Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
"Intel and Micron provide several use cases for 3D Xpoint technology, suggesting it will let retailers quickly detect fraud patterns in financial transactions and allow healthcare researchers to process and analyze larger data sets in real time"

why do they need these ******** "use cases"

surely anyone can appreciate if it's 1000x faster, 10x denser AND more durable it's going to make everything better

*cough* SKYNET *cough*
 
Interesting:

  • 1000 times faster than NAND
  • 1000 times higher endurance than NAND
  • 10 times denser than DRAM
  • Bit addressable
  • Transistor-less
  • Based on some new material
Backed by Intel and Micron, this might go somewhere, but it sounds like they've departed from standard silicon process, so it will take time to get it cost competitive.

Kind of annoying to say this is the first new memory technology in 25 years, though. That small earthquake reported in Silicon Valley is just the simultaneous grumblings of every engineer who's made a new memory architecture...

[As soon as I saw the 16GB size for initial parts, I knew the forum would explode. It's like the number of the beast around here-- people get completely irrational.]
Worrying about how much RAM an application is using, may become an issue of the past. When writing applications developers may not need to care how much memory it is using or leaking etc. Because there simply won't be any memory.
There will always be memory, but if they use the same storage pool for RAM and (P)ROM then the contention for storage is going to become even tighter. Your available memory for execution is going to start to depend on how many apps you have installed, and how many pictures you've taken. How much free space will you need for your app to run? Have you taken too many pictures since you last checked?

I suspect Apple will still keep the memory segmented, just to avoid having to educate users. Most of my family doesn't understand the difference between their 8GB of RAM and their 2TB HDD...
And that's a good use for it and will continue to be one. However, many people have speculated that "everything" (more or less) will be in the "Cloud" in the future and one will not need much on-site storage. I found that unsafe at best and slow to boot. But this obviates that idea for now.
It's a continuum along the fast/cheap tradeoff... Registers, cache, SRAM, DRAM, [xPoint], NOR Flash, NAND Flash, HDD, Cloud... If xPoint gets cheap enough, SSD may drop off the list but you're not going to carry all of Wikipedia locally.
Now, the invention of DRAM that really brought volatile memory to the computers. (And SRAM, albeit it was somewhat non-volatile - just capacitor leakage so it lasted for hours or days)
Wasn't uncommon to put in a small lithium battery to keep the SRAM nonvolatile for years...

DRAM won because it was dense. If this turns out denser (and higher yield) than DRAM we could see DRAM go away and maybe a bigger cache to hide access latencies.

Or we could see nothing change-- there have been dozens of new memory technologies that never manage to succeed. I think the usual cause of failure though is that the companies pushing the new technologies don't have the process expertise to compete with the entrenched tech. Having both Intel and Micron behind this one could make the difference.
 
intel3dxpoint-800x441.jpg

latest
500px-T-800CPU.jpg

Hmmmm....
 
  • Like
Reactions: cosmos and Nicky G
So 1000X faster? What is the clock speed on the DRAM in the A8 SOC? Because 1000X faster NAND would result in local storage that is nearly as fast as DRAM. That's absolutely insane and means that you could possibly just use your local storage as DRAM if it's durable enough as well. At the very least you could heavily cache things to disk without issue. We probably won't see this until the iPhone 8 at the VERY earliest, but hopefully Macs will see it sooner! It feels good to see some breakthrough technology from Intel.


NAND Access times per read cell can be measured in uS (micro seconds) for Reads while Program/Erase are sometimes measured in mS. (Milliseconds).

Moving from mS to uS units are measure would result in orders of magnitude results....so 1000x isn't far off.

These numbers are measured from the best possible point.

DRAM write/read Access times are around 50nS to 150nS. While SRAM is around 10nS. So this still has a long ways to go.

Kind of annoying to say this is the first new memory technology in 25 years, though. That small earthquake reported in Silicon Valley is just the simultaneous grumblings of every engineer who's made a new memory architecture...

Is definitely not the first new memory technology in 25 years.
  • There's EEPROM
  • There's Flash Memory (Nor Flash or NAND Flash)
  • There's Resistive Ram or Re-RAM.
  • etc etc.
Publications shared through IEEE and other open sources. It takes years or decades to create a viable product that is a MFG cost effective solution.
 
