On my final trip to The Black Hole, I picked up some more Tektronix
documentation. One of these was a photocopy of the manual for the
4953/4954 Graphics Tablet. This manual combines the instruction
and service manual in one with circuit diagrams and descriptions
in the back of the manual.
However, since this was a photocopy and not an original, the circuit
diagrams were not available in their original size. The person who
photocopied the manual had attempted to cover the diagrams with multiple
photocopies of different parts of the oversize page. It was difficult
to manually match up the different pages, if only just to identify the
order in which they should appear in a scanned PDF of the manual.
Instead I just fed the scans into Microsoft ICE, available as a free
download:
Microsoft ICE (Image Composite Editor)
<http://research.microsoft.com/en-us/um/redmond/groups/ivm/ice/>
The results speak for themselves. See the last three pages of the
resulting PDF:
<http://bitsavers.trailing-edge.com/pdf/tektronix/49xx/070-1791-01_4953_4954_Graphics_Tablet_Instruction_Manual_Jun_1980.pdf>
You can see that the person who photocopied the manual did miss a few
bits of the schematics, but for the most part the schematic is now
legible as a single diagram. Due to the blending that occurs at the
edges of the individual images, Microsoft ICE will sometimes produce a
grayscale image from the bilevel scans I use for making PDFs of
manuals. So a little post processing in ImageMagick reduced the bit
depth back down to 1 bit per pixel and converted the image to PDF for
combining with the rest of the pages via pdftk. The command stream
went roughly like this:
convert composite.tif -monochrome -compression Group4 -density 600x600 page.pdf
The -density argument tells the PDF file that the image was scanned at
600 dpi so that the composite page is shown at the same relative size
to the other pages in a PDF viewer.
I had previously used hugin, an open source image compositing program
that has tutorials for compositing multiple scans like this. However,
using hugin to do the same job was a very tedious, laborious and
lengthy process. With MS ICE, it was a simple matter of drag-n-drop
of the image files onto the application which it processed
automatically to find the resulting composite.
--
"The Direct3D Graphics Pipeline" free book
<http://tinyurl.com/d3d-pipeline>
The Computer Graphics Museum <http://computergraphicsmuseum.org>
The Terminals Wiki <http://terminals.classiccmp.org>
Legalize Adulthood! (my blog) <http://legalizeadulthood.wordpress.com>
Show replies by date