UPDATE: SECTORS EXTRACTED! (see bottom of this log)

----------------------

at this point I'm in way too deep to try out this trick I read recently...

If I recall/understood correctly, apparently the disk controller in IBM-compatible PCs tend to have a particular characteristic that makes it hard for them to read sectors immediately after the index-mark. So, ten-sector-per-track disks are sometimes hard for them to read.

Hmmm

I can't vouch for this, but it seems the vast majority of sectors I haven't been able to read are all sector-zero (or 10, on head 2)... is it likely they started with 0 immediately after the index?

So, there's that.

The solution...? They say the trick is to adjust a potentiometer on the floppy drive which changes the rotational-speed of the disk. Slight adjustments to make it spin slower won't affect the baud-rate dramatically, but will give the disk-controller just a bit more time to respond after that index-mark.

Well, I don't particularly like the idea of tweaking this floppy drive, since it works in the original system. But it's nice to know if it comes to that. And if I'd had any idea how deeply this path'd've taken me, I probably *would've* gone that route before going this one...

And...

"The Trick" from before... in which I've diven, dived, dove so deeply... Now at a couple thousand lines of code, seems to be locating sector 0's! So there's that.

So, basically, what I did was tell the disk-controller to extract a single sector that happened to be 16384 bytes long. It found that sector, then extracted 16384 bytes, which caused the disk to spin around a couple full revolutions... So this file I've been processing has each sector in it a couple times... Looks like I should be able to extract every sector except 18 without a hitch (and, I managed to get 18 somewhere else).

Can't really vouch for the oddnesses I've encountered. Why *so many* failures...? Dunno. e.g. Sector 14's ID was read to be 1024 bytes, in that particular go-round, but I have it checking the Data-CRC every 128 bytes, and it found a match at the more-logical 512.

Been claiming I'm planning to actually *extract* the sectors for days, now... but I think I'm really going to be doing it soon, real soon.

---------------------------------

UPDATE:

I've managed to whittle the missing-sectors list down by 2/3rds!

So... only 6 sectors I haven't been able to read-back at all, yet...

Only two which remain "read-with-errors"... not bad, considering the literal hundreds of missing sectors I started out with all those months(?!) ago.