08-25-2003, 12:00 AM
|
|
Editor Emeritus
Join Date: Aug 2006
Posts: 15,171
|
|
MRAM Coming To A Device Near You?
When you think about it, RAM technology really hasn't changed much in the last ten years or so. We've got dynamic, static, and Flash memory. DRAM is somewhat fast and volatile. Static is very fast, volatile, but expensive. Flash is non-volatile, but slow, expensive, and less reliable. Motorola and IBM think that MRAM (magnetoresistive random access memory), which is cheaper, faster, yet non-volatile, might be the answer.
"Unlike conventional high-speed memory devices, MRAM uses magnetism instead of electrical charges to store data -- making it, in a sense, a back-to-the-future technology based on the same laws of physics that enabled the creation of audio and videotape recorders as well as hard drives."
Fluffy comments aside, the implications could be enormous -- MRAM could do things from powering the next-generation of solid state drives to providing memory to PDAs and laptops that reduces power requirements. If you're curious, Howstuffworks has a useful article on the subject.
|
| |
|
|
|
08-25-2003, 01:34 AM
|
|
Pupil
Join Date: Jun 2003
Posts: 47
|
|
Interesting article about different RAM architectures, including magnetic technologies MRAM and FeRAM http://organics.eecs.berkeley.edu/~v...section7p6.pdf.
Quote:
FeRAM Operation
� Programming: Application of field switches the polarization
� Read is performed by applying a field. If a polarization switch occurs,
a large displacement current results, which can be sensed. Note that
the read is inherently destructive, so regenerative reading schemes are
used, similar to DRAMs.
MRAM Operation
� Uses a multilayer stack called a magnetic tunnel junction. This is used
in a spin valve configuration (similar to a hard GMR head), where the
magnetoresistance can be sensed and compared to a reference cell to
determine the state.
|
Looks like IBM is leveraging their hard drive research in a new area.
edit:
http://www.cypress.com/cfuploads/img...s/CY9C6264.pdf
This company's MRAM product, although just 8kB (8x8kbit) in size, mentiones 10 years of data retention, and a latency comparable DRAM devices (70ns). That might not be fast enough for a main memory, even in a PDA, but it's order of magnitudes faster than anything possible with the flash memories currently. It's also worthwhile to remember this baby can take potentially trillions of rewrite cycles compared to just 100 000 rewrite cycles a flash memory can withstand.
Very impressive stuff.
|
| |
|
|
|
08-25-2003, 01:52 AM
|
|
Thinker
Join Date: Apr 2004
Posts: 312
|
|
How would this effect...
How would this effect x-ray machines and walk through airport scanners?
__________________
JohnJohn
PocketUp.com
- iPhone3G - Blackberry Curve
|
| |
|
|
|
08-25-2003, 01:57 AM
|
|
Pupil
Join Date: Jun 2003
Posts: 47
|
|
Re: How would this effect...
Quote:
|
Originally Posted by JohnJohn
How would this effect x-ray machines and walk through airport scanners?
|
As far as I can see, this would only be affected by very strong magnetic fields. So x-rays shouldn't be a problem. The packaging of the chip has probably some physical protection against the magnetic fields.
|
| |
|
|
|
08-25-2003, 03:17 AM
|
|
5000+ Posts? I Should OWN This Site!
Join Date: Jul 2003
Posts: 5,133
|
|
Ah, so I don't have to give up my fear of magnets. I keep looking at those PDA cases with magnetic closure, and it just feels *wrong*... 
|
| |
|
|
|
08-25-2003, 03:51 AM
|
|
Philosopher
Join Date: Aug 2002
Posts: 593
|
|
This will help laptops overcome PDA's. Magnetic memory could be temporary storage for hard drives during a shutdown, therefore allowing for fast start ups of where you left off. A big reason people don't get a laptop is fast turn on, this would eliminate start-up time.
|
| |
|
|
|
08-25-2003, 04:33 AM
|
|
Thinker
Join Date: Jul 2003
Posts: 443
|
|
Quote:
|
Originally Posted by Kati Compton
Ah, so I don't have to give up my fear of magnets. I keep looking at those PDA cases with magnetic closure, and it just feels *wrong*... |
PDAs don't use magnetic storage Kati. Though a strong magnet might affect certain types of batteries.
|
| |
|
|
|
08-25-2003, 04:48 AM
|
|
5000+ Posts? I Should OWN This Site!
Join Date: Jul 2003
Posts: 5,133
|
|
Quote:
|
Originally Posted by qmrq
PDAs don't use magnetic storage Kati. Though a strong magnet might affect certain types of batteries. |
I know - but if they started to use the MRAM, they might.
|
| |
|
|
|
08-25-2003, 01:05 PM
|
|
Ponderer
Join Date: Jan 2003
Posts: 50
|
|
It would take a pretty strong magnetic field to disturb the MRAM. Like... I wouldn't go taking your device into an MRI. But then again, that's a pretty moot point in itself.
This whole technology reminds me of an old piece of equipment that was in use until... maybe 5 years ago. It was a huge Chyron character generator at a television studio my father works for. From the '70s, this thing used magnetic core memory. Sure, it's a different beast. Completely different. But they used it up until it was replaced. Almost 30 years, it ran. And the memory was never really a problem. Not that it had much... and it was huge. Used 8 inch floppies. Huge.
Now, what did that have to do with the price of beans?
|
| |
|
|
|
08-25-2003, 08:16 PM
|
|
Intellectual
Join Date: Feb 2004
Posts: 123
|
|
Blech
I'm tired of hearing and reading about technologies that will never make it to production.
Harumph! Just face it: we're stuck with DRAM and slow hard drives forever.
|
| |
|
|
|
|
|
Linear Mode
