Tony Cliff

Tutorial: Black Fountain Techniques, Part 3

For The Black Fountain, I tried three techniques that were new for me. I completed the line art entirely within Photoshop, I made value studies before beginning the clean art, and I tried using Gamut Maps to help with my colour selection. If you’ve read part 2, you won’t be surprised to hear that they mostly positive changes, though there is one technique in particular whose usefulness I found surprising.

Read on and I’ll break ‘em all down…

DIGITAL LINE ART

There turned out to be only one major downside to doing the line art digitally: it meant more time staring into the computer. The upside, of course, is that it saved me from having to print page enlargements and then having to scan ‘em back in. The scanning is a major ass-ache, especially if you don’t have a large format scanner. Or if, like me, you need to go buy a new scanner anyway because Canon doesn’t make software for the old one that works on your new Macintosh. Good job, Canon. So, saved time there, and I like the line quality. It would have been nice to be able to get away from the LCD glow, though.

VALUE STUDIES

It’s hard to underestimate how important the value studies were. They helped me with the composition, they sped things up, they made it easy to move into colour confidently, and in general I’d say they’ve made a fundamental improvement in my image-making abilities.

With the old method, I wasted a lot of time juggling values and colours simultaneously, switching back and forth between different layers of flatting. Maybe I’d paint a character, but they weren’t distinct enough from the background. Then I’d have to jump back to the background and adjust it. Then maybe the background stopped working with some other background element and it turns out that the entire focus of the panel isn’t where I want it to be. By delaying my value considerations until late in the process and thinking only of the outlines and the shapes, I had pigeonholed myself into a weak composition. There was a lot of this, struggling back and forth between different layers, adjusting everything to fit together well. Very fiddly, time consuming, and not worth the effort, at least not now that I’ve reached a level of comfort where I don’t need to lean on the flatting.

Starting with a value sketch allowed me to move very simply through the colouring process. Yes, of course, making the value sketch involves some of the same compositional back-and-forth as the previous process, but by doing it earlier and not having to deal with different layers — by being able to just throw some brush strokes at the page — it went by much more quickly and was a more intuitive process. It also meant I could be more confident with the colour; I’ve removed the guesswork from the value, so all I need to do is pick an appropriate hue. Generally, if I stuck to my value arrangement, the colours looked believable, even if the hues seemed like they’d be weird.

It was hard to fight the idea that things should all be coloured their “proper” colour. Eventually I discovered that colours would only have to have the correct sort of “temperature” relative to each other, similar to having to have a value that’s relative to the other values. For example, a white dude wearing a blue jacket: the skin tone would be “warmer” (more red/orange) than the hue I picked for the jacket. I found this was a good way to go about picking hues. “The thing I have to colour now - is its hue warmer or cooler than the other thing I just painted?” The shirt’s cooler than the skin tone, the pants are cooler than the shirt, but the grass is warmer than the pants… this is usually how I’d progress through a panel, always aiming to stick to the value study. In this way, I ended up with some strange colourations that I’m sure I wouldn’t have if I didn’t have the value study to lean on. It provided a level of confidence that made it easy to move into colour.

Since it was easier to get into colouring, it was possible to simply skip the flatting process. I still put different elements on different layers when it seemed like a good idea (often when characters had the sky as their background), but isolating each element in the scene wasn’t a necessary step before I could feel safe to start chucking colour on things. Flatting is useful as a digital masking technique, of course, but by flatting a page, you’re also building a foundation for consequence-free exploration. Everything’s isolated, so you can tweak and manipulate something on one layer without worrying about the rest of the painting. Wait - that sounds awesome, yes, but there’s a downside. It can be a crutch. I had been using it as an excuse - I hadn’t been planning my values or colours early enough, so the flatting was there to support me and keep my image looking tight while I explored with the painting.

I think this may be a poor way to work. Of course it’ll be different for everyone, but in my colouring process I was trying to perform a loose, experimental task (defining the values and colours) within a rigid framework (the carefully named and ordered layers). Once the image is flatted and the colouring is happening, it’s easy enough to switch between layers with the CTRL-right click — a little palette pops up with all the layers underneath the cursor. “I want to paint that layer now” — it’s certainly more intuitive than trying to use the “layers” palette. But still, having to think, “that object is on a different layer,” is putting speedbump in the path of all the colour-and-value thinking. It’s disruptive. You want your brain to be occupied entirely with light and surface qualities and texture and shadow, not abstract technical distinctions. By shifting the value creation to an earlier, rougher part of the process, I think I was able to achieve better results. Now I was focusing on the right thing — the painting, not the technicalities.

