MOS / CSG Commodore Semiconductor Group
Timeline
Written by Ian Matthews of commodore.ca
February 15, 2003
Last Revised Dec 30 2018
NOTE: Your are in the TEXT ONLY version of our site; click HERE to go to our full MOS Technologies / CSG page.
Introduction
At the heart of what people think of as Commodore,
was a company called MOS Technology. MOS’ claim to fame was their
development and manufacturing of the wildly successful 6500 line of
microprocessors which in addition to being used in nearly all Commodore
computers and floppy drives, was also what powered all pre-1984 Apples.
However, MOS was not started by Commodore or even by the engineers who
made it famous, like Chuck Peddle and Bill Mensch.
Started MOS as a
‘second source’ for Texas Instruments (TI) chips and even produced the
famous Atari PONG chip for limited time. A major player in the calculator
business called Commodore Business Machines quickly became MOS’ number one
customer in the early 1970’s.
During
our January 2006 interview with Chuck Peddle, we were informed that MOS
was always pronounced M.O.S. so as not to confuse it with the MOSTek which
was a competing company that was also established in 1969.
Motorola’s Refusal to Innovate
As one of the lead Motorola 6800
engineers at a time before most people had even heard the term “CPU”,
Chuck Peddle was often tasked with explaining the capabilities of a
microprocessor to large industrial manufactures like Ford. After an
education from Chuck engineers were usually highly impressed with the
potential for such a device but at US$300 ($1350 in 2018 dollars!) would
inevitably say it was far to expensive to use. Chuck asked Motorola’s
customers the price they thought it would be possible to put CPU’s into
their mass market products. $25 came up as the magic number.
As one
might guess, a companies engineers discussing a $25 ($110 in 2018 dollars)
version of a successful $300 product did not impress Motorola management.
Conversely, management’s failure to pursue obvious improvements to their
chip did not impress Motorola’s engineers. When Chuck received a formal
letter from management telling him to stop working on a cost reduced
version 6800, he saw an opportunity.
Chuck
Peddle and Bill Mensch took six other key Motorola 6800 engineers to work
for an old General Electric colleague who ran a small “fab” called MOS
Technologies. Bill Mensch explained that Chuck handled most of the
negotiation with MOS and that part of the deal was that each of the eight
were to receive “a fraction of the profits.” The key word here is
‘profits’. At this time, not many companies actually made money from chips
and as your might expect, when the group found out about the deal months
later Bill Said “they were all pissed!”
As might happen today in
your work environment, it irked the existing MOS staff to have an entire
team of highly qualified and recognized people parachuted into the
organization. Bill says that “it was painful, I had two flat tires in my
first week at MOS.” A MOS staffer of Italian decent told him “you don’t
mess with the Godfather”, referring to one of the MOS Vice Presidents who
was unhappy with the recent conversion of Motorola staff. The existing
staff wanted to work on development of a microprocessor and correctly felt
they would be largely left out of the process.
Chuck said he gave
the engineers a “..tight list…” of features to build into the chip along
with a fixed die size. To get to the $25 price they need to produce only
the instructions that its customers would required, nothing extra.
MOS 6500
Within a year the team developed the CPUs
line that would change the world; The MOS 6500 series. There are many
interesting stories surrounding the 6501 but the most amazing is that Bill
Mensch was able to take the 6501 schematics, create a layout completely by
hand (remember no-one had computers back then) and produce a working CPU
on the very first attempt. This was unheard of. Several engineers we
talked with have said that they had never seen anyone manually produce a
successful chip on the first pass. It often takes 10 or more tries to get
it right. In a January 2006 interview with Commodore.ca , Chuck said “Bill
…was like a layout savant… he can just picture an entire layout in his
head.”
MOS 6501 CPU
EXTREMELY RARE: Purchased directly from MOS’ Norristown Fab in late 1975,
This historically important CPU should not exist. Chuck Peddle, engineer
of the fabled 6501 and famous 6502, said that none of the 6501’s should
have ever left the factory because they were not for sale. It is likely
that this the sole survivor.
Note that the MOS 6501 shown to the
right is extremely rare and was purchased directly from the Norristown, PA
factory for the princely sum of $20. When we asked Chuck about it in March
2007 he said “…we stopped producing the 6501 so none really made it to
market in any numbers. We never intended anyone to buy it anyhow. It was
an in your face to Motorola… It is indeed rare.”
