Image Map Image Map
Page 1 of 3 123 LastLast
Results 1 to 10 of 24

Thread: Imagedisk to 8" keeps using 500kbps for SD disks

  1. #1

    Default Imagedisk to 8" keeps using 500kbps for SD disks

    I've got a working imagedisk system now. Whohoo.
    Adaptec controller which can do Single Density (SD).
    It writes 5.25" SSSD disks for an old FLEX system fine.

    Now I've got an 8" SA800 hooked up via DBIT FDADAP.

    I'm trying to read in a known-good standard CPM disk - SSSD, 128b/sec, 26 sec/trk, 77 tracks.
    It says it read it, but when it writes it out, it says 500kbs, and SD is suppose to be 250kbs.
    Reading the disks says 2002 sectors which is 26x77, but it also says 500k SD which appears wrong to me.
    500Kbps would be DD right? 250Kbps should be SD.
    The disk won't read in a SSSD CPM system (MDS225).

    What's the deal? Why is imagedisk stuck on 500kbs which is double density?
    Last edited by Ragnarock; February 20th, 2020 at 04:58 PM.

  2. #2


    you say known good disk, but then you mention "writing it out"

    are you trying to duplicate a working physical disk, or trying to write a disk from a known good image?

  3. #3
    Join Date
    Jan 2007
    Pacific Northwest, USA
    Blog Entries


    You have modulation confused with bitrate and cell timing.

    A standard timing cell on 8" floppies is 2 µsec. or 500KHz. This does not correspond necessarily to bitrate.

    A single-density (FM) modulated bitstream requires 2 timing cells per data bit; one for the clock and one for the data.

    A double-density (MFM) modulated bitstream requires only a single timing cell per bit, due to some clever encoding.

    To reprise my post of almost exactly 3 years ago, what this means is that the recovered data stream from a single-density/FM floppy is 250,000 data bits per second, as the clock bits merely serve as synchronizers and are discarded while reading.

    The MFM/double-density does away with separate clock bits and delivers twice the throughput--500,000 bits per second.

    In any case the basic cell timing is still 2 µsec. or 500KHz for 8" floppies.

    To encode things in MFM, however, requires a bit of memory--what you write this instant depends on what you've just written, in addition to adding a half-cell transition. In the diagram above, note that the maximum flux reversal rate is the same in both cases, which means that the bandpass needed in the read recovery circuits doesn't change. MFM is more sensitive to instantaneous speed variations (ISV) than FM, however.

    So Imagedisk has it right.
    Last edited by Chuck(G); February 20th, 2020 at 05:32 PM.

  4. #4

    Default Imagedisk for 8" CPM SSSD

    Quote Originally Posted by maxtherabbit View Post
    you say known good disk, but then you mention "writing it out"

    are you trying to duplicate a working physical disk, or trying to write a disk from a known good image?

    I have a known-good CPM disk (SS SD, 128b/sec, 77 track, 2002 sec)

    I'm trying to simply duplicate this disk
    since I haven't had luck writing .imd images from dunfields site yet.

    I can read in this CPM disk into a .imd file. No errors or anything.

    And I can write a new disk with this .imd file, no errors given.
    Final message says
    500k SD, 26 sectors of 128 bytes.
    deleted data, single-sided, single step, 2002 sectors.

    This disk wont read in my CPM system like the original will.
    I've tried a couple of different write gap values.

  5. #5


    And I can use IMGV to view the created .imd file and it does indeed show the directory starting on track 3 which I think is right for CPM.
    So it appears to have read the CPM disk.
    But the written disk is not viewable on my CPM machine like the original.
    I'm missing something basic here.

  6. #6


    Is it possible that my adaptec AHA-1542CF shown here

    Is incapable of writing SD 128 byte sectors?
    I think it is capable because I can read the original disk, write a new one, and view it on the PC,
    it just isn't viewable on my MDS225. I'm wondering if there's something wonky with RW gap or Format gap
    that the MDS is choking over. Again, this MDS reads the original CPM disk.

  7. #7


    My understanding is, correct me if I'm wrong, on 8" disks

    Single density is
    128b x 26 sectors x 77 tracks
    256k per disk

    Double density is
    512k per disk

  8. #8
    Join Date
    Jan 2007
    Pacific Northwest, USA
    Blog Entries


    The data rate is 250Kbit/second, but the clock frequency is still 500KHz. Don't let this seeming discrepancy distract you, even if you don't understand it. IMD uses the clock frequency (500KHz) on its menus. Trust me--I've done hundreds of these things.

    Try using IMD to write your image, then use the "disk to file" function to see if you can read what's been written.

    However, very few FDCs can write 128 byte MFM sectors, though they may be able to read them. Generally speaking, if they can read 128 byte FM sectors, they can write them.

  9. #9


    If I use IMD to read in a known SSSD CPM disk (26sec x 128b) x 77 tracks = 2002 sectors =256KB capacity.
    When I then use IMDA to view that disk it says:
    Data Rate on that disk is 500kbps. That doesn't sound right. Should be 250kbps for SD FM.

    Looking at the shugart manual clearly says SD FM is 250kbps data rate, DD MFM is 500kbps data rate
    which jives with what I remember from 40 yrs ago.

  10. #10
    Join Date
    Jan 2007
    Pacific Northwest, USA
    Blog Entries


    Re-read and at least try to understand what I wrote. The reference frequency for either 8" SD or DD is 500KHz--and that's what IMD is faithfully reporting. The bit-rate of SD is half that of DD--but the reference frequency is the same--and it's the the latter that IMD is reporting.

    To see where Dave got his numbers, consider table 3-6 in the PC8477 datasheet. Note that the data rate select register setting is the same for 8" FM or MFM--the FM data rate is half that of the MFM one. (The 1 Mb/sec setting is for ED drives).

    There is no error. 8" FM requires two bit times to record a data bit, where MFM requires only one.

    Got a spectrum analyzer handy? Hook one to the raw data output of your drive and look at the spectra for FM and MFM samples. You'll see that MFM shows 3 peaks, where FM only shows 2--but the maximum frequency is the same.

    When I wrote Anadisk and a few other utilities, I stayed away from using actual numbers for density to avoid confusion. There is high-density and low-density FM and MFM; internally, the drive type is used to make appropriate adjustments. So, a 360KB PC 5.25" disk appears as the same for both 1.2M and 360K drives, even though the clocks are different (1.2M drives usually spin at 360 RPM, vs. 360K drives at 300 RPM).
    Last edited by Chuck(G); February 21st, 2020 at 08:58 AM.

Tags for this Thread


Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts