I was thinking about other people’s processes that involve shading on a separate layer that’s been constrained by a mask of the shapes it’s shading, and knocked this out in about an hour.
It looks nice, but I’m not sure how I feel about the fact that it resulted in about three times as many layers as I’d normally use.
(All the ‘light’ and ‘shadow’ layers have an opacity mask of a duplicate of the layer they’re modifying; light layers are 100% screen, shadow layers are 100% multiply. Most paths in them are heavily gaussian blurred; all the shadow paths are a blue-purple gradient.)
And just to show how much some simple effects can add, here’s the same image with most of the blurs removed:
I may try experimenting with this on a few panels of Absinthe. Making myself separate “shading” from “drawing” is weird; I’m really used to considering them at the same time. I may also want to experiment with using this method for broader strokes of shading, and doing some of the subtler shading within the base layers the way I normally do. Dunno. My methods are a constant work in progress, and ultimately what matters most is a mix of working speed and how fast they feel – I’d rather do something that’s empirically a bit slower, but feels faster to my brain, because that’s more likely to keep me happily working for a longer stretch. Making sure everything works in silhouette and basic colors before going to shades probably won’t hurt, though.