memory memories

Dave Long (
Thu, 14 Oct 1999 17:59:26 -0700

Nature | 1 Oct 1949
(also in the "50 years ago" of vol 401, 30 Sep 1999)
> Except for the input and output mechanism, the EDSAC (electronic
> delay storage automatic calculator) has no moving parts, all
> computing and control operations being performed by means of
> electronic circuits. Within the machine numbers are expressed by
> trains of pulses ... Numbers expressed in this form are stored by a
> method depending upon the use of an ultrasonic quartz delay unit.
> The pulses are applied to a quartz crystal mounted at one end of a
> column of mercury, and give rise to ultrasonic pulses which travel
> through the mercury with the velocity of sound. On arrival at the
> far end they strike a second quartz crystal and are reconverted into
> electrical pulses. The time taken to traverse a column of mercury
> 5 feet long is about 1 millisecond, and the interval between the
> beginning of one pulse and the beginning of the next is 2
> microseconds; there can thus be as many as 500 pulses passing down
> the column at any one time. On emerging from the delay unit the
> pulses are amplified and reshaped, and passed back to the input of
> the delay unit. They then continue to circulate indefinitely and
> are available when required.

If you wish to get the flavor of computing a half-century ago, when
the difference between warm and cold boots was literal, there
appears to be an EDSAC simulator at:

Ivars Peterson's Mathland, "Computing with the EDSAC"
> Wilkes recalls in his memoirs, "By June 1949, people had begun to
> realize that it was not so easy to get a program right as had at one
> time appeared. I well remember when this realization first came on
> me with full force. The EDSAC was on the top floor of the building and
> the tape-punching and editing equipment one floor below.... I was
> trying to get working my first nontrivial program.... It was on one of
> my journeys between the EDSAC room and the punching equipment that ...
> the realization came over me with full force that a good part of the
> remainder of my life was going to be spent in finding errors in my own
> programs."