• Please review our updated Terms and Rules here

Bad disk image?

this is getting frustrating. Similar to the DSDD diskettes, I have the original SSDD drives in the NABU. I format diskettes, write files and I could even transfer files with 22Disk so I have all of the CP/M Plus files on the disk but as soon as I image the disk with ImageDisk and write out a copy, there are no files shown in a subsequent directory listing on the NABU. 22Disk sees them fine. I then write a file to the disk and it's the only file that shows up. I just imaged that disk and the directory for that one is roughtly at 0x3F10 of the IMD file. The NABU happily will do a directory listing of that directory but not the one at 0x1400.

I am using an earlier Rev A version of the firmware. It does not have the option to boot from a floppy. I am using this because the NABU Network software doesn't work with Rev B. The Nabu computer with floppy drive came with Rev B. I may be chasing bugs or something. I don't know why imaging and writing a disk would make any difference unless it was sompletely off but I don't specify anything except for 40 tracks, 1 side, single stepping. I did have to use ldkraemer's 22Disk definition because otherwise, there was a conflict.

I will try to figure out something else but I am afraid the Rev A firmware may have me running around in circles.
 
snuci,
No, the images that are used in B: can not be read or booted in A: since the Disk Parameter Block
information is different.

Do you know someone that has a Debian Linux Desktop with a 5.25" 360K Floppy Drive. You should be able to
read the single sided Image with Debian Linux (versions 8 thru 11) by setting the floppy parameters
in the /etc/mediaprm file.

If you have a Greaseweazle, Catweasle, Supercard Pro, or Kryoflux you can copy the Boot Floppy that boots
the system from Drive A:.

The other thing you could do is define a new 22DISK definition, and try accessing that floppy with 22DISK
to be able to backup the files.

BEGIN NAB0 Nabu PC - SSDD 5.25" 48 tpi 32 Track - 1024 x 5
DENSITY MFM, LOW
CYLINDERS 32 SIDES 1 SECTORS 5,1024
SIDE1 0 1,3,5,2,4
BSH 3 BLM 7 EXM 0 DSM 151 DRM 63 AL0 0C0H AL1 0 OFS 1
END

That would allow you to verify if the Directory is correct for the floppy in A:.
Just don't write to the floppy in A: until you are sure there are backups.

After Booting from A: will you run these commands and see what is displayed?
A:> STAT A: DSK:
A:> STAT B: DSK:

Do you happen to have a DOS Computer?

Larry
 
Last edited:
I understand what you are saying but once I create a diskette with one of the NABUPER1, NABUPER2 or NABUPER3 images I can read them with the definition (40 CYL 5x1024 sectors) above in 22Disk with no problems. I can read newly created diskettes that come from the NABU with the same definition in 22Disk. I can write files to the diskettes. I can read those diskettes in the NABU PC. When I write images to diskette, I can read them with 22Disk but the directory location appears to be wrong when I try to read them in the NABU PC. Just for diskettes that were created from an image, from my experience.

My thought was that if the Disk Parameter Block was somehow off on diskettes created from ImageDisk, it might throw off where the NABU expects the directory. The original NABUPER1 attempts to boot. Once it gets past the boot track, it hangs. If I try to look at the diskette in 22Disk, it looks fine. If I look at the same diskette on the NABU, the directory is garbage or blank. I am probably not explaining this properly but it is strange behaviour.
 
So if you write a long basic program and save it as ASCII ? SAVE "PROGRAM.BAS",A ? That would make easier to understand the sector logic. Just bunch of REM's would do it since they are numbered like: 10 REM / 20 REM / 30 REM... The sectors are 128 bytes but we don't know the block size. Don't save more than one file, unless you identify each program like 10 REM 1 / 20 REM 1 / 30 REM 1.... and 10 REM 2 / 20 REM 2 / 30 REM 3. I think 32 lines from 10 to 320 would be enough. If we figure out the exact geometry, maybe we can re-order the CP/M disks as longs every sector is there.
 
Will you post the output from each of these commands?

A:> STAT A: DSK:
A:> STAT B: DSK:
A:> NABUSHOW A
A:> NABUSHOW B

Embeded in the help files I see several different Double Density floppy's listed for access
KAYPRO, OSBORNE, CROMECO (Single Density), NABU, IBM, and DEC VT-80

If you were to format a fresh floppy in A:, and have it handy, you could Boot the system
Boot Floppy, and Execute
A:> BACKUP

You could BACKUP A: to A:, then re-image that copy of your boot floppy to see if it matches
the NABUPER1.IMD.


Larry
 
Larry,

I don't have a CP/M Plus command prompt to work with. I only have a list of "Disk Utilities" within the NABU network download software. I don't have a CP/M diskette I can boot from because I cannot make or restore from disk images and boot from them on the NABU.

