computerwriter.com

25^th Anniversary for Microprocessor

Toronto Star, Fast Forward, section front for November 17/96

© Myles White, 1996

Main Article

Computer History Timeline

Back to Star article directory

This week marks the 25th anniversary of the microprocessor, measured from November 15, 1971--the day an advertisement for the Intel 4004 first appeared in Electronic News Magazine. However, acknowledges its "inventor," Dr. Marcian E. (Ted) Hoff Jr., the development of what many see as the heart of the computer revolution wasn't so much an invention as it was an evolution. "Going back as far as 1958, there had been lots of literature talking about building a monolithic computer using vacuum tubes," he said, "but by the mid to late 60's, although many of us were trying to figure out how to build a computer on a chip, the yield from the semiconductor manufacturing process was still too low for it to be practical."

When it first started in the mid-sixties, Intel produced electronic memory components. Hoff was employee number 12 at the company assigned to work with minicomputers and in June, 1969, he was asked to liaison with a group of Japanese engineers from a company called Busicom. They'd approached Intel with a design for a small calculator--a design which called for 12 different semiconductor-based custom chips to handle various of its functions.

Hoff says he looked at the design and struggled with it for a while, but eventually decided there had to be a better way. He felt that programming through read-only memory and general-purpose registers could replace the separate (i.e., discrete) components the Busicom engineers had requested. When he presented his idea to Intel's then chairman, Bob Noyce, the boss was enthusiastic. But when the design was presented to Busicom's engineers, it almost died right there.

"They didn't want to change their design," Hoff said, "but we were able to convince them to allow us to make a pitch directly to the owners of the company. We held an off-site meeting in October. Their engineers made their pitch and we made ours. Busicom's management bought ours."

It took another nine months before a team of Intel engineers, led by Frederico Faggin, could turn Hoff's ideas into hardware. The original 4004 was a silicon-based chip measuring 1/8th of an inch long by 1/16th of an inch wide, containing either 2,108 or 2,300 transistors (it depends on who you ask--Hoff's count is 2,108 but, he says, Faggin included 192 "virtual transistors" in his count). It had about the same amount of computing power as the original ENIAC which weighed 30 tons, occupied 3,000 cubic feet of space and used 18,000 vacuum tubes.

It didn't take Intel long to discover it had something here. The only problem was that the company didn't have it. The 4004 belonged to Busicom (which was also sometimes known as Nippon Calculator). Noyce and his crew could have "retired to the Waggon Wheel bar and cried in our beer," but they flew to Japan and bought back the rights for $60,000 instead. A short time later, Busicom went bankrupt (although, says Hoff, the original founder is still around and still has the rights to the Busicom name, but the company isn't producing anything) and the rest, as they say, is history.

Before the development of the 4004 was completed, Faggin and his crew were already working on the 8008 (3300 transistors), but it would still be another three to four years before the microcomputers of the modern age would appear (see timeline).

"The first microprocessors were industrial controllers," says Hoff. No one really thought of using them in computers. Instead, they wound up as embedded controllers in things like automated gas pumps, traffic controllers and manufacturing pressure and flow meters."

One of the more naive questions I put to Hoff brought a chuckle (he was a kind man when I made dumb queries), but it started to open up how this development was evolutionary instead of fundamental. If Intel was at the forefront of microprocessor design, how did they let it get away from them to companies such as Motorola and Texas Instruments?

We tend to think (or at least I tended to think) that the microprocessor was a unified set of technologies all developed at the same time. It wasn't. Solid state physicists had discovered semiconductors (solid or liquid material, able to conduct electricity at room temperature more readily than an insulator, but less easily than a metal) in the 1920s. Today's favourite materials include chemical elements and compounds such as silicon, germanium; selenium, gallium arsenide, zinc selenide, and lead telluride. A Bell Labs team led by William Shockley used germanium to develop the first transistor in 1948 and by the 1960's, using them in integrated circuits was well under way.

The history of the computer goes back even farther. I'm not sure if anyone knows precisely when the abacus first came on the scene, but we do have some pointers to the development of today's computers and the distance we have to travel in time may surprise you.

In 1623 German scientist Wilhelm Schikard invented a machine that used 11 complete and 6 incomplete sprocketed wheels that could add and, with the aid of logarithm tables, multiply and divide.

In 1642, French philosopher, mathematician, and physicist Blaise Pascal invented a machine in that added and subtracted, automatically carrying and borrowing digits from column to column. Pascal built 50 copies of his machine, but most served as curiosities in parlours of the wealthy. Seventeenth-century German mathematician Gottfried Leibniz designed a special gearing system to enable multiplication on Pascal's machine.

In the early 19th century French inventor Joseph-Marie Jacquard devised a specialized type of computer: a loom. Jacquard's loom used punched cards to program patterns that were output as woven fabrics by the loom (and he ultimately had to flee for his life from Lyon pursued by weavers who feared for their jobs).

