H37 Disk Imaging

153 views
Skip to first unread message

Les Bird

unread,
Sep 22, 2021, 8:08:21 PMSep 22
to SEBHC
Hi all,

Do we have a good solution for creating disk images of H37 software? I have some disks that I want to image. Previously I used Dunfield's IMD but I don't have a system that can run that anymore.

Les

norberto.collado koyado.com

unread,
Sep 22, 2021, 8:32:23 PMSep 22
to SEBHC
This is something Douglas and I were brainstorming on how to add such support to the Z80 V4.0 monitor. We need feedback on how to do this so that we are all align to the same process. 

Norberto

From: se...@googlegroups.com <se...@googlegroups.com> on behalf of Les Bird <lesb...@gmail.com>
Sent: Wednesday, September 22, 2021 5:08 PM
To: SEBHC <se...@googlegroups.com>
Subject: [sebhc] H37 Disk Imaging
 
Hi all,

Do we have a good solution for creating disk images of H37 software? I have some disks that I want to image. Previously I used Dunfield's IMD but I don't have a system that can run that anymore.

Les

--
You received this message because you are subscribed to the Google Groups "SEBHC" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sebhc+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sebhc/915c7cef-de01-404f-9600-c6f56d55c1a4n%40googlegroups.com.

Kenneth L. Owen tx836519

unread,
Sep 22, 2021, 8:43:34 PMSep 22
to se...@googlegroups.com

Hi Les,

 

I have an old XT machine running ImageDisk (IMD) and have imaged and creaated bootable disks in the past.  It has been a while since I last booted that system, but I expect it to boot with no problem.

  • ken

 

Sent from Mail for Windows

 

From: Les Bird
Sent: Wednesday, September 22, 2021 8:08 PM
To: SEBHC
Subject: [sebhc] H37 Disk Imaging

 

Hi all,

 

Do we have a good solution for creating disk images of H37 software? I have some disks that I want to image. Previously I used Dunfield's IMD but I don't have a system that can run that anymore.

 

Les

 

--

Douglas Miller

unread,
Sep 22, 2021, 8:53:37 PMSep 22
to se...@googlegroups.com

What Norberto and I were brainstorming was something like H8DUTIL.SYS program but for the H37, booted off the VDIP1 USB and using image files on the USB. The issue(s) are related to what sort of image file(s) to support, whether saving diskettes to image files is possible, and how much code is required to support the image formats (must fit in H8 memory). Worst case, I suppose, is we'd have a separate utility for each image format, and possible each having a separate "save" and "restore" program.

The H8DUTIL is pretty simple, since the image file format is "raw" and the H17 routines are in ROM. Things like TD0, IMD, etc., and the H37 routines, takes up more space and requires more development (I know there's C code for some of these formats, with TD0 being complicated by an optional but messy, proprietary, compression algorithm)

So, it would help to come up with some minimal set of requirements, rather than just support "everything".

Darrell Pelan

unread,
Sep 22, 2021, 10:01:25 PMSep 22
to SEBHC

I've got a Windows 95 system I keep for that purpose. I can can convert any disks and then save them to IMG format using my Disk Image Utility. Let me know if I can help out.

  Darrell

Les Bird

unread,
Sep 22, 2021, 10:23:37 PMSep 22
to SEBHC
Ok, so we don't have a solution yet except for IMD it seems like. I think I do still have my IMD DOS machine with a 5.25 drive in storage but I'd have to go dig it out. Still it's not the solution I'd like to use.

My initial thoughts are to try and stay away from any custom hardware solutions. Douglas, the VDIP/USB wouldn't help those who need to create a disk from an image but lack the VDIP/USB hardware. I think we need something that works very much like H89LDR does, on real H37 hardware and that allows to create images and also create disks from images. We have to think about someone out in the wild who needs to cold boot their system from a disk image with only access to their H37 hardware.

I'm going to start researching this. I think since the H37 has a WD179x chip on it we can talk directly to the chip to copy track by track and then send over a serial port connection like H89LDR does. Or even using the ROM routines to do this would probably work instead of direct chip programming.

Les

Terry Smedley

unread,
Sep 22, 2021, 11:09:09 PMSep 22
to SEBHC
I'm not sure this gets you to your endpoint, but for H37/HDOS, I've used a lightly modified version of the HUG duplicator utility to write out .H8D sector images of H37 disks.  Those are then put into Glenn's JukeBox.  HDOS reads and mounts that image just like an H17 image.

Terry

norberto...@koyado.com

unread,
Sep 23, 2021, 12:03:51 AMSep 23
to se...@googlegroups.com

I'm going to start researching this. I think since the H37 has a WD179x chip on it we can talk directly to the chip to copy track by track and then send over a serial port connection like H89LDR does. Or even using the ROM routines to do this would probably work instead of direct chip programming.

 

If you can get this working, that will be great.

 

Norberto

 

From: se...@googlegroups.com <se...@googlegroups.com> On Behalf Of Les Bird
Sent: Wednesday, September 22, 2021 7:24 PM
To: SEBHC <se...@googlegroups.com>

--

You received this message because you are subscribed to the Google Groups "SEBHC" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sebhc+un...@googlegroups.com.

