In the early 80s Andy Grove approached Apple and suggested that he
be licensed to mass produce the Macintosh chip so clone makers
could have access to them. Apple said no. Actually, Steve Jobs said
no. So Grove when to IBM and they licensed him to make PC chips.
IBM chose to use an existing Intel processor (8088) in the original IBM PC (model 5150). At some point later, Intel licensed IBM (not the reverse) the rights to manufacture various x86 designs. (The most important part of this license is a patent cross licensing deal, which has allowed IBM to manufacture various x86 implementations for other companies. AMD has similar access to patents, though it took a protracted legal battle to realize them.)
That's why clones are PCs rather than Macs.
The clone market for PCs resulted from IBM's details documentation of the design, IBM's inability to prevent other companies from cloning the design due to legal restrictions from their consent decree, and most importantly, that the operating system came from Microsoft and IBM did not have an exclusive on the OS. (As hardware companies are wont to do, they underestimated the import of software.)
Any company wanting to build a PC compatible machine had access to the information necessary to do so and could easily cut a deal with Microsoft for the OS. Throw in a couple serendipitous factors that forced 100% compatibility to be required (e.g. it was IBM after all, not some smaller company) and the widespread market arose fairly naturally.
Grove at that time thought the Mac architecture was superior.
Grove may (or may not) have liked the Mac OS better than MSDOS, but Intel was in intense competition with Motorola -- x86 vs. 68k. It wasn't until 1990 or so that Intel started leaving Motorola in the dust. Intel's interest would have been in convincing Apple to use x86, not in manufacturing 68k parts (which they would not have been allowed to do).
There was a brief flurry
of Mac clones when Motorola was making the chips...
This happened after the switch to PowerPC, thus it was both IBM and Motorola making the chips. Timeframe was 1994 to 1997 or so if I recall.
[...]
A friend of mine who is highly placed at Intel says both chip
architectures are obsolete and that the big winner will be whoever
jumps first to a new level of technology.
Lots of people say so, but they have been proved wrong time and time again. It isn't so much that x86 is a great instruction set as that it isn't so bad and there is so much money behind it that people work hard to make it go fast. There is no end of rationalization for creating new instruction sets, but history proves that it is seldom worthwhile.
For example, Intel has the IPF (nee IA64, aka Itanium) architecture which is an attempt at moving to a new technology. Close to 15 years and likely 10 billion dollars (collectively across the industry) after it was started, IPF is still not significantly ahead of Power/PowerPC or x86 and to many eyes it has proved to be much harder to implement in practice. It is basically a really expensive pig.
In fact, AMD basically proved Intel wrong on this very point with AMD64 vs. IPF. Intel has been forced to implement AMD64 and likely will be forced to abandon IPF within the next five to ten years. Intel themselves bet against x86 and lost.
x86 is here to stay. It aint perfect but it could suck worse. I'm fine with this as it makes for a very large market of cheap fast processors that all run the same instruction set. The Apple/Intel move, if it happens, is interesting as the same would be true for PowerPC. There's no strong technical reason to switch, so it has to be about a business decision.
-Z-
