Who remembers CDC Plato?...run it again on a cyber at home...could it be done?

From: Rick Bensene <rickb_at_bensene.com>
Date: Thu Oct 19 20:07:52 2000

Claude wrote (excerpted)
>
> I grew up having access to the Plato system at the University near my
> home. A neighboor was a chemistry teacher there and knew I was
> interested in computers and gave me access to a bunch of stuff at the
> university and I guess CDC's Plato was the most fun.

An amazing software system for the time.
>
> Now I never the saw the hardware running Plato. I only know that it used
> a CDC cyber. So I am thinking...
>
> What Cyber ran plato?

Any of the Cyber 7x and 17x machines could run Plato, I believe,
but I think that there were some special interfaces that were used to
hook up those special terminals to the system's I/O subsystem.

>
> How big were they?(I heard of huge! Did not find any pics on the
> internet of one...)

Oh, they're big all right. Back in the late 1970's, I used to be a systems
operator at Tektronix, in Beaverton, Oregon. The Tektronix Scientific
Computer Center had a a CDC Cyber 73. It was monsterous.

First off, it had a cabinet about the size of a VAX 11/780 CPU that
was nothing but the REFRIGERATION for the system. There was a series
of pipes that ran through the system that provided the cooling.
Next, was the massive motor/generator down in the basement of the building
(the computer center was on the 4th floor of the building). The Cyber
ran on 400Hz 3-phase. The motor/generator downstairs took 220V
3-phase and converted it to 400 cycle power. I never saw the
motor/generator,
but I was told that it was about the size of a semi-truck trailer. The
power supply of the machine was another approx. 8 ft. tall, 4 ft. wide
cabinet. The CPU itself was, as I recall, 6 bays, each of which was big
enough, that when the outer doors were opened, and the internal bays
containing the electronics were swung out, a man could fit inside. The bays
were about 4 feet deep, 8 feet tall, and 6 to 7 feet wide.

The machine was truly amazing. A 60-bit word main CPU, with ten
PPU's (peripheral processing units) that performed interface to
peripherals. I think (can't remember for sure) that it had
128K-words (each word 60 bits) of core memory. One of the 'peripherals'
of the system was a Modcomp 16-bit minicomputer that handled all of the I/O
to literally hundreds of async terminals all over the Tek campus.
Characters
were represented in 6 bits...called 'Display Code'. The Modcomp
did all of the translation between ASCII and Display code for all of
those terminals.

The console of the system was glorious...a very large desk, with
two round CRT displays that used vector graphics to paint the screen.
There were a lot of fun graphics programs that could only run on the
console...things like 'eyes' that would look around and blink, a great chess
game, and a Snoopy vs. the Red Baron animation that was fun.

The 73 had two 9-track and one 7-track magtape drive, a monster
CDC chain-type line printer(that thing was FAST), a card reader, a card
punch,
and a big slew of those washing-machine-sized 10 or 12-platter removable
disc
drives (there must have been 14 or 16 of 'em, each holding something like
300 MB of 12-bit words.

This system didn't run Plato. I learned about Plato from an
engineer that worked at Tek who went to the University of Illinois,
a place where a lot of Plato development work was done.

The OS that ran on the system was called KRONOS. It was a multi-user
multi-executive OS that did both interactive and batch processing. There
were a number of RJE stations around the campus that had punch card
readers and printers that users could submit their F77 (Fortran 77)
and PASCAL jobs to remotely.

Later, Tek brought in another big CDC machine, a Cyber 175, I believe.
It wasn't quite as large as the 73, but still needed liquid cooling, most
all of the logic in the 175 was ECL, and it ran fast and hot. The console
wasn't nearly as impressive, one 'page' oriented CRT, but still vector.
I think that the 175 had 20 PPU's versus the 10 of the Cyber 73, and of
course, the main CPU was quite a bit faster than the old 73.

>
> What are my chances of fiding one?

Next to nil, I would guess. They were loaded with precious metals (gold!)
and lots of good scrap metal too. The scrappers loved 'em.

Even if you could find one in storage somewhere (it'd have to be a pretty
good-sized warehouse!) it'd be a major operation just to move the machine.
Plus, once you got it to your house, not likely you'd have any place
to put it! Not the kind of machine you can set up out in the garage.
Then...even if you had a place to put it, getting the 400Hz 208V
is going to be a challenge...probably draws some massive load. Don't
know what the current loading was

> And the plato software? (slim I guess...)
>
Very hard to say. I know that University of Illinois did a lot with
Plato...perhaps someone there has some old tapes laying around.
Might not be able to read them anymore...early Cybers used 7-track magtape,
at some ridiculously low bpi.

> If I remember well, there was a emulator for the IBMPC to run a plato
> terminal....no touch screen but still...anybody remember this?...that
> could save me from having to look for an old "plasma" terminal like they
> called them back then...
>
I remember seeing an old show called "The 21st Century" sometime in the
'70's
that was hosted by Walter Cronkite. One show was on computers. They
showed one of the plasma Plato terminals, with a child running a CAI
(Computer Aided Instruction) program. Now, you've probably got a better
time trying to find one of those old plasma terminals than you've got
of finding a Cyber laying around somewhere.

Thanks for the memories,

Rick Bensene
The Old Calculator Web Museum
http://www.geocities.com/oldcalculators
Received on Thu Oct 19 2000 - 20:07:52 BST

This archive was generated by hypermail 2.3.0 : Fri Oct 10 2014 - 23:33:17 BST