First, I am not a math specialist, I am asking a question that seems to have a very clear answer to me but it seems wrong, so I am hoping to learn why. Here goes... Long ago I read a description of Maxwell's Demon in a Douglas Hofstadter book (I think) and ever since learning about digital images I've been thinking about what I read, but using digital images instead of the video tapes that were used in the book.
I'm using a permutation formula I found for $N$ possibilities at each of $r$ locations: $N^r$. If I allow only 256 colors (so $N = 256$), which produces "OK" fidelity, and an image size of, say, 256 x 256 pixels ($r = 256^2$) then there are $256^{256^2}$ possible images that may exist. That is a very large number (Wolfram Alpha tells me there are 157,827 digits) but to me it seems very, very small because:
That set contains every possible image, from every angle, at every time (past, present and future) of every real and imaginary thing that may be represented in an image. That includes images of soundtracks, sound wave forms, musical scores, film frames, and digitized text. It includes printouts for any 3D printer models as well as the plans for every device ever invented throughout all of time. As if that isn't enough, the images themselves may be combined to create higher resolution versions of all of the same images plus other images that were too large to fit in the 256 x 256 pixel frame!
I hope I've explained this idea well enough. Most of the people I've tried to talk with about it just get glassy-eyed or tell me it is nonsense. If it is not nonsense, is this written about somewhere so that I could read an expert version and share it with my friends? Also, if it is not nonsense, how can all of time and space be so "small?" (smile)
EDIT what melts my brain is that all possible real and imaginary images across all of space and time, can seemingly be represented by a finite number of images.
If I increase the color depth from 8 to 24 bits, the number of digits roughly triples and then increasing the images to 1,000 pixels on a side gives a number with over 7 million digits. Insanely huge, but still finite and most will still be static/noise (those those might be a smaller part of a larger image, too.)