Remembering Memory

A recent visit to MIT’s campus museum reminded me about the way memory was implemented on early computers (50’s & 60’s).

“Core” memory, as it was called, was composed of a matrix of little iron doughnuts. Each tiny core had two wires passing through it, one on the x axis, and the other on the y. By passing current through these wires the computer of the day could magnetize the core in either a clockwise or a counter-clockwise direction. In this way each core could represent a 1 or a zero.

In the picture below, we see 64K words (probably 128K bytes; we don’t know the word size of the computer) of core memory.

What is remarkable of course is that the memory in the photograph occupies about 27 cubic feet of space and probably weighs over 50 pounds. The cell phone I used to take these pictures contains about one million times that much memory!

One benefit of core memory, however, is that it is non-volatile. This means that, in the event of a power failure, the cores remain magnetized and no data is lost. Modern computer technology employs RAM memory which is erased whenever the power is cut.

Your comments are welcome.

Home