Facts about the Win NT/PowerPC story

[Roz Ault - User Support Technology - ext 377 <AULT -at- FAXON -dot- COM>]

I don't want to get embroiled in an OS war, but I would like to
pass along some facts about the Win NT / Power PC story. (Incidentally,
I use both Mac and Windows; I prefer the Mac because I'm more productive
with it, but I have written plenty of documentation for DOS and Windows-
based systems as well.)

This report is from the Mac Daily Journal -- true, it's a Mac-oriented
publication, but it does get to the facts behind a lot of ill-informed
press reports (and doesn't shirk from slamming Apple when they deserve
it.)

I think what this episode really shows is that once again, Microsoft has
won the day by the power of public relations, and Apple once again
has come out looking like a loser because of not being able (or maybe
not even trying) to deliver a clear and cohesive public message.

(This is a copyrighted story, but the editor has given permission
to pass along extracts from the MDJ to limited mailing lists.)

--Roz Ault
  ault -at- faxon -dot- com
 ========================================================================
**Windows NT for PowerPC No More**

  The fun started on Friday, after Apple's press briefing (MDJ_
  1997.02.10), when Microsoft Corporation announced the phasing out
  of development work on Windows NT for PowerPC-based platforms.

  Many industry observers expected Windows NT was expected to be the
  "server platform from hell" across computing hardware. Versions
  for PowerPC, MIPS, Intel, Digital Equipment Corporation's Alpha
  RISC processor and others would create a strong undercurrent for
  an OS that was supposed to be so powerful, so robust that it would
  start knocking UNIX out of the water, and be the logical migration
  path from Windows 95 for most PC users who want more power.

  It hasn't turned out that way, by most accounts. Microsoft had
  already dropped support for MIPS microprocessors before this
  announcement, and there have never been versions of Windows NT for
  Sun workstations, an important server platform. In December, both
  IBM and Motorola, co-designers of the PowerPC chip, stopped
  funneling resources into Windows NT for PowerPC development
  because no one was buying the systems, despite having price-
  performance advantages virtually identical to Mac OS PowerPC-based
  systems at a MegaHertz-for-MegaHertz level.

  The word in the development community (though unsubstantiated) is
  that Microsoft produces Windows NT for Intel because it's
  profitable, and for DEC Alpha chips because Digital Equipment
  pretty much pays for the work involved. When IBM and Motorola
  stopped paying for Windows NT development, Microsoft stopped doing
  it. Part of the problem is that Microsoft has begun "transferring"
  Windows NT support to computer vendors, requiring (for example)
  Motorola to support its customers instead of Microsoft doing it.
  That's why some major companies, including Compaq, have removed
  bundled Windows NT configurations from most (if not all) of their
  product lines, leaving users to purchase the operating system on
  their own so the computer company doesn't have to deal with the
  high costs of NT technical support.

  Rumor (the companies involved do not confirm these stories) also
  has it that Microsoft is causing problems for vendors who make
  platforms for competing server-based operating systems -
  particularly Linux, the freeware BSD UNIX clone so popular with
  Internet service providers. A message on the developer mailing
  list "Semper.Fi" claims Microsoft is "asking" (quotes in original)
  that vendors not provide support or actively be involved with any
  efforts to move Linux to their hardware platforms. The adoption
  rates for NT are not what Microsoft or most analysts predicted,
  and dropping hardware platforms due to lack of sales isn't helping
  that any.

  The long-delayed release of PowerPC Computing Platform (PPCP)
  machines, which were to be able to run Mac OS, Windows NT and
  perhaps Sun's Solaris operating systems, have not helped
  Microsoft's position - it's difficult to write an alternate OS for
  Macintosh systems, as the MkLinux folks have shown through their
  slow but steady efforts to bring Linux to Power Macintosh systems
  (it now works with PCI-based systems; it didn't when first
  announced last May).

**What They're Saying** -- So this is all a setback for Windows
  NT, right? No, of course not; Microsoft is invincible. Instead,
  it's a setback for _Apple_, according to the press. Mind you,
  Apple has never shipped a Windows NT machine, despite repeated
  press inquiries for years about when they would do so, but
  suddenly the lack of Windows NT for future PowerPC, non-Mac OS
  machines is a huge "emotional setback" to Apple, say the pundits.

