|
|
Dr. Gordon Moore's Interview on the First Microprocessor
Gordon Moore talks about the first microprocessor in his office at Intel’s Headquarters in a clip from the documentary “Silicon Valley Story,” (1996), by journalist and film director Chiara Sottocorona.
Interview Transcript:
"The first microprocessor that we made at Intel was in response to the interest of a Japanese calculator company that wanted to make a family of scientific and business calculators. Ted Hoff thought that by using a general purpose architecture, general purpose computer architecture, the kind of thing that were being built in larger minicomputers at the time, it would be possible to build all of these calculators and he also thought that the sophistication of the technology at that time was at the point that we could probably put the entire processor on one chip. This was really a very important breakthrough. He was working with Mr. Mazor in seeing how to realize this and then we hired Dr. Faggin who we had known back at Fairchild to come in and actually lead the project to realize the processor that Dr. Hoff had conceived. And Federico Faggin and the team then succeeded in completing the development of these products so that we could ship them. Actually the first one was shipped for calculators early in 1971 and the product was finally announced generally available to other users in November of 1971.
Leadership of the MCS-4 Project - Federico Faggin
"...and then we hired Dr. Faggin who we had known back at Fairchild to come in and actually lead the project to realize the processor that Dr. Hoff had conceived."
In this interview, Gordon Moore, in his own well-pondered words, acknowledges Federico Faggin as the person who led the microprocessor project, clearly dispelling every ambiguity as to the leadership of the chip design, which is often attributed to Ted Hoff. Faggin led the project from inception in 1970 to its successful debut on the market in 1971.
Faggin was the natural leader of the project because at Fairchild, where Dr. Moore had known him before he came to work at Intel, he had developed the technology that made semiconductor memories and microprocessors possible -- the Silicon Gate Technology -- and had designed the world's first commercial integrated circuit using silicon gate (the Fairchild 3708). This technology was eventually adopted worldwide.
Conception of the MCS-4 Project - Ted Hoff
The processor that Dr. Hoff conceived"
"Ted Hoff thought that we could probably put the entire processor on one chip"
The conception of the MCS-4 architecture was not a breakthrough for the following reasons: (1) In 1969 it was well known how to architect a small computer (2) The use of a general purpose CPU at the heart of a desktop programmable, printing calculator was done in 1965 by Olivetti with their Programma 101, and the Busicom architecture already included a CPU to be partitioned in 3 chips (3) Making a CPU on a single chip was an already predicted trend at some of the most advanced semiconductor companies, but had not yet been done, although CPUs using more than one chip had been designed before the 4004 was completed.
Ted Hoff was not a chip designer and he was in no position to tell if the single chip microprocessor could actually be done, and even less to develop it or direct its development (see quotes from interviews). In fact, he thought that the design could use two-phase dynamic logic. However, such methodology required bootstrap loads which were considered not realizable with silicon gate technology without an additional masking step that would have made it uneconomical. Therefore the design would have needed static logic, which required at least twice as many transistors, rendering also the project unfeasible. Hoff and Mazor after contributing their block architecture were not involved with the 4004 design and also they could not give design directions to Hal Feeney during the 8008 project (see 8008 designers).
It was Faggin's inventions of the bootstrap load with silicon gate, combined with his invention of the buried contact that made the 4004, the 8008, and the 8080 (architected by Federico Faggin) possible.
Design and Creation of First Single Chip Microprocessor - A Novel Contribution by Federico and His Team
However, no one had yet been able to design an integrated circuit with the required complexity, speed, power dissipation, and small chip size to make a microprocessor a commercial reality. This was Faggin’s novel contribution that required much creativity in methodology, logic, circuit and layout design, as well as in process technology. The unique silicon design of the 4004, done by Faggin and his team, was the missing link to make the first microprocessor a commercial reality. Hoff and Mazor were not part of the team working with Faggin.
REFERENCES:
Silicon Gate Technology and Fairchild 3708
4004 Microprocessor Display at Opening of New Intel's New Museum (1992)
Programma 101: The incredible story of the first PC, from 1965
Wikipedia: Programma 101
-Ted Hoff was not a chip designer ref: in the book "Inventors at work" by Kenneth A. Brown (1988 Microsoft Press, p. 285). Hoff says that his role consisted in pretty much in defining the architecture and his related concepts. But then he turned his architecture over to the MOS Design Department for its conversion into actual chips.
The Designer Behind the World's First Microprocessor - Federico Faggin
Stan Mazor interview with Rob Walker in “Silicon Genesis”(2000): "Ted and I thought it [the 4004] was a little too aggressive and we weren’t so sure it could be done…"
The Busicom Calculator Engineering Prototype, with the first 4004 ever produced was sent to Federico as personal present by Busicom’s president Yoshio Kojima. It was given in grateful recognition of Faggin's leadership in meeting the new schedule he proposed to Busicom after the architecture proposal idled for about 6 months.
|
|
