Anyone interested in this might enjoy reading the book Commodore, A Company on the Edge. It chronicles much of the history told here as well as the creation of the KIM-1, Commodore PET and Commodore's acquisition of MOS Technology.
Second the recommendation. As well as the other two in the series.
(Worth knowing that if you see "On The Edge: The Spectacular Rise and Fall of Commodore", that's the previous version which covers the same time period as all three new volumes in about 1/3 the page count. "A Company on the Edge" covers roughly the "8-bit era", though of course with Commodore there was a big overlap)
#1 is the "first edition". #2/#3/#4 covers exactly the same material, just expanding on it.
Hence my "warning", as there's not really any reason to read both #1 and #2/3/4 now that all are out. Especially as #1 is sold out most places and usually shows up on Amazon at extortionate prices for used copies.
Looking forward to The Early Years, but that's also more of an "extra" for those with a very particular interest in Commodore, as it's not covering their computer business but their pre-computer office furniture and calculator business.
One question which someone might be able to help with.
The 6502 was the third CPU the team designed. Second was the 6800. But the first was also made by Motorola but as a one-off for Olivetti I think. Does anyone have any details of this CPU?
Asking because it would be interesting to see the development of their thinking.
Check out Bill Mensch's Oral history or one of his presentations on YouTube about the 6502; I think he mentions the Olivetti project that he and Rod Orgill worked on.
I'm probably missing something, but it's incomprehensible/mind-boggling to me that a processor released in 1975 would be used in the Commodore 64 and similar computers in 1982 and even later. Imagine someone today releasing a computer with a 7-year old processor! Price trumps everything else, I guess?
And the 6502 was so primitive and unpleasant to use. I remember writing some 6502 assembly code back then. You have a total of 3 8-bit registers and no multiply instruction. And only 16-bit addressing because "64 K ought to be enough for anybody".
All this nostalgia about 1980s computers is completely lost on me, I'm glad those days are over.
CPU speeds were increasing rather slowly up until the mid 90's when suddenly we went from 33Mhz to 1Ghz and higher over the course of a few years it seemed.
RAM was super expensive in this time frame, and the 6502 was conceived as a cheap industrial controller. So it was never meant to be part of a system with oodles and oodles of memory for the time, such as the VAXes and PDPs of the time.
Surprised you're not complaining about the absence of floating point, but that didn't stop Bill Gates from implementing software floating-point on the 6502 BASICs he sold to Commodore, Atari Apple at the time.
x86-64 was released in 1999. Is it incomprehensible that you're probably buying a computer with an x86-64 processor in 2022? ;)
More seriously computers are being sold today with CPUs with the same performance as those available seven years ago and are perfectly useable.
I have to disagree with you though on the 6502 being unpleasant though. For a microprocessor with 3,500 transistors it very elegant. In the hands of a skilled programmer (like eg Sophie Wilson of BBC Basic fame) it could be quite powerful for the era.
I'm still amazed that Motorola didn't react to the extreme difference in pricing between the 6502 and 6800 quicker. I do wonder what they were being told about the early personal computer boards?
They didn't have the time machine they'd need to view this from our perspective. There were no personal computers. The boards were basically just evaluation boards (which were essentially Motorola's design). They were likely focused on TI and Intel. They probably regarded MOS as IP thieves to be dealt with using legal measures.
It would be interesting to hear from someone who was there.
They certainly reacted to the pricing difference to the 6800-pin-compatible 6501, both by dropping the 6800's price and suing MOS. The 6502 is essentially a 6501 modulo reorganization of the pins.
I believe you mean Ricoh. Also, a whole bunch of glue logic was put on chip as well for the 2A03 / 2A07 that otherwise would have been pricier as discrete parts on the board.
Edit: I see they have credited that site for some of the information they have here as well.