In general, if
a PC (say) is capable of duplicating a disk, then it's
also capablie of recoridng and rewriting a disk image (this just involves
saving the data that would have been transfered from the source to
destination disks in a normal PC file of some suitable format). So this
is a simple-ish matter of programming
And without that simple-ish matter of
programming, you end up wasting way
Absolutley. FWIW, I _have_ written software related to this -- the 'LIF
Utilities for Linux'. This is an on-going project (so don't moan that the
feature you need isn't there, and it's GPLed. It's designed to handle the
HP9114 disks only, so as to transfer programs and data between older HP
handheld calculators and linux PCs.
I didn't write a program to directly copy a disk (for one thing, my PC
only has 1 3.5" floppy drive) -- I wrote programs to transfer between a
physical disk and an image file in both directions. That seemed to cover
it (just put the image file in /tmp if you don't need to keep it :-))
ON A PC, using the PC hardware, a PC CAN NOT READ
ANYTHING other than
"Western Digital" style MFM sectors on a PC, using PC hardware,
Even
then you might have problems. There is an index timing problem that
The statement that you "CAN NOT READ ANYTHING other than" is not
invalidated by the existence of ADDITIONAL problems.
Of course. I was just pointing out that you might still have problems,
and that a PC can't read all disks with 'western digital headers'. In
particular, I couldn't read disks formatted on my TRS-80 Model 4 on this
PC. I could format a disk on the PC with the same type of format and then
read/write it correctly on both machines.
means the Intel disk controllers can't read
some disks formatted by real
Western Digital chips (if the first sector header starts too soon after
the index pulse, the Intel controller won't see it). Some clone disk
controllers mighy have got this right, some certainly haven't!
In addition, there are some other minor complications, such as additonal
code needed to read 128 byte sectors, disks (such as Kaypro DS) that have
invalid content in sector headers, etc.
Oh, indeed. Linux makes this a little easier (at least in the (old)
version I'm using) in that there's an ioctl() call that allows you to do
just about anything the hardware is capable of (some of the parameters
set up the DMA controller to transfer the appropriate number of bytes,
and the _system_ takes care of allocating a buffer not crossing a page
boundary, etc, other paramemeters are sent to the disk controller chip
without modification). I have had to do it under MS-DOS (the PERQdisk
program), which was 'interesting' -- just about every other comment
relates to some misfeature of the PC :-)
Each
handles a fixed finite group of formats. Most have provision for
adding in additional formats yourself, but almost nobody ever succeeds in
doing so, and after fielding too many "which numbers do I put in to make
I
have mananged to add a new CP/M format (FTS-88 double density 8") to
22disk, I think.
It CAN be done. I had already assumed that Tony dould do it. _Most_ end
Why me? I am not a programmer. In fact I doubt I'm a hacker [1] either.
[1] In the original sense. I am certainly not a cracker.
Is it
_possible_ to add formats to Hypercross? I have the original disks
and manaals for the Model 4 version, and it doesn't say it is. If it is,
what's the trick?
Probably not. ("_Most_ have provision") I haven't heard from Mike Gingel
Sure, I understand 'most' :-). I was just asking because I have that
program, and if it is possible to add more formats I'd like to be able to
do it. If not, never mind.
-tony