Notes
I make the comic on a PC using a Wacom Cintiq interactive pen display. The Cintiq was an investment that nearly bankrupted me, but I've never regretted buying it. That said, similar results could be achieved using a standard graphics tablet; I just reasoned that if I was going to be drawing on my computer all the time, it made sense to make that process as intuitive as possible - and it doesn't get much better than the Cintiq.
As for software, I use Adobe Photoshop CS3 for all the drawing and then Adobe Illustrator CS3 to add the dialogue. Adobe's Creative Suite is expensive and out of the reach of many, but it's so good that I can't imagine working without it. If I add the fact that this site was made in Dreamweaver CS3, I know I'm getting good value out of it.
For drawing and inking, I use standard round brushes between 6 and 15 pixels in diameter. I found that I liked thinner lines for my inking rather than calligraphic effects.
The font for my lettering is called 'Jack Armstrong'. It's not a free font, but only cost about 20 US dollars.
At the document dimensions I'm working at, you need a fairly decent computer because the file sizes are huge. Most of the finished comics in Photoshop are in excess of 40Mb in size. To get this down to 60Kb or so for the web, a lot of detail is lost which can be frustrating if you've spent ages drawing something that can't really be seen in the final version. I'm learning that the hard way!
Finally, the BPelt filter that is mentioned in this guide is available as a free download here. I owe a lot to it because trying to shade the comics without it was a nightmare.
I'm including this guide for those who are curious, rather than offering it up as a shining example of how things should be done. It works for me and is the result of a lot of trial and error, but I'm always looking at ways to streamline the process. My methods are really an amalgamation of tips I've collected from other webcomic artists and owe nothing to my own ingenuity. The steps below follow the creation of comic 012.
![]()
I start by opening up a drawing template in Photoshop. It's pretty big to enable me to zoom in very close and get smoother lines on my brush strokes. This is how it looks:

The resolution is 300ppi. The 1000 pixel margin all round is just so that all areas of the comic are reasonably central on the Cintiq. Otherwise I'd be cramped in the corner of the screen. On the layers pallete you can see that I keep everything separate - this is important for later on. I start off with the 'pre-rough' layer and do some quick gesture sketches in a red colour to work out the placement of the characters. At this point I will have already finalised the dialogue, so my main thoughts are to ensure the character speaking first in the panel is in the right place. If they're not, you get the tails of the speech balloons crossing each other and looking awkward.

This particular strip features Lindsey's car. I know that I'm not going to be able to draw a decent car from memory, so I find a photograph on the internet and drop it into place on another separate layer. I will trace over this later.
When I'm happy with the layout, I switch to the 'rough' layer and change the brush colour to a blue so that it stands out from the red. I start by drawing a series of circles with the ellipse tool to represent the character's heads. In simple terms, a human head is just a sphere with a jaw hanging off it. Biologists may frown, but this is only a comic.

Although in this instance the blue layer closely resembles the red layer, I'm not slavish to it. If I feel like changing a pose completely, I will. It makes me feel manly and in control. When drawing heads, I keep things reasonably realistic by following some basic rules found in human anatomy: the eyes are halfway down the face, the bottom of the nose is halfway between the eyes and the chin, and the bottom of the mouth is halfway between the bottom of the nose and the chin. Ears go from the top of the eyes to the bottom of the nose. I don't obsess about it, but these are good guidelines. When roughing, I usually find that I draw something too big or too small - a head, a hand, a torso, whatever. When that happens I just select the area I want to change with the lasso tool and use 'Free Transform' to re-scale it. Far better to resolve these issues now than after you've inked it.
As shown in the layers pallete earlier, I keep the rough background on a separate layer to the rough foreground. This is so that I can draw lines that cut right through the characters and then erase the overlaps without disturbing anything.
By the time I've finished the roughing I've already spent about two hours on the strip. This is normal for me, but probably very slow by most people's standards. It varies depending on the complexity of what I'm drawing, but that's the average. The inking layer isn't much quicker, but at least I don't have to think about much now. I hide the red 'pre-rough' layer and trace over my blue roughs, making sure that I'm on the 'inking' layer.