  The worst example is from the _San_Francisco_Chronicle_, co-
  written by former _MacWEEK_ writer Jon Swartz who certainly ought
  to know better. The article

<http://www.sfgate.com:80/cgi-bin/chronicle/article.cgi?file=MN2840.DTL&directory=/chronicle/archive/1997/02/08>
  from Saturday contains complete misrepresentations of the
  situation. Here are just a few examples, followed by reality:

* "Microsoft said it will not make its popular Windows NT software
  for Macintosh computers." Microsoft is phasing out development on
  Windows NT for PowerPC computers, not for _Macintosh_ computers
  which, currently, pretty much run only Mac OS (with some limited
  support for BeOS and MkLinux, both of which have to work around
  serious assumptions in Macintosh hardware). Read Microsoft's press
  release
  <http://www.microsoft.com/corpinfo/press/1997/feb97/powerpr.htm>
  for yourself if you don't believe it; the words "Apple" or
  "Macintosh" are never mentioned. There never has been a Windows NT
  for Macintosh.

* "The PowerPC chip has not caught on." Sales of PowerPC-based
  computers outsell all other RISC computers, _combined_, by about
  4:1, according to Apple. That combination includes the vaunted
  Digital Alpha processor that still has Windows NT support.

* "It is an emotional blow to Apple," said analyst Jim Turley.
  "This certainly won't help Apple in corporate sales." Apple is
  putting together Rhapsody as a powerful operating system that has
  unique advantages for in-house developers as well as others; why
  would they want to sabotage that now by apparently endorsing
  Windows NT? Jim Gable, Vice President of AppleSoft Marketing, said
  that the drop in NT support is an issue for those who needed
  PowerPC-based NT access, but the impact on Apple is minimal
  because it doesn't sell PowerPC-based Windows NT machines.

  The rest of the coverage doesn't get much better. A
  _San_Jose_Mercury_News_ report
  <http://cgi.sjmercury.com/business/apple/app020897.htm> by Mike
  Langberg speculates that IBM and Motorola might be dropping
  PowerPC support soon, because they don't have the resources Intel
  does and therefore can't "keep pace," conveniently forgetting that
  the PowerPC chip's capabilities are ahead of Intel's main
  processor line. Langberg describes Intel and PowerPC chips as
  "roughly equivalent performance at about the same price," compared
  to a goal of "running twice as fast as Intel chips at half the
  cost." For the record, the current price of a 200MHz Intel Pentium
  chip is US$528 in lots of 1000; the price of a 240MHz PowerPC 603e
  chip, which is faster by all benchmarks, was US$425 at
  _introduction_ and may be less expensive today. (The Pentium chip,
  when introduced, was over US$1000 each in lots of 1000.) Roughly
  equivalent performance, maybe. Price? No way.

  The Associated Press story would have us believe that "falling"
  sales of Mac OS computers (they're not falling; only Apple's sales
  are declining, not the Mac OS market in general [MDJ_
  1997.01.14]), which can't run Windows NT anyway, are the reason
  Microsoft is getting out. Newsbytes' story was factual but again
  implies that it's Microsoft walking away from customers, not
  customers ignoring Microsoft. The Reuters story is even better,
  being the closest to purely factual of the whole lot.

  The _Wall_Street_Journal_'s article, by G. Pascal Zachary, calls
  the PowerPC chip "fizzled" and, after interviewing a PowerPC
  marketer responsible for embedded systems sales, concludes that
  embedded systems are the PowerPC's only chances for broader
  success. Zachary accepts without question that the PowerPC is
  somehow inferior to Intel chips, using phrases like "The inability
  of Power PC [_sic_] to keep pace with advances in Windows NT" and
  "Another looming question about the Power PC is the depth of
  Apple's commitment to the chip family." Again, people are trying
  to turn Apple into Just Another Clone Maker because understanding
  something different is Way Too Hard.

