Well, I am on round 3 of test printing for the forest rays. I’ll mention a few highlights here.
I got the shadow blocks carved! Here they are, still with the hanshita remnants, and after cleaning them off below.
And I tested those shadows!
Test Round 1Test Round 2
There were more test prints; above I show one example of the first two testing rounds. In the first, I used a creamy yellow as the base color for almost all but the road. This worked out well for the ground and the foreground leaves that are caught in sunlight, but the yellow as sky just doesn’t work. In the second, I used a blue base color, but I think that didn’t work out as well for the ground.
Next attempt involved using a couple of gradients on the base color; the left-hand image is early in the test printing process after using a creamy yellow gradient up, and a blue gradient down, plus the light gray on the road and rocks. The righthand image has all the impressions except for the key lines, going at top speed and not trying particularly hard to get an even impression – just trying to quickly get an understanding of how this approach might affect the final colors.
I like the way I was able to recapture some of the intensity and tonal variation in the ground while keeping the blue sky, but the sunlit leaves on the left really suffer. I think I will need to alter the base color block to carve away the sunlit leaves. Alternatively, or maybe in addition, I might isolate the bright green leaves on their own block (they are currently part of the next under-layer) so I can intensify them without affecting the rest.
One of the things that makes me scratch my head for a long time is, how do I do the color separations for a complex print? I’ve taken various approaches, including just starting to carve color blocks and keep adding them until I’m happy at one extreme, and planning them all out from the start and just going with it at the other extreme. The print I am working on now is the most complex so far, and I am being careful. I’m striving to make the best use of the wood, since if I don’t avoid it, I could end up using dozens of blocks. I also expect there will be many revisions to my plans along the way. Here’s how I have approached it so far.
First of all, I made a mockup in a drawing program that attempts to approximate the print. I used the same scan that I used for the keyblock transfer and floated it transparently on top, then made layered vector objects underneath that so I could easily change the fill or transparency, add gradations, change the boundaries etc.
With this, I could make some hypotheses about what colors I would need and how they would combine. I can’t really explain how I did that though! Just experience mixing colors I guess. (To those who are trying this at home: Don’t be paralyzed! Guess-and-test will get you there. Emphasis on the test! 🙂 )
I made a list of the colors I thought would go on each area of the print, combining to make the colors I wanted to be there. Then I did the color testing I mentioned in the last post, overlaying regions of color with a brush to test out those hypotheses.
I made a table listing all the regions of the print as rows, with the pigment to be applied as columns. These are still just guess! Then I made a ton of printouts, and colored these with colored pencil to indicate the regions I thought would be color blocks. These sorts of steps help make sure that all of the regions of the print are considered, and that I don’t end up with a big white spot because I forgot to put color there.
I used these “coloring book pages” to figure out which color regions could be combined on the same piece of wood. Those two on the left can be clearly combined if I flip one of them 180 degrees, and if the wood is long enough and wide enough that the registration cuts don’t interfere.
From when I initially made that table, I changed plans about the outlines of the trees in the distance. Not only the farthest trees, but the two at mid-distance, need to have lighter outlines than the main key lines of the image. On the right-hand image above, the pencil sketch on the right shows that I can put both the light gray of the roadway and rocks, and the key lines of the mid-distance and distance trees on the same block, if I use different sets of registration marks. On the “coloring book page”, the lines that are over-lined in green will go on the block where the light green blobs are on the right half of the shot; the gray of the road and rocks will go at the bottom. The white circles on the very right show the placement of the registration cuts – two corner cuts near the bottom, and straight cuts on the upper right.
I also decided I needed more subtle control over the fill and shading of the trees in the distance. That’s why I isolated each individual background tree twice on this last block. By placing the top two registration marks circled in orange, I can print the small tree where the upper yellow circle is; with the lower two orange-circled registration marks, the larger tree will print where the large yellow circle is. Flip that block around, and I can do the same thing again to print some shading for these trees where the green highlights are. The trick is to position the registration cuts so they won’t interfere with regions that will print.
In all, I think I can get away with 8 more blocks for the basic structure of this print; then I’ll need about three more for the special shadows that reveal the sunbeams.
How will colors combine on the paper? How can I achieve that particular shade I am looking for? Doing little experiments like this, on a sample of the paper I plan to use, is one way to find out.
Vivid Laboratories 