Sure, sorry it took a little while. I’ll include a few process snaps from the CA portrait of Stephen Gautier.
So, first thing — around one week from deadline, sometimes a bit earlier, I get headshots to work from. One thing I’ve noticed is that good headshots can make for the worst reference photos — black and white, high contrast headshots look slick but can hide a lot of important detail. Intense shadows can also distort facial proportions visually and I usually have to take extra care to not end up with lopsided noses and the like.
Depending on how soon after the headshots I receive the actual articles, I try to polish the portraits as much as I can right away to save time later on. I’ll often finalise the actual pencil drawing of the headshot before I’ve even read the article.
Unedited, straight from the scanner, this is what the pencil drawing looks like:
When I receive the articles I go into idea-finding mode. I take down key points of the article, paying special attention to anything that might work well visually. Some of the articles can be very theoretical, which is occasionally difficult from a visual communication standpoint — as I said on Twitter the other day, I’ve yet to find a working metaphor for the concept of “good design”. I try not to let my solutions get too trite or overused (i.e. lightbulb = “having an idea”) but I’ve sacrificed subtlety for the sake of getting the work done on time a few times.
This is what an idea-finding page looks like, although I usually need more than one: (Sorry for the cruddy phone picture, too lazy to scan anything else tonight.)
You’ll notice I’m a lot more verbal than visual, especially in the early stages of the idea-finding process, but it’s been useful for me to force myself into visual development earlier on — makes me come up with more varied solutions.
Eventually I (hopefully) end up with two to three ideas for each article. I scan and print the portraits I’ve drawn, then draw on top of them to come up with my final roughs. I tend to get too detailed at that stage — that’s something I need to work on, as I’d really like to speed up that process a bit. Then those drawings get scanned again and I sloppily throw some color on them just to give a rough idea of where I’m going colour-wise.
This stage looks something like this:
At that point I send them off and bite my fingernails a lot hoping no one will think they’re terrible. I hear back about which one to finalise within a day or so — I usually have two days left ‘til deadline at that point.
It’s fairly straightforward after that — I either draw the background right onto the original portrait drawing or (more frequently) on a separate sheet of paper. They’re about A5 to A4 sized and I scan them at 400 dpi because I like to work a bit bigger, then size down. Then I assemble the line art in Photoshop, clean it up a bit (I don’t ink, just do clean pencils and darken them), flat (fill with random colours) all the big areas so they’re easily selected later, and get to colouring. I started out working with only flat fills, but I’ve recently begun to start digitally painting into some areas in order to add some extra texture.
I usually have all the portraits open at the same time and jump back and forth between them. This both helps with keeping them uniform enough and me from getting bored or frustrated. When I’m eventually satisfied with the way they look I flatten my layers and paste the images into an appropriately sized background, adjusting their size to fit. Then I send them off and hope for the best.
And that’s how I make my CA portraits. I hope I managed to offer some insight without being too terribly tl;dr. If I left any questions unanswered, please feel free to use the ask box (again).