  People believe this stuff, no matter how silly it seems to the
  informed. Soundview Financial analyst Scott Randall renewed his
  short-term and long-term buy ratings for Intel stock on Monday,
  partly because Microsoft's decision not to make Windows NT for
  PowerPC systems is being reported as a victory for Intel
  technology. The decision "impressed" Randall, says Reuters.

  Although the idea of Windows NT from Apple makes less and less
  sense each day - as Rhapsody moves closer, an Apple license for
  Windows NT would be seen as a lack of confidence in the new OS -
  CNET has gotten the bee in their bonnet that this is exactly what
  Apple is trying to do. Stories at
  <http://www.news.com/News/Item/0,4,7671,00.html> and
  <http://www.news.com/News/Item/0,4,7756,4000.html?latest> both
  report that Bill Gates is trying to license Windows NT to Apple,
  and Apple is considering it. CNET's NEWS.COM site says Amelio
  talked about it at the shareholder's meeting last week, but if he
  did, it must have been in a whisper because no one else heard it.

  The stories think Apple is creating confusion, sending mixed
  messages about Rhapsody by toying with Windows NT. Guess who
  agrees? Former AppleSoft head Ike Nassi, who left Apple in
  November at the, shall we say, "polite urging" of Ellen Hancock.
  "Either Apple's going to make a play on the Intel side, or
  somebody's wasting somebody's time," said Nassi, whose
  mismanagement of Apple's engineering teams and failure to make
  timely, effective decisions is at least as responsible for the
  Copland debacle as any other individual. We therefore concede
  Nassi's expertise in the subject of wasting corporate time.

**What's Really Going On?** -- The PowerPC architecture is strong
  and growing stronger - the new "G3" architecture chips to ship
  this year start at about 250MHz (MDJ_ 1997.02.07), with potential
  for 400MHz and beyond. The X704 chips from Exponential Technology
  coming this summer start at 450MHz and go up to 533MHz, with 1GHz
  (1024MHz) possible in a year or two. Most observers didn't expect
  Intel to have 300MHz chips on the market by the end of this year.

  However, it's a very painful mistake to underestimate Intel. The
  company surprised developers and analysts alike last week by
  demonstrating a 400MHz version of the "Klamath" chip, basically a
  Pentium Pro with MMX additions to the instruction set. Intel says
  it can make the 7.5 million transistor chip with its current
  high-production fabrication process, and has seen the chip run at
  461MHz in ideal lab conditions.

  It's unlikely Intel speeds will beat PowerPC speeds, though, and
  the gap between the faster PowerPC 604e-and-beyond chips and the
  slower Pentium Pro models will only widen as the speeds get
  faster. Still, Intel's crafty marketing and sizable market share
  have pushed the old CISC architecture much further than IBM,
  Motorola or Apple thought it could, although it will hit the wall
  someday.

Article from the Mac Daily Journal, Feb. 11, 1997.
    MDJ_, The Daily Journal for Serious Macintosh[tm] Users, is
  published by GCSF, Incorporated.

  Publisher:  Matt Deatherage

  Letters:    <letters -at- gcsf -dot- com>
  Press:      <pr -at- gcsf -dot- com>

  This file is formatted as setext.  For more information, send
  email to <setext -at- tidbits -dot- com>.  A file will be returned shortly.

  Copyright (c) 1997 GCSF, Incorporated.  All rights reserved.

  All trademarks are the property of their respective holders and
  owners.

  GCSF, Incorporated.
  P.O. Box 1021
  El Reno, OK  73036-1021
  (405) 262-1399
  <info -at- gcsf -dot- com>

 TECHWR-L (Technical Communication) List Information: To send a message
to 2500+ readers, e-mail to TECHWR-L -at- LISTSERV -dot- OKSTATE -dot- EDU -dot-  Send commands
   to LISTSERV -at- LISTSERV -dot- OKSTATE -dot- EDU (e.g. HELP or SIGNOFF TECHWR-L).
  Search the archives at http://www.documentation.com/  or search and
browse the archives at http://listserv.okstate.edu/archives/techwr-l.html