4 ms·

Ah, neat. If I'm reading this right, then the real big benefit of these things would be solid state magnetic storage. The benefit of these things is they don'

Ah, neat. If I'm reading this right, then the real big benefit of these things would be solid state magnetic storage. The benefit of these things is they don't create a magnetic field while they do respond to magnetic fields. That means you can pretty tightly pack these things together without concern that they'll interact with each other. A light electric pulse could determine if the bit is a 1 or a zero and a strong pulse would flip the bit. I'm guessing that due to this nature, these things would actually have pretty long shelf lives and near infinite read/write cycles since you are, effectively, just flipping atoms around and not actually breaking structures or dumping in charge. These should mostly work with regular silicon manufacturing. The tricky part will be how tightly you can pack these things together before the reading structures start interfering with each other.

imdsm 1y ago
> A light electric pulse could determine if the bit is a 1 or a zero and a strong pulse would flip the bit. Feynman moment. Breaking it down into one sentence. Bravo!
- blks 1y ago
  [flagged]
- OhNoNotAgain_99 1y ago
  That line is worth a dollar of subscription,
  - willlma 1y ago
    I had the same thought with a recent NewScientist article. Signed up, read it, and tried to cancel to avoid the recurring fee. There's no click to cancel; I had to submit a form + request and repeatedly check my email to see if my request had been honored. I'm still waiting for the perfect pay-per-article platform to show up.
pedro_caetano 1y ago
> If I'm reading this right, then the real big benefit of these things would be solid state magnetic storage. Wouldn't this also enable a much higher resolution and better noise immunity for the entire zoo of industry sensors that are based on the Hall effect?
foota 1y ago
Note that you can actually flip the magnetic field of certain "normal" magnetic materials: https://www.nature.com/articles/s41563-022-01354-7 https://www.nature.com/articles/s41563-022-01354-7
- olddustytrail 1y ago
  That's kind of what the article is talking about. You can do it in a ferromagnetic material (like in your link) but then you've got the problem that it's a magnet and screwing up everything around it. The idea here is that you can do it in something that is overall neutral.
fc417fc802 1y ago
> these things would actually have pretty long shelf lives and near infinite read/write cycles since you are, effectively, just flipping atoms around and not actually breaking structures or dumping in charge. Didn't we already play this exact scenario out with 3D XPoint? Is there any expectation that this new type of magnet will be able to compete with flash memory on a unit cost basis? Actually I guess data retention when unpowered was one year or less but still, is that factor ultimately what led to its discontinuation? I doubt it. For long term data retention you're competing with magnetic tape which has a shelf life measured in decades and an impressively low cost per bit.
- vlovich123 1y ago
  3D XPoint wasn't based on magnetic storage and was intentionally targeting a point between NVRAM and DRAM both in terms of price and performance. BTW, I'm assuming you know but spinning disks are magnetic storage and even today are cheaper than flash memory on a unit cost basis. Generally there's a pretty clear tradeoff you make on all of these primarily around price vs performance (durability tends to be secondary until you get to tape storage). > Actually I guess data retention when unpowered was one year or less but still, is that factor ultimately what led to its discontinuation? I am unable to find any actual specific numbers for flash or XPoint. There's just a bunch of unsubstantiated rumors ranging from 3 months, to 1 year, to 5 years, to 10 years. The factors that led to discontinuation is the same as for any product - the market for it never materialized and the cost of continued investment wasn't worth it. Even if it was profitable (not sure), there is always still the opportunity cost of investing that money in a higher ROI venture. Of course, this is Intel who continually failed to sustain long-term payoffs and strategic investments.
  - fc417fc802 1y ago
    Yes, I'm well aware that it isn't magnetic. I'm not sure what your point is? The typical user doesn't care about the technology behind a given storage medium, only the performance characteristics such as cost, density, latency, cycle durability, power draw, environmental hardiness, and retention time. > spinning disks are magnetic storage and even today are cheaper than flash memory on a unit cost basis Presumably not for much longer though. Unless there has been a major upset recently? > the market for it never materialized and the cost of continued investment wasn't worth it An entirely empty statement. I was suggesting that unpowered retention time is almost certainly not what led to the failure. The point I'm making is that consumers are impossibly cost conscious in this area. We already tried the pitch of "clearly technically superior in every metric that matters, but noticeably more expensive" and the market soundly rejected it (much to my eternal dismay). So again I'll ask, is there any reason to expect this new type of magnet to be able to compete with flash memory on a unit cost basis? Because if not I assume it's doomed.

