Intel Crooks (WAS: Interesting the AMD 2500 is a slower clock than the 2400)

Thu Mar 13 19:17:22 CST 2003

I'm sorry, it's been a long time since I read the story. It appears to
have changed a bit [since about 1991 when I read it initially]. So I
guess, maybe they aren't as evil as I thought they were. Here's the
information. Sorry for the length. I couldn't find a current link. It
looks as though Intel and TI are the current recognized inventors.

Brian

[article about Ted Hoff - of Intel]

As one of the developers of the microprocessor, Ted Hoff is well-known
internationally. Microprocessors are found in virtually every
automobile, medical device, and computer in the world, and the
microprocessor industry has grown astronomically since the first unit
hit the market in 1971. Even though the possibility of putting a central
processing unit (CPU) on a single chip had been discussed throughout the
1960s, Hoff is credited as one of the first people to recognize how to
make a single-chip CPU possible. Because of this, the Economist has
called him "one of the seven most influential scientists since World War
II." By making computers available to the masses, his discovery has even
been equated with as much potential lifestyle changing influence as the
Guttenberg press.

Marcian E. "Ted" Hoff, Jr. was born in Rochester, New York in 1937.
Always interested in science, at the age of 15 he won a trip to
Washington, D.C. and a $400 scholarship from the Westinghouse Science
Talent Search. In 1958, Hoff received a bachelor's degree in Electrical
Engineering from Rensselear Polytechnic Institute in Troy, New York. He
continued his education as a National Science Foundation Fellow at
Stanford, where he was awarded an M.S. and Ph.D. in electrical
engineering in 1959 and 1962, respectively.

While working for the General Railway Signal Company during his summers
and college holidays, he developed two designs that eventually resulted
in his first two patents (he currently has 17 patents). The first design
was for a circuit that detected trains through the audio frequencies
transmitted along the railway track. The second was for an energy
absorbing circuit for lightning protection.

In 1968, Hoff joined Intel Corporation as one of its first employees.
Fairchild founders Bob Noyce and Gordon Moore formed Intel as a spin-off
from Fairchild Semiconductor. Intel believed that the semiconductor was
the wave of the future and would replace the magnetic-core memories
being used in computers. Hoff echoed this view and left his job as a
research associate at Stanford, where he was doing semiconductor R&D, to
join Intel. Intel hired him as the manager for applications research, in
other words, to find applications for the products Intel produced.

In 1969, a Japanese firm called Busicom contacted Intel about developing
custom chips for its new desktop-printing calculator. Hoff thought there
was a better, simpler way to develop the technology than what the
Japanese were initially looking for. Rather than build 12 customized
calculator chips, each with a single specific function, Hoff proposed
that Intel develop a more universal CPU chip that could also run the
calculator. The idea of a CPU on a chip had been around since the early
1960s but had not been feasible then. But Fairchild and Rockwell had
both done some preliminary work in the area and Hoff thought he could
make it work.

During the fall of 1969, Hoff worked with Stan Mazor, an applications
engineer at Intel, to develop an architecture for a chip with a 4-bit
CPU. The new CPU would work with three other chips: a read only memory
(ROM) to store the program instructions, a random access memory (RAM) to
store data, and several input/output (I/O) ports to interface with
external devices such as the keyboard, printer, switches, and lights. In
April 1970, Federico Faggin came to work at Intel from Fairchild, where
he had developed silicon-gate technology. This process allowed for the
fabricating of high-density, high-performance metallic oxide
semiconductor (MOS) integrated circuits (ICs), which was just what Intel
needed to make its chip work. Faggin's job was to translate Hoff's
architecture into silicon, and he immediately began developing the
logic, circuit design, and layout for the chips, which involved creating
a whole new silicon-gate methodology.

The chip design actually encompassed four 16-pin devices, known as the
4000 set. The 4001 was a 2Kb ROM with a 4-bit mask programmable I/O
port. The 4002 was a 320-bit RAM with a 4-bit output port. The 4003 was
a 10-bit serial-in parallel-out shift register to be used as an I/O
expander. The 4004 was a 4-bit CPU. The first chips came off the
production line between October 1970 and January 1971. In November 1971,
Intel advertised the 4004 in Electronic News as a "microprogrammable
computer on a chip." By mid-March 1971, Intel was able to ship the first
chips to Busicom.

Hoff, Faggin, and Mazor immediately pitched ideas for new chip uses to
Intel. But Intel initially thought the chips were only good for
calculators, and besides, Busicom owned the rights to it. The developers
finally convinced Intel that the 4000 series could be used in other
applications and Intel renegotiated a non-exclusivity agreement with
Busicom. In November 1971, Intel began marketing the 4000 series (then
known as the MCS-4) to the public. The original 4004 chip had as much
computing power as the first electronic computer, ENIAC (1946), which
filled an entire room.

Meanwhile, in 1969, Computer Terminal Corporation (now Datapoint)
contacted Intel about incorporating its CPU into a few chips. Hoff began
work on an 8-bit single chip for CTC. This chip, called the 1201, was
intended for intelligent terminal applications and was more complex than
the 4000 series concurrently under development. But Intel suddenly put
the project on hold while its developers worked on other contracts. The
project was not resumed until the end of 1970. CTC had also contracted
with Texas Instruments (TI) to develop a chip and, in June 1971, TI
announced that it had developed a chip for CTC. Unfortunately for TI,
this chip never worked as expected and was never marketed. But TI did
apply for and got a patent for the chip's architecture.