  • Like
Reactions: retep42 and Nicky G
Interesting:

  • 1000 times faster than NAND
  • 1000 times higher endurance than NAND
  • 10 times denser than DRAM
  • Bit addressable
  • Transistor-less
  • Based on some new material

It doesn't say NEW it says Breakthrough Advances of material. Could be new flavor of Gallium Arsenide.
 
So what done this all mean for us folk? So I can save more Netflix picture shows on my phone right? I gots to have more faster phone calls. I don't get none reception where's I's living is.
My wife has a Samsung phone and it holds hundreds of pictures and picture shows.
 
There will always be memory, but if they use the same storage pool for RAM and (P)ROM then the contention for storage is going to become even tighter. Your available memory for execution is going to start to depend on how many apps you have installed, and how many pictures you've taken. How much free space will you need for your app to run? Have you taken too many pictures since you last checked?

I suspect Apple will still keep the memory segmented, just to avoid having to educate users. Most of my family doesn't understand the difference between their 8GB of RAM and their 2TB HDD...

Although, a large portion of what RAM is used for today is for caching resources such as images, audio, textures, executables and the like, none of which will have to be loaded with this new architecture, which would mean that you would only need a very small amount of space for execution. Most likely, however, the initial implementations will simply make use of the existing swap file. On a disk this fast, that swap file will fly, and it will be dramatically easier to implement until a more efficient architecture can be designed.
 
So what done this all mean for us folk? So I can save more Netflix picture shows on my phone right? I gots to have more faster phone calls. I don't get none reception where's I's living is.
My wife has a Samsung phone and it holds hundreds of pictures and picture shows.

Although phone implementation could reap the rewards of reduce bottlenecks in bandwidth from non-RAM based components.

It's more of a integration to a system, most likely the biggest gains (for INTEL) would be their chipset performance and also their Enterprise solution groups to offload the DRAM to XPOINT before moving to NAND. If the XPOINT does create a storage solution that 3D NAND can't complete and a price/GB that is acceptable then you can completely remove the NAND component.

I don't see this coming to mobile platforms for atleast 5 years. We just started 3D NAND. I also don't see Apple taking a risk on this product until is been proven. That can be said for any mobile manufacturer.
 
  • Like
Reactions: V.K.
Although phone implementation could reap the rewards of reduce bottlenecks in bandwidth from non-RAM based components.

It's more of a integration to a system, most likely the biggest gains (for INTEL) would be their chipset performance and also their Enterprise solution groups to offload the DRAM to XPOINT before moving to NAND. If the XPOINT does create a storage solution that 3D NAND can't complete and a price/GB that is acceptable then you can completely remove the NAND component.

I don't see this coming to mobile platforms for atleast 5 years. We just started 3D NAND. I also don't see Apple taking a risk on this product until is been proven. That can be said for any mobile manufacturer.

I see, pretty cool. Thanks for breaking that down for me.
 
In 25 years?

If we're still using this type of typical storage in 25 years, then we are stupider and less innovative than we think are are
 

...Intel and Micron say that 3D Xpoint was built to create a non-volatile, high-performance high-capacity storage and memory solution that was also affordable. In addition to being faster than NAND Flash, it's also 10 times denser than the DRAM chips used in computers, which means much more data can be stored closer to the processor and accessed quickly.

...3D Xpoint is designed to complement existing NAND Flash and DRAM options because it will likely be too expensive to serve as a standalone option at launch, but in the future, it has the potential to introduce dramatic performance improvements in mobile and desktop products.
...
They said it: Too expensive. Or in other words: We will see acceptable prices in 5, 10 or 15 years.
 
