Intel (INTC) , Micron (MU) Unveil New 3D XPoint Technology

[orangecat]

Apparently there's a new storage technology faster than the fastest flash...

 

http://www.cnet.com/news/intel-and-micron-debut-3d-xpoint-storage-technology-thats-1000-times-faster-than-existing-drives/

 

According to the article...

 

"The companies claim that 3D XPoint is a major breakthrough in memory process technology, the first new memory category since the introduction of NAND flash in 1989. It's said to be extremely fast and durable, up to a thousand times faster (both in read and write speeds), and it will have higher endurance than existing NAND Flash memory currently being used in SSDs."

 

"What's more, it also has as much as 10 times greater density, leading to much more storage capacity in the same physical space, while remaining as energy efficient and affordable as existing NAND flash memory."

[Bob Jim]

Please... Make sure nobody else has posted this hours and hours before you next time...

[Aytex]

starting price at 10.4 trillion dollers

[Misanthrope]

Not even sure we can max out that much speed with current processing power anyway, but nice to know that the future still holds amazing advances.

[orangecat]

Not even sure we can max out that much speed with current processing power anyway, but nice to know that the future still holds amazing advances.

According to the article current motherboard and PCI-e aren't fast enough and will likely require an entirely new architecture to work.

[orangecat]

starting price at 10.4 trillion dollers

Hopefully it won't be like that for long seeing as it's better than typical NAND flash in nearly every way.

[AnonymousGuy]

Yeah yeah yeah let me know when it actually goes on sale.

 

I'm sure there's a lab somewhere that has made graphene nanotube jism batteries that last 4 years on a single charge, doesn't mean it's coming any time soon.

 

Mass production is always the challenge.

[Thogert]

As someone who works in the fab that produces this, mass production isn't going to be a problem.

[2FA]

Please... Make sure nobody else has posted this hours and hours before you next time...

 

ikr, I've seen at least four other reposts today.

[patrickjp93]

I meant the standard itself, AHCI ended up not being good enough for NAND based SSD, let alone SSD with the new NV memory. (I should have worded my original comment differently).

It was partly my fault. I came back from dinner and drinks with my Dad and Grandmother, hence my hesitation with the second stanza of that post. I'd had a tall apple ale, a tall Golden Monkey (Belgian Tripel), and a strong rum & coke before I'd logged on. I was a bit fuzzy. You were fine.

 

I think as a standard, it's behind what HPC could make NAND do if given the chance. With a quad-core controller of A57 cores or something of the like, you could very well design an SSD to have near 32Gb/s read and write which is the PCIe 3.0 4x limit.

[AnonymousGuy]

As someone who works in the fab that produces this, mass production isn't going to be a problem.

 

But if you're a fab tech you don't know the yields

 

Tons of Broadwell wafers were being made..doesn't mean they didn't look like someone spilled ketchup all over the wafer map

[patrickjp93]

Edit: Don't know how that ended up on both threads, but okay.

[Thogert]

I actually have very good info on the yields, as with all people in our fab that have anything to do with the product.They like the people who deal with the actual wafers to know what kind of impact we are having on them. 

 

Performance of the architecture and that kind of thing are above my pay-grade, but that's a different story.