There's a
bit of reverse engineered information on the web, but it does not
seem complete - Writing to a disk and re-reading seems like the quickest and
most reliable way to do it given that I didn't have THAT many to do. Having
the filenames auto-entered, and the comments captured and transferred makes
this approach reasonable to do.
But how can you be *sure* nothing was mangled in the conversion? If you
can avoid two analog generational losses, hey, why not...
I check that no detectable errors occured in both processes (writing and
reading), that no sectors in the final image are flagged missing, and that
the number of sectors/track in the resulting .IMD image is what I expect
for the system in question (in case a complete sector was somehow "lost").
The chances of data corruption getting past the CRC checks seems to me to
be somewhat lower than the chances of losing data do to misinterpreting an
undocumented file format based on minimal reverse engineered notes which
contain numerous instances of terms such as "unknown", "seems to" and
"may
mean" when referring to various fields in the image. The problem with data
loss due to a flawed understanding of it's format is that such loss often
goes undetected.
--
dave06a (at) Dave Dunfield
dunfield (dot) Firmware development services & tools:
www.dunfield.com
com Collector of vintage computing equipment:
http://www.classiccmp.org/dunfield/index.html