Jon Danniken wrote:
Paul wrote:
Jon Danniken wrote:
Thanks Paul, I appreciate it. One other thing I think I read about
is that SE also removes sectors marked as bad, but now I can't find
the page (of course). Fortunately the drive I'm buying only has 90
hours on it, so maybe I'll be fine.

Jon
Yeah, I read that somewhere too, that the spare sectors also get
erased. What that means is, the head will pass over all sectors, and
apply write current. But it doesn't have to verify how the write
turned out. It's not a write-verify kind of operation, more or less a
"best effort", since it's inevitable that some sectors are too
defective to be touched (like, embedded sync is missing as well). If
the embedded sync is demagnetized, then the head won't be able to
actively track position and keep the head over the track. The erase
routine would have to move on to the next location, if that happens.

Paul

I think my concern is that whatever effort the drive has put into
identifying bad sectors is going to be wiped over, re-exposing the bad
sectors as legitimate locations, instead of as marked bad. In other words,
does the record that identifies bad sectors get written over as well?

Jon

No, the state of automatic sector sparing should be preserved.
The "bad" sectors are still bad, and their replacements are still in usage.

It would go against the ATA/ATAPI sense of automatic sector sparing,
to unto that. But SCSI heritage drives support that kind of thing - the
ability for the user to override the system (I've done it, to drives at
work). AFAIK, ATA/ATAPI doesn't have that as an option. ATA/ATAPI drives
have private means to make changes to drive state - one would be some kind
of factory software, or perhaps by means of a firmware update. The drives
accept firmware updates, either temporary ones, or permanent ones. And some
drives have a three pin serial port interface, and there is some kind of
terse language used there to talk to the controller. Perhaps some command
there could do it. But using regular protocol commands over IDE ribbon or SATA cable,
probably not. The ATA/ATAPI automatic system continues to maintain the same state.

Paul