This site may earn affiliate commissions from the links on this page. Terms of use.

Mainframe computers may seem quaint at present, merely they ruled the calculator manufacture for decades outset in the 1950s. A single company dominated the industry for much of that time: IBM. One of the visitor's smaller, but critically important systems was the IBM 1401, a data processing solution beginning produced in 1959. These low-cost systems (rented for roughly $20K a month in 2015 dollars) relied on a six-bit compages and utilized ferrite core memory.

Ferrite core memory relies on wires threaded to a serial of toroid rings to fix a charge that's either clockwise or counter-clockwise. $.25 tin can be set to either ane or 0 in this fashion, though the process of reading a flake also resets it (called destructive readout). Unlike modernistic RAM, cores maintained their accuse land when ability was deactivated. And dissimilar modern SRAM, core memory was typically laid out and constructed by paw, which placed sharp limits on its ability to scale. A contempo article dives into the challenges of troubleshooting this obsolete retention blazon and the IBM 1401 itself, and information technology'due south a fascinating look at how different repairing a reckoner used to be.

The IBM 1401'due south core memory is shown above. Each eighty×l grid of wires contains 4,000 retentiveness cores. Each of these plates contains 500 bytes of memory; the 1401'southward 4K of retentivity consisted of eight base of operations plates, and in that location were 16 plates in the 1406 add-on module. All of the failing planes in this case were located in the same contiguous expanse of 4K RAM (the 1401 predates the eight-chip standard we use today; information technology used half dozen-scrap characters). As the article notes, the addresses that were failing all ended in ii, 4, and vi — merely disentangling why those addresses were declining took boosted time and some intimate noesis of the system's memory encoding scheme.

Eight of these boarsd together made up 4K of RAM. Image by Ken Shirriff

Eight of these boards together made up 4K of RAM. Image by Ken Shirriff

What's particularly interesting almost this system is the degree of documentation IBM provided — hundreds of printed sheets detailing the exact function of every carte, wire, and system in the calculator — and the sheer physicality of the various components. With just one-2 transistors per Standard Modular System card, you can see, literally, every component that hooks to every other part of the system. Even the memory is laid out in wire and available for inspection. Every component of the system is documented on paper, even if the aged blueprints are incredibly difficult to read and run to thousands of pages.

AEM 0001

Two transistors (capped in red). Image by Ken Shirriff.

Fifty-fifty by the early 1960s, it was obvious that some of these trends could not persist, but one of them — the use of ferrite core retentiveness — hung on until the introduction of DRAM in the early 1970s. The wire layouts and designs of ferrite core memory presented barriers to scaling that miniaturization couldn't overcome, since nigh core memory was handmade. Today, we can pack 4K of RAM into structures that strain the limits of the Mark One Eyeball. But it's fascinating to run across the relics of another time, when the construction and logic of computing was something anyone could observe without so much as a microscope.