17 Nov
2024
17 Nov
'24
2:15 a.m.
In order to have the "world beat a path" to a new microprocessor is has
the be sufficiently better than what is there to justify the time and
expense. Intel doesn't know how to architect a decent CPU. They just
keep kludging their previous successes.
From what I heard from some inside sources around the time that the PC
came out, IBM didn't pick the 8088 for it's speed or ease of use. Quite
the contrary, they chose it because it was slow and they could buy part
of Intel. They didn't want their $10,000 computer competing with their
bread and butter main frames.
A 68000 could address more memory than a basic IBM 360 (which could only
address up to 64K) and was faster as well. As a matter of fact when IBM
created the System 360 PC card that was an IBM 360 on a card that
plugged into an IBM Xt is used a re-microcoded 68000 to emulate the
System 360.
On 11/16/2024 7:49 PM, Wayne S via cctalk wrote:
Why did those processors not catch on?
It seems to me that hardware people had a “if we build it, they will come” mentality and
hoped other companies would adopt it and actually write software to make it useful.
Sent from my iPhone
> On Nov 16, 2024, at 17:38, Chuck Guzis via cctalk <cctalk(a)classiccmp.org>
wrote:
>
> On 11/16/24 16:24, Fred Cisin via cctalk wrote:
>
>> So, Intel went with the "quick fix" rather than the long-term good.
> Okay, I vass dere and know what we were being told by Intel marketing in
> the late 70s. The 8086 was not intended to be the eventual migration
> target for larger-scale applications. Similar claims can be made for
> the 80186--it was mostly intended for embedded applications.
>
> The thing that was supposed to be the architecture to hang one's hat on
> was the iAPX432. Intel's "Clean Slate" which was a horrible flop.
> Another "clean slate" was the i860; my i860 reference manual has a
> statement by BillG saying that MS intended to develop for that platform.
> It seems that every time that Intel tries to do development from a
> tabula rasa, they get burned. Witness Itanium/IA64.
>
> The thing that saved Intel's bacon on several occasions was their
> liberal licensing. Would we even have had the IBM 5150 if there weren't
> a pile of second sources for the 8088? My early 5150 had an AMD CPU in it.
>
> --Chuck
>
>