The 6501 and 6502
where nearly identical. The primary difference was the pin arrangement; a
6501 is pin compatible with the Motorola 6800 and 6502 is not.
The
group was not without humour. One of the important designers on this chip
was Rod Orgill (who can be seen in this picture to the right and in the
full 1975 article HERE). Bill said that one of the 6502 pins is officially
named SO (Set Overflow). “Chuck, Rod, and I know that it’s real name is
Sam Orgill… Rods dog”.
In the 1970’s, 70% of the industries
chips produced were detected as defective at the factory. This
substantially increased the cost of each viable chip. When a chip is being
laid out for etching on a silicon wafer, it drawn at a large scale and
then photo reduced over an over again (just like a photocopy reduction)
until its microscopic size will fit on the required die. Each reduction
layer is called a Mask. MOS figured out a process to repair Masks as they
are reduced. The end result was that they had a 70% success rate. This
obviously reduced the per chip cost of manufacturing and made the $25
processor a possibility.
Chuck explained that selling a
dramatically less expensive CPU was not as easy as it sounded. A few years
earlier there had been a high profile scam involving a company that
claimed it could produce mainframe terminals it would lease for just $10
per month. The company went bankrupt in a cloud of scandal after taking
millions of dollars from investors, and blamed the failure on industries
inability to produce cheap chips.
In an effort to drum up interest in the
chip they ran an advertisement stating that anyone could see and buy the
amazing $25 microprocessor at WestCon (Western Electronics Show and
Convention) in 1975. Unfortunately when MOS arrived at the show they were
told that, in an effort the keep the show ‘high brow’, exhibitors were not
allowed to sell product at their booths. Chuck quickly rented a nearby
hotel room and had his very attractive wife sit at a table with two full
glass jars of 6501’s and 6502’s. Little did most people know that all of
the chips in the bottom of those jars were defective. Image is everything.
Motorola’s Anger: A Blessing In Disguise
In June of 1975 Motorola
realized they had turned their engineers into their competition. Motorola
got mad and sued MOS for infringement of 6800 patents. Chuck said “…there
was no substance to their claims…” but it scared the old line industry
management at Allen-Bradley. “As soon as lawyers got involved, they wanted
out.” said Chuck. As a shock to everyone, Allen-Bradley walked away from
MOS and basically gave it to the existing MOS management team.
It
is interesting to note that Bill Mensch tells a more complete version of
this important part of the story. “It was not about patent infringement;
it was about intellectual property.” Those eight engineers knew an awful
lot of unpatented concepts developed at Motorola and that is what Motorola
was trying to protect. “We knew we were (infringing). The (MOS) 6520 was a
direct copy of the (Motorola) 6820.” MOS had agreements in place with
Motorola and “…We paid Motorola all along.”
A Star is Born: The
6502
MOS_6502MOS designed and manufactured two 6502
trainers call the TIM1 (Terminal Input Monitor) and KIM-1 (Keyboard Input
Monitor). They are often incorrectly referred to as kit computers the
Altair. The TIM and the KIM came fully assembled and were the world’s
first single board computers.
Like many great products, the 6502
had a humble beginning. Bill told us that he and a few others wrote a
nearly complete specification for the 6500 line “…on the back of an Arby’s
napkin!” When it was completed in the Spring of 1975, the MOS 6502
initially ran at about 1Mhz, the same as the Motorola 6800. However,
6500’s performed about 4 times the number of calculations a 6800 could.
In a 2005 book about Commodore, Bill Mensch is quoted as
saying he had the 6502 running at about 12Mhz. Remember that it wasn’t
until 1983 that Motorola released the 68010 which runs at 12Mhz and Intel
took until 1984 to release a 10Mhz 80286 chip. This was WAY ahead of its
time. In our discussions, Bill clarified this amazing story, by explaining
that the chips he had running over 10Mhz were actually early manufacturing
errors to be discarded as trash. For the fun of it, he played with these
flawed units just to see if he could get them to work. When I asked him
why he did not present these as notable engineering successes to industry
or to the Guinness Book of Records, he said “I was worried about eating…
not making records.”
