Re: Alpha remembrance day


JF Mezei wrote:
Andrew wrote:
What a truly ludicrous point. By the time Palmer came on the scene in
1992 Alpha had been a fully funded project since 1989 and was itself a
followon from PRISM. The decisions to design and build Alpha and the
decisions about what OS platforms it would support and what migration
options would be offered to existing customers had already been made.
The only thing Palmer could have done at that point was cancelled the
project.

It was under Palmer that Digital's software portfolio was severely
reduced and that many products abanmdonned/not ported to Alpha (FMS was
one of them, a decisions later reversed). Rememeber also the dropping of
VAX document/bookreader in favour of some unknown 3rd party, a decisions
that was also rescinded much later after damage had already been done.


It was under Palmer that decisions were made to refuse to use the Hudson
fab to its fullest, reserving production capacity for Alpha should it
ever take off, this, at the same time that Palmer made damned sure Alpha
couldn't grow to compete angainst Intel's 8086.


So who would Digital have sold FAB capacity to? They had no track
record of working with FABless vendors like Sun and they were in a lot
of instances major competitors for the companies who might have been
target customers to use Digitals FAB.

That and the rather grandiose plans for Alpha meant that Palmer had to
reserve FAB capacity ofr Alpha. Its rather tricky just to turn a
contracted FAB run off in order to make your own units if demand rises.

It was also Palmer that killed one of the biggest profit centres for
DEC: ALL-IN-1. Shortly after they ad announced the porting of ALL-IN-1
to Unix and Windows, Palmer announced that he had struck a deal with
Microsoft to deploy Office everywhere and abandon ALL-IN-1 and all the
messaging infrastructuire that had been so profitable to DEC.

Prior to Palmer, the technical decisions/development of Alpha were made.
Under Palmer, it was the business decisions that were made about Alpha
and it is those that crippled Alpha and killed Digital.


Sadly not true. Apart from doing Alpha at all as a design the second
business/technical decision which crippled Digital which Palmer
inherited was the decision to FAB Alpha in house. The cost of designing
Alpha and then fabricating Alpha proved to be business decisions which
bled Digital dry. Now its possible that the decision to FAB in house
was made because Digital could not persuade any of the 3rd party FAB
vendors to work with them something that would not have been entirely
suprising, but there is no evidence that Digital ever actually tried
this route.



SLATER: I don't think so. Alpha is an outstanding piece of technology.
It is probably superior architecture to the architectures that exist,
but I don't believe the differences are big enough to overcome the fact
that it is very late entering the market.

Ironic statement in hindsight isn't it ? It was in fact extremely early
into the market. But His point still has validity in that the 8086
architecture had already spread its roots.


No Alpha was extremely late into the market. MIPS, SPARC, HP-PA and IBM
POWER had all beaten Alpha to market by a wide margin. Alpha was
released in 1992, MIPS in 1985, ARM in 1986, IBM PC/RT in 1986
(replaced by POWER in 1990), SPARC in 1987, PA-RISC 1989. Alpha wasn't
even the first 64 bit RISC processor, the MIPS 4000 series which is 64
bit was released in 1991.

By the time Alpha arrived many of the RISC processors Alpha was
designed to compete against were in their 2 and or third iteration.
MIPS had gone from R2000 to R3000 to R4000. Sun had gone from SPARC v7
to SPARC v8 etc.

However, there would have still be chance for Alpha to uproot the 8086
in the 1990s, especially during the hectic "fleet replacement" cycles
for Win95/NT and later the Y2K. But by then, Plamer had already decided
to not allow Alpha to compete at the low end.

You have to remember that the early 1990's was an period where Alpha
was one of a number of processor architectures trying to unseat x86.
SPARC, POWER and MIPS were all attempting to do the same thing and had
been for some time so Alpha entered a crowded market. That and Alpha
was being pushed by Digital a company that had seriously lost its way,
its customer base and possibly most importantly attention from its ISV
partners.

The reality is that whatever Palmer did he started from the standpoint
of having been delt a really terrible hand of cards. Expecting him to
win out was as unreasonable as expecting to take over a hand at poker
and win when you have a nothing and your three oponents have 4 aces, 4
kings and 4 queens between them.

regards
Andrew Harrison

.

Relevant Pages

  • 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. ... Compare that with the confuse DEC actions (even before Palmer). ...
    (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 ... In the 3 years prior to the official announcement of the Compaq buying ... The decisions which led to Digitals downsizing were made in ...
    (comp.os.vms)
  • Re: Alpha remembrance day
    ... It died because the mistakes made by DEC prior to ... So Palmer did not ... Alpha which was a great opportunity that had been presented "ready to ... Microsoft reseller. ...
    (comp.os.vms)
  • 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
    ... 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)
Loading