On 9/7/20 1:17 PM, Fred Cisin via cctalk wrote:
BTW, XT compatibles with the Xebec (and Xebec-like) hard disk controllers would not give a "COMMAND ERROR" message for a Low-level-formatted/wiped hard disk. Probably a "1701" error.
This one's a Xebec, almost identical to one of the branded variants that IBM used, actually. It's got the four jumpers lower-right for selecting different drive types, but it also has jumpers for ROM and I/O address selection, which I don't think I recall seeing before on the boards that IBM used.
One of the ROMs is labeled as "104833A", which isn't too inconsistent with what I've seen on IBM ones (they generally seem to have a 10483xx format). The other ROM has a Mitsubishi Electric sticker though, labeled as "FXD J12" and "213". One of the functions is presumably to hold custom drive parameters, selected via the four board jumpers, but I don't know what other stuff might typically be in that ROM - I didn't want to automatically rule out it being the source of the "command error" (where other responses might be issued for the same underlying error by other firmware)
The hard drive in the system is a Miniscribe M3425; at power-on it spins up, runs the heads in and back out, then blinks its LED once - which I believe is normal, "all is well" behavior, and if there were a significant fault with the spindle speed or positioning then it would start blinking out an error code.
And, even if it were somehow "compatible" with a completely different hard disk controller, "COMMAND ERROR" doesn't sound like anybody's error message for a wiped drive. More likely to be a generic "READ ERROR"
I agree "read error" would seem more sensible, but this thing's quirky enough that I wouldn't rule anything out right now. I didn't actually try starting the system with the hard disk controller in place but the drive not attached - maybe I'll do that just to see if there's any change in behavior.
Jules
