A few years ago I spent a couple of hours tinkering with implementing Thad Walker’s scheme for printing computer generated holograms (see the link for the paper):
A recent comment has got me thinking that I should revisit the idea and the paper. For reasons which aren’t entirely clear to me now, it seems that I understand the underlying mathematics much better now (I have done a lot more tinkering with FFTs and the like) and I think that I could use FFTW to implement a much fancier and more robust version of the original program. I also have access to some equipment that might be capable of creating accurate 2400 dpi prints, which wouldn’t hurt. Stay tuned.