Full Version: Writing Macintosh Floppies
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
Pages: 1 2 3
I'm trying to archive 800K Macintosh floppy disks by creating a .scp image then writing it to a blank floppy disk to check the image.  However, I haven't been able to accomplish this, and there seem to be hints around the Internet that suggest the SuperCard Pro can't write variable bit rate Macintosh floppies.  Is this true?  If so, how should I verify the disk image?  Is PCE tools, mentioned here, the only way for Macintosh disks?
SuperCard Pro reads and writes flux.  Bitcell times can change (and do) constantly.  You should be able to use INDEX mode for copying 400K/800K/1.44MB Mac disks.

You can use a few different tools for converting Mac disks to/from .scp image format such as PCE tools and a8rawconv.

Where did you see that SuperCard Pro can't read or write variable speed formats?  I need to set the record straight.
I've made .scp images on two different drives using the following settings:

Weakbits:  Raw
Density:  Low for both Reads and Writes
Index Sensor:  Required for both Reads and Writes
Backside Control:  None
Disk Types:  Macintosh LD (default Track and Head settings for Start and End)
Preservation enabled

I tried both Index and Splice for the Copy Mode.

With one of the drives, an Epson SD-800 (dual 3.5"/5.25"), I previously made a .scp image of a copy-protected PC game on a 3.5" disk, and used it to created a working copy on a blank floppy, so I know my setup can work.  The other drive is an Imation 3.5" drive, which I hadn't previously used with the SCP, but is in good working order.

Regarding where I read certain things about the SCP, I did plenty of reading months ago when deciding whether to get a SCP or a KryoFlux, and I can't recall most of the exact sources.  I should have said that I simply couldn't find any definitive answers.  Perhaps I misread some things, and I didn't mean to suggest that anybody was out there actively smearing the SCP  Big Grin .  However, the write-up here says that some drives may have trouble with variable bit rates, and as a result, I'm wondering about my drives.

Coincidentally, those postings where written by "SomeGuy," who wrote the post on this forum that I previously referenced, which claims that Mac 400K/800K disks "don't use the index pulse."  Does that affect what I should choose under Index Sensor?  Might I have to try another drive, perhaps one on your list of compatible 3.5" drives?
While it is true that user formatted 400K and 800K Mac disks don't use the index pulse to start/stop tracks, commercially produced disks (like Apple system disks) do and so INDEX mode can be used. Otherwise you probably need to use a8rawconv to handle these disks. a8rawconv can control the SCP board directly.
Before I pursue creating some kind of sector image, I'd like to nail down for sure whether there's some way to create a duplicate disk for use on a real Macintosh, and what the procedure for that would be.  I consider this to be the ultimate test of an archival-quality disk image.

According to the manual included with the latest versions of a8rawconv (0.92, weaktest), Macintosh 3.5" disk support includes "diagnostic decode only," while other formats support bonafide reading, and only the Atari formats support writing.  Thus, I can't create a duplicate disk without write support.  Or would you rather recommend that I create a .scp image and write it to a disk, both using a8rawconv?

Furthermore, even if I do create a decoded sector image, such as a .img file created by PCE tools, I'm not aware of any way to write such an image to a real disk.  Using it with an emulator would be the only option.  The only conventional way to write a 400K/800K Mac disk is to use a real Mac with a variable-speed drive, which requires a disk image format that a program in the Classic Mac OS can use.  The only other method I can think of is to use something such as the SCP to read the flux right off the original disk, then plop it right back down onto a blank disk, which I haven't been able to do so far.

Before I consider using an emulator, I'd like to know all possible methods of duplicating the disk.  I'm perfectly fine with buying other floppy drives if that might help.  Thank you very much for your assistance thus far!
You should be able to just take any commercially produced Mac disk and use the INDEX mode to make an image or a copy. Commercially produced disks use the index pulse. You might want to check the HxC Floppy Drive Emulator software. I know that Jeff added variable bit length support for some of the CNC machine systems. There might be Macintosh format there as well. I will send an email to Jeff about that.
In that case, I guess I'll tinker with it some more, and probably end up trying another drive.  As of now, any copy I make doesn't mount on the real Macintosh, as the OS just offers to format the disk.  The original disks mount though, so I know the Macintosh's floppy drive is working.

Regarding the HxC Floppy Drive Emulator, there seems to be no trace of old Macintosh hardware or disk image support at the website.  I'd bet this has to do with the fact that the HxC hardware itself uses the standard 34-pin floppy interface, while the Macintosh uses a 20-pin interface.
A Mac high density disk is the same as IBM PC MFM format though, but uses HFS instead of FAT12/16.
Yes, I figured that the IBM 1.44 MB disk type would work for Mac HD disks, and indeed, I have successfully imaged and duplicated such a disk.

I've made some interesting progress with a commercial 800K disk (not made by Apple, though).  Index mode never works, and neither does splice mode with the default 2 revolutions.  However, after using splice mode with the maximum 5 revolutions, I got the following results with the duplicated disk in the real Macintosh:  after I insert the disk, the drive motor chugs along for around 20 seconds, then finally, the disk mounts.  However, when I tried to run the program on the disk, again the drive chugged and chugged, but then spat out a disk read error.  When using the original disk, it mounts almost instantly, and the program runs fine.

That was with the Epson SD-800.  I retried the same procedure with the Imation drive, but the disk would not mount (there seemed to be a little more chugging by the drive than usual, though).

Therefore, I'm pretty convinced that it's in fact the drives that I'm using with the SCP that are the issue.  They can't handle the variable bit rate.  The SD-800 almost could, but not quite.  I'm going to try some other drives, then report back.  However, my findings also bring up the question of whether I could write a perfect working disk if more than 5 revolutions in splice mode were available...
It's NOT the drives! FLUX itself is a variable bit rate! Every single bitcell is a slightly different time period and that can vary from <1us to over 15us, depending on the disk format. Mac disks have 4 "speed zones" with fixed width bitcells.

It's more likely that your disk drive's heads are dirty. You are going to find that you will probably need to be cleaning the disk drive heads just about every disk you intend to archive. The oxide is just literally falling off because of time, and 3.5" disks are far worse than 5.25" disks.

The number of revolutions after 2 revs doesn't matter. Because only 2 revolutions are actually used - the last 2 in the case where you have selected more than 2. You only need 2 full revolutions to make sure that 100% of the entire track is captured.

I would try cleaning your drive heads and trying it again. Smile
Pages: 1 2 3