Keeping in mind that I didn't start working for MOS technology (and therefore Commodore) until 1979, which was after the PET was already in production, I believe the following information is correct:
Commodore started out as a typewriter company in Canada and also supplied mechanical adding machines. They were one of the first companies to realize the enormous sales potential of the electronic calculator and got into the calculator business very early. I believe they were one of the first "consumer" calculator companies, producing low cost calculators (as opposed to HP which was very high end and TI, which was cheaper, but still not at Commodore's level).
MOS Technology was started by a number of ex-Motorola engineers who had worked on the 6800 microprocessor. They developed the 6502 as a cheaper, faster, simpler alternative (and were ultimately sued by Motorola for their efforts). MOS Technology was also a design house and did contract work for others, including designing and producing calculator chips for Commodore.
The KIM-1 was designed (actually by a friend of Albert's) by MOS Technology as a sales and development tool for the 6502. They figured they would sell a few to developers and tap into the hobbyist market. Actually, many companies started buying KIMs not as a development tool, but for use as a controller in end products, which MOS hadn't anticipated. Total sales were, I believe, over 30,000 units and the KIM-1 actually became a profit center. While I was there, there were a number of attempts to discontinue it because it was technically obsolete, but there were so many people who wanted to buy them that it stayed in production, even after Commodore had bought MOS and was producing the PET.
Commodore acquired MOS through less than ethical means that would best be described as "kick 'em when they're down". Commodore pretty much intended to use MOS as a captive semiconductor house for their calculators. Chuck Peddle, one of the original 6502 engineers, came up with the idea for the PET (Personal Electronic Transactor--obviously the acronym came first) model 2001. A self-contained "consumer" computer which came fully assembled, integrating a keyboard (little calculator-style keys), monochrome monitor, mass storage (cassette) and BASIC. The idea was a person could turn it on and use it out of the box without having to build it, debug it, buy accessories, hook them together, etc. There was a great deal of skepticism regarding the sales potential and far less resources were put into the PET than for the typical Commodore calculator (in other words, it wasn't like everybody was saying "yeah, this is just what the world needs" and everybody was excited about it. Mostly it was just the visionary hobbyists at Commodore who were making it happen).
The PET and the Radio Shack TRS-80 were introduced at pretty much the same time, not long after the introduction of the original Apple II. Both the TRS-80 and PET were more complete packages than the Apple II (which either required you to buy a second-source video monitor or kludge up your own RF modulator to hook to a TV set, plus you had to supply your own cassette recorder and cables). In that respect, both products were more of a consumer product or "appliance" than the Apple, but the open architecture and machine-language development tools made the Apple II the choice for most hackers, which still comprised the majority of the market at the time.
MOS continued as a merchant semiconductor house, supplying processors, memory and I/O to a wide variety of people (particulary ROMs for ATARI for use in their video games). As custom chips became more and more important to Commodore's computers (such as the VIC-20 and C-64), there was less and less capacity for other customers, although they never went completely out of the merchant business. MOS only produced semiconductors, but they may have done some watch and calculator manufacturing there at some point. The KIM-1 was subcontracted and built out of house. In the heyday of the C-64, Commodore did move all of it's production from Santa Clara, CA to West Chester, PA, but also had factories in Japan and Germany.
I have one of the original prototype KIM-1 computers, which didn't have a cassette interface or any other means of program storage. I rescued it from the trash at MOS!
Regarding misinformation:
>PET 2001 was the original model, and the prototype was in a plastic case, rumored to have been designed by Porsche(!).
Actually, the original PET case was bent sheet metal. Commodore would never have paid for plastic tooling on such an "iffy" product. The plastic cases (actually structural foam) didn't come until years later for the PET (later CBM) family. Of course the VIC-20 and C-64 had proper injection-molded plastic cases befitting mass-produced products.
I really wish the "Porsche" rumor would go away, it pops up so many times in so many different ways! The only involvement with Porsche occurred around 1981 when Commodore Germany sent us some styling sketches that Porsche had made of an imaginary Commodore computer. The design pretty much looked like a racing helmet and would have been prohibitively expensive to produce. The sketches were never more than entertainment and never went anywhere.
-------------------------------------
From: andy@shell.portal.com (Andy Finkel)
In article <48g32u$s2u@netaxs.com>, Michael Hyman <mikeh@netaxs.com> wrote:
>The following was written by a friend and coworker of mine who was also the designer of the Commodore/MOS 6581 (SID) chip. I showed him this message thread and asked if he had any comments. Here they are:
A couple minor corrections (I was there too :-) )
>Keeping in mind that I didn't start working for MOS technology (and therefore Commodore) until 1979, which was after the PET was already in production, I believe the following information is correct:
>Regarding misinformation:
> >>PET 2001 was the original model, and the prototype was in a plastic case, rumored to have been designed by Porsche(!).
> >Actually, the original PET case was bent sheet metal. Commodore would >never have paid for plastic tooling on such an "iffy" product. The plastic cases (actually structural foam) didn't come until years later for the PET (later CBM) family. Of course the VIC-20 and C-64 had proper>injection-molded plastic cases befitting mass-produced products.