This could be the next "BIG THING" if what I'm reading is correct. The key word there is NON-VOLATILE. Regular system type ram has never been non-volatile and while it's been faster than either flash storage or SSD drives, you could never swap one for the other without dire consequences. In other words, SSD storage, fast as it is can't compare to regular memory speeds and using regular memory for storage would mean you'd have to keep it refreshed (power to it) making it largely useless for storage as it would be erased the moment it lost battery power and it would be a relative power hog. Here, however, you have the potential to replace FLASH, SSD and RAM all with this new type of memory. Imagine a computer that comes with 5TB of this stuff that is both RAM and STORAGE with no differentiation between the two and now speed drops transferring things. Games would never "load" in the traditional sense again as "loading" is moving data from storage to the main system memory so the CPU can manipulate it. Here, there wouldn't have to be a difference! Even external storage of this type would be as fast as the data bus lines could possibly move it (with current technology), making all current SSDs obsolete, etc. The only issue, of course, is PRICE. I'd imagine, however, that this stuff is going to be so popular that it will change the face of the entire computing industry within a few years time, unless it has a major manufacturing issue. The article mentioned it being designed to be affordable yet throws "too expensive at first" back in your face in the same article so I imagine that will be a limiting factor at first. But long term, storage may change entirely.

I think I speak for many here when I say "thank you" for putting this whole thing in semi-layman's terms and writing it cogently so that it's easy to grasp the potential significant of this technology!!
 
1000x faster at what, writes, reads, setup or latency. We don't have all the information yet, they could be comparing it to some USB dongle flash from a decade ago.
A decade old USB dongle? Really? They specifically said current NAND tech. And although they didn't say 1000x faster at what, it's probably safe to assume what they mean, considering read/write is what's been increasing in droves lately above anything else.
 
I think I am getting closer on the materials.

Given that Micron has strong relations with Boise State University, I wonder whether these 2 patents have something to do with it...

  • Peter Mullner, William B. Knowlton, Chad S. Watson, Magnetic Shape Memory Based Multi-State Memory, Submitted March 18, 2011, (US Patent No. 20,120,236,632, Sept. 20, 2012).
  • Peter Mullner and William B. Knowlton, Multi-state memory and multifunctional device based on magnetic shape-memory alloys, (Patent No. 7,710,766 – May 2010).
Source: http://nano.boisestate.edu/about-us/publications/#patents
 
  • Like
Reactions: V.K.
so does this replace the ssd and the system memory and make it one solution? is this a shift in hardware architecture? as in making the ssd and system memory the same component on a system board?

Yep! But not really a shift, more an evolutionary step. Modern CPUs and OS already treat RAM and system storage as essential one and the same (RAM is basically used as cache for the slower disk/SSD). Removal of RAM as a component would certainly make sense.
 
Don't really care about SSD but...

...can it be used for first level caches as well?

That would be...

wow.
 
I think I am getting closer on the materials.

Given that Micron has strong relations with Boise State University, I wonder whether these 2 patents have something to do with it...

  • Peter Mullner, William B. Knowlton, Chad S. Watson, Magnetic Shape Memory Based Multi-State Memory, Submitted March 18, 2011, (US Patent No. 20,120,236,632, Sept. 20, 2012).
  • Peter Mullner and William B. Knowlton, Multi-state memory and multifunctional device based on magnetic shape-memory alloys, (Patent No. 7,710,766 – May 2010).
Source: http://nano.boisestate.edu/about-us/publications/#patents

Beautiful! "...Ni—Mn—Ga micropillars..." (patent full text e.g. under http://patents.justia.com/patent/20120236632)
So, they are using a nickel-manganese-gallium alloy according to this patent. Maybe with a twist, but this is essentially the technology, I am sure.
 
This means jack and **** without understanding how bus speed improvements will actually allow us to take advantage of such massive improvements at a single system component level.
 
"For consumers, 3D Xpoint will allow faster interactive social media"

Huh? Sounds like someone in marketing was stretching for a way to tie faster memory to the largest common denominator. I imagine it went like "It will make computers and smartphones faster... what do regular non-engineer types do on those??? How about social media? Yeah they're all doing that. Let's go with that."
 
Yep! But not really a shift, more an evolutionary step. Modern CPUs and OS already treat RAM and system storage as essential one and the same (RAM is basically used as cache for the slower disk/SSD). Removal of RAM as a component would certainly make sense.

Considering the use of high-speed RAM and slower-speed disk has been a paradigm in IT for... decades... I'd probably not call it an "evolutionary" step, and frankly would be willing to label it as a massive shift, if these two layers could converge. But considering this announcement is just a BS marketing/PR move more than anything from what I can tell, it doesn't really matter.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.