Using modern a modern laser printer and scanner how much data could be encoded into a 8 by 10 piece of paper?
I mean make a program that prints dots representing binary data on paper and then use a scanner to read this data?
blade
Here’s a WAG:
at 1200 dpi (a reasonable laser printer and scanner resolution), you’d get 1200x1200 = 1,440,000 bits, or 175.8 KB per square inch. If you used 8x10 paper, that gives you 80 square inches, or 14.062 KB, or about 13.7 MB.
If you used a color laser printer, you’d be better off, maybe even a factor of 24 better (assuming 3 primary colors at 8 bits each- you’d probably not do better than 8 bits between printer and scanner, even with 24 bit color).
So, at an upper limit, with color, I’d guess about 4MB per square inch, or about 330MB for the entire sheet.
Getting anywhere close to this rate would require some serious software and digital signal processing, due to imperfections and variations in printers and scanners. IRL, you’d probably be lucky to get 1/10 or 1/20 of the above figures.
Arjuna34
bladeohlsson, weocome to the SDMB. The following is a graphic example of what can result when we fail to take our prescription medication as directed:
All of the date in the known universe could be stored on it:
-
Collect all data in the known universe.
-
Assign a number 1-26 to each letter of the alphabet:
Space = 0
A = 1
B = 2
C = 3
D = 4
E = 5
F = 6
G = 7
H = 8
I = 9
J = 10
K = 11
L = 12
M = 13
N = 14
O = 15
The words MAD DOG becomes 131404157, and so on.
-
Feed the data into a machine which can make the necessary conversion so that all data in the known universe is translated into this numerical equivalent, resulting in one hell of a long-ass number. Let’s call this machine the Attrayantometer.
-
Place a decimal point in front of the long-ass number, turning it into a small-ass number (131404157… becomes point131404157…).
-
Instruct your computer to draw a single line on the paper, dividing it in two areas a and b in such a way that the ratio of the areas a:b is equal to the previously mentioned small-ass number. The data has now been stored.
-
When retrieval is necessary, invent a device that will measure the areas a and b with sufficient accuracy to extrapolate the ratio (the small-ass number).
-
Remove the decimal point, bringing us back to the large- ass number. Run the number backwards through the Attrayantometer, which interprets each number and assigns the corresponding letter. You have now retrieved all data in the known universe from a single line on a piece of paper.
Of course, 1314 can become MAD, or it could be ACAD. The Attrayantometer would have to be smart enough to know which letter groups aren’t real words & look for alternate combinations. Idealy it would also be able to examine the context of the decoded data in case there was a tie between two possible translation results that both end up as valid words.
Don’t everybody run to the patent office at once now…
probally the best way is to use the 2d barcodes, don’t know how much they can hold but it is a standard. and don’t forget you have 2 sides of the paper
Of course, the Attrayantometer[sup]TM[/sup] does have some limits. Assume the computer has infinite precision in its drawing ability. The success of the system still comes down to the position of that very last molecule of ink on the sheet of paper. For infinite precision, that position would have to be fixed. Wait, is that Mr. Heisenberg I spy? If the position is fixed, we know the momentum precisely. It’s zero. If we know the momentum, we can’t measure the position anymore (at least not accurately enough). Even if we allow that the atom is moving a little bit so we can measure its position, it doesn’t really help us, since know we precisely know an imprecise position! Egads!
And BTW, if you’re gonna start by assuming infinite precision, why don’t you just use 2 digits for the low letters (A=01) and avoid that nasty multiple ambiguous translation problem?
Even given infinite precision, the Attrayantometer couldn’t record the areas of a and b.
Is that Mr. Godel I spy?