Michael J. Mahon wrote:
The problem is that there is no algorithmic way of determining what is "data" on a track, fractional or otherwise, since the track may be written full of data (which may or may not look valid) just for the purpose of containing one or more "magic" nibbles at certain rotational angles in certain track positions.
But who cares? We just need to capture the bit stream. You can calculate the rotational position using the counter... Just add to the counter for arm movements and resetting stopping and resetting.
I think it falls short (as I explain below). I also suspect that *every* reasonably space-efficient proposal can be defeated by *some* disk format, and the remaining question is: How many protected titles are you willing to *not* be able to represent?
I obviously disagree, this bit stream format should encompass any formatting known or unknown.
File length would vary depending on how many entries in the track table.But what do you mean by "track length"?
Tracks are continuous circles. The length reference would tell you when to wrap around the bit stream.
Some programs do a 'nibble count'. This is done by counting bytes or 'cycles' that pass between a reread of a signature. By giving a length, the track can be read in circles with accurate timing.
Thankx, Ed