Magnetic Core Memory Reborn (2011) | Hacker News Reader

Magnetic Core Memory Reborn (2011)

corememoryshield.com

65 pointsgalaktor8 comments

Neat project.

Interestingly enough, I often use the core memory patent as an example of a true invention worthy of being issued a patent. If there's an invention that was, at the time, new, useful and non-obvious this was it.

Here's that patent: http://www.google.com/patents/US2667542

One of my favorite examples of the opposite, that is, obtaining a patent for something that is not an invention is this MIT patent: http://www.google.com/patents/US5650704

Yes, they patented a motor connected to a spring with a force sensor to control, well, force. You know the old F = kx spring formula. Not to beat MIT to death --they do a lot of great stuff that I'll never understand-- but filing for a patent on a motor with a spring? The real joke is that the patent office granted it.

Don't get me started on software patents. OT for this thread anyway.

Wikipedia cites Professor Jay Forrester of MIT as the original holder of the core memory patent. I believe this is the patent from Forrester: http://www.google.com/patents?id=MNxYAAAAEBAJ&pg=PA1&...

MIT and IBM entered into litigation over the patent in the early 1960s and Forrester eventually won the dispute.

Detailed article (also linked on the original page) http://www.corememoryshield.com/report.html

I wonder if read and write times on-par, or better than, DRAM's are achievable with such modern core memory.

Cool project, btw :-)

Most on-die embedded DRAM is one cycle away for a whole word (I'm not sure about arduino but I'll bet it's in the same ballpark). This uses a 500 ns pulse for each bit, so reading a 8-bit word would take 40 clock cycles (at 10 mHz)

No, you can't get anywhere close to modern DRAM with this type of core memory. FRAM chips [1] which use the principle of Ferrormagnetism to retain data while power is off are fairly quick however. They can replace the typical serial flash or EEPROMs you might use to save settings, and they have the benefit of unlimited write cycles and faster (than flash) write times.

[1] http://www.ramtron.com/about-us/what-is-f-ram.aspx

The sheer size of the core memory, e.g. measured in millimeters, will prevent it from having access times in comparable orders of magnitude as silicon-based DRAMs, whose dimensions are measured in micrometers.

I think it was tested at 200Kb/s throughput, but with tweaks that could be doubled

Reply on news.ycombinator.com