Still more
Aussie bundle goodies. (Or baddies in this case. (Read on...))
This time it's
Tile Mosaic by Hilditch Software and Cambridge Micro Software (Cambridge University Press).
This one came on a 5.25-inch DFS floppy disc that was very well protected -- so well protected that I had a hell of a time trying to export it to a DataCentre, and I'm still not entirely sure that the disc image I've attached below is totally error-free.
The BASIC programs that make up Tile Mosaic were hidden away on dodgy tracks on side 2 of the disc -- a disc which, mind you, also had 40 tracks on side 0, with a machine-code !BOOT that did OSWORD sector-reads and switched you to side 2 if the drive was 80-track; and it was hacked so that it looked like the floppy had fewer tracks on side 2 than it actually did, the BASIC programs being located on the "extra" tracks.
On side 2 (the "truncated" 80-track side), a machine-code program called BOOT was decrypting parts of itself and doing lots of EOR checks to make sure you hadn't tried to alter it. It then read certain sectors of the disc, depending on the value of memory location &3AA, which was initially set by the !BOOT on side 0. (Hope you're following this.) Each batch of sectors that BOOT read off the disc contained a BASIC program, which would be OLDed and then RUN. And each BASIC program in turn changed the value of &3AA to signal to BOOT which BASIC program needed loading next.
Argh.
This was the first time I'd been faced with a disc quite this evil. I disassembled !BOOT on side 0 and got a few lines into disassembling BOOT on side 2 before I gave up in confusion. EORs and BITs never did agree with me. So I switched tack and fired up the
Advanced Disc Investigator ROM instead -- which I hadn't used in anger before. (What a time to learn!)
Using ADI, I managed to find the "hidden" tracks on side 2, which had been given non-standard sector IDs: sectors are usually numbered from zero to nine, but tracks 59 and above had sectors numbered 128 to 137 instead. I think that was why my DataCentre couldn't make an image of the floppy disc: it was expecting to see standard sector IDs. I think there were also a few completely unformatted tracks thrown in for good measure (not to mention a track which had suffered genuine damage in transit, but which I fortunately managed to reconstruct using data from side 0).
So I used
Ripoff IX to clone the original disc to a blank floppy, which it managed to do with aplomb, I have to say (certainly more aplomb than had so far been displayed by me). I then used ADI to reformat the non-standard sectors on the floppy-copy so that they had normal sector IDs. And finally I transferred the data for those sectors from the original floppy (with the non-standard sector IDs) to my floppy-copy (with normal sector IDs). This was made all the more tedious by the fact that I only have a single floppy drive, so I had to keep swapping discs after every track. (Argh.) Anyway, what I ended up with was a copied disc which wouldn't actually work (because BOOT couldn't find the dodgily numbered sectors any more), but which could at least be exported to the USB stick in my DataCentre as a .SSD disc image.
Then it was just a case of hacking the BASIC programs out of the .SSD file with a hex-editor on my laptop and CHAINing them together in the right order before finally giving in to total nervous collapse.
Was it worth all the effort? I'll let you be the judge of that.*

- Screenshot of Tile Mosaic

- Screenshot of Tile Mosaic

- Screenshot of Tile Mosaic
Play online:
Download:
- TILEfin.ssd.zip
- .SSD DFS disc image of Tile Mosaic
- (15.63 KiB) Downloaded 47 times
- broken_TILE.zip
- The trail of broken images I left as I tried to hack Tile Mosaic
- (304.96 KiB) Downloaded 44 times
* The answer is no.