Pictures in the reflective area of a CD-R

This may seem like an odd question. Truth is that I was thinking of doing a program that could burn a two tone picture in the reflective part of a CD-R disc (yin-yang symbol, happy face, etc.), to be used as a decoration or souvenir.
Of course the technical implications of such an idea are overwelming me. Any documentation regarding the phisical representation of bytes in the CD program area is welcome. Some questions:

  • Is there some way to write raw data bits in the program area without using data block separation, subchannel bits, etc. (to make the picture smoother)?
  • If not (or even if so) is there a way to tell how many frames (or bytes) can be written in a full turn (360º) of the CD? Are there more sectors (or bytes) in the outter part of the CD than in the inner?
  • I have read that the bits representation in the CD is a NRZ inverted. Is this true in all cases?
  • Am I wasting my time and should stop right now thinking about an inviable project?

Thanks in advance.

Hi and Welcome!

Disc Tattoo is already invented. Supported by Labelflashequipped drives (Nec and Pioneer models) on DVD media (but not on CD).
Some ancient Yamaha CD burners supported Disc Tattoo on CD-R.


This is great news. Then I shall look forward on making it possible. Thanks for the tip.
Although my idea would be to do it without that kind of technology, just with the plain old burner… Let’s see what can be done.

Edit: for what I read, this technology discards the normal EFM patterns and makes it’s own. This has it’s pros and cons. On one hand I could (if I knew how) make my own pattern and forget about the questions above, on the other hand it seems that doing this (if at all possible) would mean working at a very low-level. Some specifications on generic CD burners at the lowest level possible?


pioneer 115d is just a [excellent] plain old burner.

For a very quick look at DiscT@2 technology you can go here.
For deeper look into DiscT@2 and DiscT@2+Labelflash go here.

it’s also possible to form an image in datastream,
that represent model of cd spiral, by utilizing user data of channel frame only,
and run it through deinterleave routine thereafter,
thus creating an cd image, that can be burned with regular software.
but you need to determine parameters of each cd.
and it’s rough - around half frame deviation.
but i guess this is as generic as it gets,
to do better you’d need to have some control over hardware.

i’ve put together a program to perform, what i described above.

ther’s actually two - 1st to do interleaving; 2nd to scale CAV image 2 CLV CD.
it’s very easy to do basic shapes this way. eg. this can be typed in notepad:


then you specify how many times to repeat this pattern.
and produced output will look something like this:

but huge drawback is, that you still need to know parameters of CD.
so basically it’s good for CD-RW only.

Are you going to continue working on this?

me? i don’t know.
to be honest, it’s not a priority - i’ve already accomplished, what i intended to do.
this program was made to test weither Sega Saturn ring could be duplicated this way.
and it can to some extent - visually, but it would not pass authentication check.
did not for me at least.
so if anyone can find some other use of this - it’s great, i would only be happy.
though one thing i can imagine -
that could be done to make it more useful and profound - added bitmap support.

also probably similar approach can be applied to DVDs.
i don’t know much about DVDs, but from what software i have seen,
i think it’s possible to get some geometry data.
so maybe it’s possible to measure DVDs more precise.
it would be great if there was an efficient way to get medium parameters without burning test discs.
(or burning just very few)
so maybe it is possible on DVDs.