The Paperless Comic: drawing on a graphic tablet
About a year ago, I acquired a Wacom Intuos graphic tablet. I used it for cleanup of scanned drawings, tracing, coloring and shading, but for a long time I didn't quite manage to actually draw with it. But then I started the not quite Daily Comic with the intention to make it paperless. As is fairly obvious from the early episodes, this didn't go so well at first.
Truth is I am a very messy drawer. I don't have a clean line in mind that I then place on the paper. If I don't have a model, I tend to scribble, and the drawing sort of emerges slowly from the scribbles (or it doesn't, and then I try again some other day).
The way I made this "work" on paper is that by pressing harder on the pencil, lines are gouged deeper into the paper and hence harder to erase. So careful application of the eraser can remove the scribbles while leaving the desired lines. There was no way to duplicate this with Photoshop. What I needed was a different metaphor.
The appropriate paradigm turned out to be another staple of pencil drawing: tracing paper. I use Photoshop Layers like tracing paper. I have a "Sketch" layer on which I scribble my initial doodle.
Then I just add a new layer on top ("Sketch2") and I set the opacity of the first layer to 42 (or any other value you find meaningful).
I then make a slightly cleaner sketch on this new sketch layer, tracing over the relevant parts of the initial scribble. I then hide the first layer and continue working on Sketch2. I repeat this process until the final clean trace, on a layer I usually call "Lines".
A really useful trick is to apply "Flip Horizontal" to the image each time you start on a new layer. Mirroring the image tends to show errors that might otherwise be overlooked.
A small note on style: while my comic shows undeniable influences of Japanese manga, anime, and games, a perhaps less apparent but much deeper and longer-lasting influence on my style is the Ligne Claire of classic Belgian sequential art of the Brussels school.
Because I'm lazy, recurring elements are drawn only once, and copy-pasted, moved, and scaled as needed.
For the not quite Daily Comic, I tend to work on a canvas of 1600x2400 pixels, which is double of the published size. This allows me to scale lineart of individual frames somewhat without it being too noticeable in the final result. I initially used a 9 point hard brush of variable thickness for drawing, but lately I switched to 5 point.
I made a template for the borders, which is the top-most layer, so I don't have to worry too much about staying within the frame while drawing.
Because the lineart is separate on a transparent layer, I can add background colors or speedline effects on layers below it.
To separate the drawing from the background, I use a "Colors" layer as described in my coloring tutorial.
Text is then added. I mostly use the "One Stroke Script" font at 36 points.
Finally speech bubbles are created on a "Balloons" layer, below the text layers but (usually) on top of the drawings. I use the Elliptical Selection and Polygonal Lasso (for speech bubble pointers), which I then Fill (on dark backgrounds only) and Stroke (inside 4 pixels).
I then duplicate the image, resize to 800x1200, and "Save for Web" usually as a 16 color PNG.