I was approached about creating a mural, or rather a large painting or wall decoration at least eight feet long. Early in the development process, it was decided that this piece of decorative art would consist of four light-weight reinforced hardboard-and-wood panels, each about two feet by three feet. These would have very heavy watercolor paper stretched across their front surfaces, and they would be mounted as a single unit (possibly with a large simple frame).
The design consisted of a number of nonsensical forms, developed procedurally in a 3D-modeling application (Rhino), seemingly floating in space and apparently “casting” shadows behind them onto the paper surface of the panels. Resolutely two-dimensional strips of gold leaf would extend across some of the panels, either masking or “behind” the procedurally-generated forms but not “casting” shadows.
I developed several approaches to transferring the computer-generated design onto the panels, eventually developing a technique for printing it n reverse at full scale on a large-format laser writer, and then transferring the carbon that composed the black-and-white images from the printer paper onto the watercolor paper surfaces of the panels using a great deal of acetone. From that point, the designs would be stained and painted using ink and watercolors, and then the gold leaf (or more likely, less-expensive “dutch metal” leaf) would be applied using water-based size.
As it developed, the approved design gradually became more-and-more hieratic-seeming and flatter, less-illusionistic. The last iteration 2b (above) deliberately seemed to mimic a (very large) hand-tinted and gilded print, with shading indicated by diagonal lines like some sort of enlarged half-toning or woodblock. This evolution called into question the whole notion of depicting “three-dimensional” computer-generated forms on a two-dimensional surface using largely traditional, non-digital techniques, and the whole project is currently halted.