Why are we still waiting?
by Alan Zisman
(c) 1997. First
published in Canadian Computer Wholesaler, July 1997
There?s a classic 1960s soul tune by Curtis
Mayfield was called
?A Change is Gonna Come?. When I started this series of columns
last
fall, that could have been the theme song. We?ve looked at a series of
new technologies, promised for PCs for 1997?DVD disks, Universal Serial
Bus, MMX processors and more.
Some of these promised technologies have arrived,
either as a flood
or a trickle. MMX processors, for example, have pretty much replaced
earlier
models? even if there?s still hardly any software that actually makes
use
of its capabilities. DVD drives are in the stores?at least the home
entertainment
versions. Presumably, the computer versions will be here ?any day now?.
As predicted, EDO memory is being replaced with faster SDRAM.
But where, for example, is USB?Universal Serial Bus?
The computer I purchased last Winter has pins on the
motherboard to
attach a USB adapter, but I haven?t done that?in fact, I?m not sure
where
I could purchase the parts. Why bother? There aren?t any readily
available
devices to plug into it.
And yet, USB devices were seemingly everywhere at the
Fall 1996 Comdex
show. What?s the holdup? What?s slowing up wide adoption of USB, DVD,
and
many of the other technological advances promised for this year?
Consumers, vendors, wholesalers, and manufacturers are
all stuck in
a classic ?chicken and egg? syndrome.
Vendors cite a perceived lack of consumer demand,
while consumers are
often unaware of promised products that aren?t yet on the shelf. Of
course,
it doesn?t always work that way? when Intel?s widely-hyped MMX
processors
didn?t make it in time for Christmas purchases last Winter, many
consumers
chose to wait? there was a big jump in sales of home PCs early in the
new
year.
At the heart of the problem, however, is the lack of
operating system
support. When so-called Class Drivers are built into the operating
system,
as with printers and modems, it?s simple for hardware manufacturers to
provide support for new models and new technologies. When these are
missing,
however, each new model requires a new driver, created from ground-up.
Microsoft, for example, promised to support USB in
Windows 95 in 1996.
It barely made that target date, shipping OEM Service Release 2.1 in
December.
Users of earlier versions, however, are out of luck. And since users of
earlier versions of Windows 95 can?t upgrade to the B version without
purchasing
a new computer, there hasn?t been much call for it. In fact, few PC
makers
include ver 2.1, perhaps fearing that it will prove incompatible with
the
next widespread Windows update.
This limits the potential market, and provides little
incentive for
hardware manufacturers to rush USB devices to market, which gives
little
reason for computer manufacturers to spend the couple of dollars to
provide
USB ports on their systems. (Intel was also several months late
providing
USB controllers to device makers?adding to the delay).
So, even though it was announced back in 1995,
wide-spread support for
USB may have to wait for next year? if not longer. The
higher-performance
FireWire bus may take before it becomes a common feature?. Device
makers
only began to receive long-promised support code from Microsoft in
April.
Let?s look at the status of some of the promised new
technologies:
? Plug and Play was promised in 1994 by Intel. It is
now supported by
most motherboards and BIOSs, and is supported by Windows 95, but not in
the current version of NT. For too many users too much of the time, it
remains more like ?Plug and Pray?. New specifications being promoted by
Microsoft and Intel as PC98 will encourage manufacturers to drop
support
for the 1984-era ISA bus? until that happens, and device manufacturers
fully support it, reliable Plug and Play will remain more vision than
reality.
? USB, as we?ve seen, may have to wait until it is
fully supported by
Microsoft in its next generation of operating systems, NT 5.0 and the
next
version of Win95, currently code-named Memphis? release dates pushed
past
1997 to? who knows?
? FireWire has been slowed down by a desire to push
the lower-cost USB
first, and by the lack of even preliminary code from Microsoft. Apple
has
announced support in future Macs, which may help get it out the door?
and
if it catches on as a high-end Mac standard, there may be more pressure
to get it on PCs. Don?t hold your breath!
? Cardbus has been promised as a high-end replacement
for PC-Cards for
notebooks? it too has been slow to catch on, though it?s available on a
few high-end models. In part, it?s been delayed because there hasn?t
been
much perceived need? it?s big advantage has been to provide support for
Fast Ethernet networking, which is not yet a mass-market item. As well,
it?s been hampered by a lack of general operating system support, as
was
the original PC-Card specification, prior to Windows 95.
PCs are changing, gradually evolving to models that
are both more powerful
and easier to setup and use. As we?ve seen, more powerful processors
continue
to be produced, and often find wide market acceptance very quickly?
consumers
easily understand that a faster processor is preferable, and have
tended
to avoid buying models released with less-powerful CPUs, even at
bargain-basement
price points.
But the heritage ISA bus has proven hard to kill, even
though it?s at
the root of continuing problems with Plug and Play. Just at look at the
failure of MicroChannel, EISA, and VL-Bus.
And until the next generation of operating system
releases offers full
support for the crop of hardware advances promised back in 1995 and
1996,
there will be only minimal support from device manufacturers or
computer
makers.
In the song, it was ?A long time coming??. But as
Curtis Mayfield wrote,
?But I know, a change is gonna come?.
(On a more upbeat note, next month, the continuing
saga of faster CPUs).