In article E1PdUCJ-0004Rh-63@beholder, John Elliott wrote:
If the discs were formatted with a WD17xx series controller, it's possible that the first sector is too close to the index hole for the PC's floppy controller to detect.
Yes, that was my thought. However, the s/w I'm using (cw2dmk) can be told to ignore the index hole and the data is no better in this mode. I'm guessing the data was dodgy when written 20+years ago, though why it's just the first sector after a track change I can only guess at. Write affected by head stepping? Who knows!
Anyway, I can now get low, real low, magnetic domain transitions on the disk low, which makes me happy. I've captured the data off all eight disks with a variety of settings and can now start writing code to parse the data. I reckon that over 99% of the data is present and correct, and another 0.9% will be there with good CRCs even if the headers are mangled. The remaining 0.1% will be "interesting".
Ian