Of course, I was also colouring outside of the lines. When I was working on DDatTL, that looked really sloppy, so I’d spend a lot of time making sure all the colour was within the right lines. For this new project, I had built a few more-natural looking brushes in Photoshop. They had rougher edges. I used one of these brushes for the majority of the project, and it had the pleasing effect that when it was all loose and rough, it didn’t look sloppy, it looked “natural”. In the end, I’ve tightened the colour up enough that it’s still pretty contained, but there are some spots where it bleeds, and it looks perfectly acceptable (to me, at least).

Additionally, since the value study would help to emphasize the focal point of the composition, I would know where to focus my time and effort in developing detail. “That’s where the reader’s supposed to be looking, so let’s make sure it’s rendered most fully.” And if an area off to the side is going to be mostly in shadow, there’s no need to develop all the little details there. If it seems to you like it’s cheating to emphasize effort on one area of the image and “slack off” on the rest, 1) stop being so pedantic, and 2) think about how much time you usually spend looking at a comic panel or page, and where within the image you spend the most time. Selective use of detail is a good compositional technique. I mean, I’m not going to say that’s a lesson I’ve fully embraced — while I’m no Geoff Darrow, I do enjoy noodling useless little details into an image. Fortunately, the rough value studies have shown me where I can ease off in that respect.

One time-saving example of this ability to scale back the detail: the trees. The value study showed me that I had a lot of dark, leafy, blobby shapes that made up the tree foliage. I decided not to include those in the line art. For the first time, I avoided drawing the tree foliage. They’re easy enough to draw, but there are generally two approaches to flatting them that would look good: I can flat it exactly as I’ve drawn it, painting into each little leaf around the edge of the shape, or I can flat it with a complete disregard for the line art, which would be stylistically inconsistent with the rest of the elements. So I left the tree foliage out of the line art (you’ll see that usually only the trunks are indicated), and I left the foliage to the natural shape and texture of the brush. I think it’s a more effective, more sophisticated look.

Finally, I’ve still got a ways to go in regards to thinking of the whole-page composition, but the value studies helped the quality of the composition within each panel a good deal. It seems as though composing the individual panels well is a good step towards having well-composed pages.

These are all benefits I discovered by making one change in the process: I forced myself to consider the image’s values at a much earlier stage. It’s almost shameful that I’m only realizing how important this is just now.

GAMUT MAPS

As for the gamut map, I’m not sure I’ve figured out how best to pick a gamut map, but I enjoyed using it. The completed panels are indeed “harmonious”. At least, more so than some of my older stuff. I’ll have to experiment more with picking a gamut at the start. I think I’d like to try some much more limited palette colour schemes. The ones I picked for this project still left me with a wide range of hues to work from, and I think I might prefer the look of pages with a less liberal gamut map. Starting with more limited palettes may also be good for learning the nuances of the system.

CONCLUSION

All of these techniques are things I’m going to stick with moving forward… except for maybe the digital line art, just as a personal preference. I hate the scanning, but it might be worth it.

Everything else is great. The gamut maps are handy, and I’ll be experimenting with more limited palettes in the future. I’m going to start thinking of the value arrangements as I’m doing my very first thumbnails (if not before). And I’m going to keep flatting, but only when there’s a distinct need for it.

The Black Fountain was — technically speaking — not completed on time. I missed the original deadline by three days. However, the deadline was also pushed back by a week for everyone, so… you know what? I’m calling this one a draw.

What worked:

• Value studies! Thinking about this stuff early on doesn’t just make the process faster, it’s a vital part of image-creation that I should have been doing all along.
• Gamut Maps! Sticking to a pre-defined palette — while it might initially seem limiting — helps keep the image from looking like a pile of technicolor barf. Like one of those dolphin/unicorn paintings. It’s a helpful limitation.
• The speed and quality of digital line art. I’m happy with the way the end-results look.

What didn’t work:

• Spending too much time in front of the computer. I don’t think doing everything digital was bad for the product, but it wasn’t great for my sanity. It’s probably worth learning to ink traditionally to be able to at least complete that step away from the monitors.
• Relying too much on Flatting. Gotta know what that image is supposed to look like before flatting. Otherwise it’s too tempting to think, “ahh, the flatting will let me keep changing that until I get it right or the sun burns out, whichever comes first.” Man up!
• Trying to paint all on one layer. I’m using Photoshop, I might as well take advantage of its features. (It’s a fine balance).

(Source: tonycliff.com)

1. bumbershoots-and-ladders liked this
2. sarasilkwood liked this
3. paranoiacat liked this
4. calibratethecommander liked this
5. avid-amelioration liked this
6. clockworkaubergine liked this
7. bitter-prayer liked this
8. eles-eyes reblogged this from tonycliff and added:

   I’m currently trying...‘level up’ artistically before diving

9. iwillnothangmyselftoday liked this
10. timlai liked this
11. msalaniusramblings liked this
12. pepillogrito liked this
13. tonycliff posted this