Keith corrected me on this which is ok since I had an Old (brain) fart! It's nice to see someone who knows or remembers, and I find this reminder very interesting and educational. On Mon, 2004-05-03 at 19:23, Keith Lofstrom wrote: > You wrote: > > For those who don't know what core memory is, it's miniature ferrite > > beads strung together with three wires through them to read/write > > [magnetize and demagnetize], and reset if I remember correctly. I have > > seen 16 MB of core memory, it was the size of a full size tower case. > > Some corrections of fact, which I will share with you privately. > You can research and post; it is always better to correct yourself > than have someone else do it publically. > > The three wires in a core memory are X select, Y select, and sense. > The cores work by pushing an X current and a Y current through an > intersection; two currents pushed the hysteretic toroidal magnetic > core past a threshold, either causing it to flip magnetic polarity > (generating a little voltage blip on the sense line) or not, depending > on previous magnetization. After flipping a core during a read, you > had to schedule a reverse set of currents on the same X and Y wires to > put the core back. The size of the arrays were limited by the number > of cores you could put on a sense wire before accumulated noise made > reading unreliable. > > A friend in high school wanted to build a core memory, and wrote a letter > to Honeywell. They sent him a one pound jar of 15mil cores. All out > of spec, as near as we could measure. > > What you probably saw was a 16*K* byte memory, the boards from the late > 60s were typically around 16K bits, 8 boards to a byte, and about as big > on a side as a tower case with the small scale integration driver chips > around the sides of the core fabric. A 16MB machine would have been > gargantuan, indeed; I would guess 100 or more 19inch racks, 8 feet tall, > and probably costing on the order of $50M in current dollars. > > Keith -- jludwig <wralphie@xxxxxxxxxxx>
