I’ve been meaning to recycle this article for a while, and I had a few minutes to work on it lately. It’s more or less a copy/paste of an art tutorial I wrote up for the player forums for YoHoHo Puzzle Pirates!, a game that I still think well of, even if it’s in its sunset years.
I tend to sketch with ballpoint pens, and paint in Photoshop. This tutorial covers taking what I think of as a rough sketch, and turning it into a 150×150 pixel “avatar”, but some of the techniques work elsewhere. I do seem to be missing some of the original art, sadly, but the original article is still up on the YPP! forums over here:
For a “Reader’s Digest Condensed Version”, please continue, and as always, I’m happy to answer questions. I’ve added a few asides here and there, always in italics.
=====================================================
It’s come up a few times, and I’ve wanted to do a Photoshop tutorial since before my YPP days, so here’s a whirlwind tour of my methodology of avatar art. It’s actually a bit generalized, but this is how I wind up doing most of my avatar art.
1. Draw something cool in my sketchbook. I do this with a ballpoint pen, most of the time. It’s personal preference… as is the definition of “cool”. This particular monkey is actually a component of an avatar I did for Phillite. He works as a standalone critter, though, so I’m reusing him for this project. (Which also means that, as might be expected, I ask that the art in this thread not be used elsewhere.)
2. Scan it in to Photoshop, usually at 600 dpi. This gives me room to play with effects. I usually shrink it down once it’s all painted the way I like it, but I like working big. It gives me more freedom to try big, sweeping brushstrokes, and more precision in tweaking. I bought a cheap Memorex scanner on sale for $40 years ago, and it’s been fantastic.
By the way, if you’re serious about computer art, do yourself a favor and get a tablet. Wacom Bamboo tablets are a great entry level product. The software doesn’t matter all that much, since paint.net, GIMP and ArtRage are free and will suffice (Clip Studio Paint and Affinity work as fairly low cost powerful single-purchase alternatives as well), and some tablets come with software. I use Photoshop Elements 2 because it’s what I have handy. I also use Painter on occasion, but that’s an indulgence. The tablet, though… that’s almost essential.
Image may be NSFW.
Clik here to view.
3. Use Photoshop’s Levels modifier to clean up the sketch. I make a duplicate of the scanned layer, just in case I need the original for some reason, and apply Levels (Ctrl-L) to the duplicate. Pulling in both end knots a wee bit cleans up most of the static that came from the scan.
Image may be NSFW.
Clik here to view.
4. Since my sketches tend to be a little rough, I need to do some Rubber Stamp surgery to clean up a bit. The Rubber Stamp tool takes data from a source part of the image, and replicates it elsewhere. You Alt-click to define the source, and then “paint” the duplicate, winding up with this sort of effect, here duplicating the alternate arm’s thumb:
Image may be NSFW.
Clik here to view.
5. Rubber Stamp to clean the drawing, like this, cloning in the blank paper/background into the areas that should be clean on the drawing… it may take a bit of work and several clone source points, chosen each time with the Alt-Click:
Image may be NSFW.
Clik here to view.
6. I then make a new level (on which I’ll be painting), and move the clean sketch to the top of the stack, and set the level blending type to Multiply. This lets me treat it as an outline, and paint the color in underneath.
Image may be NSFW.
Clik here to view.
7. Start painting on a layer underneath the drawing. I don’t paint on the drawing layer. All coloring takes place on layers between the drawing and the white background layer I’ve set up. This gives me the ability to tweak the painting independent of the background and the sketch. This use of layers is one of the huge strengths of Photoshop (or any program that uses layers), and why working digitally can be a very different animal from traditional art.
Image may be NSFW.
Clik here to view.
8. The base color for the monkey is in, carefully covering his space. Now, it’s time for another layer for the shadowing.
Image may be NSFW.
Clik here to view.
9. The shadow layer is just a bit of paint that’s darker than the base color. It’s painted in a bit roughly at first…
Image may be NSFW.
