On 01/09/10 22:03, Dennis Boone wrote:
My understanding is that there's enough variation
in the low level
formatting between controllers that MFM drives aren't very portable.
That could be interpreted as timing differences between models, or as
timing variations in different instances of the same model, or even
actual structure differences. I don't know which was meant by whatever
source told me this long ago.
There are format differences. Different CRC implementations, different
encoding rules (gap lengths and so on), slightly different ways of
encoding the data (LSB-first, MSB-first, bit-sequential, staggered),
interleaving, ......
The fun part is figuring out what those differences are, then 'teaching'
the software how to identify them, and make sense of what's there. Same
as I did with MagScan (which is still tied up in Licensing Hell and very
close to being rewritten from scratch).
Even assuming it is true, it doesn't necessarily
kill your concept,
though it might make implementation a hell of a lot more exciting.
My life would be FAR more boring without all these fun and interesting
puzzles :)
I'm not above "blackbox" reverse engineering if it proves necessary.
Change one byte, see what happens, that sort of thing.
--
Phil.
classiccmp at philpem.me.uk
http://www.philpem.me.uk/