On Thu, 16 Jan 2014, Al Kossow wrote:
If you're designing a file format, you should look
at this:
http://www.fadden.com/techmisc/file-formats.htm
In addition, over the years, there have been several threads here about
it, something on the order of
Re: Let's develop an open-source media archive standard
as well as numerous other related threads.
The intended PURPOSE of the archive is critical.
Is the intent storage, with recreation of working diskettes?
or use of the data on other systems?
or analysis and repair of damage?
Basically, for SOME uses, including copy-protected disks, it is necessary
to archive every flux transition, since such disks may rely on bits
between the sectors, sector sequence relative to index, and other
weirdnesses. For 5.25" "double density" ("360K" TYPE of
diskettes) that
is 6K - 9K per track
For "non-standard" but not copy-protected disks, the sector headers may be
crucial, but inter-sector gaps aren't. Such things as Kaypro invalid head
number fields, etc. Figure on 5K-6K per track.
For "standard" types of formats, the data within the sectors is all that
is needed, but sector boundaries should be preserved, since sector
sequence and file boundaries need to be re-creatable. assume 320K - 440K
per disk (again, for "360K" type)
For "standard" formats that are completely understood, and for which there
is no reason to expect weirdness, then the file contents is all that is
required. Note: that means a level of understanding of the format
sufficient to recreate a usable disk from the file contents, including
such things as knowing the special position requirements of system files,
knowing that CP/M system tracks are not reflected in the file DIRectories,
etc.
--
Grumpy Ol' Fred cisin at
xenosoft.com
http://www.xenosoft.com/FPUIB