Actually, there was a Pet prototype carved out of wood; I rescued it from the Santa Clara warehouse when it was closed down, and brought it back east. I kept it in my office until a C= president grabbed it for his office. I believe during the Commodore shutdown it got donated to the Smithsonian.
> >I really wish the "Porsche" rumor would go away, it pops up so many times in so many different ways! The only involvement with Porsche occurred around 1981 when Commodore Germany sent us some styling sketches that Porsche had made of an imaginary Commodore computer. The design pretty much looked like a racing helmet and would have been prohibitively expensive to produce. The sketches were never more than entertainment and >never went anywhere.
There was one mock-up made by Porsche, of the design with the wide, thin band holding the monitor. I remember unpacking it at the Commodore offices after a show.
--------------------------------
From: mikeh@netaxs.com (Michael Hyman)
I have nothing to add about the first Commodore computers, but I can give a little trivial information about MOS Technology. I worked for a company located across the street from MOS for 4 years in the 1980's. (This was in Valley Forge, PA - about a mile from the national park.) The MOS facility had quite a reputation for environmental and hazardous chemical problems. From what I understand, most of the ground water in the area is (and has been for many years) contaminated with chemicals and solvents dumped by MOS into the ground. To this day, there are airators (sp?) scattered around the corporate center - installed specifically to help clean up the ground water problems which MOS created. The local fire department has orders NOT to enter the MOS building in case of fire. (MOS is the only building in the area with this restriction.) I can recall at least one publicized incident of MOS employees passing out en-masse after a ventilation problem there.
Our own company's parking lot was periodically roped off, and large drilling equipment brought in to test the underlying ground for chemical contamination and ground water monitoring.
I've also heard stories of inexplicable car paint problems of MOS employees' cars parked in their parking lot. (But there may be other explanations for this... :)
> Actually, as I remember it (no, I wasn't working there), the lawsuit was about the 6501, which was much closer to the 6800 in pinout (and perhaps, in internal architecture as well). The 6502 was the result of modifications to the design to get out from under the lawsuit.
My former MOS/Commodore friend replies:
This is essentially correct, although the 6502/3/4/5 were designed at the same time as the 6501, not as a re-design. MOS wanted a 6800 hardware compatible part, along with a line of "improved" parts. The 6501 was drop-in pin compatible with the 6800. The Motorola lawsuit forced the 6501 to be dropped. The 6502 is functionally identical to the 6501, but had an internal clock generator and was not pin compatible with the 6800 (a number of 6800-specific bus-control signals were eliminated).
> Lovely collector's item. Lots of wire-wrap jumpers soldered on for changes, or a fairly clean board?
The board is completely clean and works fine. Rather than locating the 1 Kbyte RAM on the board at zero page, the RAM in the 6530 chip was located there and the 1 Kbyte RAM was located at $8000. This meant that the stack and zero page variables were limited to 64 bytes! Fortunately this was changed before the KIM-1 went into production.
Someone else (???) wrote:
> I thought they started out in office furniture. I have a Commodore Desk and File cabinet at work. Anybody know what the proper order is. Was it furniture then the typewriters and then computers and then bankruptcy or were the typewriters first.
I don't know whether Commodore was in the office furniture business before the typewriter business, but Commodore Canada was in the furniture business while I was there. Most of our office furniture came from them. That was another reason that the PET case was bent metal, they made it themselves.
----------------------------
From: brain@mail.msen.com (Jim Brain)
In article <48g32u$s2u@netaxs.com>, mikeh@netaxs.com (Michael Hyman) wrote:
> The following was written by a friend and coworker of mine who was also the designer of the Commodore/MOS 6581 (SID) chip. I showed him this message thread and asked if he had any comments. Here they are:
> MOS Technology was started by a number of ex-Motorola engineers who had worked on the 6800 microprocessor. They developed the 6502 as a cheaper, faster, simpler alternative (and were ultimately sued by Motorola for their efforts). MOS Technology was also a design house and did contract > work for others, including designing and producing calculator chips for Commodore.
William Mensch, Jr (One of the Motorola employees who left, was hired by MOS, implying that the company was in existencelong before. The move came in 1975 or 1976 (I don't have notes handy), and MOS must have been making calc chips for some time. The reason for the move from Motorola to MOS is complex, but not overly so. The 6800 was based on the DEC PDP-8 architecture, and the PDP-8 was old by 1975, but well used. The Designers wanted to design around the newer PDP-11 architecture, but Mtorola said no. So, after the 6800 was done, MOS wooed the designers away with the promise of being allowed to develop a PDP-11 based CPU, the 6502. The suit was really over the 6501, which was pin compatible with the 6800, but there were other points.
---------------------------------
From: brain@mail.msen.com (Jim Brain)
> In article <48g32u$s2u@netaxs.com>, Michael Hyman <mikeh@netaxs.com> wrote:
> Actually, as I rember it (no, I wasn't working there), the lawsuit was about the 6501, which was much closer to the 6800 in pinout (and perhaps, in internal architecture as well). The 6502 was the result of modifications to the design to get out from under the lawsuit.
Bill Mensch states that the two chips were developed at the same time, and offered for sale starting in the September 8th issue of Electronic Engineering Times, Pages 34-35. The 6501 was $20.00, and the 6502 was $25.00
