I am trying to burn a CD/DVD/Blu-ray Disc so that each 2pi segment of the spiral track alternates between:

  1. a steam 0's, or blank, untouched media.
  2. and 1's (an approximately circular groove).

So under a microscope, but at lower magnifications, the disk would almost look as if it had concentric circles etched into it. How could I go about doing this?

Some problems that jump out at me are the facts that a different amount of data is recorded on optical media per revolution, and that disk identification and partition maps are written at the head, which would be the center, so there would need to be some kind of an offset, but these are the least of my problems I'm wagering.

I don't need it to be perfect by any means, so the metadata and the trail having extra info I'm not too worried about, as long as the spirals are matched up in 2pi segments. Don't care about the synchro bits that are written every few bits too much either, though I'd prefer it if they weren't there. I'm pretty sure you can force disable the check bits? As I am not looking to read from this disk at all (the ideal image would not have a valid PMAP/FS at all) so none of the error correcting/synchronizing data is needed.

The only think I can think of is writing straight 1's to the disk by dd'ing to the blank disk, but I'm not sure if that writes any additional info. It would produce a spiral groove with space between itself, but it doesn't solve the problem of taking the geometry of the disc into account and alternating spirals. (I'd really like to be able to do this, so I can adjust the space between the grooves).

I realize that there is absolutely no practical reason to do this, since this will result in an unreadable volume. I am basically trying to use the CD/DVD burner I have laying around as a precision laser engraver that can create diffraction gratings of various d values.

*At this point please accept my apologies for the incoherent rambling question, I'm not sure if I even know what I am talking about, and this strips away all of the abstraction that I've grown used to, so much of what I say may not make sense, any edits to improve readability would be much appreciated. Also, please forgive me if this is not the right SE to ask on, honestly I'm confused as to what subject to file this question under. *

  • TL:DR You're trying to generate a 'picture' not data??
    – Tetsujin
    Commented Jan 8, 2017 at 19:31
  • I guess you're right, the technique would be identical. Though it may be worth pointing out that the "picture" I'm trying to generate is very regular (as in predictable) and can be expressed as a vector function. But basically, yeah, I need some way to map the physical location on the optical media to the storage blocks in the file system. Writing raw binary streams to the disk and calculating where to flip the bit using geometry seems much more reliable to me, but is probably impossible short of writing a custom driver for the drive :/ Commented Jan 8, 2017 at 20:28
  • I'm curious as to why you want to do this?
    – DavidPostill
    Commented Jan 8, 2017 at 20:45
  • 1
    It might be a pipe dream, but I'm hoping this could be a cheap way to make high-quality radial diffraction gratings with varying d values. Obviously for CD's it would be useless for anything in the visual spectrum (maybe except for the one where I "one" the disc out (if that's even a phrase) but if I can expand the technique to use on DVD's and Blu-rays, the scale would be small enough for colors we can actually see. Commented Jan 9, 2017 at 14:12


You must log in to answer this question.

Browse other questions tagged .