Clik here to view.
10. Then the Gaussian Blur filter gets applied, to soften it up (I usually do this, as illustrated, on a copy of the shadow painting layer, just in case I need to go back a step and tweak it):
Image may be NSFW.
Clik here to view.
11. This makes for a nice rounding effect, and even gives a nice “reflected lighting” subtlety to the larger areas, like the monkey’s torso. (The dark side of most objects in real space is tempered a bit by reflected light, which this neatly simulates.)
Image may be NSFW.
Clik here to view.
12. The Gaussian Blur pretty much obliterates the subtle shadows in the hair, so I make a new layer, and start painting in new, detailed shadows. These are brushstrokes, like the main shadow layer, but I don’t use the Gaussian Blur on these. I just use the Smudge tool to push things around the way I like them. Here’s a close shot on the hair in progress:
Image may be NSFW.
Clik here to view.
and the tail:
Image may be NSFW.
Clik here to view.
and I sharpen up the cast shadow under the chin with a few additive strokes:
Image may be NSFW.
Clik here to view.
13. Erase around the edges of both shadow layers. It’s a subtle thing, but this shows how the Gaussian Blur pushed the color out of the outlines. I prefer to keep things clean, so I erase the blurred bit. Of further note, looking at this from 2019, this edge cleanup can also be accomplished by putting all of the color layers into a layer group, and adding a layer mask to that group that simply masks off anything not inside of where you want colors. This lets you create the edge cleanup for all of the color layers with a single operation, which is a great update to the workflow. Photoshop Elements 2 had neither layer masks nor layer groups, so this is a bare-bones tutorial. The fuller releases of Photoshop give more tools to work with, including “Smart Objects”, which I’ll revisit in a different tutorial.
Image may be NSFW.
Clik here to view.
14. Now for a highlight layer. I do this the same way I did the shadow layer, just with a different color, and from a different direction. In other words, paint,
Image may be NSFW.
Clik here to view.
blur,
Image may be NSFW.
Clik here to view.
and make a secondary highlight layer for detail work, then erase around the edges to be clean:
Image may be NSFW.
Clik here to view.
15. Since monkeys in YPP have a two tone look to them, with the belly, feet, hands and face a different color, I make a new layer to try to get this effect.
Image may be NSFW.
Clik here to view.
16. Paint the relevant parts in a lighter color, then change the layer Blending options to get the desired effect. I settled on Soft Light. This allows me to paint in a second color tone, without losing the shading and hair effects I’ve made so far. I’m using a subtle secondary tone here, and you can do more with color shifting by using a different paint color and layer compositing effects like Hue (instead of Soft Light) that shifts the color underneath while maintaining the shading:
Image may be NSFW.
Clik here to view.
Image may be NSFW.
Clik here to view.
17. Close to being done, it’s time for little tuning. I decided that the monkey’s belly needed a bit more dimension, so I added a bit to the shadows:
Image may be NSFW.
Clik here to view.
18. Finish by painting the sword on a few new layers, using similar effects for shading:
Image may be NSFW.
Clik here to view.
Add a layer for his eyes and nose…
aaand he’s done!
Image may be NSFW.
Clik here to view.
Since this was done at 600 dpi, it’s not really ready for an avatar. It comes out to be this big, useful for seeing detail:
Image may be NSFW.
Clik here to view.
After rescaling the resolution, a middle sized version looks like this:
Image may be NSFW.
Clik here to view.
And the avatar might look like this:
Image may be NSFW.
Clik here to view.
It loses a lot of detail at that scale, so this methodology isn’t always appropriate. It’s how I work because I like to have my art around at high resolution if I need it for my portfolio, especially if I need to print it out. Working high and reducing as necessary winds up looking a lot better than working small and magnifying it if necessary.
I would also usually go back and flatten some layers, erase the edges, throw in a background and/or a border… but that’s about it.
Thanks for stopping by! I’m happy to answer any questions.
-Silver