Birth of the Microchip

For a good 50 years before the advent of the transistor, electronic devices were built with vacuum tubes. A vacuum tube served as both an amplifier for the electric current passing through it as well as the essential “on/off” fuction required for the logical functions essential to computers (think binary and boolean). Unfortunately, vacuum tubes were large, fragile, and power hungry devices that got incredibly hot. Engineers had to keep the limitations of vacuum tubes in mind when designing circuits to prevent hardware from melting due to the incredible heat pouring off of these tubes.

Fun Fact: The warmth and glow of the vacuum tubes also constantly attracted moths and other insects to the delicate innards of the computer causing short circuits. Maintenance of these systems required constant “debugging”, coining the popular term still used today to refer to the process of fixing a computer.

The Tyranny of Numbers

The transistor was invented in 1947 by a group at Bell Labs and quickly replaced vacuum tubes as the superior technology. Transistors provide the same key functions as vacuum tubes–amplifying a signal and rapid on-off switching–by moving electronic charges along controlled paths inside a block of semiconductor material like silicon or germanium. Transistors are lighter, smaller, and over 20 times faster than vacuum tubes. With the transistor entering the scene, the potential of circuits was finally fully unleashed.

There are four primary components which can be combined in a variety of ways to make a circuit:

  • The Resistor - restricts the flow of electricity to provide the circuit designer with precise control of the current at any given point
  • The Capacitor - absorbs electrical energy and can then release when triggered
  • The Diode - blocks the current under some conditions and allows the current to flow under others
  • The Transistor - turns current flow on and off, sending digital signals over the circuit

Withe the addition of the transistor to this set, a circuit could now theoretically be as large and complex as you could want, made of almost any combination of the above components. The more complex the task the circuit is performing, the more components it needs. A circuit can easily require thousands if not millions of these components to perform the type of complex tasks that were only able to be read about in science fiction in the 1950s. The key, however, is that the circuit must be complete and unbroken in order to effectively pass the electronic current on to the next node in the circuit. If any one of these components in the circuit fails, the circuit is broken and the entire system fails.

So while on paper engineers could dream up the most complex and beautiful circuits their minds could imagine, the act of manufacturing these systems was complex, arduous and rife with bugs. At the time there was no machine capable of automatically building out these complex circuits, each individual component had to be sodered by hand to a wire connecting it to the next component. The industry quickly realized that this technology was limited by what was referred to as the “tyranny of numbers”. The more components you add to a circuit the more advanced your product, but also the more likely your product would be to never get off the manufacturing floor due to constant component failures and set backs. The computers we have today which have millions of components and interconnections, would be impossible to mass produce with these methods.

Fun Fact: Women were hired in droves to do the meticulous job of wiring these circuits because the “size of their hands were more suited to the job”. Ugh.

The Microchip

Thus a race to solve the “tyranny of numbers” problem began throughout the electronics industry. The solution was found in 1959: connecting multiple devices on a single piece of silicon in order to make interconnections part of the manufacturing process to drastically reduce the cost, weight and size of each element. This solution hinged on two major ideas: first that all parts of a circuit could in fact be made out of the same material, silicon, and integrated onto a single chip, and second that all the wires connecting these components could then be printed directly onto the chip during production. The cost and unreliability of handwiring all of these components could then be entirely circumvented.

References

The Chip, T. R. Reid, 2001

ABOUT LAURA KAPLAN

Throughout my 10 year career I have worked as a web developer, systems administrator, software engineer, security analyst and now cybersecurity engineer. I currently develop software applications to automate security vulnerability and compliance scanning and reporting for a multinational financial institution.