Subject: Re: Looking for MFM endec description
X-Newsreader: Microsoft Outlook Express 5.50.4807.1700
X-MIMEOLE: Produced By Microsoft MimeOLE V5.50.4910.0300
Date: Tue, 08 Oct 2002 15:43:31 GMT
NNTP-Posting-Date: Tue, 08 Oct 2002 11:43:31 EDT
Organization: Cox Communications
wrote in message
> I can't seem to find a definitive description of an MFM
> I'm looking at the bits on the edge connector of a drive, and I'm just
> not "getting it".
Modifying Guzik's description slightly,
Current Next MFM
User Bit User Bit Code
0 0 01
0 1 00
1 X 10
Decoding is the reverse process. Note that it's pretty easy: if you can
identify the first and second bits in each two bit MFM code group, the user
bit is the first MFM bit. If you get it wrong, and read the second bit in
the MFM code group, then all your data turns to garbage.
As for what you'll see at the edge connector, it's a little more
First, there's a sync field at the start of a sector, which will be either a
3T (100100...) or 4T (10001000...) pattern, long enough for the PLL to lock.
That's followed by a sync mark, which tells the endec that valid data is
coming (it's been too long - I can't remember what the sync mark pattern
was, but it's only a few bits long, and it's supposed to be something that
can't be confused with the 3T or 4T sync field patterns). The sync field and
sync mark will be present at the start of every sector. Following the sync
mark the data begins, coded in MFM. At the end of the data there's an area
of garbage that's a buffer between sectors. Further, there are actually two
groups of sync field/sync mark/data/garbage for each sector. The first one
contains the sector ID, which you need to read to know where you are. The
second one is where the user data is actually written.
It really helps if you have something to trigger on when looking at the
waveform, or have something like a DSO with a large memory depth. If you've
got a hard sector drive, then there's a hole in the disk and a
light/photodetector that generates a trigger pulse once each rotation. As I
recall, that signal is sent to the ISA card along with the raw data - if you
have that signal, it makes triggering much easier.
-- Mike --