[Info-vax] HP wins Oracle Itanium case

[Richard B. Gilbert]

On 8/22/2012 7:41 PM, David Froble wrote:
> Richard B. Gilbert wrote:
>> On 8/21/2012 1:44 PM, David Froble wrote:
>>> Bob Koehler wrote:
>>>> In article <k0u9du$bko$1 at dont-email.me>, David Froble
>>>> <davef at tsoft-inc.com> writes:
>>>>> If DEC had positioned VMS to perform all these jobs, and I'm talking
>>>>> 1990 or perhaps even before, including proper marketing and pricing,
>>>>> would it (VMS) have been feasible, or are you saying VMS could not do
>>>>> the job?
>>>>
>>>>    Too late.  VAXen were way behind in performance, people left, and
>>>> did
>>>>    not come back.  DEC still thought they were selling hardware.
>>>>
>>>
>>> I don't buy that.  There are many jobs that don't require bleeding edge
>>> performance.  There are some jobs that highly value reliability.
>>>
>>> VAX died because DEC stuck a knife in it, even as the low end systems
>>> were still selling well.
>>>
>>> People left because DEC quit responding to customer needs.
>>
>> Don't you think that the Alpha processor had something to do with the
>> decline of the VAX?  I couldn't help noticing the great speed speed of
>> the Alpha.
>>
>> I know that my then employer jumped on the Alpha bandwagon.  We were
>> well aware that we needed greater speed than the VAX could deliver.
>> We purchased a pair of Alpha 4100 processors and found the speed we
>> needed and some to spare!  We had to buy more RAM than the VAXen needed
>> but the Alphas got job done and done FAST!
>>
>> The switchover worked and solved our performance problems.  For the
>> first time in a long time we had more speed than we really needed.
>>
>>
>
> Ok, so you needed the performance.  That doesn't mean everyone did.
>
> A while back someone posted about process control in a steel mill.
> Computers with sensors controlling a line that got up to speeds of 60
> MPH.  (That impressed me.)  I'm betting much older and slower computers
> could provide the required performance.  What was needed was
> reliability, replacement parts, etc.
>
> DEC had all types of customers.
>
> When DEC stopped the N-VAX production line, they claimed they had build
> a stockpile designed to last at least 5 years.  Didn't come close.
> Where did all those CPUs go?  The only place I can figure is customer
> demand.  When they were gone, that was the end of VAX sales.
>
> The question that occurs to me, and I doubt it can be answered, is how
> long could DEC continue to mfg and sell N-VAX CPUs?  No more R&D, just
> keep the line running.  It's called a "cash cow".  Instead of milking
> it, they butchered it.
>
> Don't get me wrong, Alpha was great, maybe the best there has ever been.
> Another great question is how would Alpha perform with a 32 nm die size?
> Multiple cores per chip.  Low voltage.  Cooler temps.  Though I have
> heard of problems getting it to run correctly on smaller die sizes.
> You'd want to think that such problems could be solved.  Some people
> didn't want to pay to solve them.

How much would they have had to pay?  How long would it have taken to 
pay the costs of development?  They *MIGHT* have put up their own money 
if they saw a potential future profit.  What was the probability that 
they would lose their shirts?

As has been said, in another context, "Follow the money!"