This article became once first published as “Marcian E Hoff.” It seemed in the February 1994 area of IEEE Spectrum. A PDF version is obtainable on IEEE Xplore. The photos seemed in the authentic print version.
Nonetheless for Hoff, the microprocessor became once merely one blip among many along the tracing of his long fascination with electronics. His ardour for the self-discipline led him from Unusual York City’s aged electronics stores to elite university laboratories, during the extra special early years of the microprocessor revolution and the tumult of the video recreation exchange, and in the kill to his job at the brand new time: high-tech non-public look.
Slightly early in his childhood Hoff figured out that the most efficient solution to undoubtedly feel less like a shrimp one—and a shrimp extra extremely effective—became once to grab how things work. He started his explorations with chemistry. By the age of 12 he had moved on to electronics, building things with draw ordered from an Allied Radio Catalog, a shortwave radio kit, and surplus relays and motors salvaged from the rubbish at his father’s employer, Frequent Railway Label Co., in Rochester, NY. Then in high college, working mostly with 2ndhand draw, he constructed an oscilloscope, an success he parlayed into a technician’s job at Frequent Railway Label.
Hoff returned to that job all the scheme through breaks from his undergraduate experiences at Rensselaer Polytechnic Institute, Troy, N.Y. Plenty of summers started with Hoff coming into the Frequent Railway laboratory to acquire the researchers’ two simplest oscilloscopes broken. He would restore the negate-of-the-art Tektronix 545s, then circulate on to extra attention-grabbing stuff, like inventing an audio frequency railroadinsist monitoring circuit and a lightning security unit that gave him two patents old to he became once out of his childhood.
The most easy ingredient about the job, Hoff recalled, became once the catch admission to it gave him to attract that had been beyond the budgets of most engineering college students in the l950s—transistors, as an instance, and even the staunch-launched energy transistor. He did an undergraduate thesis on transistors aged as switches, and the money prize he obtained for it swiftly went for a Heathkit scope of his indulge in.
Early Neural Networks
Hoff beloved the engineering applications at Rensselaer, but now not the narrow focal level of the school itself. He wished to expand his perspective, each intellectually and geographically (he had never been larger than about a miles west of Niagara Falls), so selected California’s Stanford University for graduate college. Whereas working toward his Ph.D. there, he did review in adaptive systems (which at the brand new time are known as neural networks) and, along with his thesis advisor Bernard Widrow, racked up two extra patents.
“He had a toy insist transferring encourage and forth below laptop regulate, balancing a broom stick. I saw him as a kooky inventor, a wrathful scientist.”
His Intel colleague Mazor, now practicing supervisor at Synopsys Inc., Mountain Assume, Calif., recalled meeting Hoff in his Stanford laboratory.
“He had a toy insist transferring encourage and forth below laptop regulate, balancing a broomstick,” Mazor stated. “I saw him as a kooky inventor, a wrathful scientist.”
After getting his diploma, Hoff stayed at Stanford for six extra years as a postdoctoral researcher, continuing the work on neural networks. At the starting put, his neighborhood made the networks trainable by the expend of a tool whose resistance changed with the quantity and route of new utilized. It consisted of a pencil lead and a share of copper wire sitting in a copper sulfate and sulfuric acid solution, and so that they known as it a memistor.
“One result of all our work on microprocessors that has repeatedly soft me is that we got laptop systems far from these [computer center] of us.”
The neighborhood quickly got an IBM 1620 laptop, and Hoff had his first journey in programming—and in bucking the system. He had to deal with officials at the campus laptop center who belief all laptop systems should always peaceable be in one area, speed by consultants who handled the containers of punched playing cards delivered by researchers. The premise that a researcher should always peaceable program laptop systems interactively became once anathema to them.
Ted Hoff: Predominant Stats
Marcian E. (Ted) Hoff Jr.
Date of birth
Oct. 28, 1937
Better half, Judy; three daughters, Carolyn, Lisa, and Jill
BS, 1958, Rensselaer Polytechnic Institute, Troy, N.Y.; MS, 1959, Ph.D., 1962, Stanford University, California, all in electrical engineering
First electronics job
Technician, Frequent Railway Label Co., Rochester, N.Y.
Supreme shock in occupation
Media hysteria over the microprocessor
Books now not too long prior to now read
Introduction to Nuclear Reactor Theory by John R. Lamarsh; A Compiler Generator by William M. McKeeman, James J. Horning, and David B. Wortman
Of us most revered
Intel Corp. founders Robert Noyce and Gordon Moore, Intel chief govt officer Andrew Grove
Favourite eating locations
Postrio and Bella Voce in San Francisco, Beausejour in Los Altos, Calif.