imdsm 1y ago

> A light electric pulse could determine if the bit is a 1 or a zero and a strong pulse would flip the bit. Feynman moment. Breaking it down into one sentence. Bravo!

blks 1y ago
[flagged]
OhNoNotAgain_99 1y ago
That line is worth a dollar of subscription,
- willlma 1y ago
  I had the same thought with a recent NewScientist article. Signed up, read it, and tried to cancel to avoid the recurring fee. There's no click to cancel; I had to submit a form + request and repeatedly check my email to see if my request had been honored. I'm still waiting for the perfect pay-per-article platform to show up.

pedro_caetano 1y ago

> If I'm reading this right, then the real big benefit of these things would be solid state magnetic storage. Wouldn't this also enable a much higher resolution and better noise immunity for the entire zoo of industry sensors that are based on the Hall effect?

foota 1y ago

Note that you can actually flip the magnetic field of certain "normal" magnetic materials: https://www.nature.com/articles/s41563-022-01354-7 https://www.nature.com/articles/s41563-022-01354-7

olddustytrail 1y ago
That's kind of what the article is talking about. You can do it in a ferromagnetic material (like in your link) but then you've got the problem that it's a magnet and screwing up everything around it. The idea here is that you can do it in something that is overall neutral.

fc417fc802 1y ago

> these things would actually have pretty long shelf lives and near infinite read/write cycles since you are, effectively, just flipping atoms around and not actually breaking structures or dumping in charge. Didn't we already play this exact scenario out with 3D XPoint? Is there any expectation that this new type of magnet will be able to compete with flash memory on a unit cost basis? Actually I guess data retention when unpowered was one year or less but still, is that factor ultimately what led to its discontinuation? I doubt it. For long term data retention you're competing with magnetic tape which has a shelf life measured in decades and an impressively low cost per bit.

vlovich123 1y ago
3D XPoint wasn't based on magnetic storage and was intentionally targeting a point between NVRAM and DRAM both in terms of price and performance. BTW, I'm assuming you know but spinning disks are magnetic storage and even today are cheaper than flash memory on a unit cost basis. Generally there's a pretty clear tradeoff you make on all of these primarily around price vs performance (durability tends to be secondary until you get to tape storage). > Actually I guess data retention when unpowered was one year or less but still, is that factor ultimately what led to its discontinuation? I am unable to find any actual specific numbers for flash or XPoint. There's just a bunch of unsubstantiated rumors ranging from 3 months, to 1 year, to 5 years, to 10 years. The factors that led to discontinuation is the same as for any product - the market for it never materialized and the cost of continued investment wasn't worth it. Even if it was profitable (not sure), there is always still the opportunity cost of investing that money in a higher ROI venture. Of course, this is Intel who continually failed to sustain long-term payoffs and strategic investments.
- fc417fc802 1y ago
  Yes, I'm well aware that it isn't magnetic. I'm not sure what your point is? The typical user doesn't care about the technology behind a given storage medium, only the performance characteristics such as cost, density, latency, cycle durability, power draw, environmental hardiness, and retention time. > spinning disks are magnetic storage and even today are cheaper than flash memory on a unit cost basis Presumably not for much longer though. Unless there has been a major upset recently? > the market for it never materialized and the cost of continued investment wasn't worth it An entirely empty statement. I was suggesting that unpowered retention time is almost certainly not what led to the failure. The point I'm making is that consumers are impossibly cost conscious in this area. We already tried the pitch of "clearly technically superior in every metric that matters, but noticeably more expensive" and the market soundly rejected it (much to my eternal dismay). So again I'll ask, is there any reason to expect this new type of magnet to be able to compete with flash memory on a unit cost basis? Because if not I assume it's doomed.