• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Bake or broil SSDs?

VirtualLarry

No Lifer
What happened to that discovery that flash memory could "heal" itself if heated? Where are the drives with a "bake or broil" setting, and a 1Mil. write lifespan?
 
Same reason we don't have a lot of things.
No sense of urgency.
If we needed electric cars, we'd have widespread development, sales and use of them.
 
Last edited:
Is this some kind of phase change memory technology? Like based on the memristor?

No, standard NAND, but with a "heater" built-in. According to the article link that was posted here, when the NAND ran out of write cycles (3000), all that had to be done was heat the NAND to a certain temp briefly, and it would be "healed", and good for another 3000 cycles.

It sounded like very promising technology, to give us nearly limitless writes on NAND.

Perhaps the industry doesn't want that, they WANT SSDs to "wear out", so that they can sell us another one.

If they didn't, people could buy one of these "super NAND" SSDs, for each of their rigs, and potentially use it for the rest of their lives.
 
No, standard NAND, but with a "heater" built-in. According to the article link that was posted here, when the NAND ran out of write cycles (3000), all that had to be done was heat the NAND to a certain temp briefly, and it would be "healed", and good for another 3000 cycles.

It sounded like very promising technology, to give us nearly limitless writes on NAND.

Perhaps the industry doesn't want that, they WANT SSDs to "wear out", so that they can sell us another one.

If they didn't, people could buy one of these "super NAND" SSDs, for each of their rigs, and potentially use it for the rest of their lives.

The hard drive industry has survived with infinite write cycles.
 
The hard drive industry has survived with infinite write cycles.
But imagine what they could potentially do to their bottom line if they could cap that?

The gaming industry is trying to make it difficult or even illegal to resell a used game. Why? Certainly not for the benefit of consumers - it's purely for their own benefit.

Unfortunately, the economic driver of competition, while generally a great boon, can also prod companies to compete on lowering the bar.
 
SSDs are already expensive and would just get even more so. Why would you add this heating tech if probably 99.9% of buyers are fine with 3000 write cycles?
 
interesting that everyone has that 3000 write cycles to talk about............My SSD says MTBF 2,000,000 hours ...........and that is just the failure, not the life........................maybe just Marketing http://ocz.com/consumer/vertex-3-sata-3-ssd/specifications
Business part MBTF can often be interpreted as an expected near-term failure rate (IE, MBTF/quantity = yearly replacement expectation). Outside of that, it's mostly a number made high enough to look good, but not so high that it's obvious it hasn't come from any meaningful testing.

It makes real sense for things like brushed motors, bearings that handle unbalanced loads, capacitord, and so on, but anything as mechanically and electrically complex as a HDD or SSD...no way.
 
What happened to that discovery that flash memory could "heal" itself if heated? Where are the drives with a "bake or broil" setting, and a 1Mil. write lifespan?

Over the last many decades, I can't remember a single invention (from memory), reportedly being invented in labs somewhere, which later hit the market.

There are all sorts of practicalities, and reasons why.

They often take so long to be ready for the market, that capacities/speed/reliability/low-cost of the competing ones, make it too late to sell.

There can be all sorts of technical difficulties to solve, before putting it on the market.

Huge amounts of R&D money have to be made available, somehow.

There can be huge patent issues.
E.g. The main competitor, holds on to several similar patents, and would have to agree, effectively halting/delaying proceedings.

Then there is the market.
It would probably cost more to make oven-refreshing-flash, quite possibly a substantial amount more.
So, if the market is happy to pay $100 for xx-Gb ones, standard flash SSD. Life = 10,000 erase/write cycles.

How much more would the market pay for xx-Gb flash SSD, with life = 100,000,000 erase/write cycles ?

I think most people would buy the cheaper, 10,000 life ones, so without capturing the bulk of flash sales, and with (probably) significant or huge manufacturing and R&D costs, it would be difficult to justify on business grounds.

Looking back over the years, I think I can remember when the CD was invented (just about), which did become a market success.
 
Business part MBTF can often be interpreted as an expected near-term failure rate (IE, MBTF/quantity = yearly replacement expectation). Outside of that, it's mostly a number made high enough to look good, but not so high that it's obvious it hasn't come from any meaningful testing.

It makes real sense for things like brushed motors, bearings that handle unbalanced loads, capacitord, and so on, but anything as mechanically and electrically complex as a HDD or SSD...no way.

Not least because the mtbf has zero to do with the number of write cycles that the NAND can withstand without degrading to the point that it needs to be replaced/heated/whatever.
 
Back
Top