Then there was Babbage's Difference Engine, designed in the early 1820s by British mathematician and scientist Charles Babbage. He never finished it, but it was intended to be a machine with a 20-decimal capacity that could solve mathematical problems. Babbage also made plans for another machine, the Analytical Engine, considered to be the mechanical precursor of the modern computer. Although the Analytical Engine was never built, its key concepts, such as the capacity to store instructions, the use of punched cards as a primitive memory, and the ability to print, can be found in many modern computers.

Herman Hollerith, an American inventor, used an idea similar to Jacquard's loom when he combined the use of punched cards with devices that created and electronically read the cards. Hollerith's tabulator was used for the 1890 U.S. census, and it made the computational time three to four times shorter than the time previously needed for hand counts. Hollerith's Tabulating Machine Company eventually merged with other companies in 1924 to become International Business Machines or IBM (heard of them?).

In 1936 British mathematician Alan Turing proposed the idea of a machine that could process equations without human direction. The machine (now known as a Turing machine) resembled an automatic typewriter that used symbols for math and logic instead of letters. Turing intended the device to be used as a "universal machine" that could be programmed to duplicate the function of any other existing machine. Turing's machine was the theoretical precursor to the modern digital computer.

In the 1930s American mathematician Howard Aiken developed the Mark I calculating machine, which was built by IBM. This electronic calculating machine used relays and electromagnetic components to replace mechanical components. In later machines, Aiken used vacuum tubes and solid state transistors (tiny electrical switches) to manipulate the binary numbers. Aiken also introduced computers to universities by establishing the first computer science program at Harvard University. Aiken never trusted the concept of storing a program within the computer. Instead his computer had to read instructions from punched cards.

At the Institute for Advanced Study in Princeton, Hungarian-American mathematician John von Neumann developed one of the first computers used to solve problems in mathematics, meteorology, economics, and hydrodynamics. Von Neumann's 1945 Electronic Discrete Variable Computer (EDVAC--note the picture is of ADIVAC, based on the EDVAC design) was the first electronic computer to use a program stored entirely within its memory.

John Mauchley, an American physicist, proposed an electronic digital computer, called the Electronic Numerical Integrator And Computer (ENIAC), which was built at the Moore School of Engineering at the University of Pennsylvania in Philadelphia by Mauchley and J. Presper Eckert, an American engineer, based on some concepts developed by John Atanasoff, a physics teacher at Iowa State College. ENIAC was completed in 1945 and is regarded as the first successful, general digital computer. It weighed more than 27,000 kg (60,000 lb), and contained more than 18,000 vacuum tubes. Roughly 2000 of the computer's vacuum tubes were replaced each month by a team of six technicians. Many of ENIAC's first tasks were for military purposes, such as calculating ballistic firing tables and designing atomic weapons. Since ENIAC was initially not a stored program machine, it had to be reprogrammed for each task. More pictures available at the Computer Museum Network.

Eckert and Mauchley eventually formed their own company, which was then bought by the Rand Corporation. They produced the Universal Automatic Computer (UNIVAC), which was used for a broader variety of commercial applications. By 1957, 46 UNIVACs were in use.

In the late 1960s integrated circuits, tiny transistors and other electrical components arranged on a single chip of silicon, replaced individual transistors in computers. Integrated circuits became miniaturized, enabling more components to be designed into a single computer circuit.

In the 1970s refinements in integrated circuit technology led to the development of the modern microprocessor, integrated circuits that contained thousands of transistors. Modern microprocessors contain millions. For example, the Intel Pentium Pro, contains 5.5 million transistors; the UltraSparc-II, by Sun Microsystems, contains 5.4 million transistors; the PowerPC620, developed jointly by Apple, IBM, and Motorola, contains 7 million transistors; and the Digital Equipment Corporation's Alpha 21164A, contains 9.3 million transistors.

Dr. Ted Hoff doesn't work for Intel any more. After a brief stint at Atari, he's become chief technical officer and a consultant for a small California firm called Teklicon, specializing in patent research. He says he couldn't have anticipated what his microprocessor would become and there have been surprises--such as the amount of progress in miniaturization that has occurred (the first 4004 used gateways 10 microns wide compared to today's .35 microns). But he's even more delighted over the social impact microcomputers have had and continue to have.

"Who'd have thought that whole industries could have grown up around specialized personal computing markets such as desktop publishing, network access, multimedia authoring, graphics accelerators and MPEG players?

"And the amount of personal control that computers give people is amazing. I mean how does a totalitarian regime maintain technological efficiency and maintain control at the same time? Once the population gets access to computers and telecommunications, the world opens up through the Internet and control vanishes."

Back to top

Back to Star article index

Microprocessor/Computer History at a glance

1623 - Wilhelm Schikard's sprocket "computer" could add, multiply and divide using 11 complete and six incomplete sprocketed wheels

1642 - Blaise Pascal develops a machine in that adds and subtracts, automatically carrying and borrowing digits from column to column. Later, Gottfried Leibniz designs a special gearing system to enable multiplication on Pascal's machine.

1800's - Joseph-Marie Jacquard invents an automated loom which uses punch cards. Mobs of weavers chase him out of Lyon, but his designs are still used today for some fabrics.

