[Info-vax] Out with Hurd, in with OpenVMS

[John Wallace]

On Aug 23, 12:22 pm, m.krae... at gsi.de (Michael Kraemer) wrote:
> In article <fc922f13-6278-445f-9810-0b30b889a... at a36g2000yqc.googlegroups.com>,
>
>
>
> John Wallace <johnwalla... at yahoo.co.uk> writes:
> > On Aug 23, 1:24=A0am, Michael Kraemer <M.Krae... at gsi.de> wrote:
>
> > There was more to VME on Alpha than military although obviously
> > defence stuff was a large part of the VME market in general. CSS was a
> > part of DEC that got a lot of revenue from military stuff, though
> > whether it was always as much as the books appeared to show is
> > debatable. There's a book for someone to write about some of the
> > things that went on in CSS.
>
> > There was at least one German outfit (Atlas Elektronik GmbH?) that
> > designed and built their own VME/Alpha system(s) but I can't remember
> > the details or easily find a reference.
>
> > VMS was not the intended market for the VME kit, VxWorks and Tru64
> > were, although some customers in some markets would have loved to have
> > a VME-native Alpha card running VMS, rather than the hassle of
> > integrating a PCI-based (or, earlier, TURBOchannel-based) Alpha with a
> > VME adapter from DEC or elsewhere.
>
> Alpha was the last choice I would consider for embedded stuff.
> But in military one obviously has enough resources to make pigs fly.
> Probably DEC simply wanted to test if this will pan out in civilian
> too, but I guess the inertia of the market is as large as in PC desktop
> space.
>
> > "the Vobis Alpha PCs were exposed the same way as the other consumer
> > PCs of that German retailer."
>
> > Maybe they were, but Vobis were largely a general PC builder and the
> > general PC market quickly found out that although MS had told Palmer
> > they were supporting non-x86 systems, MS didn't actually deliver what
> > they promised (who else remembers the Advanced RISC Computing
> > consortium? NT used to run on MIPS and PowerPC too).
>
> Nevertheless this would have been the only sector
> where they could get the alpha out of the 100000 chips per year limbo.

"Alpha was the last choice I would consider for embedded stuff. But in
military one obviously has enough resources to make pigs fly."

Rather dependent on your definition of "embedded". Low cost low power
embedded (e.g. consumer electronics) is a market both Windows and x86
struggle with, despite Gates's attempts with WinCE set top boxes
inside BT (I suspect this was a CEO-to-CEO deal, not one approved by
the engineers). ARM generally rules in the consumer electronics
market, and for very good reason.

There were embedded sectors where Alphas were beginning to enjoy some
success. Embedded Alphas in high performance printer controller
(modern equivalent of a typesetting machine) where the power
consumption in watts is largely irrelevant and the availability of
commodity software is largely irrelevant, but the available CPU power
is critical was one such market.

Another similar market was the telco switching market (and the GSM HLR
etc market). I seem to remember (but cannot quickly find) announcement
of a deal where Ericsson decided to use Alpha in its core switching
product.

It helped in the "pre-press" market that there was some *marketing*
going on from DEC, that (a) identified this as a relevant sector for
Alpha (b) made sure relevant programmes were delivered. There is more
to marketing than bringing a product to market and then advertising it
(or more often with DEC, not advertising it).

There were other "embedded" Alpha markets too, such as specialist
hardware for the signal processing and HPC sectors.

Pigs may not fly very often, but military radars certainly do (and
some similarly serious radars also go on board ships, and battlefield
trucks, etc). These radars need serious numbercrunching capabilities
both for doing stuff that these days would be done by DSP and for
doing things like track following and display and general data
management, and the availability of a two battle proven and proven
secure OSes plus an embedded kernel were relevant in that market.
Alpha was in many ways way ahead of the competition, especially in (or
connecting to) a field-proven form factor. But we're wandering into
CustomSystems territory here, and it is a different world, where the
volumes are relatively low and the divisional profitability depends
somewhat on the book-keeping practices. Part of the reason the initial
VME/Alpha boards were not supported on VMS was the question of costs:
who would pick up the cost of qualification and support?

"<Vobis> would have been the only sector  where they could get the
alpha out of the 100000 chips per year limbo."

None of these were ever the Vobis-style market, but that should not be
a complete surprise. However just because these sectors are largely
invisible to most folks does not mean the total quantities could not
have exceeded the 100K per year you mention. And in principle if you
can charge a substantial premium price for the higher end product (as
with Xeon and with high end consumer CPUs) the lower end product can
be made to look more attractive, or the point of profit comes earlier.

Anyway, this time I really do need to be somewhere else now.