Motorola may or may not have had a solid legal
case but they definitely had something that MOS did not, money. It did not
take long for MOS to kill the 6501. The law suite dragged on a for a few
years and MOS eventually settled the claims with a $200,000 payment to
Motorola.
The Calculator Wars
1973 COMMODORE MINI-SLIDE
RULE CALCULATOR – From a December 1973 advertisement in Playboy… um uh
ummmm yaaa well, this was the what I was looking for in the magazine… I
was shocked to find naked women… I suppose I will have to take one for the
team and scan all the old Playboys for Commodore materials. Its a tough
job but I will sacrifice myself for Commodore.
In 1975, Commodore
had a huge inventory of Texas Instruments based calculators when the
market began to collapse. Because Commodore sourced their TI chips from
MOS, MOS was in financial trouble. Then the unthinkable happened. Texas
Instruments started retailing their own brand of calculators at a prices
less than Commodore manufacturing cost. The November 1975 New Scientist
magazine reported “Commodore is struggling to survive. Two weeks ago the
firm reported its end of year results, which showed a $4.3 million loss on
sales which were up 12 per cent over the year to $55.9 million.”
COMMODORE PET 2001 WOOD
PROTOTYPE DISPLAY: The Text on the board says: first truly affordable
appliance computer to be manufactured first truly integrated personal
computer housing a computer logic board, a keyboard, and a screen in one
chassis first mass produced personal computer with operating system in ROM
designed by the worlds famous MOS 6502 microprocessor designed and
developed by MOS Technology Inc (A subsidiary of Commodore Business
Machines Inc) first model shown in Consumer Electronics Show Chicago
January 1977
Commodore’s founder and CEO, Jack Tramiel, convinced
Commodore’s Canadian financier, Irving Gould, that vertical integration
(owning all of the parts of production) was the only way Commodore could
survive. Soon after, the September 1976 edition of New Scientist noted
“Commodore, quoted at $60 million on the New York Stock Exchange, has
acquired 100 per cent of the equity of MOS Technology Inc of Pennsylvania
in exchange for a 9-4 per cent equity stake in Commodore. MOS Technology
is privately owned and valued at around $12 million.”
Commodore continued the KIM-1 and Jack Tramiel personally approved the
development and production of Chuck Peddles unified computer, the PET.
In an effort to start sales of the 6502, MOS staff ran a quick
tour of the US, dropping into see major manufacturing companies like Ford.
On the trip Chuck was told that two young guys working in their garage
wanted some help using the 6502. Steve Jobs and Steve Wozniak were working
on the first Apple and Chuck was happy to lend a hand even though he did
not design the 6500 line for computer use. “…Not in a million years… it
was supposed to go into industrial and consumer products.” Little did
Chuck know that the computer business would quickly become the mass market
consumer product he was targeting.
Over the
previous few years Chuck had met with hundreds of computer enthusiasts,
educational institutions and main frame corporate users. He learned that
“…what people wanted was a computer that looked like a terminal.” He
bought a little book on how to build your own television, written by the
legendary Adam Osborne and contracted out the construction of a wood case
to house the computer. Yes, that’s right, the prototype was made of wood!.
Using a motherboard based on the 6502 processor Chuck designed and built
the worlds first computer which would later be named the Commodore PET.
Explaining how experimental this was, Chuck said “The first time we turned
it on, the image was upside down… we got Adam’s book out to figure out how
to turn it…”
In a January 2006 interview, Chuck told a great story
about trying to sell the CPU’s to Atari. Atari was so worried about
industrial espionage that they sequestered their top engineers in a remote
facility on a dirt road, several hours from Los Angeles. Chuck packed
himself and his wife for brief holiday and made plans to stop in to see
this top secret Atari think tank on the way. Atari was working on three
new options for their new machine one of which called for a 6502, an IO
chip and a custom chip. Atari wanted him to produce the 6502 and the IO
chip for just $12! By this time, Chuck estimated that MOS’ production cost
on two of those chips was just $4 and so it was easy to agree; the Atari
400 and 800 were announced in December of 1978.
Engineer Exodus &
The Rise of 16Bit
Shortly before the Commodore take over, frustrated
6502 co-designer Bill Mensch left the company. It had been made clear that
“Commodore was going into Game Systems… stopping microprocessor
development” I was the head of Microprocessor development… (and) I saw the
writing on the wall”.
