Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!columbia!rutgers!princeton!allegra!ulysses!sfmag!sfsup!jeffj
From: jeffj@sfsup.UUCP (J.S.Jonas)
Newsgroups: comp.terminals
Subject: Re: Brain-damaged Terminal Contest
Message-ID: <880@sfsup.UUCP>
Date: Tue, 2-Dec-86 12:29:30 EST
Article-I.D.: sfsup.880
Posted: Tue Dec  2 12:29:30 1986
Date-Received: Wed, 3-Dec-86 06:19:55 EST
References: <3078@cbosgd.ATT.COM> <58000001@smu>
Organization: AT&T Information Systems, Summit N.J.
Lines: 33

> People who think ASR-33s or Flexowriters epitomize bad hard-copy
> terminals have led a sheltered life. I once saw the console
> typewriter for an ICL 1900 clone, made in Poland under license.
> That's what it was, a typewriter, complete with a platten that
> went back and forth and printing hammers!

Ummm, perhaps your memory doesn't go back far enough, 
but I once played with OLD flexowriters that were just that --
typewriters with solenoids for each key.  Yup - the keyboard was not
switches but little levers like a manual typewriter.
The baudot code corresponded with keystrokes like shift up/down,
ribbon red/black.  I guess that they took a manual typewriter
and added solenoids and switches.  It was fun to see the little
hammers jumping up at 10CPS, with the carriage swishing back and forth.

Even sillier was the General Precision LGP-21 computer to which it was
attached.  It had NO CORE - just a fixed head disk (not drum).
The accumulator, accumulator extension, instruction address register
and instruction register were recirculated on the outermost track by
4 heads placed evenly around the circumference.  The baudot codes
corresponded to the 5 bit opcode (B for 'bring' - load the accumulator,
S for 'save' - store the accumulator, A for 'add', etc...).
The  display was an oscilloscope which displayed the square waves
for the accumulator, instruction register and instruction address
register. I guess that it was a serial machine -- there weren't enough
transistors to support parallel byte manipulation.
The manual mentioned a card that was used to find the optimal data addresses
to use so that execution was done with the least disk rotation.
(card meaning a cardboard device for manual use, not a PCB).
That's why SOAP was used by bigger machines.

	Jeff 'historian for a quarter of a decade' Skot
	{ihnp4 | allegra | cbosgd} attunix ! jeffj