search

davidgiven / fluxengine

PSOC5 floppy disk imaging interface
MIT License
358 stars 69 forks source link

SCP (from GreaseWeazle) to Flux to Victor 9k #123

Open drdpj opened 4 years ago

drdpj commented 4 years ago

Hello - it's having a bit of a hard time converting this to anything useful. I converted from SCP to flux, then tried to read as victor 9000. It was captured on an 80track 1.2MB drive, so was spinning at 360rpm. Single sided, so a 600kb victor disk, MS-DOS format.

v9kdump.zip

davidgiven commented 4 years ago

Interesting --- I see bad sectors down near track 0:

H.SS Tracks --->
0. 0 .B.B.........B..................................................................
0. 1 B...............................................................................
0. 2 .BB.............................................................................
0. 3 ..BB............................................................................
0. 4 B.BB............................................................................
0. 5 BB.B.................B..........................................................
0. 6 BB........................B.....................................................
0. 7 B..B.........B.B.....B..........................................................
0. 8 B.BB...B...B.BB.................................................................
0. 9 BBB.............................................................................
0.10 B............B.B................................................................
0.11 BB...........B.B................................................................
0.12 B..B...................................................................XXXXXXXXX
0.13 B..B..........B.............................................XXXXXXXXXXXXXXXXXXXX
0.14 B..B...........B................................XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
0.15 BBBB..........BB......................XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
0.16 B.B.........B..B.......B.B.XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
0.17 BBBB...B.....B..XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
0.18 B...XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

I'd be inclined to think it's just not a very good disk, but IIRC you said that you knew it was good? How? Do you have a real Victor 9K to test on? That suggests that my decoding could be improved (which is always true).

Also, given the triangular disk format, the MS-DOS disk format will have to be tweaked to work there. Do you know anything about this?

drdpj commented 4 years ago

It does indeed work on a real V9k. I'm not sure what MS-DOS format looks like on victor disks in truth as I've never managed to get one copied across. Do you have a copy of the technical manual?

drdpj commented 4 years ago

dos31.zip Here's another one - this is an original factory duplicated disc (dos 3.1).

davidgiven commented 4 years ago

If you have access to a V9K, then what woud be invaluable is a file system disk image of one of these produced on the machine itself --- that would allow me to compare the data produced by FluxEngine with the known good data and figure out precisely where the bit errors are (and therefore find out why they're there).

(Also, wow! Aren't they super rare?)

drdpj commented 4 years ago

I guess they are fairly rare. I got mine back in the early 90s. It's full of 6522s (thanks Chuck :))! - the issue is getting a file system image off the blasted thing. It is still working, I just checked it. The screen is a bit wobbly. I wonder if some of the old capacitors are feeling a bit unhappy. I think I need a null modem cable though to have any hope of getting anything off it...

drdpj commented 4 years ago

One more... Here's a blank formatted disk from it (again, single sided). This should be more predictable! v9kblank.zip