Mark Garlanger

unread,
Sep 23, 2021, 12:45:53 AMSep 23
to se...@googlegroups.com
I believe that I've used the FC5025 USB board for soft-sectored disks(if not it should be easy to add support for it), but normally use a WinXP system I can boot to DOS and use Dunfield's IMD software to image the soft-sectored disks. Unlike hard-sectored disks, with soft-sectored disks, density, the sector size, and the number sectors per track, are not constant, so that information really needs to be saved with the disk image, and many disks would not be able to be converted into H8D format.

Mark


norberto...@koyado.com

unread,
Sep 23, 2021, 1:38:27 AMSep 23
to se...@googlegroups.com

For IMD to run you have to run old MSDOS on a computer with a 5.25” floppy drive with an old floppy controller; correct?

Mike Loewen

unread,
Sep 23, 2021, 1:51:42 AMSep 23
to se...@googlegroups.com

FreeDOS works just as well for IMD.

On Wed, 22 Sep 2021, norberto...@koyado.com wrote:

> For IMD to run you have to run old MSDOS on a computer with a 5.25? floppy drive with an old floppy controller; correct?
> To unsubscribe from this group and stop receiving emails from it, send an email to sebhc+un...@googlegroups.com <mailto:sebhc+un...@googlegroups.com> .
> To view this discussion on the web visit https://groups.google.com/d/msgid/sebhc/824eb5eb-9c0a-4531-90f9-657f6a1d98fbn%40googlegroups.com <https://groups.google.com/d/msgid/sebhc/824eb5eb-9c0a-4531-90f9-657f6a1d98fbn%40googlegroups.com?utm_medium=email&utm_source=footer> .
>
> --
> You received this message because you are subscribed to the Google Groups "SEBHC" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to sebhc+un...@googlegroups.com <mailto:sebhc+un...@googlegroups.com> .
> To view this discussion on the web visit https://groups.google.com/d/msgid/sebhc/01d801d7b030%24022532b0%24066f9810%24%40koyado.com <https://groups.google.com/d/msgid/sebhc/01d801d7b030%24022532b0%24066f9810%24%40koyado.com?utm_medium=email&utm_source=footer> .
>
> --
> You received this message because you are subscribed to the Google Groups "SEBHC" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to sebhc+un...@googlegroups.com <mailto:sebhc+un...@googlegroups.com> .
> To view this discussion on the web visit https://groups.google.com/d/msgid/sebhc/CAAjkm79jnJ2TBwsu5t27JW8qCfBhM9ZQdK9L%3DFYe3-uiuO7AUA%40mail.gmail.com <https://groups.google.com/d/msgid/sebhc/CAAjkm79jnJ2TBwsu5t27JW8qCfBhM9ZQdK9L%3DFYe3-uiuO7AUA%40mail.gmail.com?utm_medium=email&utm_source=footer> .
>
> --
> You received this message because you are subscribed to the Google Groups "SEBHC" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to sebhc+un...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/sebhc/01ef01d7b03d%2438fd0a80%24aaf71f80%24%40koyado.com.
>

Mike Loewen mlo...@cpumagic.scol.pa.us
Old Technology http://q7.neurotica.com/Oldtech/

Glenn Roberts

unread,
Sep 23, 2021, 5:25:31 AMSep 23
to se...@googlegroups.com
I resonate with Terry’s and Les’ comments. IMO for saving/sharing disk images we need a solution that only requires a working H8/H37 or H8/H37 setup, not specialized hardware or 25 year old Windows boxes. The simplest solution is to save these images as a stream of sectors (H8D) which HDOS and CP/M already know how to read. The reworked HUG DUP utility sounds perfect!

Sent from my iPad

> On Sep 23, 2021, at 1:52 AM, Mike Loewen <mlo...@cpumagic.scol.pa.us> wrote:
>
> 
> To view this discussion on the web visit https://groups.google.com/d/msgid/sebhc/alpine.LFD.2.21.2109230151200.49374%40cpumagic.137clearview.org.

Les Bird

unread,
Sep 23, 2021, 8:09:10 AMSep 23
to SEBHC
Terry,

That's a good potential starting point. Prior to making H8DIMGR I had modified HUG SDUP to send data over the serial port instead of to a destination drive but from what I remember, and the reason I made H8DIMGR, the SDUP only worked with HDOS disks and I needed a solution to do HDOS and CP/M. Is the HUG DUP that you mention the SDUP utility or is this something different? I did find SDUP37 and started looking at the source code for it as a possible starting point.

Les

Terry Smedley

unread,
Sep 23, 2021, 9:25:48 AMSep 23
to SEBHC
Les, 

I used SDUP37 as the base, and my only use has been for HDOS images.  

Does the CP/M disk structure have a place to store key physical disk parameters like the HDOS volume label?  I have a very vague recollection that the BIOS had to dance a bit to figure out what the physical media looked like.  I've read the comments that H8D doesn't work for other OSes because the disk structure parameters aren't carried, and that it doesn't permit exact duplication of physical structure like sector interleave.