There was a desire among MOS engineers to
design a 16 bit version of the 6502 but Commodore’s management was
apathetic and would not fund the project. In the end Commodore never
produced a 16 bit 6500 but Bill Mensch who still retains ownership of the
WDC, did. The 65C816 is now in over 5 Billion (yes, that’s Billion with a
“B”) devices. That chip can run in both 8 and 16 bit modes (hence the 816
designation). From pacemakers in your chest, to dashboard controllers in
your car, to the famous Super-Nintendo, Bill’s low voltage, highly tested
6500 version is ubiquitous.
The original 8 bit
6502 and many MOS derivatives are legendary. It was put into everything
from the Apple I and II, to the VIC and C64, to the original Nintendo Game
System. The 6502 was also used in many of the original arcade video games
like Defender, Battlezone, and Asteroids. Bill Mensch’s Western Design
Center holds most of the 6500 related patents and still are responsible
for their production.
Commodore’s CEO, Jack Tramiel started to
“play” with the lives of some of the MOS key staff shortly after the take
over and many of the key players left the company. Most notably, Chuck
Peddle left and took the equivalent to the Chief Technology Officer
position with Apple, before returning to MOS a few months later.
Commodore set up the Moore Park Research Center in California for Chuck
and other important West Coast based engineers. Just a few years later, in
a moment of anger, Jack ordered it closed. Most of the MOS Engineers,
including Chuck Peddle, refused to move to Commodore’s Pennsylvania
headquarters. By the mid 1980’s Commodore had only about a dozen certified
engineers still working for them world wide, most of them at MOS.
The Slow Decay of the 80’s
Even though it had renamed MOS
to “Commodore Semiconductor Group” (CSG) shortly after the acquisition,
all MOS production used the old MOS logo until 1989.
Using MOS’
engineers and facilities, Commodore was able to produce prototype chips in
days rather than months and at practically no cost. Other companies, like
Atari, Apple, and Osborne would have to spend tens of thousands of dollars
and wait weeks or months for new chips. MOS was critical to Commodore’s
fast paced style of business in the 1970’s and early 80’s. Commodore
engineers could get a chip produced without so much as formal paperwork.
When Bill Herd (the lead Plus/4 and C128 engineer) quit and went to work
for a small chip design shop, he was shocked that he could not simply
order up new prototype chips.
In addition
to providing millions of chips to the electronics industry and computer
manufacturers, MOS / CSG produced nearly all of the chips used in
Commodore computers and floppy drives. The notable exceptions to that
statement were the Motorola 68000 used in the Amiga line and all memory.
Chuck explained that “…MOS tried to produce memory but just weren’t any
good at it”.
In the early 1980’s Commodore hired dozens of notable
engineers including Dave Haynie and Bill Herd to work in their in house
“fab” but Commodore still refused to make significant investments in
research and development. Other similar sized companies would have
hundreds of engineers.
Under Jack Tramiel R&D was a secondary
concern. After Jack left in 1984 Commodore’s money man, Irving Gould,
dramatically reduced they already spartan R&D budgets in an effort to
increase short term profits. Commodore could have easily licenced Bill
Mensch’s 16 bit version 6502 and they did consider it for the Amiga but in
the end managements desire to squeeze ever more profit out of existing
technology left MOS / CSG to languish.
Commodore 65 DX Prototype – The C65 was never released and officially did
not exist according to Commodore before it went bankrupt in 1994
By
the mid 1980’s MOS’ best days were history and they were fully integrated
into the Commodore structure. With a few very important exceptions MOS
staff seemed content to manufacture what they had always manufactured. By
1985 they were definitely looking to the past and had forgotten that
theirs was a fast paced, dynamic industry.commodore_c65_csg_4510r5
Even when Commodore began its struggle for survival their management team
just did not see that they were living in an R&D business. In a near last
ditch effort to save the company in 1990, Commodore started work on a
secret project called the Commodore 65 and had MOS / CSG develop a new
chip called the CSG 4510 which was little more than a slightly enhanced
8bit 6502 with two integrated 6526 I/O adapters.
