27 Jun
2012
27 Jun
'12
12:37 p.m.
Instead of tying in at the MFM signal level, why not
replace the drive
controller chip with something that looks like it at the register level?
_What_ disk controller IC? Not all classic computers use a single-chip
disk cotnrolelr...
A case in point is nthe PERQ 2T2 and 2T4. The disk controller on that is
part of the EIO (ethernet and input/output) board. That board has a
couple of hundred ICs on it, so makign a compuete repalceemnt board is a
lot of work (and it would ruing the history of the machine). The disk
controller itself is based roudn an AM2910 sequencer running firmware in
bipolar PROMs. THere's a fair amount of logic, a 16 bit FIFO, and so on.
Then ther'es the 'DIB' (disk interface board) that sits on top of the
drive and handles the acutall MFM encoding/decoding.
Oh, and the recoding format is non-stnadard. There are 4 words of
filesystem pointers in the sector header, for example.
While I undertand how this thing works -- roughtly -- I'd rather not
have to modify any part of it. A thign that fits in place of the drive
would be a lot easier (and, more importantly, would involve les
modification to the PERQ).
FWIW, I would not consider a ready-built thing to do this. I dont; buy
electornic devices any more. I want a PCB, a parts list and a source
listing. That way I can make sure it's built properly.
-tony