MRAM – The Birth of the Super Memory

[iddo]

Hello All,
We have just published a new article on the future of Freescale's MRAM memory. Although at the current 25$ per 0.5MB it is still far from mainstream computing future lithography might bring the price to a reasonable level in a few years time - allowing for ultra fast computer memory which is non-volatile.

It will be interesting to hear what you think of MRAM as the future of computer memory.

Iddo

 

[stephen1960]

Hello All,
We have just published a new article on the future of Freescale's MRAM memory. Although at the current 25$ per 0.5MB it is still far from mainstream computing future lithography might bring the price to a reasonable level in a few years time - allowing for ultra fast computer memory which is non-volatile.

It will be interesting to hear what you think of MRAM as the future of computer memory.

Iddo

I have yet to read the article but I don't believe the volatility of main memory is a detriment. Is it? But if it is ultra fast, we can all smile at that.

Perhaps I will look into this new memory, I do try to keep up so I'm not entirely without a clue.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
I read the post. Right now, Freescale's MRAM memory is too slow for main system memory and much too expensive but they are targeting some special use areas.

Now then, as in every new technology it is the ideas, the interesting parts and bits of the theory and implementation of such that is of real value and it seems to me there are some very interesting parts and bits here.

In particular is the tunneling concept put to use in the write process, no electrons are moved across the dielectric in the usual sense of the word and thus no power is needing dissipation. This is exactly what is needed in our computer components. It is the moving to and fro of electrons (a type of lepton, see http://www.particleadventure.org/index.html) that gives rise to the inductance issues and the power dissipation issues.

In brief, a wonderful and excellent concept and implementation worth looking at. I suspect that we, as enthusiasts, won't see much of MRAM in it's current state of evolution but I do hope to see it grow into something cool and wonderful.