The source .gb files come from http://zakros.ucsd.edu/~arobert/hanziData.html (found by Pablo Duboue).
I converted them to UTF-8 and then Big-5 using the utf7 conversion software from the same site. I then imported them into Excel, and printed them out under various sortings to Adobe PDF.
The "frequency" column gives the relative frequency of the character or word. A character with frequency 0 is among the 10% most common characters. Frequency 9 means that it is among the 10% least common.
Note that these .PDFs are about 2.5MB each.