Terry

Darrell Pelan

unread,
Sep 23, 2021, 10:03:28 AMSep 23
to SEBHC
Yes, and the 5.25 drive should be a 1.2MB drive to handle all H37 formats

Darrell Pelan

unread,
Sep 23, 2021, 10:08:01 AMSep 23
to SEBHC
Yes.
   /*
        Disk type: byte 5 in sector 0 on H-37 disks (starting from 0) to define disk parameters
        Allocation Block size: number of bytes in an the smallest block used by CP/M on the disk. must be a multiple of 128 (0x80)
                AB numbers start with 0. The directory starts in AB 0.
        Directory Stat: start of directory entries in bytes
        Allocation Block Number Size: number of bytes used in directory entry to reference an allocation block
        Dir Size: number of bytes used for the directory
         0040 =         DPEH17 EQU 01000000B
         0060 =         DPEH37 EQU 01100000B
         0008 =         DPE96T EQU 00001000B
         0004 =         DPEED EQU 00000100B
         0002 =         DPEDD EQU 00000010B
         0001 =         DPE2S EQU 00000001B

        */
        // 0.Disk type, 1.Allocation block size, 2.Directory start, 3.Allocation block byte size, 4.dir size, 5.interleave, 6.Sectors per Track, 7.Sector Size,
        // 8.# Tracks, 9. # heads, 10 Skew Start Sector (0 based)
        public int[,] DiskType =
        {
            // 0          1             2     3           4    5    6   7       8      9  10
            {0xff, 0x800, 0x4800, 2, 0x2000, 6, 18, 0x200, 80, 2, 10}, // 0 1.44 MB Small Z-80 
            {0x6f, 0x800, 0x2800, 2, 0x2000, 3, 5, 0x400, 80, 2, 4}, //   1 800k H37 96tpi ED DS
            {0x6b, 0x800, 0x2000, 2, 0x2000, 3, 16, 0x100, 80, 2, 12}, // 2 640k H37 96tpi DD DS
            {0x67, 0x800, 0x2800, 1, 0x2000, 3, 5, 0x400, 40, 2, 4}, //   3 400k H37 48tpi ED DS
            {0x23, 0x800, 0x2000, 1, 0x2000, 1, 8, 0x200, 40, 2, 0}, //   4 320k Z100 48tpi DD DS
            {0x62, 0x400, 0x1000, 1, 0x2000, 3, 16, 0x100, 40, 1, 12}, // 5 160k H37 48tpi DD SS
            {0x63, 0x800, 0x2000, 1, 0x2000, 3, 16, 0x100, 40, 2, 4}, //   6 320k H37 48tpi DD DS
            {0x60, 0x400, 0x1e00, 1, 0x800, 3, 10, 0x100, 40, 1, 3}, //   7 100k H37 48tpi DD SS
            {0xE5, 0x400, 0x1e00, 1, 0x800, 4, 10, 0x100, 40, 1, 0}, //   8 100k Default H17 48tpi SD SS
            {0x00, 0x400, 0x1e00, 1, 0x800, 4, 10, 0x100, 40, 1, 0}, //   9 100k Default H17 48tpi SD SS

        };


    /*
    H37 disk identification at byte 6 on the first sector
    MSB = 6 for H37
    LSB
    Bit 4 1 = 48tpi in 96tpi drive
    Bit 3 0 = 48 tpi drive, 1 = 96 tpi
    Bit 2 1 = extended double density, used in conjunction with bit 1 (0110B)
    Bit 1 1 = double density, 0 = single density
    Bit 0 1 = double sided, 0 = single sided
    */

Glenn Roberts

unread,
Sep 23, 2021, 10:08:55 AMSep 23
to se...@googlegroups.com
So my statement that CP/M could work with something like H8D may be wrong.. my lack of experience with this is showing…

Sent from my iPad

On Sep 23, 2021, at 9:25 AM, Terry Smedley <terry....@gmail.com> wrote:

Les, 

Les Bird

unread,
Sep 23, 2021, 10:36:20 AMSep 23
to SEBHC
Darrell, I did see that info in your source code listing. Good to know that. As long as we replicate that data in the disk image it should work.

Terry, as for interleaving, it hasn't really been a problem for CP/M H8D disk images. As long as we copy the sector order in the disk image (as we are doing for CP/M H8D) it should be fine as all the CP/M H8D disk images work without any issues when copied to a floppy and run on a real Heathkit computer.

Les

Norberto Collado

unread,
Sep 23, 2021, 11:57:45 AMSep 23
to se...@googlegroups.com

So, if I transfer from a 5.25” H37 floppy to a 3.5” floppy diskette and create an exact copy using HDOS or CP/M, then I should be able to use any Win10’s imaging tools to create an image; correct?  

Douglas Miller

unread,
Sep 23, 2021, 12:00:37 PMSep 23
to se...@googlegroups.com

Right, but the trick would be if one is accessing the disk image on the host, e.g. via cpmtools. H8D images contain no metadata, so there's no way to tell what order the data is in. There would need to be an enforced convention on interleaved in order to be able to use the images reliably outside of the save/restore paradigm.

