Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > alt.folklore.computers > #151652 > unrolled thread

Univac Military Computer (1960)

Back to article view | Back to alt.folklore.computers

Contents

  Univac Military Computer (1960) hancock4@bbs.cpcn.com - 2015-09-21 08:27 -0700
    Re: Univac Military Computer (1960) "Osmium" <r124c4u102@comcast.net> - 2015-09-21 10:41 -0500
    Re: Univac Military Computer (1960) Quadibloc <jsavard@ecn.ab.ca> - 2015-09-21 14:21 -0700
      Re: Univac Military Computer (1960) Jon Elson <elson@pico-systems.com> - 2015-09-21 23:13 -0500
    Re: Univac Military Computer (1960) isw <isw@witzend.com> - 2015-09-21 22:00 -0700
      Re: Univac Military Computer (1960) Gene Wirchenko <genew@telus.net> - 2015-09-22 18:09 -0700

#151652 — Univac Military Computer (1960)

bitsavers has a neat 65 page brochure from Univac advertising its St. Pual based military computers.

http://bitsavers.informatik.uni-stuttgart.de/pdf/univac/military/Univac_St_Paul_1960.pdf  
(note--large file, 46 meg).

Cool photos of consoles, circuit boards, manufacturing, missiles, maintenance gear, etc.

Also description of the Univac airline reservation control system.

On the last page is a map of Univac's plants in St. Paul.  Later I'll compare that to google maps to see what's there today.

(The last company I checked out was once near LAX, today the site is a LAX runway.  So it goes.)

[toc] | [next] | [standalone]

#151653

<hancock4@bbs.cpcn.com> wrote:

> bitsavers has a neat 65 page brochure from Univac advertising its St. Pual 
> based military computers.
>
> http://bitsavers.informatik.uni-stuttgart.de/pdf/univac/military/Univac_St_Paul_1960.pdf
> (note--large file, 46 meg).
>
> Cool photos of consoles, circuit boards, manufacturing, missiles, 
> maintenance gear, etc.
>
> Also description of the Univac airline reservation control system.
>
> On the last page is a map of Univac's plants in St. Paul.  Later I'll 
> compare that to google maps to see what's there today.
>
> (The last company I checked out was once near LAX, today the site is a LAX 
> runway.  So it goes.)

Fair warning: Plant 4 Building 1 was reduced to a pile of rubble, carted 
away, and replaced by a new and shiny similar building.  The Univac plant 
was built in the early 1960's and destroyed about 2010. 

[toc] | [prev] | [next] | [standalone]

#151680

From the brochure, I learned something interesting about the front panel of the 
Univac II. It had what seemed like enormous square Nixie tubes in it - but one of 
them, which represented the different phases of a computer cycle, was clearly 
custom made; it displayed the lower-case Greek letters beta and gamma to indicate 
phases of a cycle. (Probably also alpha at least.)

John Savard

[toc] | [prev] | [next] | [standalone]

#151698

Quadibloc wrote:

> From the brochure, I learned something interesting about the front panel
> of the Univac II. It had what seemed like enormous square Nixie tubes in
> it - but one of them, which represented the different phases of a computer
> cycle, was clearly custom made; it displayed the lower-case Greek letters
> beta and gamma to indicate phases of a cycle. (Probably also alpha at
> least.)
Another technology that was used had light bulbs that illuminated a piece of 
film that was then projected on a frosted screen.  You could get any image 
onto the film, words, arrows, digits, etc.  I think they typically had 12 
bulb positions, so you could get 10 digits, decimal point and minus sign.

Another scheme had multiple sheets of lucite that were edge lit.  Only the 
sheet that was illuminated showed its pattern.

Jon

[toc] | [prev] | [next] | [standalone]

#151700

In article <56de8784-3b69-4dfb-ac25-ae35b9044581@googlegroups.com>,
 hancock4@bbs.cpcn.com wrote:

> bitsavers has a neat 65 page brochure from Univac advertising its St. Pual 
> based military computers.
> 
> http://bitsavers.informatik.uni-stuttgart.de/pdf/univac/military/Univac_St_Pau
> l_1960.pdf  
> (note--large file, 46 meg).
> 
> Cool photos of consoles, circuit boards, manufacturing, missiles, maintenance 
> gear, etc.

The brochure mentions "NTDS", Univac's inter-computer bus (at least for 
it's Navy computers).

It was a terrible thing to interact with. The parallel version needed 
well over a hundred signal wires, and (not only for that reason) 
couldn't be used over even moderate distances. The serial version (using 
a pair of RG-11 coax cables running  at 10 MHz) was even worse. The 
little bit of Univac documentation we could get showed it to be a 
poorly-thought-out, generally bad design.

In the late '60s or early '70s, working for a small contracting firm 
building specialized data-collection gear for the Navy, I designed an 
NTDS serial-parallel converter box so computers with only parallel 
interfaces could talk to other Univacs more than a few feet away. You 
put one of the boxes at each end, hooked them to their respective 
computers with a couple of giant cables, hooked the boxes together with 
a couple of RG-11s, and Bob's your uncle.

Nasty protocol:

"Can you take a 36-bit data message?"

"No; sorry. Busy right now."

"Well, HERE IT IS ANYWAY!"

In the demo for acceptance, my box would not work with a Univac 
serial-parallel converter at the other end. The Univac guy said "Well, 
too bad. Guess you guys have some more work to do."

My boss said "how about you try with Univac boxes at both ends so we can 
scope out what we're doing wrong?"

The Univac guy really didn't want to do that, but pressure from the Navy 
personnel (who were paying our bills) made him do it. It failed too.

Then my boss said "Just for grins, let's try our boxes at both ends". 
Univac guy *really* didn't want to do that, but he had no choice.

The channel worked just fine.

Isaac

[toc] | [prev] | [next] | [standalone]

#151741

On Mon, 21 Sep 2015 22:00:40 -0700, isw <isw@witzend.com> wrote:

[snip]

>In the demo for acceptance, my box would not work with a Univac 
>serial-parallel converter at the other end. The Univac guy said "Well, 
>too bad. Guess you guys have some more work to do."
>
>My boss said "how about you try with Univac boxes at both ends so we can 
>scope out what we're doing wrong?"
>
>The Univac guy really didn't want to do that, but pressure from the Navy 
>personnel (who were paying our bills) made him do it. It failed too.
>
>Then my boss said "Just for grins, let's try our boxes at both ends". 
>Univac guy *really* didn't want to do that, but he had no choice.
>
>The channel worked just fine.

     OTOH, a problem getting a printer to work with a Convergent
Technologies system was due to CT following the interface spec more
closely than most.

Sincerely,

Gene Wirchenko

[toc] | [prev] | [standalone]

Back to top | Article view | alt.folklore.computers

csiph-web