A CDE Definition
One integrated circuit (one chip). As you trace the evolution of all electronic devices (computers, calculators, phones, etc.), more circuits are combined into one chip all the time. What took hundreds, even thousands, of chips in years past fits on a single chip today. See chip.
Five Years Apart
One Transistor Per Chip
(1) A bit in a spreading signal. See chip rate.
(2) (C.H.I.P.) World's first $9 computer. See C.H.I.P..
(3) (CHIP) (Children's Health Insurance Program) See healthcare IT.
(4) A set of microminiaturized, electronic circuits fabricated on a single piece of semiconducting material. The driving force in this industry and officially called an "integrated circuit" (IC), unpackaged ICs look like tiny "chips of aluminum." While most chips contain only digital circuits, some are analog only, and some are mixed analog and digital (see mixed mode). Digital chips are designed for use as processors, memory and controllers in computers and myriad consumer and industrial products.
Before placed in their housings, raw chips are approximately 1/30" thick and from 1/16" square to the footprint of a postage stamp. Small chips hold from a handful to tens of thousands of transistors; large ones can contain billions. It is actually only the top one thousandth of an inch of a chip's surface that holds the active circuits. The rest is substrate. Although chips may be formed from other materials, silicon is the primary element. See silicon.
Types of Chips by Function
Logic Chips - Completely Fabricated
A logic chip processes data. A general-purpose logic chip, called a "microprocessor," follows instructions in software. Since software is easily changed, microprocessors are the most flexible logic chip. See microprocessor.
A special-purpose logic chip, called an "application specific IC" (ASIC), performs a fixed set of steps that cannot be changed. ASICs are typically smaller and faster than microprocessors and less expensive when manufactured in large volume. See ASIC.
Logic Chips - Partially Fabricated
All logic chips start out in a semiconductor fabrication facility. However, there are categories of logic chips that are only partially finished at the plant and programmed to completion by the customer, who is the circuit designer. See PLD.
Memory chips store data and instructions either temporarily or permanently. RAM chips are the computer's main memory and are either DRAM (fast) or SRAM (faster), but both are volatile and lose their content without power. Firmware is a category of memory chips that holds its content without power. See RAM, ROM, EEPROM, flash memory and early memories.
A "microcontroller" (MCU) is a single chip that contains all the components of a computer, including the processor, non-volatile ROM or flash memory, volatile RAM memory, I/O control unit and timing clock. More than a billion microcontrollers are used in myriad products every year. See microcontroller.
Analog/Digital and Signal Processing Chips
"A/D converter" and "D/A converter" chips convert signals from the outside world (audio, video, voltage, etc.) to the digital world. A related chip is a "digital signal processor" (DSP) that performs fast instruction sequences commonly used in such applications. See A/D converter, mixed mode and DSP.
How the Chip Came About
In 1947, the semiconductor industry was born at AT&T's Bell Labs with the invention of the transistor by John Bardeen, Walter Brattain and William Shockley. The transistor, fabricated from solid materials that could change their electrical conductivity, would eventually replace the bulky, hot, glass vacuum tubes used as electronic amplifiers in radio and TV and as on/off switches in computers. By the late 1950s, the giant first-generation computers gave way to smaller, faster and more reliable transistorized machines. See transistor.
Drs. Bardeen, Shockley and Brattain
The original transistors were discrete components; each one was soldered onto a circuit board to connect to other individual transistors, resistors and diodes. Since hundreds of transistors were made on one round silicon wafer and cut apart only to be reconnected again, the idea of building them in the required pattern to begin with was obvious. In the late 1950s, Jack Kilby of Texas Instruments (TI) figured out how to make capacitors and resistors from the same semiconductor material. Subsequently, Kilby, along with Robert Noyce of Fairchild Semiconductor, created the integrated circuit, a set of interconnected components on a single chip.
Since then, the number of transistors that have been put onto a single chip has increased exponentially, from a handful in the early 1960s to millions by the late 1980s. Today, a billion transistors take up no more space than the first transistor. See integrated circuit.
From Tube to Transistors in a Chip
LOGIC AND MEMORY
In first- and second-generation computers, internal main memory was made of such materials as tubes filled with liquid mercury, magnetic drums and magnetic cores. As integrated circuits began to flourish in the 1960s, design breakthroughs allowed memories to also be made of semiconductor materials. Thus, logic circuits, the "brains" of the computer, and memory circuits, its internal workspace, were moving along the same miniaturization path.
By the end of the 1970s, it was possible to put a processor, working memory (RAM), permanent memory (ROM), a control unit for handling input and output and a timing clock on the same chip.
Within 25 years, the transistor on a chip grew into the computer on a chip. When the awesome UNIVAC I was introduced in 1951, one could literally open a door and walk inside. Who would have believed the equivalent electronics would some day be built into a wristwatch?
To learn how the chip is made, see chip manufacturing. To understand how circuits process data, see Boolean logic.
Transistors to Gates to Circuits
Before/After Your Search Term
Terms By Topic
Click any of the following categories for a list of fundamental terms.