Favourite motion photographs
2001, Dr. Strangelove
“If it undoubtedly works, it’s shapely”
Fiddling with electronics; attending operas and concerts; going to the theater, physique browsing in Hawaii; walking his Alaskan malamutes
“The most easy motivation is self-motivation”
IEEE, Sigma Xi
Stuart Balantine Medal of the Franklin Institute, IEEE Cledo Brunetti Award, IEEE Centennial Medal, IEEE Fellow
“One result of all our work on microprocessors that has repeatedly soft me,” Hoff informed IEEE Spectrum, “is that we got laptop systems far from these of us.”
By 1968 pupil hostility to the manager over the Vietnam Battle became once rising and lifestyles for researchers on campus who, like Hoff, relied on executive funding became once taking a stare as if it would also catch discouraged. Hoff had already been contemplating the potentialities of industrial jobs when he got a phone name from Robert Noyce, who informed him he became once starting a brand new company, Intel Corp., and had heard Hoff will seemingly be attracted to a job. He asked Hoff the put the semiconductor constructed-in circuit exchange would acquire its subsequent boost home. “Memories,” Hoff spoke back.
That became once the reply Noyce had in thoughts (Intel became once launched as a reminiscence producer), and that year he employed Hoff as a member of the technical crew, Intel’s 12th employee. Engaged on reminiscence abilities, Hoff quickly got a patent for a cell to be used in MOS random-catch admission to constructed-in circuit reminiscence. Titillating on to alter into supervisor of applications review, he had the first buyer contact of his occupation.
“Engineering of us tend to possess a undoubtedly haughty perspective toward advertising and marketing, but I found out you be taught a huge quantity in case you preserve your eyes and ears open in the self-discipline.”
“Engineering of us tend to possess a undoubtedly haughty perspective toward advertising and marketing,” Hoff stated, “but I found out you be taught a huge quantity in case you preserve your eyes and ears open in the self-discipline. Attempting to attract shut what complications of us strive to resolve is terribly priceless. Of us encourage in the lab who don’t possess that contact are working at a problem.”
From 12 Chips to One Microprocessor
One neighborhood of customers with whom Hoff made contact had been from Busicom Corp., Tokyo. Busicom had employed Intel to manufacture a space of personalized chips for a low-charge calculator and had despatched three engineers to Santa Clara to work on the chip designs. Hoff became once assigned to gaze after them, getting them pencils and paper, showing them the put the lunchroom became once—nothing technical.
Nonetheless the technical half of Hoff’s thoughts has no off-switch, and he swiftly concluded that the engineers had been entering the irascible route. Twelve chips, every with larger than 3000 transistors and 36 leads, had been to address assorted aspects of the calculator common sense and controls, and he surmised the packaging by myself would charge larger than the focused retail trace of the calculator. Hoff became once struck by the complexity of this shrimp calculator, when put next with the simplicity of the PDP-8 minicomputer he became once for the time being the expend of in one more project, and he concluded that a easy laptop that can also address the functions of a calculator is also designed with about 1900 transistors. Given Intel’s developed MOS direction of, all these, he felt, can also fit on a single chip.
Marcian E. “Ted” Hoff
The Busicom engineers had no staunch interest in dumping their make in desire of Hoff’s unproved proposal. Nonetheless Hoff, with Noyce’s blessing, started engaged on the project. Quickly Mazor, then a review engineer at Intel, joined him, and the two pursued Hoff’s solutions, rising a easy instruction space that is also utilized with about 2000 transistors. They showed that the one space of instructions can also address decimal addition, scan a keyboard, retain some degree out, and produce other functions that had been allocated to separate chips in the Busicom make.
In October 1969, Hoff, Mazor, and the three Jap engineers met with Busicom management, visiting from Japan, and described their divergent approaches. Busicom’s managers selected Hoff’s methodology, partly, Hoff stated, attributable to they understood that the chip can also possess diverse applications beyond that of a calculator. The project became once given the within moniker “4004.”
Federico Faggin, now president and chief govt officer of Synaptics Inc., San Jose, Calif., became once assigned to make the chip, and in 9 months came up with working prototypes of a 4-bit, 2300-transistor “microprogrammable laptop on a chip.” Busicom got its first shipment of the devices in February 1971.