Darrell Pelan

unread,
Sep 23, 2021, 12:14:30 PMSep 23
to se...@googlegroups.com
Which imaging tools? The ones I use on windows 10 all depend on an image, not a physical disk.

You received this message because you are subscribed to a topic in the Google Groups "SEBHC" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/sebhc/PbejZUeQgUU/unsubscribe.
To unsubscribe from this group and all its topics, send an email to sebhc+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sebhc/7395B1A9-FAB8-4C0F-864B-AA697CB14EE7%40koyado.com.

Darrell Pelan

unread,
Sep 23, 2021, 2:53:13 PMSep 23
to SEBHC
I think it would be simpler to store the data in IMG format which isn't interleaved. If you need to create a physical disk, you could write the disk with the appropriate interleave. I'd be tempted to stay with the Flash Floppy Gotek using IMG files, then interleave is no longer an issue.

norberto.collado koyado.com

unread,
Sep 23, 2021, 2:54:39 PMSep 23
to se...@googlegroups.com
I was thinking on using the same tool I have to image the Z67-IDE+ CF cards. I can do a quick test tonight on my home laptop. I will try to image H37 3.5" floppy media for HDOS and CP/M. Then move image to a new diskette and see if it boots. 


From: se...@googlegroups.com <se...@googlegroups.com> on behalf of Darrell Pelan <pel...@gmail.com>
Sent: Thursday, September 23, 2021 9:14 AM

Douglas Miller

unread,
Sep 23, 2021, 3:56:34 PMSep 23
to se...@googlegroups.com

I would advise caution on using IMG. I believe it is a Windows-only proprietary format - at least I've had great difficulty using images of that type in the past. I would say we need to stick to open formats like IMD.

Darrell Pelan

unread,
Sep 23, 2021, 4:09:38 PMSep 23
to SEBHC
IMG format in this context is nothing more than a byte for byte image of each sector on the disk. It is not interleaved. I've had good success using it with my Disk Image Utility program and transferring data between my PC, H89, Z100, and SmallZ180 using the Flash Floppy Gotek. You can also use the Disk Image Utility program to convert IMG to IMD for use with HxC Gotek flashed units.

geneb

unread,
Sep 23, 2021, 5:42:36 PMSep 23
to se...@googlegroups.com
On Thu, 23 Sep 2021, Douglas Miller wrote:

> I would advise caution on using IMG. I believe it is a Windows-only
> proprietary format - at least I've had great difficulty using images of that
> type in the past. I would say we need to stick to open formats like IMD.
>

IMG is (should be) a literal raw dump. It's basically the equivalent of
doing "dd if=/dev/fd0 of=filename.img" on a unix box.

g.

--
Proud owner of F-15C 80-0007
http://www.f15sim.com - The only one of its kind.
http://www.diy-cockpits.org/coll - Go Collimated or Go Home.
Some people collect things for a hobby. Geeks collect hobbies.

ScarletDME - The red hot Data Management Environment
A Multi-Value database for the masses, not the classes.
http://scarlet.deltasoft.com - Get it _today_!

Douglas Miller

unread,
Sep 23, 2021, 6:09:11 PMSep 23
to se...@googlegroups.com
OK, so IMD is superior then as it contains enough information to
reconstruct the actual diskette format. Otherwise we're just guessing at
the format based on image size. Although, IMD also requires more code to
handle. We might want to choose a different moniker as there already are
.img files that mean something very different.

Glenn Roberts

unread,
Sep 23, 2021, 6:57:08 PMSep 23
to se...@googlegroups.com
And that’s also essentially what H8D is. A linear stream of the bytes on the disk…

Sent from my iPad

> On Sep 23, 2021, at 5:42 PM, geneb <ge...@deltasoft.com> wrote:
> --
> You received this message because you are subscribed to the Google Groups "SEBHC" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to sebhc+un...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/sebhc/alpine.LRH.2.21.2109231441290.4319%40sidewinder.deltasoft.com.

Mark Garlanger

unread,
Sep 23, 2021, 9:23:25 PMSep 23
to se...@googlegroups.com
I strongly agree with Doug, that we need a format like IMD, which provides all the details on sides/density/tracks/sectors/sector size. The slight extra code, is definitely worth knowing the details of the disk.

Mark

--
You received this message because you are subscribed to the Google Groups "SEBHC" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sebhc+un...@googlegroups.com.

Les Bird

unread,
Sep 23, 2021, 9:28:10 PMSep 23
to SEBHC
My plan is to copy the H8D format which is a sector for sector, track for track copy of the disk. We can write back in exactly the same format and it'll work for HDOS or CP/M and does not require extra hardware like a GoTek or HxC. And of course if you want you can write a conversion utility to convert it to whatever format you want.

I would like to use the H37 extension but I think we're using that for something else that Mark created? Or is it safe to use that extension? So we'll have H8D and H37.