1820's - Charles Babbage designs the Difference Engine and Analytical Engine. Although neither is completed, their key concepts of storing instructions, use of punch cards and ability to print can be found in many modern computers.

1890 - Herman Hollerith borrows ideas from Jacquard's loom to produce a tabulator used in the US census. Hollerith's Tabulating Machine Company merges with others to form IBM in 1924

1920's - Solid state physicists throughout the world begin investigating semiconductor materials

1936 - Alan Turing proposes a machine resembling an automatic typewriter using symbols for math and logic instead of letters. Turing's machine was the theoretical precursor to the digital computer.

1930's - Howard Aiken develops the Mark I calculating machine, built by IBM.

1945 - John von Neumann develops EDVAC, the first electronic computer to store programs within its memory.

1945 - ENIAC (Electronic Numerical Integrator And Computer) developed by a team at the Moore School of Engineering at the University of Pennsylvania. The same team goes on to Rand Corp., to develop UNIVAC (the Universal Automatic Computer).

1948 - William Shockley leads a Bell Telephone Laboratories team to develop the transistor, which will eventually kill off the vacuum tube and form the basis for modern digital computers.

1965 - Semiconductor pioneer Gordon Moore predicts that the number of transistors contained on a computer chip will double every year. Moore's Law has proven to be somewhat accurate. The number of transistors and the computational speed of microprocessors currently doubles approximately every 18 months. Components continue to shrink in size and are becoming faster, cheaper, and more versatile

1969 - Intel employee, Dr. Ted Hoff, meets with a group of Japanese engineers and convinces their company, Busicom, to buy an idea of his to integrate 12 discrete custom chips into one.

1971 - Intel advertises the 4004 microprocessor in Electronic News Magazine. 2300 transistors

1972 - Intel 8008 microprocessor, 3300 transistors.

1974 - Motorola 6800, 4000 transistors, destined for use in automotive and industrial applications

1974 - Intel 8080 microprocessor, 8-bit, 2 MHz, 6000 transistors. Ten times the performance of the 8008

1975 - MITS (Micro Instrumentation Telemetry Systems) of New Mexico releases the Altair 8800. Sold primarily in kit form, the Altair was based on the 8-bit Intel 8080 microprocessor, had 256 bytes of random access memory, received input through a bank of switches on the front panel, and displayed output via a row of light-emitting diodes (LEDs)

1978 - Intel 8086, 16-bit, 29,000 transistors, 5, 8 and 10 MHz. Ten times the performance of the 8080

1979 - Intel 8088, 8-bit, 29,000 transistors, identical to 8086 except for 8-bit internal operation.

1979 - Motorola 68000, which winds up in the first Apple Mac in 1984. It and others born from it go on to be part of products such as the Atari ST, Commodore Amiga and NeXT. Only the Mac survives.

1981 - IBM releases the PC which underwhelms the CP/M community, but manages to wipe it out anyway. It uses the 8088.

1982 - Intel 80286, a 16-bit processor with 134,000 transistors. The heart of the AT (advanced technology) PC

1984 - Motorola 68020. 200,000 transistors, 32-bit processor

1985 - Intel 80386 DX, a 32-bit processor with 275,000 transistors.

1987 - Motorola 68030.

1988 - Intel 80386 SX, identical to the 386 DX, but with 16-bit external operation.

1989 - Intel 80486 DX, 32 bit processor with 1.2 million transistors. Copies of the 386 from Advanced Micro Devices (AMD) begin to appear on the market and seriously erode Intel's share.

1990 - Motorola 68040 - 1.2 million transistors

1991 - Intel 80486 SX, identical to the 486, but lacks an internal floating point unit (i.e., math coprocessor). Intel's answer to the AMD 386.

1993 - Intel Pentium. Would have been called the 80586, but the company couldn't copyright a number. For a short time was known as the P5. 3.1 million transistors, 64-bit operation. Cyrix, AMD and others including IBM and Texas Instruments directly attack Intel's 486 market with their own versions.

1994 - Intel settles its suit with AMD and allows direct copies of the 486. Apple, Motorola and IBM combine forces to produce the PowerPC RISC processor line

1995 - Intel Pentium Pro. Would have been the 80686 and was known briefly as the P6. 64-bit processor optimized for applications with 32-bit code base. 5.5 million transistors.

1996 - By now, Motorola has produced several generations of its 600 series PowerPC processors, rivaling Intel's internal clock speeds. The Power Macintosh line is in full swing.

1997 - MMX - a set of multimedia code extensions for the Pentium that will directly handle 'natural data types' such as animation, sound, telecommunications, video accleration and 3-D rendering. Will first appear in Pentium processors, but expected to be throughout Intel's product line by late 1998.

1997 - or early 1998. Merced is the code name for what would have been the 80786 or P7. Jointly developed by Intel and Hewlett Packard, this new processor is designed for applications using a 64-bit code base. Microsoft has already announced they're working on a version of Windows NT to support it.

This document is protected by copyright. If you want to use it for your own amusement, okay. If you want to quote it or excerpt from it for any reason for publication in any medium, you have to ask me. Fair enough?

Back to Top

Back to Star Article Directory