Faggin recalled that when he started imposing the microprocessor, Hoff perceived to possess misplaced interest in the project, and now not continuously interacted with him. Hoff became once already engaged on his subsequent project, the preliminary make of an 8-bit microprogrammable laptop for Pc Terminals Corp., San Antonio, Texas, which, architected by Pc Terminals, became once named the 8008. Hoff repeatedly “had to attain very chopping-edge work,” Faggin informed Spectrum. “I will also uncover a tension in him to repeatedly be at the forefront of what became once occurr ing.”
In these early Intel days, Mazor recalled that Hoff had a assortment of solutions for projects, many of which, even though now not commercially winning, proved prescient: a RAM chip that might perhaps well well act like a digital digital camera and draw shut an image in reminiscence, a video recreation with transferring spaceships, a tool for programming erasable programmable ROMs, and laptop-aided make instruments supposed for common sense simulation.
The Intel advertising and marketing department they estimated that sales [of microprocessors] can also total supreme 2000 chips a year.
Meanwhile, the microprocessor revolution became once gearing up, albeit slowly. Hoff joined Faggin as a microprocessor evangelist, making an strive to persuade of us that long-established-purpose one chip laptop systems made sense. Hoff stated his toughest sell became once to the Intel advertising and marketing department.
“They had been barely adverse to the premise,” he recalled, for various reasons. First, they felt that every person the chips Intel can also catch would plod for various years to 1 company, so there became once shrimp level in advertising and marketing them to others. Second, they informed Hoff, ‘‘We possess diode salesman obtainable struggling like crazy to sell memories, and also you wish them to sell laptop systems? You’re crazy.” And at splendid, they estimated that sales can also total supreme 2000 chips a year.
Nonetheless note went out. In Would perhaps also fair 1971 an article in Datamation journal mentioned the product, and the next November Intel produced its first advert for the 4004 CPU and positioned it in Electronic Knowledge. By 1972 experiences about the miracle of what started being known as the microprocessor started performing commonly in the press, and Intel’s competitors adopted its lead by launching microprocessor merchandise of their indulge in.
Hoff never even belief to be patenting the microprocessor. To him the invention perceived to be apparent.
One step Hoff didn’t draw shut at that time became once apply for a patent, even though he had already efficiently patented various inventions. (Later, with Mazor and Faggin he filed for and became once granted a patent for a “reminiscence system for a multi-chip digital laptop.”)
Having a stare encourage, Hoff recalled that he never even belief to be patenting the microprocessor in as of late. To him the invention perceived to be apparent, and obviousness became once belief to be grounds for rejecting a patent utility (even though, Hoff stated bitterly, the patent area of commercial for the time being seems to push apart that rule). It became once apparent to Hoff that if in twelve months a laptop is also constructed with 1000 circuits on100 chips, and if in the next year these 1000 circuits is also put onto10 chips, in the kill these 1000 circuits is also con structed on one chip.
Slightly than patenting, Hoff in March 1970 published an article in the lawsuits of the 1970 IEEE Global Convention that acknowledged: “An fully new methodology to make of very shrimp laptop systems is made doubtless by the mountainous circuit complexity doubtless with MOS abilities. With from 1000 to 6000 MOS devices per chip, a entire central processor might perhaps well well be fabricated on a single chip.”
Nonetheless in December 1970, an honest inventor outside the cliquish semiconductor exchange, Gilbert Hyatt, filed for a patent on a processor and mentioned that it became once to be made on a single chip. In 1990, after a tall assortment of appeals and extensions, Hyatt became once granted that patent and started gathering royalties from many microprocessor producers. Presently, even though history traces at the brand new time’s microprocessor encourage to Hoff, Mazor, and Faggin, the precise rights to the invention belong to Hyatt.
The Invention of the Codec
Whereas the microprocessor has proved to be his most successfully-known success, Hoff does now not stare it as his supreme technical step forward. That designation he reserves for the single-chip analog-to-digital/ digital-to-analog coder/decoder (codec).
“Now that work became once an thrilling technical grief,” Hoff recollected with some glee, “attributable to there had been so many who stated it couldn’t be done.”
The project became once kicked off by Noyce, who seen the phone exchange as ripe for new abilities, and urged Hoff to acquire an awfully grand product for that market. Studying phone communications, Hoff and various different other researchers saw that digitized inform transmission, then being aged between central locations of work, depended on the expend of advanced costly codecs that tied into electromechanical switches.
”We belief,” Hoff informed Spectrum, “we can also combine this, the analog-to-digital conversion, on a chip, after which expend these circuits as the premise for switching.”