Les

Les Bird

unread,
Sep 23, 2021, 9:30:07 PMSep 23
to SEBHC
Mark,

The WD1797 has a full track read/write command which will read/write entire tracks at a time including all the header and gap information. I could possibly use that to image the disks. Might be easier to do it that way.

Les

Darrell Pelan

unread,
Sep 23, 2021, 9:43:28 PMSep 23
to SEBHC
I used the H37 designation in the modifications I made to the H8DUtility 2.2. However, I made an error in always starting with sector 1 which was different than the exact copy provided in the IMD format. As as I know, there is no reason to not use H37. I moved away from it when I wrote the Disk Image Utility to focus on IMG and IMD files.

Darrell

On Thursday, September 23, 2021 at 9:28:10 PM UTC-4 Les Bird wrote:

Douglas Miller

unread,
Sep 23, 2021, 10:03:56 PMSep 23
to se...@googlegroups.com

The read track command is not as simple as it might sound. Especially for diskettes that have been written (since formatting), the byte stream will get out-of-sync and you end up having to basically simulate the synchronization in software in order to get useful data out of the sectors (except it's worse because you've lost the clock bits). Also, in order to format a disk with that data you cannot just send the bitstream to the 1797 but you have to translate it back into the special characters for the various sync bytes, as those have to be written with missing clock bits.

For last-chance recovery of data, and for analyzing an unknown diskette format, it is useful. For general data, probably not so much.

I had written a utility to use this command back in the day. It would read the track and display the data in hex. But I had to add bit-shift operations (left/right) in order to bring the data into "focus", especially for sectors that had been written since formatting as the re-write of the sector data could be off by any number of bits from the original formatting.

geneb

unread,
Sep 23, 2021, 10:04:23 PMSep 23
to se...@googlegroups.com
On Thu, 23 Sep 2021, Mark Garlanger wrote:

> I strongly agree with Doug, that we need a format like IMD, which provides
> all the details on sides/density/tracks/sectors/sector size. The slight
> extra code, is definitely worth knowing the details of the disk.
>
Just for bang for the buck, you can't beat a Grease Weazel. It's got a
funny name, but it does a hell of a good job imaging media. It will write
it out too. My only complaint is that it can't do hard-sectored media, so
I have to use a different device.

Douglas Miller

unread,
Sep 23, 2021, 10:10:28 PMSep 23
to se...@googlegroups.com
The way I'm looking at this is that it's not about personal data
archiving, but a scheme that everyone can use to share and create
diskettes. That's why I was leaning towards getting a consensus on a
format such as IMD, which can be the most universally useful (e.g. lots
of utilities can work with IMD, including cpmtools and many emulators).

geneb

unread,
Sep 23, 2021, 10:26:30 PMSep 23
to se...@googlegroups.com
On Thu, 23 Sep 2021, Douglas Miller wrote:

> The way I'm looking at this is that it's not about personal data archiving,
> but a scheme that everyone can use to share and create diskettes. That's why
> I was leaning towards getting a consensus on a format such as IMD, which can
> be the most universally useful (e.g. lots of utilities can work with IMD,
> including cpmtools and many emulators).
>

Ah, got it!

Darrell Pelan

unread,
Sep 23, 2021, 11:16:58 PMSep 23
to SEBHC
If we use IMD, do we need to make sure the make sure sectors aren't compressed like IMD does for sectors filled with the same byte? Knowing each track is stored as the same number of bytes makes adding information easier if you want to be able  add a file to an image. 

Douglas Miller

unread,
Sep 24, 2021, 7:53:33 AMSep 24
to se...@googlegroups.com

I would assume that adding files to an image is done on a host, not the H8. In that case the entire diskette image could easily be placed in memory - if necessary - and modified uncompressed, then written back as an IMD with compression.

I don't think it's advisable to try and splice data into an IMD, it seems oriented more towards writing in it's entirety/sequentially. But, I've not implemented much using IMD, perhaps Mark or others have more experience. The utility I wrote to convert my simulator images to IMD does that track-by-track, sequentially.

--
You received this message because you are subscribed to the Google Groups "SEBHC" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sebhc+un...@googlegroups.com.

Les Bird

unread,
Sep 24, 2021, 8:40:35 AMSep 24
to SEBHC
Ok Darrell, I'll use the H37 extension.

Thanks

Les Bird

unread,
Sep 24, 2021, 8:42:18 AMSep 24
to SEBHC
Douglass,

Thanks for the info on track reads. I could see how that makes sense, having to bit shift. I'll not pursue this option then.

