doug@terak.UUCP (Doug Pardee) (11/20/85)
> IBM chose Intel for business, not technical, reasons.
Hold on a moment. There seems to be a common belief that International
Business Machines management decided one day to build a Personal
Computer, one which would take over the small computer world, and that
they chose the 8088 as the CPU for that machine.
It ain't so. A version of history which is closer to the truth:
A small start-up company in Boca Raton, Florida decided that they wanted
to get into the booming microcomputer market. They designed a machine
based on an 8088. At that time, I doubt that they even *considered* a
68000 because
a) the 68000 was a 16-bit CPU, meaning that they would need a
16-bit bus and 16-bit peripherals (if you don't think that this
was important to them, then why didn't they use the 8086?);
b) the 68000 cost many times more than the 8088;
c) the Motorola folks believed (at that time) that 68000's shouldn't
be used for measly little microcomputers, but rather for $50,000+
systems, and wouldn't supply samples or support to small outfits;
d) the only compilers and operating systems available for the 68000
cost more *per copy* than the whole machine was going to sell for,
and given Motorola's stance this wasn't likely to change soon.
At about the same time, IBM management was deciding that they needed to
have a new "bottom end" computer. Nothing fancy, they weren't really
going to sell many -- it was just for image (the same way the 370/195
was for image). Customers like to see a "full spectrum" of computers
available.
IBM bought the Boca Raton start-up, and started selling the "PC". This
was *not* what we think of as a PC today. It was a 16K, cassette-based
computer.
But contrary to IBM's plan to have a low-volume 16K cassette-based
computer, a whole lot of businessmen bought the thing and found out that
they could add disk drives and a lot of memory, and it could do some
useful things. Software writers jumped in the fray, and PC sales took
off. IBM deserves great credit in that they did not try to limit the
PC's sales in order to protect their other products such as the
Displaywriter. (If you think that this would be perfectly obvious,
consider that IBM later purposely made the PCjr unexpandable and with
a useless keyboard in order to protect the PC, and that Apple has done
much to try to kill off the Apple ][ which they think is cutting into
their ability to sell Macs).
IBM never planned to build "the PC that took over the world." And they
didn't pick the 8088. Don't blame/thank IBM for the commercial success
of the Intel-based architecture; Motorola did it to themselves with
their snooty attitude about the "proper" applications for the 68000.
--
Doug Pardee -- CalComp -- {hardy,savax,seismo,decvax,ihnp4}!terak!dougjosh@polaris.UUCP (Josh Knight) (11/25/85)
In article <918@terak.UUCP>, Doug Pardee (doug@terak.UUCP) writes > > IBM chose Intel for business, not technical, reasons. > > Hold on a moment. There seems to be a common belief that... > ... > A small start-up company in Boca Raton, Florida decided... > ... > At about the same time, IBM management was deciding that they needed to > have a new "bottom end" computer. ... > ... > IBM bought the Boca Raton start-up, and started selling the "PC". This > was *not* what we think of as a PC today. It was a 16K, cassette-based > computer. > > But contrary to IBM's plan to have a low-volume 16K cassette-based > computer, a whole lot of businessmen bought the thing... > ... > IBM never planned to build "the PC that took over the world." And they > didn't pick the 8088. Don't blame/thank IBM for ... > ... In article <397@ukc.UUCP>, Michael Fischer (mf1@ukc.UUCP) writes: > If one remembers, IBM didn't choose anything. When the IBM micro > was announced, IBM officals were falling all over themselves to deny that > they would support the machine at all. > > I'm sure I'll be corrected if I have my micro-mythology wrong, but I recall > that the IBM-PC was created by a friend of the son of the District Manager > of IBM in Boca Roca Florida, who managed to get IBM to reluctantly put > their name on it as an experiment. Even if this "origin myth" is wrong, > the IBM-PC was an outside product sold to IBM. ... > ... Both of these "histories" of the IBM PC are basically wrong. The development of the PC was internal to IBM, but done in an "Independent Business Unit". This is supposed to be as close to an independent company as possible, given that IBM provides all the financing and staff. The PC development effort was headed up by a long time (joined in 1959) IBMer, Philip D. Estridge. He was was killedin an airline crash last August (along with his wife and several other IBMers, the crash made the business section of the New York times), after he had left what had become IBM's Entry Systems Division to become IBM vice president for manufacturing. ESD (sometimes jokingly called Estridge Systems Division before his untimely death) is now a significant part of IBM's business, and now includes things that weren't part of the original PC effort. I don't know whether there is an "official" history of the development of the PC, but I'm pretty sure that the above "histories" bear no particular relation to reality. In article <790@umd5.UUCP> Ben Cranston (zben@umd5.UUCP) writes: > What makes you think technical considerations had *anything* to do with it? > Is our memory so short we forget the big cross-licensing agreement that IBM > and Intel signed ten years ago? Intel got access to all IBM's magnetic > bubble work (:-) and IBM got access to the 80xx design and masks. And I > think a hefty piece of Intel too. The acquisition of a significant fraction of Intel's outstanding stock (my memory is about 10% with an option to purchase more up to 30%, but I'm hardly sure about this) occurred substantially after the PC was established not, as suggested, "ten years ago". As usual, any opinions (expressed or implied) or errors are mine, not my employer's. -- Josh Knight, IBM T.J. Watson Research josh at YKTVMH on BITNET, josh.yktvmh@ibm-sj on CSnet, ...!philabs!polaris!josh
frodo@wcom.UUCP (James Scardelis) (11/27/85)
> A small start-up company in Boca Raton, Florida decided that they wanted > to get into the booming microcomputer market. They designed a machine > based on an 8088. At that time, I doubt that they even *considered* a > 68000 because [followed by a whole bunch of stuff] Well, that isn't the way I heard it...the way I heard it goes like this: IBM decided to get into the microcomputer market. As with mainframes, they decided to take the market over, based on what the IBM name means to corporate MIS people. IBM formed several semi-independent groups within itself that could appear to be start-up companies, each with the mission of making a personal computer by a given date. The group that won was the Boca Raton group, whose machine was based on the semi-16 bit 8088. The logic behind this choice was the fact that almost all software at the time ran on the 8 bit predecessor, the 8080, and could be converted at the assembly language source code level by translator programs to run on the new machine, ensuring a supply of application software shortly after the PC's introduction. The 8088 was chosen over the 8086 because of the 8 bit data bus, that among other things, made 64 K machines possible. The Intel 8088/8086 series was also an easier sell to IBM management because IBM was already experienced with the 8086 -- in the Displaywriter. The Boca Raton group became the Entry Systems Division, which continued to function somewhat independently of IBM corporate until recently, when the absorption of ESD into the IBM corporate structure was completed. -- Jim Scardelis, SA {vax135,ihnp4}!wcom!frodo #include <favorite disclaimer>