Again, if I can see an opportunity to improve on the rough drawing I will, but most of the time I follow it pretty faithfully. The brush size depends on how far away the character is. In the shot above, I've put the car on a separate layer and reduced its opacity slightly so that it doesn't dominate the panel too much. I'm working at sizes far in excess of the finished strip, so I have to remember that a lot of detail is going to be lost. Lindsey's keys here don't really show up in the final version, but it's nice to know they're there. To give you an idea of the dimensions at this point, I'd have to zoom out to just 12.5% to fit the whole comic on the monitor. As with the roughing, I keep the inked background on a separate layer. For straight lines I'll usually use the line tool, and for complex objects like the car I'll draw a path with the pen tool and apply a stroke to it.
Colouring comics takes a long time but leaving them as line art lacks depth. I compromise by using shades of grey, yet even this would be a painfully slow process if I hadn't picked up the following tip from Jeph Jacques who draws the webcomic Questionable Content. It involves using a third party filter for Photoshop (a free download) to apply random colours to all the enclosed areas of the artwork (see side panel above for details). Firstly though, I hide the blue rough layers so that I'm left with just the inking layers, the panel layer and the white background. Then I hit Ctrl + Shift + Alt + E (Command + Shift + Option + E on a Mac) to merge all the visible contents onto a separate layer. This is the best keyboard shortcut ever!
The BPelt filter only works with pure black and white lines, so to kill the anti-aliasing I bring up the levels dialog and drag all the sliders to the right as follows:

This leaves me with a hard edged, pixillated* version of the artwork in which subtle details are lost, but it doesn't matter because we'll be losing the lines altogether in a minute. Now I apply random colours by going to Filter > BPelt > Multi fill. After a spot of processing (this is a big file!), the random colours appear. At this point I will look for any unwanted 'bleeds' - that is, areas that I want to shade differently but have got the same random colour because I haven't closed the lines properly. If I find any problems, I undo the filter and close any gaps using the pencil tool (not the brush because it has anti-aliasing). When I'm happy I apply the Multi fill filter again and then straight away apply the other BPelt filter, 'Flatten'. Below is the strip before and after the flatten filter has been applied.

As you can see, the Flatten filter removes all the black lines and leaves us with all the separate areas of colour. Now I drag this layer so that it's underneath the 'panels' layer:

I can now shade this layer by selecting areas with the magic wand tool (tolerance set to zero) and filling them with shades of grey from my swatches pallete, which I have customised to include all shades between white and black in 5% increments. So that the comic doesn't look entirely flat, I try to be creative in my shading by using gradients, lighting effects and shadows. The characters are usually shaded in a single tone, unless they have darker hair, because shading individual articles of clothing would take more time than I have available to me. Or perhaps I should just say that it's an artistic choice. Below is how the shading layer looks after I've finished - I've hidden the inking and panel layers to make it clearer, but I always have these visible when I'm shading.

That's it for Photoshop - now I need to apply the lettering in Illustrator. The Photoshop file is far bigger than I need for importing into Illustrator, so after saving it at full size I save a further temporary copy at just over 2000 pixels square (after trimming off the margins). In Illustrator I open up my lettering template, which is just a black frame with a set crop size, and use the 'place' command to add the resized comic. Now I just lock this layer and use separate layers for the dialogue and speech balloons.

Lettering isn't always as easy as it looks. Sometimes you struggle to fit all the words in without overlapping something important, and other times the words leave a lot of gaps within the balloon. I avoid hyphenating words unless it's absolutely necessary because I think it interrupts the flow. Maybe that's just me. I draw the balloons with the rounded rectangle tool. As this tool doesn't have a keyboard shortcut of its own by default, I changed the preferences to give it one since I use it all the time. It's really worth using as many keyboard shortcuts as possible - all those wasted seconds add up! I let the balloons overlap the panel edges if necessary because it looks better that way. After drawing the balloons, I add the tails separately using the pen tool. When these are all in place I lock all the other layers, use Ctrl + A to select all balloons and tails, and merge them together with the pathfinder pallete. And that's it! I just have to save the Illustrator file (in case I decide to tweak the dialogue later) and then use Ctrl + Shift + Alt + S to bring up the 'save for web and devices' dialog, where I've saved a preset for the correct size and quality settings. When saved, it's all over. I can exhale, relax for thirty seconds and then start to worry about the next strip.
Script: 30 mins - 1 hour
Pre-rough: 15 mins
Rough: 2 - 2.5 hours
Inking: 1.5 - 2 hours
Shading: 1 - 2 hours
Lettering: 30 - 45 mins
![]()
* 'Pixillated' is apparently the correct spelling, although 'pixellated' would be far more logical. Why must certain spellings make us look like we can't spell, necessitating annoying footnotes like this?