Les

dwight

unread,
Sep 24, 2021, 11:18:54 PMSep 24
to se...@googlegroups.com

Physical sectors are only about speed of access. When something like CP/M has a table to interleave, it doesn't care about the physical sectoring, other than speed.
In the simplest case, The H8D can be copied to a disk. To optimize speed, a second disk can be formatted from that disk or other and then the files sys copies to the new disk.
The problem with any setup is if there is a sector size issue.
I don't recall for sure but I believe the H8D format reads the disk in the sector header order. It doesn't care about physical sector order as it request sector in the headers sector number order. The data can then be transfered in any desired order ( as I recall, I have a specific command to do just that ). If one is not sure which order to place it in, I'd recommend the formating of an additional blank disk to copy the transfered disk data to on the H8/H89.
I do think it would be a good idea to keep track of the sector size information but it you have such a disk image, that information is on that disk someplace or HDOS wouldn't know what to do with it. Since the H8D is in the sector disk order it shouldn't be that difficult to extract it from the image.
Since I do not have a H37 disk setup and don't intend to get one any time soon, I'd think it is not all that difficult for one of the group to look into it.
Dwight

Kenneth L. Owen tx836519

unread,
Sep 25, 2021, 4:48:17 PMSep 25
to se...@googlegroups.com

Well, it looks like I spoke too soon on the booting of the XT machine.  When Julia started going to the hospital about every two months, I had to set most of my activities aside and care for her.  So the machine has been sitting for several years.

 

When I tried to boot, I discovered the CMOS memory battery had died.  I replaced the battery, I tried to boot and got memory errors.  I removed and reinstalled the memory sims and memory checks OK now, but the system will not boot.  It is a DECpc 433dxLP with 8 MB ram and Quantum ProDrive ELS had disk.  I am trying to reestablish the CMOS memory settings to boot from the SCSI HD but I got this machine free (being discarded by original owner on upgrade to new machine) so I don’t have much in the way of documentation.

 

This machine is not HEATH, but was the machine I used to make images (ImageDisk application for IMD image files) of my H-37 disks to support the H8 and several H-89 computers.

 

I am hoping that someone on this group may know something that will help me get this accessory machine booting to support the Heaths with H37 disk images.

 

  • ken

 

Sent from Mail for Windows

 

From: Kenneth L. Owen tx836519
Sent: Wednesday, September 22, 2021 8:43 PM
To: se...@googlegroups.com
Subject: RE: [sebhc] H37 Disk Imaging

 

Hi Les,

 

I have an old XT machine running ImageDisk (IMD) and have imaged and creaated bootable disks in the past.  It has been a while since I last booted that system, but I expect it to boot with no problem.

  • ken

 

Sent from Mail for Windows

 

From: Les Bird
Sent: Wednesday, September 22, 2021 8:08 PM
To: SEBHC
Subject: [sebhc] H37 Disk Imaging

 

Hi all,

 

Do we have a good solution for creating disk images of H37 software? I have some disks that I want to image. Previously I used Dunfield's IMD but I don't have a system that can run that anymore.

 

Les

 

--

You received this message because you are subscribed to the Google Groups "SEBHC" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sebhc+un...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "SEBHC" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sebhc+un...@googlegroups.com.

Steven Hirsch

unread,
Sep 25, 2021, 5:01:39 PMSep 25
to se...@googlegroups.com
On 9/25/21 4:48 PM, Kenneth L. Owen tx836519 wrote:
> Well, it looks like I spoke too soon on the booting of the XT machine.
> When Julia started going to the hospital about every two months, I had to
> set most of my activities aside and care for her. So the machine has been
> sitting for several years.
>
> When I tried to boot, I discovered the CMOS memory battery had died. I
> replaced the battery, I tried to boot and got memory errors. I removed and
> reinstalled the memory sims and memory checks OK now, but the system will
> not boot. It is a DECpc 433dxLP with 8 MB ram and Quantum ProDrive ELS had
> disk. I am trying to reestablish the CMOS memory settings to boot from the
> SCSI HD

Most ISA SCSI boards have an on-board boot rom. Try setting both primary and
secondary hard drives to NONE in the BIOS. There may also be a way to disable
the on-board IDE hardware.

With the IDE out of the way, you should get a configuration splash screen from
the SCSI adapter followed by a boot attempt.

Kenneth L. Owen tx836519

unread,
Sep 25, 2021, 7:32:22 PMSep 25
to se...@googlegroups.com

Hi Steven,

 

Both of the IDE hard drives are set to “not installed” (as prescribed for SCSI boot).  I don’t get anything except “no boot drive” on restart.

 

  • ken

 

Sent from Mail for Windows

 

From: Steven Hirsch
Sent: Saturday, September 25, 2021 5:01 PM
To: se...@googlegroups.com

--

You received this message because you are subscribed to the Google Groups "SEBHC" group.

To unsubscribe from this group and stop receiving emails from it, send an email to sebhc+un...@googlegroups.com.

Steven Hirsch

unread,
Sep 25, 2021, 8:16:37 PMSep 25
to se...@googlegroups.com
On 9/25/21 7:32 PM, Kenneth L. Owen tx836519 wrote:
> Hi Steven,
>
> Both of the IDE hard drives are set to “not installed” (as prescribed for
> SCSI boot). I don’t get anything except “no boot drive” on restart.
>

What vendor and model is the SCSI controller? The system should detect the
adapter ROM and make controller initialization part of the boot process.

Do you see any indication that it knows the SCSI adapter is in the system?
Maybe a message about firmware version or vendor?

