Re: Alpha remembrance day

In article <44E6087E.FA9BDDEC@xxxxxxxxxxxx>, JF Mezei
<jfmezei.spamnot@xxxxxxxxxxxx> writes:


Well, considering it was the first "mainstream" 64 bit architecture, I
wouldn't say that Alpha was late. If at all, it was early to the market.

even if it were true (Andrew corrected you on that one), so what ?
It is just now, more than a decade later that people start really looking
at 64 bit stuff.
In 1992, there simply was no use for 64 bit integers or 64 bit pointers.
(64 bit FP already existed for decades on other architectures).
My home DEC 3000 (of 1993/4) has room for as little as 192MB.
My home alphastation 500 (of 1996) offers just 512 MB (and still is partially
empty because modules aren't cheap/easy to find).
So what's the practical use for 64 bit addressing if one can't even
fill 32bit ? Certainly you do not suggest to use swap space for large matrix
multiply, do you ?

DEC was late only in lowering prices for its VAX lines to stay in
business until Alpha came along.

DEC was late with alpha about 2 to 3 years, plain and simple.
Even in 1992 they still sold so-called "alpha-ready" VAXstations
(I sold one of those veterans on eBay a couple of weeks ago),
just because alpha wasn't ready.
And to be honest, inspecting their interior makes me wonder
how that "64-bit" CPU would fit into a 32-bit box.


At the time, intel was still just a toy controller maker and there
weren't that many high tech FABs that were commercially available. For
DEC, having its own FAB was a way to be to market before any others by
building a next generation FAB now. Had DEC succesfully commercialised
its FAB business, it may have had enough business to pay for upgrades.
In the end, DEC didn't commercialise that business and it lost big money
and certaintly coudn,t afford to build a next generation FAB. DEC
vacating that market left the door opened for IBM and Intel to
commerciallise their FAB business. And because of the extremely high
costs of FABs, fewer and fewer FABs can be built, and those who stayed
in teh business now get everybody else's business.

At that time, it was already out-of-fashion to own a fab,
except if it was a company's core business to make chips,
like Intel, Motorola, TI, etc. Fabbing a handful of alpha chips
certainly wouldn't qualify DEC as chip making expert.

Only the fittest survive. DEC didn't exercise its FAB enough and was
eliminated from the race.

They entered the race when it was already over.


Again, read his book. The strategy was to break the company up. It was
Gerstner that immediatly cancelled this and set the company on a course
where it woudl have its own strategy and compete.

IBM had a product strategy at least. Their RISC/UNIX line started in 1989/90
(not counting their RT nice try), and when it hit the streets IBM
always acted as a hungry company that wanted to grab a large portion
of the UNIX market (which was one of the fastest growing segments back then).
Compare that with the confuse DEC actions (even before Palmer).
A bit mainframe imitation, a bit of VMS, a bit Unix, maybe, later NT,
and does somebody remember that flop called ACE ?


Digital was #2 computer maker for a very long time, and I believ it
still was at the time Palmer took the helm.

#2 doesn't say much without knowing the distance to #1.
And you forget about HP, which certainly was "larger" than DEC,
at latest during the 90s. But I've the impression that people
here still live in the 80s anyway.

DEC was downsized out of
existance uder Palmer, partly due to "suggestions" by Pfeiffer.


When Palmer took over DEC was already about to die.
There wasn't much that he could do to raise money on the short term except
selling parts of the company that weren't absolutely core competency.
.

Relevant Pages

  • Re: Alpha remembrance day
    ... to fill the shoes that Alpha had been expected to fill. ... reserve FAB capacity ofr Alpha. ... Palmer took steps to prevent Alpha from growing, ... "Competing with Wintel" why would you think that Alpha was ever going ...
    (comp.os.vms)
  • Re: Alpha remembrance day
    ... wouldn't say that Alpha was late. ... having its own FAB was a way to be to market before any others by ... its FAB business, it may have had enough business to pay for upgrades. ... When he got in, IBM had already contracted ...
    (comp.os.vms)
  • Re: Alpha remembrance day
    ... used Hudson to build those chips. ... to fill the shoes that Alpha had been expected to fill. ... reserve FAB capacity ofr Alpha. ... Palmer took steps to prevent Alpha from growing, ...
    (comp.os.vms)
  • Re: Alpha remembrance day
    ... having its own FAB was a way to be to market before any others by ... its FAB business, it may have had enough business to pay for upgrades. ... Palmer was acting on the requests of Pfeiffer to scale down Digital ... When he got in, IBM had already contracted ...
    (comp.os.vms)
  • Re: Alpha remembrance day
    ... AMD wanted to use the FAB. ... The topic, which your attention span couldn't embrace, was that others wanted to use the FAB, but DEC wouldn't take the business, wanting to keep the FAB ready for more Alpha business. ... by any other vendors. ...
    (comp.os.vms)