• Please review our updated Terms and Rules here

Bad disk image?

snuci

Veteran Member
Joined
Nov 22, 2012
Messages
1,551
Location
Richmond Hill, Ontario, Canada
I have a Nabu PC that I've gotten disk images for. It runs CP/M 3.0. I've been trying to write out the disk image and it seems I am successful as I used the original 180k SSDD drive on a PC to write it out with IMD. I can read it back, dump both images and they compare correctly. I'm not sure if this is a 100% foolproof method but I think I am writing out and reading whatever the image is properly. It does not work. On the Nabu PC it starts to boot, doesn't go beyond track 1 (I think) and then continues to spin with the drive light on forever. If I put in another CP/M disk, I get a Boot Error.

There is no definition in 22disk for this and it's obscure enough that it makes sense but I can't seem to get a proper manual definition made. I am not a registered used of 22DISK so this is unsupported (I missed that registration window). Anadisk says there is data but no ids for each sector and keeps trying to add a sector of 105. Just for clarity, this disk is 5 sectors x 1024 bytes per sector SSDD. Just so I am not going crazy (because I have spent a few days on this computer), does the directory below look correct? If it is, I will soldier through and try to sort this out but I am thinking it just may be a bad disk image. My image PC is a P2 with Windows 98SE and the image is in IMD format. I used the original drive from the Nabu PC after trying a 360K drive and a 1.2MB drive with appropriate settings. Adaptec AHA-1542CF controller right now but I ordered an AHA-1522 just in case. I think I have tried everything so I just want to confirm I am not starting with a bad image.

This is a dump of the image with IMDU /B. It just doesn't look right according to the 22DISK example but maybe CP/M 3.0 is different with directory structure?

rawdisk.jpg
 
A quick look at the images doesn't show any obvious errors. What a bizarre little system, though! A single directory block with slots for 32 file entries for the whole floppy. Now, I'll admit that the CP/M 3 stuff takes up a lot of space (the whole disk is only about 200KB), but 32 directory entries does seem to be a bit thin. I'll have more when I get a bit of time.

Was this some sort of training system?
 
It is an interesting system. This was a device that was for the Canadian Nabu Network. It was a "PC" that would normally download it's OS from the cable network via a separate box (the default configuration). More info on the Nabu Network here: https://en.wikipedia.org/wiki/NABU_Network

The Nabu PC also had the option to have dual local disk drives that could boot CP/M. The one I have is paired with a Nabu 1100 development server that is S100 based (haven't fired that up yet) so I believe it was used for development/testing. I'm not sure how many Nabu PCs actually had the disk drives but it would have been uncommon. Mine did not come with the network box but I did get one afterwards just to complete it.

Thank you for looking into it and please take your time. There is no rush.

Here is what it looks like (with the dual drive in pieces). The on-screen message is different without the FDC card. It can also boot from the network if that is connected and present but the network box is not attached. It's about the same size as the PC and stacks on top.

NabuPC-floppy_not_working med.JPG

From the tech specs:

This design specification defines the Internal Operating
Software (IOS) for the NABU Personal Computer (NPC), a lowcost,
expandable personal computer. It is unique because it
is capable of communicating on one-way, hybrid and two-way
cable systems and telephone networks, as well as operating in
a stand-alone mode, depending on which options are selected.
When used in association with a CATV network the NABU P.C.'s
prime function is to run software downline loaded from the
cable head-end.
 
Based on your initial data, and assuming this is a 5" floppy drive, then this looks to be a 40-track single-sided double-density floppy (200K). I am reasonably confident that the CP/M DPB is:


Code:
spt: dw  40
     db  3,7,1
dsm: dw  194
drm: dw  95
av0: db  11100000b,0
cks: dw  24
off: dw  1

Not sure how that translates to 22disk.
 
Based on your initial data, and assuming this is a 5" floppy drive, then this looks to be a 40-track single-sided double-density floppy (200K). I am reasonably confident that the CP/M DPB is:


Code:
spt: dw  40
     db  3,7,1
dsm: dw  194
drm: dw  95
av0: db  11100000b,0
cks: dw  24
off: dw  1

Not sure how that translates to 22disk.
In addition, there appears to be a logical sector skew being used, of "2", meaning first sector of directory is 1, next sector of directory is 3, etc. Most likely, sector translation table is 1,3,5,2,4 (1K sectors, 5/track).
 
Thank you durgadas311. I took the clue from Chuck(G) that there were 32 directory entries and part from you with the "av0: db 11100000b,0" line and was able to build around that. My 22Disk definition is as follows:

BEGIN NABU Nabu PC CP/M 3.0 SSDD 5.25"
DENSITY MFM,LOW
CYLINDERS 40
SKEW 1
SIDES 1
SECTORS 5,1024
SIDE1 0 1,2,3,4,5
BSH 3 BLM 7 EXM 0 DSM 195 DRM 31 AL0 11100000b AL1 0 OFS 1
END

Here are the results. On first glance, it looks good but I will have to do some further checking.

screen.jpg

Sadly, that is not good for the Nabu PC as that would mean that there is something wrong with it and I checked all but one chip on the Floppy disk controller board. I know the floppy drives are good because one of them was used to read this disk.

