Litton Industries Monrobot XI computers available

From: Rick Bensene <rickb_at_bensene.com>
Date: Fri Feb 28 10:11:00 2003

The big part of the difference in speed between the PDP-8 and the
Monrobot XI is that the PDP-8
used random-access magnetic core memory as the main memory and a
parallel architecture, and the Monrobot machines used magnetic drum and
a bit-serial architecture.

In drum-based architectures, all operations, including instruction
fetch, operand fetch,
and (not sure with the Monrobot XI, but some machines did have all
internal registers
also located on the drum) register access had to wait for the drum to
rotate to the
proper location in order for the data to be read. In many cases, the
master timing
for the machine was generated by a dedicated pre-written clock track on
the drum.
Thus, the speed of the machine was limited by the rate of rotation of
the drum and
the density of the clock track.

Along with all of the waiting time for the drum, the logic of the
machines of this ilk was
serial, which mated well with the fact that the data comes streaming off
the drum in bit-serial form (in most cases). Serial logic is
inexpensive...but slow
(observe the performance differences between a PDP-8/L and a PDP-8/S).

Another key difference between the early PDP computers and devices like
the Monrobot
machines is that the PDP was designed as a general-purpose computer,
with expandability
and a generalized I/O architecture to support a wide range of
interfaces.
Machines such as the Monrobot series were quite fixed in their
architecture. There
was no way to add more memory, or interface to more than a few external
devices.
The architecture was quite fixed.

The Monrobot series of machines started out as early desk-sized
sequence-controlled
electronic calculators, with decimal arithmetic, and math-oriented
functions.
These machines were originally designed for automating accounting and
bookkeeping operations.
Some made their way into engineering and scientific persuits, though.

The machines had little in the way of bit manipulation functions. As
time and the marketplace went on,
the Monroe machines (and others of the time, such as those made by Smith
Corona Marchant (SCM),
and Friden) began to take on more computer-like
functionality...switching from decimal-based
arithmetic to true binary, adding boolean logic functions, and providing
more sophisticated
branch and looping functionality...moving the machines from the realm of
sequence-controlled
calculators to true computer capabilities.

Rick Bensene
The Old Calculator Web Museum
http://www.geocities.com/oldcalculators
Received on Fri Feb 28 2003 - 10:11:00 GMT

This archive was generated by hypermail 2.3.0 : Fri Oct 10 2014 - 23:35:57 BST