Biohazard
During the 1970’s Allen-Bradley’s MOS had installed a large underground
cement tank to store the extremely hazardous waist chemicals produced
during the chip manufacturing process. By the early 1980’s this tank had
cracked and leached chemicals into the ground water. Most of the
surrounding residential neighborhood used piped city water but several
used well water.
The problem was initially ‘hushed up’ and only a
small number of CSG managers were aware of the problem. In 1981 Commodore
excavated some of the contaminated soil and in 1984 “household carbon
units were installed at residences where at least 1 part per billion of
VOC was detected”. By 1989 the US Environmental Protection Agency has
started a serious investigation into the the problem.
An
Unceremonious End
Commodore produced hundreds of
millions of chips in the MOS Norristown fab but in 1992 facing serious
financial problems Commodore put the troubled facility into Bankruptcy
protection. Oddly, even though CSG was bankrupt Commodore maintained the
equipment in an effort to keep it functional.
gmt-microelectronics-mos-csgImmediately after the 1994 bankruptcy of
Commodore International the plant was sold to a group of its former
managers for $5.3 million which included $1 million in expenses for things
like EPA liens. It was renamed GMT Microelectronics and at the height of
that organizations success in 1999 its 180 employees produced and sold $21
million in product. Two years later the EPA would force the famous
Norristown fab to close and GMT’s assets were liquidated.
In
November of 2005 an EPA study shows the site had been clean for five years
and as you can see from the satellite shot on the left, Google Earth shows
that the building still exists today.
In the simplest of terms, MOS
was Commodore and Commodore was MOS. MOS was a precursor to Commodore and
it was a bell-weather foretelling Commodores future. Together they soared
and together they crashed.
Technical Notes
The 6500 line just
keeps getting bigger under Bill Mensch. He plans to start production of a
new line of 6500’s using RISC (Reduced Instruction Set) called “Terbium”
(65th element in the periodic table) in early 2007. It will be competition
for the ARM7 32Bit RISC processor.
COMMODORE
CHESSMATE: Console game from late 1970’s, Germany’s Doc Phura explains,
You have to use a real chess board to play the game. You enter the move
you have chosen. The MOS6504-driven Chessmate answers by displaying its
move by means of chess notation codes.
In September of 1983 Jim
Butterfield wrote a nice summary of the 6503, 6504, 6505, 6506, 6507,
6509, 6510 (used in the Commodore 64), 6512, 6513, 6514, and 6515 chips
for Compute magazine. You can read his write-up clicking HERE.
The
Technical Manual for the 6500 and 6502 are available on our Manuals page
or by clicking HERE. The Technical Manual for MOS / CSG’s 6509, is
available in the same location or by clicking HERE .
There are
KIM-1 schematics, manuals and history on our KIM1 page.
Click HERE
for schematics and instructions to Build your own MOS / CSG KIM1 with
components you can still find in 2003.
A complete Commodore time
line is available on our site HERE.
MOS also manufactured the
Commodore ChessMate game.
Click HERE for the huge list of 6502
based computers.
MOS / CSG Timeline
August 1974
Eight
Motorola employees including Bill Mensch and Chuck Peddle start MOS
Technologies Inc.
June 1975
mos-technologies-inc building
signMOS Technology announces the MC6501 at US$20 and soon after the MC6502
at US$25. This was truly breakthrough pricing; the Intel 8080 costs about
US$150
MOS was able to produce the chips at this low price because
they increased their yield dramatically by reducing flaws in their chips
‘masks’ before starting production. Click HERE for details.
Motorola sues MOS Technology over the similarity of the 6501 and 6502
processors to the 6800. In an out-of-court settlement, MOS Technology
withdraws the 6501 from the market.
About the same time MITS ships
one of the first PCs, the Altair 8800 with one kilobyte (KB) of memory, as
a $397 mail-order kit. The machine and company have nothing to do with MOS
but it gives you an idea of what was happening at that time.
The
6502 was not backward compatible with its Motorola 6800 inspiration but it
was pin compatible and ran up to 4 times faster.
Summer 1976
MOS
Technology Inc. announces the KIM-1 Microcomputer System, with 1-MHz 6502
CPU, 1KB RAM, 2KB ROM monitor, 23-key keypad, LED readout, cassette and
serial interfaces, for US$245.
It was called a “Computer Trainer”
but was largely designed as a demo machine for the MOS 6502 processor.