Besides reducing the charge of the systems to the phone company, such chips would enable companies to produce shrimp branch exchanges that handled switching electronically.
Hoff and his neighborhood developed a multiplexed methodology to conversion wherein a single converter is shared by the transmit and catch channels. They additionally established a assortment of different tactics for conversion and decoding that Hoff saw as now not being apparent and for which he got patents.
With that project’s completion in 1980, after six years of effort, and its transfer to Intel’s manufacturing facility in Chandler, Ariz., Hoff became an Intel Fellow, free to pursue whatever abilities sharp him. What sharp him became once returning to his work on adaptive structures, combining the ideas he had wrestled with at Stanford with the energy of the microprocessor in the carrier of speech recognition. After a year he constructed a recognition system that Intel marketed for various years.
A major buyer for the system became once the auto exchange. Its inspectors aged the systems to succor them review out a automobile as it at splendid left the meeting line. When an inspector successfully-known out loud loads of complications that wanted fixing, the system would rapid him for extra data, and log his responses in a laptop.
From Intel to Atari
Though his area as an Intel Fellow gave Hoff an even quantity of freedom, he found out himself getting bored. Intel’s success in microprocessors by 1983 had grew to alter into it into a chip seller, and other companies had been designing the chips into systems.
“I had repeatedly been extra attracted to systems than in chips,” Hoff stated, “and I had been at Intel for 14 years, at a time when the moderate close at an organization in Silicon Valley became once three years. I became once previous due for a circulate.”
Another time, Hoff had now not long previous beyond sharp by leaving Intel when a brand new job came to him. Atari Inc., Sunnyvale, Calif., then a booming video recreation company owned by Warner Communications Inc. and a predominant user of microprocessors, became once purchasing for a vp of corporate abilities. In February 1983, after discussing the scope of the guidelines that Atari researchers had been pursuing, Hoff latched onto the chance.
Intel from the open up had a structured, extremely managed custom. At Atari, chaos reigned.
Intel from the open up had a structured, extremely managed custom. At Atari, chaos reigned. Under Hoff had been review laboratories in Sunnyvale, Los Angeles, and Grass Valley, Calif.; Cambridge, Mass.; and Unusual York City. Researchers had been engaged on characterize telephones, electronic aids for joggers, laptop controls that gave tactile feedback, graphical environments an a lot like at the brand new time’s digital actuality, digital sound synthesis, developed non-public laptop systems, and tool distribution by the expend of FM sidebands.
Nonetheless Hoff had barely had time to be taught about the entire review projects below scheme old to the video recreation exchange took a successfully-publicized tumble. Without stable within controls, Atari became once unable to be taught how successfully its games had been selling at the retail level, and distributors had been returning a entire bunch of hundreds of cartridges and recreation machines. Hoff started receiving orders for crew cuts monthly.
“It might perhaps perhaps had been one ingredient if I had identified I had to slash encourage encourage to, negate, one-quarter the size of my neighborhood,” he informed Spectrum. “Nonetheless when every month you’re thinking that you simply would also fair should always slash encourage one more chunk, morale undoubtedly drops.”
In July 1984, whereas Hoff became once at his 30th high college reunion, Warner equipped Atari to Jack Tramiel. Hoff then had to catch a name from convincing Tramiel that he can also play a job in a narrowly focused company bored with funding futuristic review, and allowing Warner to clutch out his contract. He selected the latter.
Having a stare encourage, loads of the of us that had been at Atari in as of late now stare them darkly. Nonetheless Hoff remembers his year there as an delectable and in the kill precious journey. “Presumably I gaze at it extra positively than I should always peaceable,” he stated, “but it completely grew to alter into out to be a staunch transition for me, and the lifestyles I even possess now is a undoubtedly good one.”
“Everytime you’re engaged on one area, there is repeatedly one more area over right here that seems extra attention-grabbing.”
He now spends half of his time as a professional and half of pursuing technical projects of his indulge in devising—a readout tool for machine instruments, loads of forms of frame grabbers, pattern recognition, and tactics for analog-to-digital conversion. This variegated schedule is better for him. He has repeatedly felt himself to be a generalist, and has had effort focusing on staunch one abilities.
“It’s easy for me to catch distracted,” he stated. “Everytime you’re engaged on one area, there is repeatedly one more area over right here that seems extra attention-grabbing. Nonetheless now it’s extra seemingly that my indulge in projects catch delayed, in area of things valuable to other of us and their employment.”
