Ruminations on the electronics industry from David Manners, Senior Components Editor on Electronics Weekly.
After the 4004: the 8008 and 8080. By Federico Faggin
The fifth in our Friday weekly series on the invention of the microprocessor is an account of how, after getting the world’s first microprocessor, the 4004, into silicon, Dr Federico Faggin set about his next task at Intel – designing the first 8-bit microprocessor, the 8008.
“It was architecturally a better machine than the 4004 – the 8008 was the ancestor of the Pentium Pro” says Faggin.
In 1969, Victor Poor vice president of Computer Terminal Corp (CTC), had asked Intel to integrate onto a few chips about a hundred logic circuits which made up the processor of CTC’s intelligent terminal called ‘Datapoint 2000′.
“Hoff estimated that Intel could integrate it all onto a single chip” remembers Faggin.
That chip could have been the first microprocessor because Intel started work on it before the 4004. However the CTC project had been shelved when Faggin’s co-designer Hal Feeney was re-assigned to a memory chip design.
“At the end of 1970 Intel resumed the CTC project under my direction” says Faggin. Feeney came back to help.
As things turned out CTC didn’t use the chip (which it internally called the 1201) because it turned out to be too slow, and needed too many support chips (about 20), and was uneconomic in the 1971 recession which brought down the cost of the logic components which the chip was designed to replace.
CTC agreed to let Intel use the 1201′s architecture in return for dropping its development charge and Intel put it on the market as the 8008.
The 8008 achieved some distinction as the CPU for what is arguably the first PC.
“The first thing you could call a PC was a French machine – the MISTRAL – which used the 8008 and was sold in 1972 or 73 as a general purpose desktop computer”, says Faggin.
The birth of the 8008 also gave rise to one of the other claims on the parentage of the first microprocessor.
CTC had asked Texas Instruments to make the same chip as a second source. In June 1971 TI ran an advertisement in Electronics captioned ‘CPU on a chip’ describing the CTC chip.
“Surprisingly,” comments Faggin, “TI patented the architecture of the 1201 which was CTC’s architecture with Intel’s inputs, but the TI chip never worked, and was never marketed.”
“An invention requires reduction to practice it can’t just be an idea” explains Faggin the idea of a CPU on a chip had been around since the mid-60s. Fairchild had developed a 1-bit serial CPU architecture as had Rockwell. TI might have been the first company to announce the microprocessor, but making it work was the trick.”
For the same reason he dismisses the claim of Gilbert Hyatt to have the first microprocessor.
“While Hyatt was said to have built a breadboard prototype of his microprocessor architecture no single chip implementation was ever produced. The idea was not reduced to practice”, he says.
“What Hyatt, TI, and others failed to do, Intel did: it made the first microprocessor work at low-cost and in volume production”. says Faggin, “it took vision, guts and lots of work to bring to market a product that was so different, but Intel did it – taking a big risk at a time when it was still small and could ill afford to fail.”
Faggin’s next trick after the 8008 was “the one that opened the floodgates” – the 8080.
“The 8080 was my architecture” says Faggin. He got the go ahead from Intel management in the summer of 1972 and by November had completed the architecture.
Shima was recruited from Busicom to work on the design team along with Faggin and Stan Mazor.
Faggin was able to use the new n-channel MOS process (the 4004 and 8008 had used p-channel MOS) which had been developed for the 4Kbit DRAM which considerably increased speed. The 8080 could execute 290 0 operations a second (the 4004 could do 60 0 and integrated 6 0 transistors (the 4004 had 2 300 Unlike the 8008 which required twenty support chips the 8080 required only six.
It took a year to put the 8080 into silicon and the first production run was in December 1973. Intel introduced the chip to the market in April 1974 at a price of $360. The response was so great that the first five months of shipments repaid the 8080′s development costs.
“The 8080 really created the microprocessor market”, says Faggin it was immediately used in hundreds of different products.
One of the hundreds of applications was the Altair computer made by Ed Roberts’ MITS company in 1975, which is often cited as the first PC, and for which the then teenaged duo Bill Gates and Paul Allen wrote the BASIC interpreter.
The 8080 was Faggin’s swansong at Intel. “I worked very hard at Intel but I was not a founder so while I got some financial success from stock options I didn’t have a lot. So I thought I’d learn a lot faster and grow more if I tested my own wings. I wanted to get more financial satisfaction out of my work explains Faggin, Intel in those days was not really committed to microprocessors – they were committed to memories. My work wasn’t appreciated. Also Intel was getting big and Grove would clearly become the next CEO. Andy’s style is kind of tough. And I want my own way too!”
NEXT FRIDAY, August 15th: Designing the 8080, by Masatoshi ShimaTags: Altair, computer terminal, CTC, french machine, Intel, invention of the microprocessor, memory chip, mistral, texas instruments