Chuck Peddle developed the PET concept and tried to sell it to Apple’s
Steve Jobs but Jobs did not offer enough money for it
Bill Mensch
leaves MOS for consulting work in Arizona.
September 1976
Commodore announces it is buying MOS Technologies for US$60 Million so
that it can become an almost completely self contained company. Texas
Instruments had provided processors for many Commodore calculators in the
years previous and as Commodore made the business more successful, TI saw
the opportunity. TI stopped selling chips to Commodore and started selling
Calculators directly. Commodores supplier became Commodore’s competitor
and Jack Tramiel vowed this would never happen again. Vertical Integration
was the answer.
Click HERE to skip to the end of this page for a
nice 1976 write up on Commodores Purchase of MOS
MOS / CSG designs and
produces chip after chip feeding 100% of Commodore product like the VIC20,
Commodore 64, Commodore 128, Plus/4 and many more
1977
MOS
co-founder Bill Mensch starts the WDC Western Design Center and produce
proprietary versions of the 6502. About 2 BILLION 65C02 processors have
been integrated into everything from cell phones to kitchen stoves since
the 25 years since. (Unrelated to MOS / CSG)
1978
Chuck Peddle
leaves Commodore to work for Apple Computer and within months returns back
to Commodore.
Commodore stops producing calculators
1983 – July
15th
Nintendo releases 6502 based ultra low cost Famicom Computer in
Japan for just US$65. In the US this device would be sold as the massively
successful Nintendo Game System and would ship with the original Donkey
Kong.
1984
Commodore Germany hires a film crew to produce a 30
minute promotional video explaining how MOS makes chips. They explain the
process from silicon to final processors and provide a tour of the
Norristown Pensylvania Chip Fab
1985 – 1994
Commodore starts to produce machines with Motorola chips (like the Amiga
1000, 500, CDTV…) but MOS / CSG is still a critical part of the Commodore
machine
1989
EPA adds CSG’s Norristown Pennsylvania fabrication
facility to its list of hazardous waste sites needing cleanup after
discovering extensive contamination from leaking underground storage tanks
at the facility
http://www.antiquetech.com/companies/GMT%20Microelectronics%20Corporation.htm
1992
Under financial duress, Commodore closes its CSG
Norristown, Pennsylvania fab
Although the Commodore Semiconductor
Group declared bankruptcy, the company continued to maintain the equipment
and systems at the plant, ensuring that the facility would not fall into
disrepair and would remain attractive to a potential buyer.
http://www.antiquetech.com/companies/GMT%20Microelectronics%20Corporation.htm
April 29th 1994
Commodore goes bankrupt and the MOS / CSG
remnants goes up for sale.
CSG is sold for US$4.3 Million to GMT Microelectronics
Corporation (Great Mixed-signal Technologies).
December 1994
US
Environmental Protection Agency opens discussions with GMT about the plant
2001
GMT experienced financial difficulties and the EPA shutdown
GMT operations in 2001. GMT ceased operations and its semiconductor assets
were liquidated.
http://www.antiquetech.com/companies/GMT%20Microelectronics%20Corporation.htm
2002
“The EPA is overseeing the cleanup of the Commodore
Semiconductor Group site. Construction of the groundwater extraction and
treatment system began in the fall of 1999. In February 2000, pipelines
and underground wiring were installed, pumps were installed at each of the
extraction wells, and the treatment building was constructed. The
treatment process equipment was installed in May 2000. Preliminary
start-up and testing of the system began in August 2000. The system
started operations in September 2000 to remove volatile organic compounds
(VOCs), the chemical components of solvents and degreasers, from the
groundwater using several filtration and evaporation techniques. The
treated groundwater is discharged to the Audubon Water Company’s
distribution system for public water. see:
http://epa.gov/reg3hwmd/npl/PAD093730174.htm and
http://epa.gov/reg3hwmd/super/PA/commodore/pad.htm
2013
See the pictures below showing that exterior of the building
is in superficially good condition but the inside has been fully gutted.
There are not even offices on the second floor and the basement has a
substantial amount of water (no-doubt very contaminated).
Much of
this list is taken from
http://www.islandnet.com/~kpolsson/comphist/comp1975.htm
CLICK HERE TO RETURN TO OUR HOME PAGE AND INDEX