DECsystem-10/DECSYSTEM-20

[Aron Insinga <ainsinga@infomation.com>]

I had to dig up a few URLs for someone else, and
they are very slightly connected to Project DELTA,
so be prepared to hit your mail reader's delete button.
(Fans of tcsh might want to go see #2!!)

<http://www.ai.mit.edu/people/tk/pdp6/pdp6.html>
Tom Knight's web page has large PostScript files for the PDP-6
schematics and print set, and some nice pictures.
This was the ancestor of the PDP-10, which came out in 1964.
(When UNIX was new, DEC suggested that it be ported to the PDP-10,
and the reply from Bell Labs reportedly was *something like*
"We aren't interested in a 1964 architecture.")
It had problems due to the design switching in mid-stream
from Ge to Si transistors, and the limited pin-out of DEC System Modules
requiring certain bit-slice modules having pins on both ends with a flaky
cable used like a 2nd backplane, and hand-soldering large machines
is tough to do.  So it inspired DEC to use wire-wrap backplanes for later
machines, and Dave Gross' program that generated the wire-wrap
machine's card deck had to be run twice on the -6 to be sure that the
-6 ran correctly for that long; its run time was about the -6 MTBF.
Dave Gross said that it was he who coined the term "kludge tower"
to describe DEC's CAD system as not just a kludge, but kludges
piled on top of kludges.)

<http://www.inwap.com/pdp10/paper-t20.txt>
Dan Murphy's history of TOPS-20 has escaped the
world of VAX NOTES behind DEC's firewall.

<http://www.xkl.com/>
The place where you can buy a DECSYSTEM-20 clone today.
Some of the people who made a pile of money at CISCO
wanted to do something that they thought would be really fun.
(For sufficiently strange values of fun! :-)

- Aron