Philip Pemberton wrote in reply to Jules Richardson:
...
  Time the pulses from the drive... interesting idea.
Would certainly make MFM
 coding and decoding easier. I'd be tempted to add an Overlapped Read mode -
 it reads from the first index pulse to the next, but waits a few microseconds
 after the last index pulse before stopping. That way you can read stuff like
 Amiga floppies (which IIRC don't use the index pulse).
 You could even go as far as adding a counter so that reading didn't stop
 until there had been N pulses - which would enable you to read hard-sector
 discs as well. Mark the positions of the IPs in a buffer too (so you can
 figure out where you need to start looking for sync signals and such).
  It still bugs me that there doesn't seem to
be a central place which documents
 what support / code / docs are actually out there. I don't want to spend $$ on
 a board without knowing what software exists to support it and what I'd need
 to write myself, but finding that out was difficult last time I looked... 
 Same here. The Catweasel looks nice, but AFAICT there's no low-level
 programming documentation, just a load of source code and people screeching
 "UTSL! UTSL!" 
Register level description of the MK3 card:
        
http://www.schoenfeld.de/inside/Inside_CWMK3.txt
That file is 3 to 4 years old.
 > Personally I hate paying for extras that I
don't want - I'm like that with any
> technology (e.g. it's why I haven't upgraded my phone in many years)
...
> But yeah, I'd love something that just did the track buffering side of it,
> providing it could be made cheaper than a catweasel. 
I realize that for many people they have to watch every penny, but I
would guess they would be in the minority on this list.  For most I
imagine the value of their time to write the software for the card is at
least 10x the cost of the card.
My biggest complaint with the card is that it doesn't come with a 50 pin
connector.  Ideally it would have a wiring block just before the 50 pin
connector to accommodate the usual 8" peculiarities.  I would have
gladly paid another $30 to have a compact and ready solution to the
problem vs having to wire up my own kludge to do this.