Re: Alpha remembrance day

Bill Todd wrote:
Andrew wrote:

...

Why keep repeating the same sorry tired mantra ?

I'm sorry that you seem to find the truth tiresome, moron - but as you
ought to be able to divulge from some of the material to which you
responded, you're not an audience that I care about.

Which truth are you refering to exactly, yours or reality ?


Alpha grew initially because it was new, it was quick and with things
like the VEST program it was relatively easy to move apps from VAX/VMS
to Alpha/OpenVMS. It died because the mistakes made by DEC prior to
Alpha's introduction created a climate in which it could not attract
enough software or market share to sustain it after the honeymoon of
the initial introduction.

Sure, Andrew - the honeymoon period explains it all, just like it did
with Itanic's initial acceptance (not) - despite the ease with which
IA32 software could run there without even something like VEST being
necessary.

Ohh come on Bill even you must grasp the weakness of this point.

When Alpha was introduced it clearly led VAX in terms of performance
and price so there was a real incentive to move. It also led all the
other CPU architectures in terms of per CPU performance so there was
also an incentive to move onto the Alpha platform from outside.

Roll forward a decade and Itanium had none of the speed advantage that
Alpha had in fact you could buy a faster system running Alpha from the
same vendor HP and if you wanted IA32 compatibility you could buy a
faster system from anyone.

Itanium never had a honeymoon period because it never won out on any of
the dimensions that cause customers to buy systems. The Merced launch
was a complete flop mainly because the one thing that Itanium was
supposed to have going for it, CPU performance it did not have at all.

However no one not even a ex Sun employee like myself would claim that
Alpha wasn't that fastest CPU on the planet when it was launched.

Alpha's problem was that because of DEC's mistakes it had no supporting
ecosystem.


Alpha sales initially went up to a pretty reasonable 10-12% market
share, fell in the latter half of the 1990's to below 5%

For the reasons I suggested, of course - reasons based on actual events
rather than on your 'honeymoon' fantasy.


So what are the actual events you would like to introduce ????

Come on Bill

and then
started to rise.

Yet another blow to your fantasy, which offers no explanation whatsoever
for such a turn-around.

Really how about Compaq finally having some one ot two OK platforms to
support Alpha ? The ES40 was introduced in 1999 as was the DS series
these modesl at least gave Compaq a competitive low to mid range (ish
in the case of the DS).

The demise of Pyramid, Sequent and to an extent SGI also gave Compaq
access to customers without a supplier.

By the time Alpha did start to claw back share it was
too late,

Too late for *what*, idiot? Too late to rise? But you just admitted
that they *were* rising, despite at best luke-warm support from Compaq.


Again you miss the point spectacularly.

Market condidtions changed, customers moved from make to buy, ISV's
like Oracle who used to support 10's of hardware platforms took an axe
to their platform list. NT began to increase market share, Linux on x86
began to increase market share.

Why would any ISV want to support the so so Alpha Platform when the
resources required to do this could be re-deployed to support Linux and
NT on x86.

Having 10% market share at the beginning of the 90's would have given
Digital at least the right to talk to ISV's about port and support, at
the end of the 90's that didn't even get them an appointment.


consolidation, industry standard platforms

Surely an ex-Sun bigot like you considers Unix to qualify as such a
standard, and (not entirely by coincidence) Tru64 on Alpha was growing
far faster than its Solaris, AIX, and HP-UX competition during that period.

And you call me an idiot. Of course it was growing faster it had got to
such a low base that a one percent increase in market share was a 20%
market share growth. Sun, IBM and HP duking it out in the 25-35%
territory would have had to grow share at 7-8% to match Tru64 market
share growth.

But, again, that doesn't fit the fantasy you're attempting to pawn off,
so by all means let's not deal with it.


Yawn

The only fantasy here is the Bob Palmer killed Alpha one, the reality
is much more complicated as it generally is and DEC's management before
Palmer had more to do with killing Alpha than Palmer.

...

When
you are 3rd or 4th in a market which has 5 or 6 vendors things look
much better than when you are 4th out of 4.

Don't assume that your audience is as incompetent as you are, Andrew.
When you enter a market later than the competition, it takes time to
catch up (especially when your early steps are as compromised by
ambiguity as Alpha's early Unix steps were). There's nothing wrong with
being 4th out of 4 as long as your growth rate promises a better
position in the future and as long as you're profitable. Both were true
of Alpha and Tru64.

You appear to be making the assumptions.

regards
Andrew Harrison

.

Relevant Pages

  • Re: Little Alpha I learned on is up...
    ... Little Alpha I learned on is up... ... to do the same kind of tasks a more mainstream platform can do. ... What kind of hardware service and support is available to customers ...
    (comp.os.vms)
  • Re: Little Alpha I learned on is up...
    ... Little Alpha I learned on is up... ... to do the same kind of tasks a more mainstream platform can do. ... What kind of hardware service and support is available to customers ...
    (comp.os.vms)
  • Re: Alpha remembrance day
    ... adequate application support in their own right, ... Since Alpha didn't even ... that it is very late entering the market. ... This lost DEC market share to Sun, HP, IBM and SGI/MIPS. ...
    (comp.os.vms)
  • RE: It is almost certain now, INTEL will have 64bit x86 !!
    ... > Microsost isn't after performance, ... > is why windows on Mips Power and Alpha were canned, ... dropped Windows as a platform. ... support the platform, then neither could Microsoft. ...
    (comp.os.vms)
  • Re: Alpha remembrance day
    ... anything it gave Alpha a lower starting point from which regaining lost ... market share might have been easier than beginning with more of it. ... Without the carrot of NT/Alpha platform sales to host Exchange etc it ...
    (comp.os.vms)