[Intel-devsys] IMAGEDISK for MDS 8" SSSD disks

Bill Beech (NJ7P) nj7p at nj7p.info
Thu Feb 27 18:00:55 MST 2020

Roger et al,

A very interesting thread!  Roger, please document your Imagedisk box.  
I am trying, in my spare time, to get one running, as well.

I have downloaded both Roger's images.  They are identical, as I suspect 
he used the original to write the test disk.  What processing went on 
between these versions other than a file name change?

I note they are not bootable and contain trash on the first and second 
tracks.  I also see the expected 6 sector skew in the data in the image.

Is it possible that the sectors written on the test disk are written in 
the wrong order?  Is it possible to read the entire track 2 with all the 
gap and flag data on another machine to see what is actually on the disk?


On 2/27/2020 9:32 AM, Herb Johnson wrote:
> Well, I am distracted by a friend's serious illness today. And I don't 
> wish to be drawn into arguing  with Jon Hales on speculative matters 
> and certain matters of opinion. Jon otherwise makes good points about 
> possible sources of difficulty. I'll focus on positive and actionable 
> suggestions.
> It's possible the various drives in use, have some variance, as Jon 
> suggests. Those can be resolved in two ways. One, use different drives 
> on the same computer and controller, format and write to and read from 
> diskettes, and confirm those diskettes are all cross compatible. 
> That's my rule-of-thumb standard for testing drives, which I've sold 
> to others, without receiving complaints. Two, obtain an alignment 
> diskette, attach an oscilloscope, and perform the tests supported by 
> the diskette; those are computer-independent.
> I now read carefully, Roger Arrick's actual message of 2/26. It's good 
> to have experiential facts.
> 1) Roger suggests his MDS225 with its own drives, can read foreign 
> CP/M 8" SSSD diskettes "fine". That can be tested by sending him an 
> arbitrary number of such diskettes. He says "it DIRs fine". That means 
> tracks zero to several are readable.
> I suggest running a CP/M file checksum program; that will force 
> read/verifies to whatever tracks have files. The CP/M UG and SIG/M 
> have a CRC or CRCK program. I"ve had "reads fine" but "checksum 
> errors" occur under various circumstances.
> Or, one might use "DU" or "DUU" (same distributions) and tell it "read 
> the next sector" from start to end. That tests sector and track 
> checksums but only reports persistent failures (or you notice the head 
> steps slower and sector-reports are slower). My point is: flux 
> physical density increases and disk media velocity decreases, on the 
> inner vs outer tracks. One should read the whole diskette. Results 
> could be diagnostic.
> 2) Roger says, "My imagedisk PC [and its drives] creates 8" SSSD CP/M 
> diskettes which can be read by other vintage systems". He cites a ISA 
> controller, known for compatibility. Again: his total system can be 
> tested by sending his *bulk erased*, then formatted and written 
> diskettes, to be read by several other systems; in the ways I suggested.
> Bulk erasing eliminates prior format information including 
> consequences of misaligned writes. These matter more on 5.25-inch 
> media due to 40/80 track differences.
> 3) Roger offers an IMD file imaged from a "successfully read" CPMUG 
> diskette. He says when he writes this file onto an 8" SSSD diskette on 
> his PC system; said disk a) cannot be read on his MDS225+drives system 
> but b) can be read on a local non-MDS+its_drives system.
> In principle he can postal mail his MDS+drive written 8" SSSD 
> diskettes to other people, see if they can read such diskettes, and if 
> not (they might) determine the source of the failure.
> As I and Jon I think suggested: Roger may wish to attach his Xerox 8" 
> disk drives to his MDS225 system; attempt to read the various 8" SSSD 
> CP/M diskettes he's produced and those produced by others; attempt to 
> write new diskettes; and then reattach his "MDS225 drives" and read 
> the Xerox-drive-MDS225 diskettes. Those results may be informative.
> In summary: Roger may choose to obtain or provide diskettes to/from 
> other people, to establish compatibility and or find errors. And Roger 
> may swap drives on his system, to see if "errors follow the drive". 
> One can posit that a particular system (computer+drives) is more or 
> less compatible or forgiving of drive format or disk media, than some 
> other system (computer+drives). That's also a choice.
> I have a Heath/Zenith Z-100 (Z-121) S-100 system which operates a pair 
> of 8-inch drives (I forget brand/model). I can run CP/M and MS-DOS 
> (but not IMAGEDSK); Zenith provides a MS-DOS based diagnostic 
> including a disk drive/controller diagnostic. I'm pretty confident, my 
> 8-inch drives can read any number of diskettes from any number of 
> sourced diskettes in 8" SSSD CP/M format. I have several brands and 
> models of 8-inch drives I can attach, and I've serviced drives for 
> years with that system.
> So if Roger chooses to contact me privately, we can arrange to 
> physically swap various 8" SSSD CP/M diskettes. I'm not offering to 
> run oscilloscope tests at this point, but a few swaps and the other 
> testing I suggested, may be useful to him. If Roger wishes to buy an 
> 8" alignment diskette, I can offer some to him.
> regards,
> Herb Johnson

More information about the Intel-devsys mailing list