TDI P-System

[Chr$]

I have these, recently purchased from a seller in Germany. He had 10 in total but I was only able to nab 5 of them, so there are another 5 of a similar flavour out there somewhere.

Upon further investigation I found the page for TDI on the QLwiki:

https://qlwiki.qlforum.co.uk/doku.php?i ... re_limited

And TDI seem to have also produced their own 68000 based computer that used the p-system. See from p74 onwards of Practical Computing (Jan 1985):
https://worldradiohistory.com/UK/Practi ... -S-OCR.pdf

Interesting that the p-system used container files, each of the files on the carts seems to be around 100k. Unfortunately, as the files almost take up the whole cart, they are a bit more prone to corruption I would imagine, and all the 6 carts have corruption (often only very low numbers of affected sectors, so perhaps the corruption is not terminal as psystem files must also contain space, right?!).

I have no idea which TDI product these carts were originally part of. I have no paperwork or packaging. For posterity, as these are listed as MIA on the wiki, here are some details:

1. front label: TDI PASCAL P-System ASM, top label: ASSEMBLER, LINKER, UTILITIES, cartname: asm, contents: one 'psystem' file 102848 bytes in length.
2. front label: TDI FORTRAN P-System FORTRAN, top label: FORTRAN COMPILER, FILER AND EDITOR, cartname: fortran, contents: one 'psystem' file 105920 bytes in length.
3. front label: missing, top label: FORTRAN CONFIGURATION OPTIONS, cartname: fmisc, contents: one 'psystem' file 102336 bytes in length.
4. front label: TDI P-System SOURCES, top label: DRIVE SOURCES, cartname: sources, contents: one 'psystem' file 104896 bytes in length.
5. front label: TDI FORTRAN P-System FBOOT, top label: FORTRAN BOOTSTRAP, cartname: fboot, contents (see image):

Image1.png (2.17 KiB) Viewed 1972 times

I'll have a word with Simone to see if he has dumps of these carts somewhere. I'd also be happy to send the dumps to anyone who is knowledgeable enough to have a look at the contents of the psystem files, to see if they really are corrupt, or not!

[RWAP]

I have sent you a PM.

I have a set of the manuals for the Fortran and Pascal (I think) there are a lot of manuals needed for this software. Most of my cartridges have corruption too unfortunately.

[Chr$]

I have sent you a PM.

I have a set of the manuals for the Fortran and Pascal (I think) there are a lot of manuals needed for this software. Most of my cartridges have corruption too unfortunately.

It'll all go stratospherically over my head, but I suppose it may enable me to read the contents of the container files. On some it really was only a case of 1 or 2 bad sectors.

[M68008]

Not sure about the p-system file format, but if you want to send me the dumps, I can at least take a look at the mdump sectors, perhaps it's possible to see if partially-read damaged sectors look empty or not.

With mdump, you can also alternate the src drive to maximize the probability of success. (And of course all the usual hacks like pushing or slightly pulling the cartridge while it's running may also help).

Finally, I have some code to merge dumps, so I could try to merge the ones from Chr$, Rich and Simone. If they are the same version of the software but have different errors, the merged dump would contain more good sectors.

[Chr$]

Not sure about the p-system file format, but if you want to send me the dumps, I can at least take a look at the mdump sectors, perhaps it's possible to see if partially-read damaged sectors look empty or not.

With mdump, you can also alternate the src drive to maximize the probability of success. (And of course all the usual hacks like pushing or slightly pulling the cartridge while it's running may also help).

Finally, I have some code to merge dumps, so I could try to merge the ones from Chr$, Rich and Simone. If they are the same version of the software but have different errors, the merged dump would contain more good sectors.

I'll email you the dumps now Daniele.

[M68008]

I'll email you the dumps now Daniele.

All the cartridges have a few bad sectors. One sector in the first one is also completely missing from the dump... the sector header must be corrupted.
Unfortunately, the partial reads of the bad sectors don't look empty, in the sense that they contain varied data rather than a single value or short repeating pattern.

I'll try the Merge function, but at the moment it only works for cartridges with identical sector maps. If that's not the case, I'll need to add some logic.

[Chr$]

I'll email you the dumps now Daniele.

All the cartridges have a few bad sectors. One sector in the first one is also completely missing from the dump... the sector header must be corrupted.
Unfortunately, the partial reads of the bad sectors don't look empty, in the sense that they contain varied data rather than a single value or short repeating pattern.

I'll try the Merge function, but at the moment it only works for cartridges with identical sector maps. If that's not the case, I'll need to add some logic.

I have some more dumps of the TDI carts, will send them to you. Hopefully the bad sectors are in different places.