A US scientist has developed a way to store binary data on dots 6nm in size - possibly leading to a one-square-inch chip holding 2TB of data. Jay Narayan is a materials science and engineering professor at North Carolina State University (NCSU) and works with nanodots, which are single crystals free of any defect. In a NCSU …
How many hours worth
of HD porn is that on a 1" square thingy you can put in your pocket?
The short answer
What has it got in its pockets!?!
Yet another thing to lose!
Every time I see a headline about a newer, smaller storage device that can hold god awful amounts of data, I can only think: which train is it going to be left on?
There's a reason why large amounts of data should be stored in an inconveniently large manner.
"""Every time I see a headline about a newer, smaller storage device that can hold god awful amounts of data, I can only think: which train is it going to be left on?
There's a reason why large amounts of data should be stored in an inconveniently large manner."""
Just pile as many of these things as you can into a standard 3.5" drive (2.5 for laptops, etc.)
Or take one of these little deals and wrap it in loads of plastic. Problem solved. Higher density storage is /always/ a good thing.
Easy to solve
Just ask the specialists at Dell to store it in a cardboard box. Or six.
Write Only Memory.
If the laser device is going to read / write to a target 6nm across, its wavelength has to be substantially less than 6nm. Now that is an interesting laser ...
6nm wavelength laser
While lasers of less than 6nm (such as an X-ray) are already available (if not quite practical for this application yet), does the wavelength really matter, or the abilty to focus the point of the laser to <6nm?
EVERY wavelength has a point where its focus is 0nm; Once every length of the wavelength as a matter of fact. The problem is that as you near that point, the power of the laser edges closer to 0... so a <6nm laser (somewhere between ultraviolet [10^-8] and x-ray [10^-10] - http://www.google.com/search?hl=en&rls=com.microsoft%3Aen-us&q=nanometer&aq=f&aqi=g10&aql=&oq=&gs_rfai=) *would* be most beneficial as that means the full power of the laser could be used, instead of attempting to focus it to a point where the wavelength starts to peter out, probably in some sort of standing wave.
It may be more economical to use an electron generator (like the old Triac flip/flop tubes), as that would easily be <6nm (as used on electron microscopes). Then, as always, its just down to a convenient power supply.
Or find a clever way to ensure the photons behave as particles, which would be more towards the 10^-10 size, no matter the wavelength...
That or calibrated so somewhere along its wave, it is able to predictably make contact with the proper 6nm bit, perhaps at a peak or trough.
"EVERY wavelength has a point where its focus is 0nm; Once every length of the wavelength as a matter of fact."
The laser is 2foot square and requires 1tw of electric.
Pointless invention for everyday use to say the least. It is great the data is in such a small 2 dimensional area of 1square inch, but the laser size, alignment and movement might present a little problem dont you think! We wont be seeing this anytime this century.
If there's one thing we've learned from the last century, it's that things always stay the same size and never get smaller or more efficient.
Yet another memory development...
...we won't see on the mass market anytime soon. I'd feel much more excited if some new form of fast nonvolatile memory were announced as ready for mass production and introduction into the general market. Until then, please don't keep my hopes up like this.
Sub wavelength imaging *is* possible
Look up confocal near field microscopy.
However I'm not sure if it could do wavelength/50 (300 micrometers is in the UV but sources and detectors do exist for it) imaging. You can bet the sensor will be *lots* bigger than raw bit.
The bottom line with *all* these clever technologies is this. You either have a *shared* readout device which spreads the cost of mfg across *all* the bits it reads (like a hard drive read head) using a simple to fabricate storage structure and some precision mechanical stuff and a fair bit of electronics (like a hard drive) *most* of which is pretty straightforward (and relatively cheap) *or* you have a lot more sensors shared across a lot fewer bits like the sens amplifiers/ row of
with *no* mechanical motion but a *much* more complex fabrication and alignment sequence (with corresponding differences in unit price and size).
I recall when people were saying magnetic bubble memories would kill hard drives. Faster, higher density, no moving parts.
Didn't quite happen that way.
- Updated Zucker punched: Google gobbles Facebook-wooed Titan Aerospace
- Elon Musk's LEAKY THRUSTER gas stalls Space Station supply run
- Windows 8.1, which you probably haven't upgraded to yet, ALREADY OBSOLETE
- Mounties always get their man: Heartbleed 'hacker', 19, CUFFED
- Android engineer: We DIDN'T copy Apple OR follow Samsung's orders