On Sat, 29 Jan 2005, Randy McLaughlin wrote:
I have tried to get a quorum on comp.os.cpm but
there are only a few with
any opinions.
We all agree that 3.5" drives are a viable way to go. Many prefer to
never
change from 8" drives.
I have a variety of CP/M systems with a variety of storage mediums.
The point is for those of us that use 3.5" HD drives should try and come
to
a consensus for a "standardized" format.
Questions such as sector size and numbering come to mind.
Why not just stick with 8"?
About 25 years ago, I asked Gary Kildall what the standard format was for
5.25" CP/M.
He replied "eight inch single density".
I said, "but what about machines with 5.25 inch drives? What is the
standard format for them?"
He replied "eight inch single density."
If you are going to create a "new" CP/M format, then I would recommend
using the same PHYSICAL specs as PC and Mac: 80 tracks per side, 2 sides,
18 sectors per track, 512 bytes per sector. That will simplify
significantly the code needed to be able to write programs for
transferring files back and forth.
But the easiest ones to implement would be to match the format of their
existing formats. 1.2M 5.25" drives were originally designed to be an
almost drop-in replacement for 8" drives, with specs that look almost
identical to the host machine. If you're intent on using 3.5" HD, then
the easiest to implement (2 * 80 * 15 * 512) would waste a sizable amount
of space, but who NEEDS more than that?
--
Grumpy Ol' Fred cisin(a)xenosoft.com
One of the "standard" formats recommended by Teac is 2 * 80 * 18 * 512 for
3.5" HD which I personally lean toward. As I see it there are more storage
devices that use 512 byte sectors because of the great and all seeing OZ
(aka Willy boy).
The Imsai Series 2 SuperIO boots from Flash ROM but I have recommended to
Howard Harte to put an offset of two tracks so other systems can use
compatible formats.
I agree 8" SSSD is the only standard that exists for CP/M, I am only trying
to create a defacto-standard for systems coming into the 3.5" world. A
format of consensus if possible, I am beginning to doubt that a quorum of
interested people can be developed.
Randy