I have this menu in the network downloaded software and there is no easy way to get to a command prompt unless there is a "back door" somewhere. If any of this doesn't make sense, I will happily make a video and show you how it works.

cpm_utils (Medium).JPG

I have done the following:
1. I have take NABUPER1.IMD and created a diskette copy.
2. I have used the 22Disk definition that was originally determined through this thread to copy the CP/M files from NABUPER1.IMD to my hard drive.
3. I have formatted two diskettes via the menu above. Works great.
4. I taken these two diskettes (one was blank and the other had a "TEST.DOC data file on it to see if that would mess up along the way) and I copied the NABUPER1 files on my PC via 22Disk to these physical floppies.
5. I then read them back via the NABU disk drive on the NABU with no issues.

Here is a DIR via the Disk Utilities menu of the A drive and B drive.
dir_a (Medium).JPG

dir_b (Medium).JPG

In trying to STAT according to your instructions, I cannot add the DSK: keyword as this utility doesn't allow it.
stat_error_dsk (Medium).JPG

Unfortunately, it doesn't have much more info than this so here is the STAT B: (remember A has a TEST.DOC so it has more disk used)
stat_b (Medium).JPG

Here is the SHOW command for A and B
show_a (Medium).JPG

show_b (Medium).JPG

The trouble starts when I try to image a disk via ImageDisk and try to write the image back via ImageDisk. When I do this, the disk I just showed you will not show any files when I do a DIR. If I write a single file to it, I will only see the single file name in a DIR.

As per your instructions, I can happily run BACKUP on the Disk Utilities menu and backup diskettes I created from A: to B:. But that doesn't get me anywhere because I don't have a diskette with a boot track. I have taken these diskettes with all of the NABU CP/M files from the NABUPER1 disk image and imaged it via ImageDisk then used a hex editor to write the boot track to this disk. However, the problem lies with writing this disk from IMD file to physical disk and reading back on the NABU.

If I did NOT have to image a disk (and restore it, trying to use the new copy) or try to boot from the NABU CP/M Plus disk that was restored from an image, I would have zero issues. Where the trouble starts is trying to use a diskette copy created from ImageDisk or any other tool. I also have used an Applesauce device that works with IMD files and this exhibits the same behavior.

I will put an example in a separate post so I don't confuse things.
 
In this example, I did the following steps:
1. I have formatted a diskette on the physical NABU hardware via the Disk Utilities menu as shown in the last post. Let's call this diskette "OLD".
2. I have used 22Disk on a PC to write the files from NABUPER1.IMD (retrieved via 22Disk with the NABU definition) and placed them on this "OLD" diskette
3. I have then been able to access these files on "OLD" via DIR. STAT is correct and it is one of the ones shown above as a STAT example. The files are all there.
4. Note that I have no means of running any of these files as there is no option on the NABU Disk Utilities menu to run a diskette based executable file.
5. I have taken that floppy diskette and imaged it on a PC with ImageDisk.
6. I then took that IMD file and written it to a new floppy diskette. Let's call it "NEW".
7. I can view the directory and access the files via 22Disk and the NABU definition on the "NEW" diskette.
8. I take this diskette to the NABU and run a DIR on it and there are NO FILES. It is like a blank diskette according to the NABU DIR command.
9. I write a file called TEST.BAS to "NEW" diskette.
10. I do a DIR on "NEW" and I see TEST.BAS.

Here are some screen shots.

Here is the "NEW" diskette in 22Disk showing the directory AFTER writing "TEST.BAS" to it. Notice there is no "TEST.BAS" file here but the NABUPER1 files.
22disk (Medium).JPG

Here is this same disk moved from my PC to the NABU and running the DIR command:
NABU_DIR (Medium).JPG

Here is the NABU STAT command. This number is wrong on all counts. The disk is almost full and the TEST.BAS is a very small file (certainly not 84k or whatever the exact difference is.
NABU_STAT (Medium).JPG

Here is the SHOW on that disk. Same bytes remaining and same disk geometry as the drive but it now thinks that this is an Osbourne disk! The previous example of stat showed "NABU Double Density Disk".

NABU_SHOW (Medium).JPG

Here is the BADNABUE.IMD file I used: https://vintagecomputer.ca/files/Nabu/disk_images/BADNABUE.IMD

This is an expanded IMD so I can view it via HexEdit. You will find the original full directory at 0x1487 (ofset because of the IMD description). Should be 0x1400. You will find the "TEST.BAS" directory entry at 0x3F46. NABU sees this directory entry here, probably because it thinks it's an Osbourne diskette? I am not familiar with where the directory is on those.

This is the problem.
 
