Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!lll-crg!ames!ucbcad!ucbvax!decvax!decwrl!amdcad!amd!intelca!mipos3!omepd!psu-cs!reed!percival!bucket!rickb
From: rickb@bucket.UUCP (Rick Bensene)
Newsgroups: comp.terminals
Subject: Re: Re: Brain-damaged Terminal Contest
Message-ID: <183@bucket.UUCP>
Date: Mon, 24-Nov-86 01:31:07 EST
Article-I.D.: bucket.183
Posted: Mon Nov 24 01:31:07 1986
Date-Received: Tue, 25-Nov-86 21:21:22 EST
References: <1447@kitty.UUCP>
Organization: Rick's Home-Grown UNIX; Portland, OR.
Lines: 46
Summary: A Flexowriter "computer"...

> 
> 	Didn't some of the Friden Flexowriters use a paper tape code that
> was NOT ASCII?
> 	I remember once seeing a Flexowriter "computer" (I am serious).
> This consisted of a Flexowriter sitting on top of a cabinet approximately
> 24" wide by 24" deep by 30" high.  Inside the base cabinet was the guts
> of a mechanical calculator (probably also a Friden) which had a zillion
> solenoids and contact switches, in addition to a number of relays.  This
> device apparently interfaced the calculator to the Flexowriter, and was
> definitely a mass-produced product.

I don't go back far enough to know of a relay version, but I was associated
in my very younger days (next-door nieghbor who baby-sat  my brother and I
when we were quite young) with some people who owned a company that
calibrated daily tanks.  The job involved taking calibrated amounts of
water, and measuring the depth of the water in the tank using accurate
'dip sticks', and then from the experimental data, use interpolation to
generate a chart which would give the number of pounds of milk in the
tank for any dipstick reading.  Anyway, back to the point...these folks
had a very interesting machine the back room of their house which
was a Friden Flexowriter with a box bolted on the back of it which was
as wide and tall as the Flexowriter, and added about 9 inches to its
depth.  The box was full of boards with transistors all over them, and
had a small array of core memory.  The box was capable of simple math,
mainly addition and subtraction of fixed size, fixed point numbers.
It was 'programmable' by reading a punched tape into core, but had
very limited capabilites.  When they first purchased the box, I recall
the Friden people being out there for days on end trying to figure out
how to program the box to make it properly do the job.  Turns out that
the job was THE MOST complex program they had ever written for the
machine, and they had to go through some real gyrations to make it
fit.  I later wrote about 100 lines of BASIC on a North*Star
Horizon that did the job infinitely faster, and in a much friendlier
environment.  It was an incredible machine for its time (late 1960's),
but hopelessly outdated just a few years after it was purchased.
I would be willing to bet that the relay-filled box mentioned in
Larry L.'s posting above was the precursor to the transistorized
version that so intrigued me in my younger days (probably responsible
in part for my intense interest in computing machines as a youngster).


-- 
Rick Bensene
..tektronix!tekig4!rickb  (work)
..tektronix!reed!{omen, percival}!bucket!rickb (home)
USMail: 1815 N.E. 148th, Portland, Oregon  97230