19 Aug
2010
19 Aug
'10
11:05 p.m.
Ian King wrote:
comparison. It is feasible for one
person to operate and use a VAX; the question in the context of a VAX-11 is an economic
one, not a question of operational overhead.
-----Original
Message-----
From: cctalk-bounces at classiccmp.org [mailto:cctalk-
bounces at classiccmp.org] On Behalf Of Chuck Guzis
Sent: Tuesday, August 17, 2010 3:39 PM
To: General Discussion: On-Topic and Off-Topic Posts
Subject: RE: PDP-1 as minicomputer [was RE: OT - sort of]
On 17 Aug 2010 at 15:14, dwight elvey wrote:
This is probably one of the most problematic questions I try to answer, far too often:
why is X a minicomputer and not something else? I think of a minicomputer as a machine
designed to not require an operator staff, designed for direct interaction between the
user and the computer (rather than batch operation), and designed for a *relatively* small
resource footprint (the PDP-1 required a LOT less resources than a 7090). There is a lot
of fuzz around all of these factors. Is a VAX-11 a minicomputer? It takes three-phase
power and was intended to serve as a large timeshared system. There is a mechanism to run
batch jobs. There was a role defined for the "system manager" who maintained
various aspects of the OS environment. However, at the same time IBM's product line
consisted of machines that still cost ten times as much, could NOT be meaningfully
maintained without an operations staff and were fundamentally designed for batch
processing. The VAX-11 was "mini" by IBM had a series of small computers they called
minis that used
water cooling. These where about the size of three standard desk,
end-to-end.
Is it safe to say that there were no minicomputers called as such
before 1960, simply owing the etymology? The BMC Mini auto dates
from 1960 or so and the miniskirt, from 1966.
With the PDP-1, I wonder if the the "mini" applied more to cost than
physical size.
Consider that when IBM transistorized the 709, they did NOT build a smaller computer
based on the increased functional density of the transistor. They built a large computer
with more capability. In contrast, the PDP-1 was intended to be small enough that it
wasn't a perceived economic absurdity for it to be used single-user.
For counterpoint, the PDP-10 was never intended to be anything other than a
mainframe-class machine. Its capabilities, capacity and footprint all served as direct
competition to System/360. So DEC != minicomputer, although they were the leading
manufacturer of that class of machine for many years - indeed, until the microcomputer
began overshadowing the mini.
This is an interesting question, and I hope we can have some meaningful dialogue about it
- and maybe even stay on topic. :-) -- Ian
Am I just all wet thinking that there was a class (admittedly with
rather fuzzy edges) of machines that could logically be called midi's?
I'll admit I really did not keep up with all that was going on in the
computer industry from say 1960 through maybe 1980 (more especially in
the marketing arena) to really know if anyone actually used that
descriptor for *any* system or not. I'm really thinking more
conceptually than anything else I suppose, and thinking about about what
folks like *us* (yeah, I know that covers a bunch of ground) might have
called the various systems. Of the limited set of machines I've
encountered, I always thought of the IBM 1401, the SDS-910, the SCC-660,
and the Datacraft 6024 as midis, at least in some senses, just based on
their size, weight, power and air conditioning requirements, etc. I had
very little contact with the IBM 1410, but I think if you'd asked me at
the time I would have said it was too "big" to be a midi. I might have
waffled on some 1401 installations too mostly based on the number and
size of peripherals. Only when I got to the Microdata 8xx computers
(can't remember the year just now, but certainly well past 1964 -
probably 1970 or later) did I think I was working with a mini.
Later,
Charlie C.