I took the fact that the NABU PC thought that the imaged diskettes were Osborne (Osbourne) diskettes and did the following:

1. While using the REV A formware, I formatted a floppy on the NABU PC. NABU thought it was a NABU floppy with SHOW A:.
2. Used 22Disk on a PC with the "OSB2" type to write all of the NABUPER1 CP/M files to this diskette. 22Disk sees the files.
3. Read this diskette on the NABU PC. No files were found.
4. Created an image of this Nabu created diskette with ImageDisk (Cylinders 40, Sides 1, Double-step Off).
5. Saved it as an expanded IMD to be able to see and work with it.
6. Wrote a new floppy with this image. 22Disk sees the files as an OSB2 diskette.
7. Read the diskette in the NABU PC. I can see the files! SHOW A: thinks this is an Osborne diskette (as expected).
8. Created an image of this diskette with ImageDisk (Cylinders 40, Sides 1, Double-step Off).
7. Saved it as an expanded IMD to be able to see and work with it.
8. Took an RAW dump of NABUPER1 (bootable diskette) so I could work with it.
9. Wrote 0-3FF, 400-7FF, 800-BFF, C00-1000 and 1000-1400 onto the NABU/OSB2 diskette to try to make it bootable.
10. Used ImageDisk to make a new floppy.
11. Read it in the NABU and the files are all there.
12. Put in the REV B firmware with the menu option to boot from diskette.
13. Tried to boot...

No luck :(

The image which works great in the NABU PC with REV A and is considered to be an Osborne 2 diskette is here: https://vintagecomputer.ca/files/Nabu/disk_images/NABO2BOT.IMD

I believe I am out of options.
 
I need to IMGdisk my personal CP/M licenses and send them to you. That has put floppy extraction on the critical path. I'm done gathering NABU stuff anyway. I've got all the good stuff out of my basement and shed.
 
I will mail what I consider to be a copied "CP/M System" from my stash of floppies. I hope it will work for you. Check your mail and send me your address.

Leo
 
Hey guys, before I dig into the images, can you explain exactly what you're looking for? If it's a problem with the NABU, I don't want to touch it, btw.
 
@snuci Do you have a different older floppy controller you could try on the IMD end , maybe there is a slight timing issue?
 
Chuck(G) said:
Hey guys, before I dig into the images, can you explain exactly what you're looking for? If it's a problem with the NABU, I don't want to touch it, btw.
Click to expand...

I guess I don't know where the problem is. The NABU can read NABU diskettes but when I take the diskette out, image it, then write a new disk and insert that disk with no other changes, the NABU thinks it's an Osborne diskette. That sounds like it might be the NABU? I need to do some more reading on the Disk Parameter Block perhaps? I am just confused and am not sure you can help Chuck(G). Maybe it's best to wait for Leo's CP/M floppy. He is going to mail it to me.

Klyball said:
@snuci Do you have a different older floppy controller you could try on the IMD end , maybe there is a slight timing issue?
Click to expand...
It's a good thought bur I am using an Adaptec AHA-1520A controller. It's supposed to be very good, compatibility wise, hence why I am using that and not the on-board floppy controller of the computer I am using. I am running Windows 98 but I switch to DOS to do the imaging as it doesn't work too well through the Win98 GUI.

We have working floppies and as long as they are created as an Osborne diskette, they can be imaged and restored at this point. I REALLY want to try the REV B of the NABU firmware but unfortunately, it doesn't work with the current NABU Network Adapter software. It looks for a PAK file on start-up that doesn't exist on the NABU Network.
 
No, not s far as I know. It starts to load files from the NABU Network like a REV A ROM but prior to the main menu, it looks for something else that is not there. If it was the first file it was dying on, I would understand that it might be referenced in the EPROM. Unfortunately, I can't do anything after that other than reboot and do it all over again.
 
So I'm wondering if using a fluximager, such as kryoflux or greaseweazle, would make any difference in the ability of the nabu to recognize the disk after image and writing it back to disk. With a flux transition image you shouldn't need to worry about how the disk is laid out when writing things back.
 
At the present time, the NABUPER1, NAPUPER2 and NAPUPER3 images are ImageDisk images. Flux imaging will likely help but not a lot of people have the hardware for that. The issue really is what we have to start with and they are .IMD files.
 
GreaseWeazels are pretty inexpensive considering it's one of the best ways to work with floppies.

github.com

Purchase a Greaseweazle

Tools for accessing a floppy drive at the raw flux level - keirf/greaseweazle
github.com github.com

I have one, but I'm in Perth, Australia, so it's probably easier for you to order one!

Regarding conversions, I've found the HxC floppy emulator app is the best to go from basically any format to almost any other.
 
You must log in or register to reply here.
Back
Top