MRAM faster than DRAM?

I read that MRAM is many times faster than DRAM and used in an sattelite.

formatting link

If MRAM is so fast then why not use it in Computers? Can we not use MRAM in Digital Cammera and other devices to make them faster?

Bye Sanny

Reply to
Sanny
Loading thread data ...
Reply to
Brendan Gillatt

I think the main reason it's used in satellites is that it's rad-hard (ie, unlike a normal memory cell, it is much less likely that cosmic rays will corrupt its contents).

I have used it and yes, it's fast, dunno if it's faster than DRAM etc though! What the article says is is, it is faster than *Flash* for writing. I figured it was essentially as fast as SRAM, but nonvolatile (ie, the data stays there when the power goes off).

MRAM has been for sale for many years under the name FRAM. For info about FRAMs look at

formatting link
a small American company who have been making them for over a decade. The main problem with their product is that the memories don't come in large sizes - they haven't had the financial resources to make (8Mbit? Whatever the current Flash sizes are) so they've had to aim for "small, specialist, cost-no-object" niches. The sizes are too small for camera memories.

FRAMs have been around a long time and Hitachi, for example, were very interested in them a decade ago and promised an imminent release of some products. Then one of our periodic semicinductor industry contractions occurred and they decided to focus elsewhere (DRAMS I think). Now FRAMS are back in fashion - Texas is into them too, I suspect they've licensed Ramtron's technology - but they tend to be called MRAMs by other companies, I am not sure if there is a subtle difference between MRAM and FRAM or if this is simply a marketing ploy.

What a lot of people want is FRAM built into microcontrollers in place of Flash. The first company to market that will clean up a lot of applications.

--
Nemo
Reply to
Nemo

FRAM (or more commonly referred to as FeRAM) is a type of memory storage technology that is distinctly different from MRAM. It's not marketing.

MRAM is potentially a better technology than FeRAM, but FeRAM technology is far more mature at present.

formatting link

Reply to
davewang202

I stand corrected. I see now why I made that error, but it's irrelevant to the newsgroup.

Useful article, thanks.

I'm not quite clear why you say MRAM is potentially better than FRAM. I think the two technologies have similar speeds, and as far as I can see there are proven FRAM products out there in the marketplace, but MRAM ones are not available yet. From the Wiki articles, *theoretically* it looks like MRAM can be denser... but maybe FRAM can too, and it's a proven commercial product which is gradually improving.

--
Nemo
Reply to
Nemo

You'll forgive me as I will tend to be somewhat pedantic about correctness of nomenclature and details on memory devices - I taught a class on memory systems, and I also co-authored a book on the topic. (Doesn't mean that I don't make mistakes of my own of course - plenty of those to go around)

The statement is about the fundamental characteristic of the memory storage element - in terms of scalability etc.

A very concise, accurate, and up-to-date comparison of the various memory technologies can be found in the 2007 ITRS roadmap.

formatting link
(page 11)

MRAM technology senses current - or more accurately magnetic state differences in each cell. FeRAM technology senses charge levels on a ferrorelectric capacitor. MRAM promises to have much better scalability in the long run, more able to utilize advanced feature sizes, as stated in the ITRS roadmap.

OTOH, it takes more energy to do writes into an MRAM cell - even papers that proposes to lower MRAM cell write current levels still have energy levels that's much larger than existing FeRAM cells.

David

Reply to
davewang202

OK - I've used serial FRAMs (er, FeRAMs) so thought I knew what I was on about, but you have out-geeked me 8)

A pleasure to be corrected by a real authority, and so politely. Thanks. I am downloading the PDF you pointed at now to study it further.

In the kinds of applications I work with (read often, write occasionally), a higher write current isn't relevant, so I'd jump at MRAM's if they were cost effective.

--
Nemo
Reply to
Nemo

about, but you have out-geeked me 8)

downloading the PDF you pointed at now to study it

higher write current isn't relevant, so I'd jump at

I like them for their durability. When used with serial interface you can't destroy them if your program goes nuts and writes to a location in a loop. EEPROM won't survive that for very long.

Another big plus is that writing takes a lot less time.

Mark

Reply to
TheM

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.