IBM bus was stalled; Intel product might share the same fate
Remember IBM Corp.'s blowout promotion of the Micro Channel Architecture bus back in
the mid-1980s? MCA was touted as revolutionary in its speed advantage over the old
Industry Standard Architecture bus and in automatic installation of accessory cards.
I bought one of the first IBM PS/2 Model 60 computers and was quite happy with it, but
I never bought another MCA-bus PC. There was no real advantage in acquiring proprietary
accessory boards. IBM's 10-pound, external 514-inch floppy drive wasn't a big selling
Everyone knew MCA was an IBM attempt to lock up the desktop market with a proprietary
bus. Although it worked pretty much as advertised, the bus just wasn't hot enough to woo
significant numbers of buyers away from the nonproprietary ISA that, interestingly enough,
IBM also had invented.
Today Intel Corp. is poised to try the same thing with a proprietary motherboard and
bus for the new Pentium II. The Dual Independent Bus (DIB) is supposed to offer some
performance improvement over the current PCI bus, but not much. DIB appears to be a move
to make a standard more profitable.
Again, consider AMD. The also-ran chip maker has drifted in Intel's wake for years,
failing to capture much market share with its marginal performance improvements. Now Intel
is looking back to see who might be gaining on it.
The microprocessor giant, unsuccessful in court attempts to block clone chips, has
decided to concentrate on the bus instead. AMD of Sunnyvale, Calif., and Cyrix Corp. of
Richardson, Texas, only managed to capture what they have of the chip market because their
chips fit the same bus architecture as Intel's.
For PC builders to re-engineer the entire bus to accommodate these bit players would be
out of the question economically.
So Intel does have a chance to rid itself of competitors. Down the road, this could
hurt PC buyers by removing the significant downward pressure the clone chips have exerted
on Intel's prices.
If AMD can bring out a really hot chip in the 200- to 300-MHz range by fall, it might
force Intel to return to a standard bus architecture.
The Pentium II is being marketed to high-end users, but experience shows that fast new
chips rapidly trickle downward. Software demands always have driven the need for speed and
probably always will, so the new architecture is something for all users to be concerned
about, not just those at the high end.
You might think Intel is too big to beat, but remember that IBM invented the PC and was
a powerful force back when it introduced MCA. Who's still specifying PS/2s today?
I'm not necessarily taking sides against Intel, but it does seem to me that AMD and, to
a lesser extent, Cyrix have held down chip prices to users' advantage-even users with
Intel inside. I hope the two companies can keep going.
Meanwhile, in the Macintosh world, Apple Computer Inc. chairman Gil Amelio has been
trying to convince skeptical software developers to support the Rhapsody operating
system-a very big deal for Apple though perhaps not so big for most users.
If leading developers fail to salute the new Mac OS, there really won't be much reason
to buy Apple computers anymore.
I wouldn't want to sink development money into yet another generation of Mac software
at this point in the company's decline, but it might take as long as six months to see
whether the big developers feel the same way.
Despite the hype about handwriting recognition improvements, I notice that the latest
MessagePad 2000 is advertised with a keyboard attached. I do like its looks, but I can't
tell you how well it works, because no one has ever responded to my requests for a review
The future of the Newton is questionable because Apple is spinning off its handheld
division as its own business unit.
John McCormick, a free-lance writer and computer consultant, has been working with
computers since the early 1960s. E-mail him at [email protected].