Because of economic downturns, CTC was forced to renege on its contract
with Intel. In exchange for canceling CTC's development costs, Intel was
able to commercialize the 1201 and found a ready market in Seiko of
Japan. The 1201 went on to become the 8008, a precursor of the 8080
which changed the face of the computer industry and secured Intel's
success. In 1996, it was estimated that approximately forty billion
dollars worth of microprocessors was sold, and Intel owned a big slice
of that market.

Unfortunately, the development of the microprocessor has not proceeded
without controversy. On July 17, 1990, after a 20-year battle, Gilbert
P. Hyatt, an Orange County inventor, received a U.S. patent indicating
that he invented the microprocessor and that he had filed for the patent
in 1970, one year before Ted Hoff or TI. According to Hyatt, his patent
for a "Single Chip Integrated Circuit Computer Architecture" established
him as the inventor of the microprocessor. Hyatt claimed to have created
his microcomputer in 1968 as part of his company, Micro Computer. Two of
the people providing financial resources for his company were, Dr.
Robert Noyce and Dr. Gordon Moore, the future founders of Intel. In
1971, Hyatt's company went out of business after Hyatt refused to assign
the financial backers rights to his patents.

According to Hoff and Faggin, they never heard of Gilbert Hyatt until he
received his patent. And since no one kept records on their developments
for the previous 20 years, it became difficult to go back and
reconstruct who did what and when. It took Hyatt 20 years to get his
patent because of ongoing legal continuations (at least seven) and two
rejections that were subsequently overturned by the U.S. Court of
Appeals for the Federal Circuit.

Hyatt also contended that Hoff, Faggin, and Mazor did not receive a
patent for a microprocessor, but instead received a patent for a coder
circuit on a RAM chip that happened to be in a microprocessor
environment. As for Texas Instruments and Gary Boone, Hyatt claimed they
patented a calculator chip, not a microprocessor like his.

In response to Hyatt's allegations, Hoff stated that Intel's patent
filing was somewhat casual and that they put their time into making the
4004 work rather that the patent process. He also contended that the
number of transistors required to make Hyatt's processor work wouldn't
have been practical at the time. Faggin believed that Hyatt's patent was
more like a prophecy than an actual product. Although Hyatt's patent
described the integration of ROM, RAM, and CPU onto a single chip, the
technology did not exist at the time to do it.

Texas Instruments decided to challenge Hyatt's patent, and in April
1991, petitioned the U.S. Patent and Trademark Office to reinvestigate
the award, an action that the Patent Office calls an "interference."
Meanwhile, Hyatt was signing patent-licensing agreements with companies
such as Philips, Sony, and NEC, reportedly worth millions of dollars. In
June 1996, the Patent Office reversed its decision and officially
announced that Gary Boone, formerly of Texas Instruments, was the
inventor of the single-chip microcontroller, not Gilbert Hyatt. The
Patent Office determined that Hyatt's original patent applications in
the early 1970s didn't include the description of a computer-on-a-chip.
This was added to the original in December 1977, six years after Boone's
patent was awarded.

Even though they haven't been involved in the legal issues, both Hoff
and Faggin believe Hyatt's patent would not stand up in court. Many
industry insiders question whether Hyatt's patent really provided enough
information that "one ordinarily skilled in the art could reproduce the
product," a major component of the patent process. Most agree that the
credit for developing a working microprocessor belongs to Hoff, Faggin,
and Mazor at Intel and Boone at TI.

After leaving Intel in 1983, Hoff spent a short time as vice president
for R&D at Atari, then joined Teklicon, Inc., where he is vice president
and chief technologist, working on patents. Teklicon is a litigation
support consulting organization that provides the services of engineers
and scientists to companies needing this type of expertise. Hoff is
involved in building models of things to verify patents as well as
testifying in patent trials.

In 1980, Hoff was named the first Intel Fellow, the highest technical
rank in the company. In June of 1996, Ted Hoff, along with Federico
Faggin and Stan Mazor, was inducted into the National Inventors Hall of
Fame. Other honors include the Franklin Institute's Stuart Ballantine
Medal, the IEEE's Cledo Brunetti Award and its Centennial Medal, and the
Rensselaer Polytechnic Institute's Davies Medal. He is currently a
fellow of the IEEE and a member of the American Association for the
Advancement of Science.

Hoff modestly believes that if his group at Intel hadn't invented the
microprocessor in 1971, someone else would have. The time was ripe, the
need was great, and all of the information was there. But Ted Hoff did
do it, and in doing so, changed the world. New product markets directly
created by the microprocessor are estimated at over five hundred billion
dollars a year. The father of the microprocessor deserves his fame.

> -----Original Message-----
> From: Parker, Ron [mailto:rdparker at butlermfg.com]
> Sent: Thursday, March 13, 2003 12:07 PM
> To: kclug at kclug.org
> Subject: Intel Crooks (WAS: Interesting the AMD 2500 is a slower clock
> than the 2400)
> 
> 
> > From: Brian Densmore [mailto:DensmoreB at ctbsonline.com]
> 
> > ...
> > additionally Intel's
> > founders are crooks who stole a patented idea from their boss and
> > started their own company.
> > ...
> 
> Care to elaborate or point me to an article WRT this?
> 