"PC99 System Design Guide", Microsoft, 1998, 0-7356-0518-1,
%C 1 Microsoft Way, Redmond, WA 98052-6399
%I Microsoft Press
%O U$49.99/C$71.99/UK#46.99 800-MSPRESS fax: 206-936-7329
%P 576 p.
%T "PC99 System Design Guide"
Come gather round children, all you not-quite-dry-behind-the-ears
young techies, and I will tell you tales of "IBM compatible," VAX
equivalents, ISA, AT, Microchannel, OSI, MPC, and many more. (Let's
not forget BOB.) I will tell you fairy tales of client/server, open
systems, interoperability, and plug and play. I will tell you of the
architectures of yesterday, whose bleached bones you can still find in
the sedimentary files of old standards documents.
No, you're probably correct. I don't think there will be any bidding
war for the movie rights.
Once more unto the standards committee dear friends, once more.
Microsoft has presented a new architecture. Like MPC (and its
children) before, this is loosely based on the computer you probably
want (as in, "the computer you want always costs $5,000") with a few
prods in the "right" direction.
(Is there any point to this type of exercise? I can sympathize with
the attempt to provide some guidance to beleaguered consumers who are
at the mercy of all-too-often-ignorant salespeople, but the plain fact
is that a} if a configuration supports useful and/or fun applications
people will buy it, regardless of standard, and b} any "buying"
standard tends to have a lifespan of about six months. I've seen very
few references to the PC99 standard, and many of those were by people
trying to show off detailed knowledge of the industry. In any case,
NT 5, one of the two operating system platforms that PC99 was intended
for, is now apparently never to be seen. Will PC99 last? But that
has nothing to do with the book itself.)
There are two chapters of general overview, and then outlines of the
requirements for PC99 basic, workstation, entertainment, and mobile
systems. Part three goes into bus specifications for USB (Universal
Serial Bus), IEEE 1394, PCI (peripheral component interconnect), ATA
and ATAPI (AT attachment and ATA packet interface), SCSI (small
computer systems interface), and PCMCIA (personal computer memory card
international association). Part four looks at specific systems, such
as I/O ports and devices, graphics, video, monitors, audio, storage,
modems, networks, printers, and digital images.
The presentation is oddly humourless. Or rather, since you don't
really expect standards documents to be a barrel of laughs, it takes
itself rather seriously. Early overviews of the architecture, for
example, keep referring to IEEE 1394. This is actually a bit of a
problem, since no details are given of the standard, and the tables
never bother to point out that a reasonable, if terse, description is
included in chapter eight. However, only one reference, buried deep
in the index, admits that IEEE 1394 is more widely known by a somewhat
less formal name--Firewire. Other "design requirements" assume an
admonitory tone with brain damaged hardware engineers: "3.6. All
expansion slots in the system are accessible for users to insert
Given that this is supposed to be a standard, a significant amount of
material is confusing, and even contradictory. For example, page five
states that only printers are allowed to use the old COM or LPT ports,
and even that is considered declasse. On the other hand, item 3.32
allows COM ports. Similar confusion surrounds ISA (Industry Standard
Architecture) slots, which are expressly forbidden in a number of
places, but tacitly allowed in others.
copyright Robert M. Slade, 1999 BKPC99SD.RVW 990403
====================== (quote inserted randomly by Pegasus Mailer)
rslade@... rslade@... slade@... p1@...
I'd explain it to you, but your brain would explode
eGroup home: http://www.eGroups.com/group/techbooks
- Simplifying group communications