Re: Hardware musings

[russotto@grace.speakeasy.net (Matthew Russotto)]

In article <tKOdnRHlve84vKrYnZ2dnUVZ_rOdnZ2d@comcast.com>,
Michael J. Mahon <mjmahon@aol.com> wrote:
>
>Unfortunately, I don't think it's an easy problem to take image
>data for all quarter tracks, together with sync info and soft bit
>info, and turn it back into a physical disk that, when imaged,
>would create the same file that produced it.
>
>A relatively simple problem is that to control automatically even
>the approximate track length in bits requires writing with a variable
>clock (quite doable with a microcontroller).

I think it's the figuring out what's on there in the first
place which is the problem.  You'd probably need to read each quarter track,
and sample (at much higher frequency than the Disk II data rate) the
_analog_ value of the data from the drive head (not just the field
transitions but the field strength).  This would tell you with high
confidence which quarter-tracks actually held data, and give you
track length data and even "holes" where data wasn't written.  Anyone
up to building such a thing?

>Part of the problem is that data read from a "track overlap" region
>will be an artifact of the micro-variations in rotation speed of the
>drive on which it was written, and there is no way to control that
>on the writing drive (floppy drives don't have a way to read adjacent
>data as new data is being written).

Fortunately, you only have to be able to duplicate what disk
duplicators of the day were capable of duplicating reliably.

-- 
  There's no such thing as a free lunch, but certain accounting practices can
  result in a fully-depreciated one.