After I did the Mona Lisa processing and found that I could get much better images than I ever expected, it occurred to me that I could take poster data files that were composed of overprint lines and process those back to numbers. From there I could run them through the Mona Lisa display routines.
In the mid-80's I was able to have a friend convert the old 9-tape IBM tape reel data to Apple ][ 3.5" floppy disks.
From there I wrote two programs, one to collect the unique sets of overprint characters that made up each pixel unit, and the other to recognize the overprint character sets and output a file that substituted a number (actually a character, since there would be 17 levels of gray, I used @=0, A=1, ... P=16). I also added Page (panel) and Line numbers to the datasets. Why? The poster data files were intended to be printed on mainframe lineprinters with the linefeed set to 8 lines per inch instead of the usual 6. Each line on those printers consisted of 133 characters. The first character would be the "carriage control" character and would signal with a space, " ", to advance a line and then print the following 132 characters. A plus,"+", character, told the printer not to advance, and then print the 132 characters. A one, "1", said to do a formfeed (new page) before printing the the rest of the line.
Therefore, lines were limited to 132 characters wide. So some of the posters were intended to be assembled out of multiple panels laid side by side.
Instead of reprocessing each poster into a true XY grid, I left the panel orientation and programmed the display routine to reset the drawing location back to the top and offset by 132 pixels. So each line of the output file would start with a one character Page/Panel number, followed by a three digit line number, followed by the 132 pixel information characters. The last line of the file would begin "9999".
This scheme worked fairly well, but it turned out on some posters, the panels did not line up at the top and therefore some posters required fudge factors when it came to displaying the additional panels.
So without further ado, here are the converted poster images.
I think I may have to revisit Einstein, the grayscale seems a bit off. I had to make a guesstimate when I assigned the overprint sequences to grayscale assignments and may have goofed on this one.
These two, LilCat and LilDog have Big Cat and Big Dog counterpart files. I still have to reprocess those files.
There was a Big Girl, but somehow that file did not get converted to the PC from the Apple ][ files properly. But look, you can see the folds of the fold-out!
One of the widest files, approximately 6 foot in the original overprint version. The widest one I had did not get converted to the Apple ][ files, it was of a PSA jetliner flying past the Bay Bridge in San Francisco. That one covered a wall!
This actually looks MUCH better than the old overprint posters. Back then, you could barely make out what the image was supposed to be.
Recently I was given a file with the color original. You can see how the overstrike image is elongated. In the 8 lines per inch with 10 character per inch pitch, the elongation must have been much worse on the printouts.
Copyright © Curt Rostenbach 2007-2009