Faggin for one is now not taken aback that such honest work appeals to Hoff. “He never became once the gregarious kind,” Faggin stated. “He beloved introverted work, the thinking, the understanding of new things. That is what he is staunch at. I repeatedly became once impressed how he became once in a area to visualize an architecture for a brand new IC, virtually on the gap.”
“He comes up with belief after belief, scenario after scenario. I mediate if he wished to, Ted can also take a seat down and crank out a patent a month.”
Acknowledged Gary Summers, president and chief govt officer of Teklicon Inc., Mountain Assume, the consulting company that employs Hoff at the brand new time: “He comes up with belief after belief, scenario after scenario. I mediate if he wished to, Ted can also take a seat down and crank out a patent a month.”
“There might perhaps be absolute self belief in my thoughts that he is a genius,” Mazor acknowledged. Summers readily concurred.
Hoff’s first project after Atari became once a informmanaged song synthesizer, which gave off the sound of a chosen instrument when someone sang into it. Hoff’s supreme contribution to the project became once a system that ensured that the rising notes might perhaps well well be in tune, or at the least harmonically complement the tune, even when the singer strayed off key. He scored one more patent for this methodology, and the intention became once equipped swiftly during the Sharper Image catalog, but never became a mountainous success.
Hoff peaceable contributes once in a whereas to product designs. At Teklicon, alternatively, the put he is vp and chief technical officer, most of his consulting is completed for attorneys. Hoff has a particular combination of long journey with electronic make and long-standing pack rat habits. His home workshop contains about eight non-public laptop systems of assorted makes an d vintages, 5 oscilloscopes, including a vintage Tektronix 545 scope, 15000 ICs inventoried and filed, and shelves loaded with IC files books dating staunch encourage to the 1960s.
“If my washer breaks down, I name the repairman. Most suave engineers would possess the replacement instruments and install it. Ted is in a position to examining the motive the instruments failed in the first area, redesigning a larger instruments from general principles, carving it out of picket, casting it at his home, and dynamically balancing it on his lathe old to installing it.”
When a criminal legit shows him a patent disclosure, even one a long time extinct, he can make a choice whether or now not or now not it would also then had been “decreased to expend” and whether or now not it equipped enough data to permit “one of standard capability in the art” to expend the invention. Then he can produce a mannequin proving his conclusion, the expend of vintage draw from his assortment, and recount the mannequin in court as an professional test. This mannequin-building can catch very general. On Spectrum’s talk about with, Rochelle salt crystals that Hoff attempted to develop for a new court demonstration littered his workshop ground, subsequent to metal-working instruments that he makes expend of to produce circumstances for his devices.
Hoff sees this capability to catch down to basics as one of his strengths. “I reveal things to valuable principles,” he stated. “Of us that don’t demand the assumptions made going into a venture most continuously stay up solving the irascible area.”
Mazor stated, “If my washer breaks down, I name the repairman. Most suave engineers would possess the replacement instruments and install it. Ted is in a position to examining the motive the instruments failed in the first area, redesigning a larger instruments from general principles, carving it out of picket, casting it at his home, and dynamically balancing it on his lathe old to installing it.”
Doing precise detective work appeals to Hoff for one more motive: it offers him an excuse to hunt for attention-grabbing “antique” draw at flea markets and electronics stores.
Hoff can now not talk about the specifics of patent circumstances he has been alive to with. Plenty of now not too long prior to now had been in the video recreation home; others possess alive to loads of IC companies. In a assortment of circumstances, Hoff became once confident that his aspect became once staunch, and his aspect peaceable misplaced, so he felt shrimp shock when the microprocessor patent became once granted to Hyatt. (After the award became once made, even though, he did take a seat down with Hyatt’s patent utility and attempted to make a working microprocessor fixed with Hyatt’s disclosures. He found out various incongruities—like a clock charge most suited to bipolar abilities with common sense that can also supreme be rendered in MOS abilities, and common sense that required far too many transistors to positioned on a chip, proving in his thoughts that the award became once wrong.)
Seeing someone else catch credit for the microprocessor, in particular in new media experiences, “is stressful,” Hoff informed Spectrum, “but I’m now not going to let it bother me, attributable to I do know what I did, I do know what the entire other of us on our project did, and I do know what extra or less company Intel is. And I do know that I became once the put the motion became once.”
Editor’s point out: Hoff retired from Teklicon in 2007. He for the time being serves as a think for the Collegiate Inventors Competition, held each year by the Nationwide Inventors Hall of Reputation. This day, his predominant technical interests surround vitality, water, and local weather switch.