1 Jul
2010
1 Jul
'10
7:54 a.m.
I'd strongly suggest reading
*Computer Engineering - C. Gordon Bell, J. Craig Mudge, John E. McNamara*
On line at:
http://research.microsoft.com/en-us/um/people/gbell/Computer_Engineering/
As it covers a lot of the history leading up to the 11/70 and VAX as
well as other machines
from 1957 to about 1977. it's straight from the
horses moth info and
also provides the
time lines and a good look at why and what.
It's important to point out that PDP11 was aimed at low to mid size
applications of the time
and VAX was for the bigger jobs and demanding apps. it was the successor
to the PDP-8
for a large segment of computer tasks and it was to be cheaper and
perform better. There
was overlap in that, a great deal of it both in performance and
timeframe. The 11/70 was
for practical cases concurrent with with the VAX11/780 and the 11/74 was
after that. When
you consider the early VAX system also offered compatibility with PDP11
it didn't take a genius
to see that there was customer base market fragmentation over PDP11 or
VAX. The
problem for DEC and the customers was that there were two paths one
being upgrade with
performance and lifespan ahead and the other cheaper but near the end
for how much more
performance it could yield. For many parts of DEC this was a conundrum
as PDP11 was popular
and well established by 1975 but it was clear by 1980 it's ranks would
be relegated to lower
cost medium performance due to technology growth where a machine
embodying more
advanced concepts would have extended growth path. This was not new to
the industry
but DEC was like many in the middle of it trying to figure out the next
direction and those
there were at most starting a third decade since PDP-1 and knew that
change was the only
constant.
It's important to recognize that computer systems in 1978 were more
costly than a car for small
ones and for the midsize PDP11 more than a house, and VAX was in the
price range of a entire
block of houses. In 1978 that dynamic range was 4000 to 1,000,000 plus
dollars. That and the
rapid advancement of computer performance over short periods made it
clear if you weren't
looking ahead you were headed for the wall. Even the PC market
exhibited this and its time
scale was considerably compressed in comparison.
My experiences span 1969 with my first contact with an pdp-8i (BOCES
LIRICS Tymeshare)
with the works through PDP10, PDP11, 8008, 8080, Z80 to current ARM9.
Thats a lot of change.
Allison
On 07/01/2010 09:51 AM, allison wrote:
On 06/30/2010 06:28 PM, William Donzelli wrote:
Well you know the VAX11/70 announcement date it's easy to work
backward by
18months to two years to get to a design start date. Since the
announcement was
introduced on October 25, 1977 we are looking at late 1975. There is
much overlap
in he time lines.
Also the supposed picture of the 70mp is the same cab style and colors
of the
post 780 era.
... and
it was a faster number cruncher than VAX-11/780.
The VAX had higher potential as the new reigning super
minicomputer. It
wasn't long after that I'd seen a VAX-11/782, 785 and VAXclusters.
I have
to think that these later VAX machines and clusters were
certainly well in motion, even if unannounced, by the time this whole
11/70mp / 11/74 project came out. In a business perspective, it makes
a lot of sense why these weird-11s were not marketed.
I know almost nothing technically about these, so
I ask - were these
multi-processor machines testbeds for ideas later seen (or not) in
other machines? Were they pet projects (DECBob)?
VAX SMP machines and later and Clusters of DEC36bit machines may have
been.
Allison
--
Will