CDC 6600/Cyber 73 Memories - WAS: Harris H800 Computer

[Rick Bensene]

Mark J. Blair wrote:

> I also seem to remember an operator's console with two round CRTs on
it,
> but I might have fabricated that memory from whole cloth.
> 
I think that you were remembering the console of one of the Control Data
6000/Cyber-70 series computers that you may have seen somewhere.  This
series of Control Data machines were famous for their consoles with two
large, round, green-phosphor monitors that used vector drawn-characters
(generated by one of the Peripheral Processors).  Most of the normal
system screens were all text, but there were some special programs
written (including a nice graphical chess game, a little program that
would put up eyes on the screens that would look around and blink, and
some others that don't come to mind at the moment.

I operated one of these systems
 (a Control Data Cyber 73 --
http://bitsavers.trailing-edge.com/pdf/cdc/cyber/brochures/Cyber70_Mod73
_Feb71.pdf) 
at Tektronix in Beaverton, Oregon, from 1977 through around 1980.   It
had two CPUs, 131K (60-bit words) of core, ECS (extended core storage),
and 20 Peripheral Processors, a combined punched card reader/punch,
something like 12 "washing machine" type  multi-platter cartridge disk
drives (that held something like 300 MB each), two 9-track tape drives
and one 7-track tape drive.   It also had a big chain printer that was
noisy, but pretty fast.  

 The machine had a channel interface to a Modcomp communications
processor (with communication maintained by one or more Peripheral
Processor programs), that provided serial I/O to terminals scattered all
over the company by some kind of serial concentrator that I can't
remember.  There was also a big modem pool for dial-in use.     The
system ran a locally-modified version of the Kronos Timeshared operating
system.  The system was used primarily by engineering departments
(research and product development) for CAD and CAD software development,
circuit simulation (SPICE), cross-assembling microprocessor code, and
mathematical modeling.    

The machine was an all-transistor design, based on the CDC 6600
processor.  It was liquid cooled, and had a large cooler unit that sat
with the machine that cooled the coolant (water) and circulated it
through the chassis, venting  the heat (which was substantial) through a
special venting system.  I remember the CDC Field guys talking about
horror stories when there were leaks in the cooling system.  We never
had any problems while I was there.

One day I was at the console when one of the big high-voltage rectifier
tubes that were in the console decided to short.
I was watching one of the system monitor displays, and suddenly I saw
the display collapse into a single very bright horizontal line.  I noted
that the other display also did the same thing.   I also heard a funny
noise that sounded kind of scary, so I started to push my wheeled chair
away from the console, but not soon enough to avoid a shower of sparks
and even some molten metal that spewed out from the console.   I had a
few small burns on my arms, and one little blob of molten metal burned a
hole in my pant leg.   One of the other operators in the machine room
managed to hit the power switch for the console and shut it off.  Then
the fire suppression alarm went off indicating that the Halon was going
to dump soon, so he ran back and hit the override since the sparking and
smoke had settled once the power was off.  Despite this, the fire
department showed up (the fire suppression system in the computer room
had a direct line to the fire department), and we had to tell them it
was a (semi) false alarm.

The machine kept running just fine, and we were able to keep tabs on it
with a serial terminal hooked up to the machine that had a program
running that kind of emulated the console displays.   The CDC guys were
there very quickly, and ended up having to replace two (IIRC)  big
rectifier tubes, and one burnt up power resistor.  When they powered it
up, the screens came up just as they were before the event occurred, and
all was well.   

I really enjoyed those days.   The machine was really cool, and I have a
lot of great memories of those times.
The Living Computer Museum (http://www.livingcomputermuseum.org)  in
Seattle, WA, has rescued a smaller version of a system like this based
on the 6500 processor that is undergoing restoration.  

Sorry for changing the subject (but at least I updated it in the
Subject: line).

-Rick
--
Rick Bensene
The Old Calculator Museum
http://oldcalculatormuseum.com