Kenneth L. Owen tx836519

unread,
Sep 26, 2021, 10:29:12 AMSep 26
to se...@googlegroups.com

Hi Steven,

The SCSI drive is connected to a connector on the main board, so the SCSI controller must be part of the DEC main board.  There is no 3rd party controller and both IDE and SCSI drives connect to the same connector on the main board.  According to the service manual SCSI is enabled by setting the ‘IDE drives’ to ‘not installed’ and ‘IDE Hard Drives’ to ‘disabled’.

 

Pressing F1 for setup displays three pages:

                (on computer)                                                   (as listed in service manual)

Page 1:

System Time:                     hh:mm:ss                                            “

System Date:                     dd mmm yyyy                                    “

Language:                           English                                                  “             

Diskette A:                          3.5”, 1.44 MB                                     “

Diskette B:                          not installed                                       “

Hard Disk 1:                        not installed                                       “

Hard Disk 2:                        not installed                                       “

Base Memory:                   640 KB                                                  “

Extended Memory:         7168 KB                                                “

Video Card:                         VGA or EGA                                        “

Num Lock:`                          On                                                          “

 

Page 2:

Password:                           not installed                                       “

Parallel port:                      Enabled as LPT 1                               “

Serial port 1:                       Enabled as Com 1                             “

Serial port 2:`                     Enabled as Com 2                             “

Diskette Drive:                  Enabled                                                “

IDE Hard Drives                 Disabled                                               “

Boot from diskette A      Enabled                                                “

Boot from IDE Disk C       Enabled                                Boot from hard disk C     Enabled

User definable drives     48 & 49                                                 “

Exchange diskettes         Disabled                                               “

 

Page 3:

CPU Speed:                        Fast                                                        “

Post write                            Disabled                                               “

Primary cache:                  Enabled                                                “

Secondary cache:             Enabled                                                “

Shadow Bios ROM           Enabled                                                “

Shadow Video ROM        Enabled                                                “

Monitor Type                    640 x 480 @ 60 Hz/NI                    “

 

This morning, on applying power the hard disk spins up to speed and steps in several tracks then seeks track 0.  There was no other activity and no display on the monitor.  Removing and reseating the two power supply connectors to the main board, things started working like yesterday: ‘ No Boot Drive’ message and ‘enter Setup’ message.

 

I’m stumped!  Any Ideas?

  • ken

 

Sent from Mail for Windows

 

From: Steven Hirsch
Sent: Saturday, September 25, 2021 8:16 PM
To: se...@googlegroups.com
Subject: Re: [sebhc] H37 Disk Imaging

 

On 9/25/21 7:32 PM, Kenneth L. Owen tx836519 wrote:

--

You received this message because you are subscribed to the Google Groups "SEBHC" group.

To unsubscribe from this group and stop receiving emails from it, send an email to sebhc+un...@googlegroups.com.

Steven Hirsch

unread,
Sep 26, 2021, 12:08:41 PMSep 26
to se...@googlegroups.com
On 9/26/21 10:29 AM, Kenneth L. Owen tx836519 wrote:
> Hi Steven, The SCSI drive is connected to a connector on the main board, so
> the SCSI controller must be part of the DEC main board. There is no 3rd
> party controller and both IDE and SCSI drives connect to the same connector
> on the main board. According to the service manual SCSI is enabled by
> setting the ‘IDE drives’ to ‘not installed’ and ‘IDE Hard Drives’ to
> ‘disabled’.

> Boot from IDE Disk C Enabled Boot from hard disk C Enabled
^^^^^---This may be part of the problem.

Also, check jumpers and termination on the SCSI drive.


> This morning, on applying power the hard disk spins up to speed and steps
> in several tracks then seeks track 0. There was no other activity and no
> display on the monitor. Removing and reseating the two power supply
> connectors to the main board, things started working like yesterday: ‘ No
> Boot Drive’ message and ‘enter Setup’ message.
>
> I’m stumped! Any Ideas? - ken

If all else is set correctly that hard drive may well not have a boot loader
and/or system on it.

Kenneth L. Owen tx836519

unread,
Sep 26, 2021, 1:19:09 PMSep 26
to se...@googlegroups.com

Hi Steven,

The only thing I have found that changed things was removing and reinstalling some connections.  I will do a wholesale remove and reinstall of ALL jumpers and connectors.  Seems like not running to generate some cleaning heat has allowed oxidation of contact surfaces.

  • ken

 

Sent from Mail for Windows

 

From: Steven Hirsch
Sent: Sunday, September 26, 2021 12:08 PM
To: se...@googlegroups.com
Subject: Re: [sebhc] H37 Disk Imaging

 

On 9/26/21 10:29 AM, Kenneth L. Owen tx836519 wrote:

--

You received this message because you are subscribed to the Google Groups "SEBHC" group.

To unsubscribe from this group and stop receiving emails from it, send an email to sebhc+un...@googlegroups.com.

Les Bird

unread,
Oct 1, 2021, 8:59:17 PMOct 1
to SEBHC
Good progress on the H37 imager. I have a program that can read a disk of any type and send to H8D Utility 3. I'm doing direct WD179x access. The good thing about it is that it'll image just about any disk if it is soft-sector and the WD1797 can understand it.

