elements put into the ultra-pure substrate.
The original substrate for transistors was germanium. The move to silicon, one of the most readily available chemicals on Earth, made transistors cheap to manufacture and eventually made computers ubiquitous. Comparatively recently, for special applications, such compounds as gallium arsenide have come into use.
Until integrated circuits became practical, transistor computers were made of individual transistors (as well as individual diodes, resistors, and capacitors), all referred ot collectively as components. They were supported on etched-circuit cards ("Printed circuit boards"). Such construction was referred to as "discrete component". Component density was, comparatively, quite low, and computers of this period comprised floor-mounted cabinets typically having a volume of a few cubic yards/meters.
The Integrated Circuit
[edit | edit source]Integrated circuits(also known as microchips) are very small electronic circuits consisting of resistors, transistors and other components. The electronic components are etched onto a semiconductor. Silicon is the most widely used semiconductor for making ICs. The first integrated circuit was made by Jack Kilby on 12 September 1958 using germanium, however the design included external gold wires and was thus not a true monolithic(in a single chip) integrated circuit. The first monolithic integrated circuit was made by Robert Noyce in 1959 and was made from silicon.