Well. Some year this has been, huh?
I'm sorry this post has taken so long. I finished the first page of the comic months ago and I've been wrestling with a longer postmortem, but I guess I don't have it in me so I'll keep this quick and maybe not bore you.
First, the comic has a name now: Necessity. I'm not thrilled with it but I'm glad I don't have to keep saying “Sorceress Comic” anymore.
Coloring started out fun but got tedious. I used just two layers: One for flat colors and one for shadows. I did all the flats first, then for each color, picked one or two shadow colors. I selected each shape with the magic wand tool, then moved to the shadow layer and used the pencil tool to shade it.
This wasn't a terrible setup. Since I can select two colors and swap between them, and use my tablet's stylus's eraser end, the shading work wasn't bad. The problem was all the switching back and forth between the flats and shadow layers, and between the magic wand and pencil tools. It ate up a lot of time and energy, and more than once I forgot to switch layers after selecting a new area so I had to do it over.
For future pages, I'm going to try something like Mary Cagle's technique for coloring Kiwi Blitz. She posted it a few years ago to her Tumblr, which no longer exists, but the post is on Archive.org. In short: Use a multiply layer and shade everything in a color suiting the environment or mood, then use adjustment layers if necessary.
I actually wrote about this tutorial on DeviantArt not long after Cagle posted it, but by the time I actually had a comic to color, I completely forgot it. Remembering that journal would have saved me some trouble: It mentions another approach closer to the one I took, and I'd described it as, “similar to the techniques I've tried before, and it looks to me like it's a lot more flexible, but also more time-consuming, and for artists without a good handle on lighting (like *cough* me) it appears to have a higher risk [of] giving different objects different lighting setups.”
I'll also be paying more attention to values. Converted to grayscale (via a blank white layer with its mode set to color), page 1 looks like this:
Not awful, but not as readable as I'd like. I'll have to do some tweaking.
I also had some trouble with the inks. My process was finishing the pencils on the layout paper before taping it to the back of the Bristol board and inking it over the lightbox. That works well with regular paper, but poorly with Bristol. I had a hard time seeing what I was inking, which led to a bunch of mistakes. (In the background of panel 2, you can see a mountain I mistook for a tree when inking and didn't notice until after I started coloring.)
Fixing that will mostly boil down to practice, but I'm also considering changing the process. I didn't trace in pencil on the Bristol before inking because I wanted nothing on the Bristol but Ink. That didn't really pan out: I had to use pencil on top of white paint/ink in order to fix errors anyway. Using a pencil at the lightbox stage would alleviate this issue, but would also add another step. I don't want to do too much new work on the Bristol because fixing mistakes is more work when I don't have the final pencils for reference. So I'll have to do some experimenting.
I don't have a really solid short-term game plan. I'm going to keep experimenting, and I'm also going to start working on the page 2 layouts, which I can do without worrying about the later stages. I do have a short-term goal for the game project, but that will get its own post when I have something to show.