[rescue] cd-rom rot

[Carl R. Friend]

On 2/22/20 12:23 PM, Robert Toegel wrote:

> Well, the only reason I mentioned it was that I once read that when they
> upgraded Hubble, the computer was upgraded to a 486 and we were beyond that
> on Earth.  The story was that the more modern chips were much denser so a
> cosmic rays striking a newer chip would more likely change something than
> would happen in a 486.

    One cannot reasonably compare computers used in space-flight to those
used in terrestrial conditions.  For one, the radiation environment in
space is vastly worse than here here what with us being protected by
a nice dense atmosphere.  Secondly, one must recall that "failure in
space-flight is not an option" because of the irreparability of same;
if Hubble failed again it would instantly become space junk.  It's best
to trust rad-hard components in that scenario if they are capable of
executing the mission.

    Study up on the control gear and the architecture of the Voyager
probes, both of which are still alive and well as they approach their
50-year anniversary of launch.

> Just wondered if the same logic applied to floppies all else being
> the same (not cosmic rays but just random noise or loss of magnetism).

    Something happened to the manufacture of floppy disks sometime in the
1990s that turned them from reliable storage mechanisms to trash.
This has nothing to do with density or radiation; the manufacturers
changed something and the medium ceased to be useful.  I have to pick
and choose *old* floppies (3 1/2, especially) to actively use on my
systems here (I have a couple of systems that can now *only* boot from
floppy, which means I need *reliability* above all else).  Anything
newer than about 1995 should be regarded as "highly suspicious" and
put through a thorough bad-sector scan, several times, several days
apart, before being accepted into service.

    CD rot is something entirely different, and depends on the type
of CD (pressed, Burn once, R/W) and how fast, in the case of the
writeable type, it was written.  Pressed CDs typically fail due to
the oxidation of the aluminium reflective layer, and usually this
is down to a crap job of lacquering the back-side or the use of
inks that attack the lacquer and let air (oxygen) in to attack the
delicate aluminium layer (this is what causes the "transparent CD
effect").  In the case of writeables, we're usually worried about
things like dye permanency, and this is why disks burnt at high
speed seem less permanent than those done slowly -- the dyes have
more time to change at slow speeds than fast, and may revert over
time.

    On the plus side, like Dave, I have lots of very old floppies that
perform admirably to this day, mostly of the 8" ilk, but also quite
a few 5 1/4s.  I'm increasingly giving up on 3 1/2s.

    Cheers!

-- 
+------------------------------------------------+---------------------+
| Carl Richard Friend (UNIX Sysadmin)            | Boylston            |
| Minicomputer Collector / Enthusiast            | Massachusetts, USA  |
| mailto:crfriend at rcn.com                        +---------------------+
| http://users.rcn.com/crfriend/museum           | ICBM: 42:20N 71:43W |
+------------------------------------------------+---------------------+