On Mon, 2003-01-27 at 06:26, Anthony DeRobertis wrote: > On Mon, 2003-01-27 at 05:07, Alexander Hvostov wrote: > > > > You missing a major factor about flash memory, they have a limited > > > number of write operations. > > > > Wasn't that problem resolved a long time ago? > > No. But if you don't do a large amount of writing (noatime should help, > so could keeping most everything mounted ro, using ramdisks for /tmp, > etc.) and use write leveling (CF does, I believe) it should last a > while. Well, I suppose if flash becomes a viable alternative to the Winchester drive, this will need to be improved. What is write leveling? I seem to recall something about Wichesters also having a limited number of writes, but I can't remember... Another thing: holographic storage technology is being worked on. I think it can withstand shocks similar to flash. Perhaps that will be our solution. > And the power savings over a hard disk are quite impressive... No spindles, no servos, no electromagnets. > Of course, if you want to use a hard disk, might I suggest some padding? You'd be hard-pressed to come up with a good enough padding to protect a Winchester from the shocks that semiconductor memory can (and, in mobile devices like cameras and PDAs, does) shrug off. > > "After a memory cell has been rewritten x times, it will no longer hold > > what was written." > > "After a memory cell has been erased x times, it breaks." There you go. Alex. -- PGP Public Key: http://aoi.dyndns.org/~alex/pgp-public-key -----BEGIN GEEK CODE BLOCK----- Version: 3.1 GCS d- s:++ a18 C++(++++)>$ UL+++(++++) P--- L+++>++++ E---- W+(+++) N- o-- K+ w--- !O M(+) V-- PS+++ PE-- Y+ PGP+(+++) t* 5-- X-- R tv b- DI D+++ G e h! !r y ------END GEEK CODE BLOCK------
Attachment:
signature.asc
Description: This is a digitally signed message part