Thanks to both of you for the clues. Looks like it's back to the drawing board for me.
 
@durgadas311 There definitely is no logical sector skew. There is a single directory sector. As I said, a bizarre little system, but after snuci's explanation of its application, that does make sense. I do see some erased file entries on the second image, which looks like someone tried to run RMAC or ASM on a test file.
 
@durgadas311 There definitely is no logical sector skew. There is a single directory sector. As I said, a bizarre little system, but after snuci's explanation of its application, that does make sense. I do see some erased file entries on the second image, which looks like someone tried to run RMAC or ASM on a test file.
Have a look at the disk image, NABUPER1.IMD. You can look at the contents of the HELP.HLP file, which starts right after the directory. The only way the data makes sense is with a sector skew of 2.

Also, while the current used directory only spans one sector, the total directory size is clearly 3 sectors by looking at the data on track 1. This also shows the skew, since track 1 only contains directory and the first blocks of HELP.HLP.
 
Gotcha--completely missed the e5 blank directory block at 1C00. It was before my first cuppa... :(

Snuci, take note:

SIDE1 0 1,2,3,4,5 should be SIDE1 0 1,3,5,2,4
DRM should be 3*32-1 = 95.
 
This is what I used:
Code:
BEGIN NAB1  Nabu PC - SSDD 5.25" 48 tpi
DENSITY MFM, LOW
CYLINDERS 40 SIDES 1 SECTORS 5,1024 
SIDE1 0 1,3,5,2,4
BSH 3 BLM 7 EXM 0 DSM 194 DRM 95 AL0 0E0H AL1 0 OFS 1
END
Probably isn't going to help with your non-booter
 
Last edited:
This is what I used:
Code:
BEGIN NAB1  Nabu PC - SSDD 5.25" 48 tpi
DENSITY MFM, LOW
CYLINDERS 40 SIDES 1 SECTORS 5,1024
SIDE1 0 1,3,5,2,5
BSH 3 BLM 7 EXM 0 DSM 194 DRM 95 AL0 0E0H AL1 0 OFS 1
END
Probably isn't going to help with your non-booter

Thank you. I will use these settings and see if I can get things going. I do think this is yet another project but if/when I get it working, it will be a conquest. Apparently nothing is ever easy :)

Much appreciated.
Santo
 
NOTE! The "SIDE1 0 1,3,5,2,5" should be "SIDE1 0 1,3,5,2,4". Otherwise, it certainly won't do the right thing as far as non-directory blocks (file data blocks).
 
Sorry if this is a bit off topic, but @snuci your picture of the FD controller board got my attention. Can you post pictures of it here? The NABU technical manual includes a schematic of the FD controller board but it shows a circuit using Western Digital’s FD1793 and a phase lock loop using WD1691.

The picture I’ve found (so far) shows a FD controller with Western Digital’s FD1797PL-02 and FDC9229BT. I’d be interested in getting any information or photos of your FD controller (both sides please?) Thank you in advance.
 
@1985MR2

I've made a specific post about the Nabu PC FDC controller pictures. I've had a few requests and am trying to focus the development discussion to a thread here: https://forum.vcfed.org/index.php?t...abu-pc-floppy-disk-controller-thread.1241006/

I don't remember and FDC schematic in the manual. Can you send me a copy to see if I have the same one?
Thank you @snuci for creating a new post with the pictures. As requested, here is the link to the FD schematic in the T1100 technical guide (see pages 98-159) https://vintagecomputer.ca/files/Nabu/Nabu-1100-Technical-Guide.pdf

The schematic in the Technical Guide is for the AFC1100 Floppy Disk controller made by Andicom Corporation. Instead of using the FD1797PL-02 and FDC9229BT chips that are on your controller, it used the FD1793B and WD1691 chips. If you look at D J Sures' video
, at the 6:05 mark, you'll see another variant of the floppy controller called the "HCC FD CONTROLLER 42-9002062". That one appears to use different chips from your controller and it doesn't appear to match the one in the schematic. So my search goes on for the documentation on the NPC FD CONTROLLER 9002344.
 
The AFC1100 board is the S100 board in the Nabu 1100. It's a different board altogether. You can see the Andicom board on my 1100 here; https://vintagecomputer.ca/nabu-pc-and-development-server/

DJ Sures also showed that board or one of the other Andicom boards from the Nabu 1100. The one you posted I've also seen from his video and I would be curious what the cermic chip is. With a 37 pin connector and no power for the floppy drive unit, it looks like it might be an 8" drive board similar to the one on the Nabu 1100 that specifically fits into the Nabu PC. Can't tell for sure though.
 
Back
Top