4

I was wondering how the data on a floppy disk physically looks like. With data stored on a magnetic disk obviously that isn't visible. The Wikipedia article on floppy disks however shows a beautiful visualization of the magnetic information on floppy disk recorded with CMOS-MagView, a high-resolution and precise measuring and visualization system for magnetic materials.

Below visualization is obviously not a full disk scan but considering we had a full disk scanned and visualized in that format: What information could be extracted from visualized magnetic information? Could it be visualization be decoded into data? What does it tell us about the quality of the material? Could we tell what formatting type was used on this floppy disk? Are the vertical stripes the separation of tracks and the horizontal thin stripes sectors? Or are we looking at (a part of) one sector or cluster?

Lastly, has CMOS-MagView or a similar system ever been used as a (non-intrusive) data recovery method?

CMOS-MagView visualization of magnetic information on floppy disk

6
  • 3
    That's not a "stretched/flat visualization", those tracks are curved, but since it is zoomed in so much it is hard to see. Now I understand how flat-earthers get their start.
    – Glen Yates
    Jan 14 at 16:54
  • @GlenYates I cannot stop laughing about your flat-earthers comment. :') Although I have luckily never been associated with that. But, I agree that some sub questions need rephrasing or removal. I will edit the question.
    – Bob Ortiz
    Jan 14 at 16:59
  • 1
    Ha, yeah, I didn't mean to imply that you were a flat-earther, just that it is their inability to realize just how big the earth really is and how 'zoomed in' we are, making local phenomena appear flat, thus leading to their belief.
    – Glen Yates
    Jan 14 at 17:07
  • 1
    It took 160ish years to decode and playback the first phonautographs (made circa 1860) and those were visual representations of audio waves - so with sufficient scanning resolution, and a healthy dose of patience, I suspect any magnetic media could be fully decode-able from a visual scan.
    – Geo...
    Jan 14 at 18:22
  • 1
    I saw this guy scanned images of vinyl records round 20 years ago and converted them into digital audio files. Jan 15 at 19:53
9

What information could be extracted from visualized magnetic information?

The very same as with a magnetic head: flux changes - and from there everything else, like headers, sectors and the data within.

Could below visualization be decoded into data?

It can be decoded into a bitstream, as it seems to hold one. But I doubt that the shown section is long enough to regain useful information.

What does it tell us about the quality of the material?

Nothing, it tells about magnetic flux changes. That is what magnetic properties are there at the moment.

Could we tell what formatting type was used on this floppy disk?

As mentioned, it's the same information a head can read, thus the same information a format can be made of.

Are the vertical stripes the separation of tracks

They are obvious the gaps between tracks (running to down) as there is no coherent magnetization but random domains.

in a stretched/flat visualization

This seems not to be stretched in any way, but visibly curves, so it's tracks. Most likely of a hard disk platter, as for a floppy the gaps would be way broader in comparison to the track width. In addition fluxes on a floppy may have less sharp borders (for older drives at least)

and the horizontal thin stripes sectors? Or are we looking at (a part of) one sector

They are magnetic changes in vertical succession. Keep in mind magnetic changes are not bits/bytes, but encoded bits/bytes. The colours and 'thickness' of stripes is not real, but a side effect of visualization and timing of magnetic fluxes. From the type of magnetic encoding that should be on a disk we can infer what it must look like:

  • Sync data usually carries sequences with less changes to enable pickup of the clock, thus will look like wider stripes with more visible 'colour'

  • Data in contrast is usually of random bit sequence encoded with the maximum number of changes get density, thus must be looking thinner and more equal.

  • Headers are short data blocks

  • Sectors are long data blocks.

Taking this, the shown part may cover the area two sectors and gap/synchronisation/header data inbetween. Rotational sequence seems to be bottom to top showing

  • End of a sector (darker thin stripes)
  • Sector gap with sync data (yellowish seemingly wider stripes)
  • Header data (short sequence of dark/thin stripes)
  • Header gap with sync data (yellowisch wider again)
  • Next sector (again darker thin stripes)

or cluster?

Clusters are no units of magnetic recording, they are logical structures used in storage management. Only sectors (ordered in tracks) can be found on a magnetic media

Lastly, has CMOS-MagView or a similar system ever been used as a (non-intrusive) data recovery method?

I would guess so. Then again, it's a huge effort to reach something that can be done by simply using a drive to read the platter. Cost will be quite high. Even with destroyed drives recovery usually done by taking the platters out, careful cleaning and mounting them in a new drive to finally read what is hopefully left.


P.S.: In Ye Olde Days it was common to read tapes, but as well disks visually by applying magnetic fluid. The human brain works quite well to decode this - at least once trained to see the diffeence. Been there, done that_))

7
  • You're fast. Thanks. I just changed the question a little bit based on the comment.
    – Bob Ortiz
    Jan 14 at 17:06
  • @BobOrtiz Well, the question is quite simple and easy. More concerning, that it is not a question about retro computing, but magnetic storage basics. It should not be asked here.
    – Raffzahn
    Jan 14 at 17:22
  • 2
    Years ago I knew someone who worked in a lab at a then well-known hard drive manufacturer. They had a very flexible system in the lab to map out surfaces. One of their common things to throw on the machine for fun was the paper tickets from the local metro. It did not take them long to figure out the coding sequence (and know how to rewrite them for an arbitrary amount of credit). The city never responded to their outreach attempts, trying to encourage a bit of security on them.
    – Jon Custer
    Jan 14 at 18:06
  • 2
    @Raffzahn Two reasons I think it should be here: (1) the answer could be different for modern magnetic storage where everything is smaller; (2) the subject matter experts are here.
    – wizzwizz4
    Jan 14 at 18:16
  • 1
    @Raffzahn About the railway control computer systems? Why not?
    – wizzwizz4
    Jan 14 at 19:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.