So far I've created images of:
1. HDOS DS 40 TRACK
2. CP/M DS 40 TRACK
3. IBM-PC DS 40 TRACK (LOL)

I'm not storing any meta data yet - once I figure out how to extract that info (sectors per track, etc) I'll do that. The chip has a READ ADDRESS command that will give me this info, I just haven't explored that yet.

What it's doing right now is using the READ SECTOR with MULTIPLE RECORD feature of the WD1797 which, given a starting sector will read to the end of the track - so starting at sector 1 (for WD1797 sectors start at 1 not 0) to the end - which is reading all the sectors on a track and storing the data in a buffer. It doesn't care how many sectors are on the track, could be 16, 10 or any number. To reconstruct the disk it'll be important to know this so the tracks can be formatted properly but as of now I can read and save the data off of a disk.

I'm only supporting MFM (double-density) right now but FM (single-density) is also planned.

More to come as I progress...

Les

glenn.f...@gmail.com

unread,
Oct 2, 2021, 7:22:45 AMOct 2
to se...@googlegroups.com

Nice progress Les!

Les Bird

unread,
Oct 2, 2021, 12:22:46 PMOct 2
to SEBHC
More updates...

I can now get the sectors per track, sector size and number of sides of a disk.

Sectors per track I can get from the WD1797 chip after reading a full track (IN FD.SEC). Sector size I can get from doing a READ ADDRESS function of the WD1797, 128, 256, 512 or 1024 are supported and reported when doing a READ ADDRESS on a disk. Number of sides is detected by attempting a read of a side 2 and checking for errors from the WD1797.

I've seen the following results:
HDOS = 16 sectors per track, 256 bytes per sector
IBM-PC = 9 sectors per track, 512 bytes per sector
ZENITH MS-DOS = 9 sectors per track, 512 bytes per sector

So now I'm trying to determine how I want to append this to the disk image. What my plan is right now is to do this:
1. Save all data sectors (all bytes written to disk) in read order
2. Append the following text to the end of the disk image: "SECTORS=xx TRACKS=xx SIDES=xx MFM" or " FM"

I want to add this to the end so it doesn't offset the data in the disk image and it can be ignored if not needed.

Les

Douglas Miller

unread,
Oct 2, 2021, 1:04:01 PMOct 2
to se...@googlegroups.com

One other piece of information is the physical sector order on the tracks. That's much harder to discern. The 1797 "read multiple sectors" command retrieve sectors in logical order, so it will not return that. The only way I can think of right now is to do a "read track" and then try and parse the formatting. Or else leave it to the user to manually specify.

Les Bird

unread,
Oct 2, 2021, 1:49:18 PMOct 2
to SEBHC
Douglas,

That's interesting. When doing a read multiple sectors where does it get the "logical" order information from? I thought it would just do a physical order read starting with sector 1 and ending with sector x. By doing that it should maintain the "logical" order of the operating system. I'm not using ROM routines, just doing direct commands to the 1797, so it shouldn't be doing a logical order table look up.

Maybe I'm not understanding this though.

Les

retrogear

unread,
Oct 2, 2021, 2:27:42 PMOct 2
to SEBHC
I am a new member to this newsgroup and this is a helluva thread. I can image almost anything but I don't own early Heath machines. I have a Zenith Z-100 with  8" and 5.25" floppies.
I understand it's floppy controller is a WD-1797. I have a PC setup with a SCSI card and 8" drive that can image anything IMD. Of course I can do any 5.25" 96 or 48 tpi as well.  I would love to assist / test any way I can.

Larry G

Douglas Miller

unread,
Oct 2, 2021, 2:59:53 PMOct 2
to se...@googlegroups.com

Yeah, it's not always obvious what the 1797 does for a command. Some of the better, original, documentation for the WD chips included flowcharts for each command, and these were instrumental in understanding what they do. I recall those flowcharts getting "A+" marks for authentically representing what a command does.

The read multiple sectors command starts with the sector number in the sector register, as I recall, and then increments that number after each sector is found. It searches the track until it finds an address that matches the sector number, then reads the sector data, then increments and repeats. When it fails to find a sector (I think after seeing two index holes - i.e. two revolutions), it stops (so, the "normal" end to that command is "sector not found"). At that point, the sector register will be +1 the last sector found. So, it is reading logical sector order, 1, 2, 3, ... If the track was formatted with a skew factor, as was commonly done to improve performance, then you can't tell that from the read multiple sectors command.

Darrell Pelan

unread,
Oct 2, 2021, 4:39:12 PMOct 2
to se...@googlegroups.com
Impressive progress, Les!

You received this message because you are subscribed to a topic in the Google Groups "SEBHC" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/sebhc/PbejZUeQgUU/unsubscribe.
To unsubscribe from this group and all its topics, send an email to sebhc+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sebhc/f052478c-3510-b625-b1a9-4fa72abbe414%40gmail.com.

Les Bird

unread,
Oct 2, 2021, 7:10:03 PMOct 2