tonycliff
patreon
Chapter Five Flatting Rambles (Style, Speed, and Composition)
🕑 Added 2025-03-25 19:08:08 +0000 UTC
I've been flatting, flatting, flatting, and it's all going at a respectable pace, but whoof whose idea was it to draw trees in outline again?
Last week I went to do a talk with the Illustration students at our local Emily Carr University of Art and Design. Someone asked about my animation training and whether or how it's influenced my illustration. At first I thought it was obvious, that all of traditional animation's focus on outline had set up shop in my mind. But I was drawing like that before animation. So when I painstakingly fill in each of those outlined tree-shapes, I think about what I'd do differently and remember how much inertia I'm working against (it's a lot).
That's a lot of little wobbly nubbins to fill in (above).
It makes me start to wonder (for the Nth time, I'm certain) if I should have done something differently, stylistically, maybe rendered all that foliage in a way that doesn't demand so much from this part of the process. But I didn't, and I can't change it now, and either way, with this approach, everything feels of-a-piece and consistent.
It also makes me wonder, "would one of those other apps have helped me?" But I am not optimistic. With Clip Studio I would be fighting the tools every time it struggled with a gap or a fuzzy spot in my line art, and I'd be spending time correcting the app's mistakes. With Krita, I'd still have to finesse all the gaps and I'd be spending more time layering everything once it had done its automagic. I'll stick with what I know, and lean into this little bit of wisdom I saw from Tonči Zonjić:
There are a lot of "pencil miles" in my flatting method, but it also "gets to done" in a straightforward manner, and as I work from foreground elements to background, it gets faster.
- - - - -
One of the Emily Carr students asked, "how do I work so fast, especially with the level of detail I include," and it made me laugh.
One evening, when I was in animation school, I was sitting at my drawing desk, half way through cleaning up an animation, I asked my teacher — Keith Ingham — "how do I draw faster?"
He looked at me in a way that said, "consider what I'm about to say," and asked, "why would you want to do that?"
I might have asked follow up questions, and tried to badger him into a useful answer, but I don't think I got one. He stuck to his message. Why would I want to do that?
I think about that so often. Mostly because I still can't decide if it's fair and true, or if it's one of those things a teacher says that's supposed to sound wise but is just teacherly bullshit.
In the spirit of Keith's response, I will leave the topic there.
- - - - -
The topic of "composition" also came up with the Emily Carr students. My mind went immediately to midcentury illustrator Austin Briggs, whose work Steve Lieber had shared on bsky recently.*
I don't even know what to say about Briggs' compositions, I just wanted to bring him up so I had an excuse to post these images. Not just for the composition, either — look at that lovely drawing quality. It's so full of life. He also knows exactly where we're looking, and where to steer us, and he simultaneously doesn't burn a lot of energy in places where we're NOT looking, but never skimps on giving us what we need. Or that's how it feels to me, at least.
EDIT: I should say, these images were all snatched from this Illustration Art blog. Also make sure to check out personal long-time favourite Leif Peng's ("Today's Inspiration") Flickr album of Austin Brigg art.
I find composition (in the art-skills sense) hard to talk about because while everything in art can be boiled down to, "if it works, it works; if it doesn't, it doesn't," composition is especially that way. To confound the issue, there are reliable guides like the rule of thirds — so reliable that modern cameras have overlays built into their viewfinders. Sometimes, application of these guides can be helpful. Sometimes, it can be confounding. You'll see people overlay the snail-shell diagram of The Golden Ratio on artwork to explain a piece's success, but most of the time I suspect this is post-hoc rationalization (we're good at seeing what we want to see).
And then there's this story. I wanted some travel photos to put on my office wall. I spent a ton of time scouring my archives, selecting some favourites, staring at them, whittling those down, staring at them some more, and adjusting them before having them printed out nicely. Despite all my time staring and adjusting and questioning myself, it wasn't until I got them home, printed out and laid for review on my drawing board that I realized three of the photos were essentially the same photo. Different places, different subject material and lighting, but the exact same composition: landscape images, with a focal subject on the one-third line, with foreground elements curving toward the subject and sweeping away in the background. For whatever I know about composition — and I do have the audacity to think it's one of my strengths — three identically-composed photos had made it past my critical eye.
I shared my surprise and disappointment about this on Twitter,** and Tonči (the same Tonči as above) said something along the lines of, "sometimes we take photos because they look like photos." That also strikes me as some is-it-wise-or-is-it-teacherly-bullshit bullshit but now I think about it every time I touch my camera or consider composition.
Nebulous as it may be, I suspect composition is where we will have a distinct edge over "generative AI" for a good while. I'll admit I don't keep up with the stream of "AI" hot air*** being blown into the atmosphere, so I might be out of date, but I haven't seen a generative model compose anything half as nicely as your average art student, never mind at the level of Austin Briggs.
Wherever anyone has a lick of half-decent taste, there will still be a market for old-fashioned artistry (though I'll admit the first half of that sentence is a hell of a qualifier).
- - - - -
Before I go! The new Mike Mignola book — BOWLING WITH CORPSES — is excellent!
- - - - -
Okay, I'm off to apply my good taste to flatting the rest of Chapter Five. My mind wanders a lot during flatting, so hopefully I can get it to provide some answers on the subject of composition. If so, you'll be the first person I share it with.
Another theme that reoccurred in my talk with the students was how grateful I was for the times my work gave me fun collaborative opportunities, like spitballing on animation boards or problem-solving broken scripts with friends. I know this Patreon isn't exactly the same thing, but I'm still glad to be able to share, say, thoughts on composition — okay, let's be real, enthusiasm for Austin Briggs — with you, and for the thoughtful comments you post in return.
Everyone please keep being lovely, I'll be back with more ramblings and/or finished coloured pages, whichever happens first (you know which will happen first).
Until then,
I remain,
wobbly nubbins,
TC
* I was loosely aware of Briggs' work, but I didn't know how extremely he was one of my favourites. See, this is why I don't think I'll ever quit social media, because one day you're poking at your phone, not expecting much, and are delivered the joy of being reminded about Austin Briggs.
** I'd look up the link for the exact wording, but I think we've both quit Twitter and honestly I don't even want to go look.
*** I don't think I've talked about "AI" here. Here's the TL;DR from my humble soapbox:
Sixteen-year-old Me would have lost his absolute marbles over it — with mad enthusiasm — and I try to keep this in mind.
There are some great uses for natural-language processing, like helping me make useful spreadsheets (sixteen-year-old Me would not have cared about this aspect).
"AI" slop is making internet search useless.
Early on, I was excited about the possibility that I might be able to feed my drawings to a model and have it draw rough layouts for me, but I don't think that reality has materialized.
"Generative AI" makes ugly images and is amoral on an artist-rights-respecting level and an energy-consumption level.
I have not intentionally used any of the big names in "AI" — initially this was because I suspected it was a "first one's free" situation, and I didn't want to get used to a tool that would later be held for ransom, but now it's both that and because I refuse to participate in the hype.
On balance, for me, "AI" stands for, "Ahh, I'll pass." 👍😎👍
Comments
CE Murphy
Those Briggs pieces are so *vibrant*. They're incredible. Thank you for sharing them with us! Your posts are always really interesting. :)
kaitou
Looking at AI from my SW Engineering background, my first observation was: this isn't even beta (test) no less production quality. It is NOT ready for prime-time. Not only did they steal their data for their LLMs, they didn't even vet it; ergo, loads of errors. I avoid it.
