This kind of drive will be OK for most casual users, but not for long-term, archival use: Flash or other electric-field storage designs imply that the bits stored are defined as 'i' or '0' by the strength or polarity of an electric field.
This means that there is a nonminimal potential difference involved. Over time, by design imperfection or quantum tunnelling, the electrons will leak away, reducing the bit potentials to intermediate states. By contrast, magnetic media store both logic levels in domain configurations which are at energy minima. The bits don't decay except over archeological time intervals.
Intel should publish the expected lifetime of data stored on these drives. One might guess that data on a solid-state, nonmagnetic drive would last at least ten, maybe 20, years -- but, who knows?
You would not want to store birth certificate images or family photo images on a solid-state drive. For example, I have a print of one of my great uncle's baptismal photos, from the 1890's -- I would expect that a copy on my hard drive should last at least a century, so that I could pass it down for generations.
If I am getting this right, a few days back I read that Samsung also had launched SSD models built with consumer-class NAND flash. The new SSDs from Samsung have capacities of 100GB, 200GB and 400GB. Are they based on similar technology used by Intel? Has anyone made techno-commercial comparison between these SSD products from Samsung and Intel?
Blog Doing Math in FPGAs Tom Burke 8 comments For a recent project, I explored doing "real" (that is, non-integer) math on a Spartan 3 FPGA. FPGAs, by their nature, do integer math. That is, there's no floating-point ...