New Storage Device is Very Small, as in 12 Atoms

nytimes.com

37 points·robg··6 comments

12 atoms is the quantity used to store a single bit (~100 atoms / byte) near absolute zero. As usual that important nugget was buried in the middle of the article; just reading the first 3 paragraphs, one would think they had fit an entire floppy into 12 atoms.

I'm not sure how you got that impression. I only skimmed the article, and picked up the one bit per 12 atoms detail, thinking, "Well, it's not quite memory diamond, but it's as close as we've ever come."

I think it's worth noting that this was an experiment done as part of a research paper. The HN title basically implies that it's something ready for production. Nonetheless very impressive.

Extrapolating this, wouldn't that mean we could possibly get as much as 6.02e23 / (12 + C) bytes per mol of storage material?

Where C is the number of atoms required to communicate with that segment of memory. It might be (6.02e23 / 12) + C, even!

That's a terrifying 5,569,618.84 petabytes for what could amount to a few hundred grams of materials (assuming it's a compound, and not just iron (at approx 56 grams a mol)).

Perhaps even many thousand petabytes in a storage space the size of a paperclip.

The potential is incredible; and it makes me worry: what will developers waste that space on next?

The article is about arranging atoms on a surface with a scanning tunnelling microscope. You you don't seem to be counting this required surface it in your "storage material". And that's still ignoring all the machinery needed to actually arrange/detect the atoms.

The reading material was considered with the C at the end of the calculation; but even still, if it takes 100 times as much equipment to read and write as it does to store, those are still ridiculous numbers. That was my point.

Reply on news.ycombinator.com