Over the years I’ve put up a few dozen comics and cartoons on this site. They’re fun, they’re quick, and they amuse me greatly. I’m not a terrific artist by any means, barely a decent cartoonist, but I’ve got a few skills.
(I know, it’s a terrible picture but it was hard to get enough contrast with the light pencil marks.)
That’s my sleazy Private Investigator following and photographing one of the other characters from the story.
I’m also going to say here that I suck at drawing women, even ones as cartoonish as this one. Getting her even half way decent was the longest part of this drawing process for me. When sketching it’s okay for guys to be square, women are all curves and if they don’t come out right they look terrible.
Once I get the pencil more or less right I go over it with a good pen, let it dry, then erase the pencil. The pen lines need to be good, thick and dark. And make sure they are all connected, no gaps, otherwise it’ll be a lot more frustrating later.
The next part is the trickiest, getting it from a physical drawing into a digital medium. This took me a lot of trial and error. I’ve got a program with my scanner that lets me mess with the settings. I scan them in as black and white, medium dpi (dots per inch), and set the threshold low. The threshold is how ‘dark’ something has to be before the scanner sees it. If the threshold is too high then it will pick up the faint remains of pencil lines, impressions, etc. It has to be set low enough to only pick up the dark pen otherwise it’ll be a mess and almost impossible to get a clean result. That’s also why the pen lines need to be thick and dark.
If done properly, it will end up being a black and white image with only the clean lines, easy to color in and edit.
At this point I’ll use a photo editing program to set the size of the image to 1800 x 2400 pixels. That’s pretty close to standard for most ebook readers.
Then I input that image into TwistedBrush, a free program that works kind of like photoshop. I’m terrible with this program so don’t ask me any questions about it (or photoshop), I just figure out how to do the bare minimum to get it to do what I want.
First, I’ll add basic color to the image. There is a trick here with TwistedBrush, you can adjust the threshold for filling in colors. Even if a whole section appears white, it usually has various shades of white. You can set the threshold at 8% and it will fill in all the colors within 8% of the one you click on (or whatever other percentage you set), so it’ll get most of the other shades of white.
Coloring is why you don’t want gaps in your pen lines, suddenly half your screen will color in and you’ve got to figure out where the holes are.
Here is the same drawing after scanning with basic colors added. Some knowledge of color shading helps a lot. The tops of things are usually lighter than the sides, like the railing the PI is leaning against. More subtle, the top of her dress is a couple shades lighter than the bottom. And the further away from the ‘camera’ the darker the colors get.
Having the basic colors only, there is already a sense of three dimensions even though is still looks pretty flat.
What I do next is add another layer to the image. This is one of the cool features to programs like TwistedBrush, it adds an invisible layer over the base image. You can then draw on it, do whatever you want without damaging the base. This is also useful for integrating multiple levels of an image. My base is ‘flat’, all one piece, but it could have been drawn as a background, middle ground, and foreground, or whatever, then integrated by using the different layers. That’s a bit more tricky but can be very useful.
For this example, I added a layer over the base to do shading and highlights. This is to make the image look less flat.
You might also notice that I cropped the image to tighten the focus and tweaked the color balance.
In our three dimensional world almost nothing is a single color. The side away from the light will be darker, the other side lighter, with a gradient in between. So I went around and darkened the edges of the shapes and lightened others, giving it the illusion of depth. Ie. below her breasts, her face under the hat, under his arms, etc.
Keep in mind, this isn’t a masterpiece, just something I was slapping together for a cheesy cover. It’s not even close to perfect, realistic shading is far deeper, I just wanted to give it a hint of depth. And the best part is that I didn’t damage the underlying image, I could mess with shading and lighting without messing with the original.
I also added random blobs of other colors to give everything some variation and, again, to make it less flat.
Here’s the ‘final’ product:
A lot happened between the last image and this one. I added another couple layers over the first two for text and more shading. To make the text ‘pop’ I darkened the top and bottom of the image on the next layer, then added another layer for each of the parts of the text (so I could move them independently). For the title I found a cool, free, text generator online that made a variety of styles then colored them in TwistedBrush. Each word of the title is actually two layers, each slightly off set with the darker one behind the lighter one to give them more dimension.
One the image was ‘done’ I saved it as a JPEG and put it into a photo editing program. I use Fast Stone which is cool because it has a lot of features for adjusting sharpness, color, hue, contrast, etc. It lets me tweak everything to get it just right. Sometimes I’ll also use it between other steps, that’s why the colors in each of the above images changed slightly as I tweaked them.
It’s not perfect but it’s a fun, working cover and I had a good time making it. The whole process took me about four hours from start to finish. I’m sure anyone with decent experience on computers and with graphics programs could have done a better job faster but I’m not motivated enough to educate myself.
And hopefully it’s been entertaining to see this process from start to finish.