It seems like bit 14 is stuck high:I'd also guess IC45 is dead. It's missing the high byte.
It is getting commands which are lower byte only, aiui.
You still might need to try a different device or Wizzo, but wizzo is unlikely to fix half the data missing.
- Cylinder count: 40732 - 32768 = 7964
- Head count: 32784 - 32768 = 16
- Sector count: 32831 - 32768 = 63
If you look at the Identify Drive dumps from https://www.os2museum.com/wp/identify-ancient-drive/ -
Code:
$ curl https://www.os2museum.com/files/atadump/cp342.bin 2>/dev/null | hexdump -C00000000 5a 0a 26 03 00 00 04 00 13 41 69 02 1a 00 32 00 |Z.&......Ai...2.|00000010 0c 00 00 00 31 42 59 4c 34 33 20 20 20 20 00 00 |....1BYL43 ..|00000020 00 00 00 00 00 00 00 00 01 00 01 00 04 00 35 30 |..............50|00000030 35 38 20 20 20 20 6f 43 6e 6e 72 65 50 20 72 65 |58 oCnnreP re|00000040 70 69 65 68 61 72 73 6c 20 20 30 34 42 4d 2d 20 |pieharsl 04BM- |00000050 43 20 33 50 31 34 20 20 00 00 00 00 00 00 00 00 |C 3P14 ........|00000060 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|*00000200I'd concur with IC45 (74HCT573) or the connections to bit 6 being iffy -- but if that doesn't fix it then (a) everything I said above is backwards, and (b) the problem is IC500 (74HCT245).
When I fixed my A4000 (which had this exact fault) I ended up swapping both chips, mostly because I got the bytes the wrong way around...
I also had an issue where the drive wouldn't format, and that turned out to be IC46. That's a particularly pernicious fault, as the IDE Identify will work fine, but every other byte sent to the hard drive will end up being corrupted.
Statistics: Posted by philpem — Wed May 07, 2025 4:47 pm