Ideas Made of Light

#29: THD by Ian Miller

THD, by Ian Miller. Copyright Ian Miller.
This analysis copyright 2010 by Scott M. McDaniel

The Image


Last week I told how Ian Miller pointed me to Albrecht Dürer. He had a table at the first Illuxcon and had some of his fantastic ink drawings on display. It was great to be able to get up close to see just what was going on, because unlike most comics inking you couldn’t really see the individual lines until you got up close. For example, I remember looking at Castles for quite a while.

I’d like to thank him for providing a high resolution version of THD for the analysis. I first saw it on his web site, and he was kind enough to send a version I could use to take detail images. THD is an ink drawing, but then he colored it, I’m guessing with watercolor. I’ll go into the color choices and linework, but I’ll also touch on the composition, concept, and ways he guides our eyes through the picture.

The Composition

Before diving into the linework and detail images, let’s look at the design of the image. Great execution in any medium doesn’t matter if the picture itself doesn’t work. The first thing to do is pull the color out and look at the arrangement of lights and darks.


The first thing I looked at in the image was the face of the… thing… underground. Often what we look at first is the area of highest contrast, but there are two reasons I saw the face before the tree above it. First, as people we’re primed to see faces, and our eyes go to them before most other things. Second, I saw it on Miller’s web site on the artwork page. It’s the middle thumbnail in the second row. Because I’d already seen that part of the picture, once I clicked through I was primed to look for and recognize that part it.

More of the picture is dark than light. The exception is the sky around the tree that’s above ground. There’s still plenty of variation, but that tree is the thing with the clearest silhouette. In the version below I started at the tree since it’s the object of highest contrast. To show how values (brightness) guide us around the picture, I’ve plotted the course that my eyes take from there.


In this version, I’ve simplified the greyscale version down even further. We see only black, white, and mid-level grey. Even broken down that much we clearly see what’s going on. Starting at the tree (1), we follow its course down and right, along the ground. As we get to the right side of the picture, a closer tree (2) frames the image and guides our eye down. The tree’s root system sets up the first loop, in the middle of which resides the face in the roots (3).

As we spin out of that loop we enter another one created by the root systems and rock formations (4) of the ground. The center of this loop is empty save for the claw that probably (?) belongs to the creature we just looked at. Coming out of that loop, the downward slope of the ground takes us over to the lower left part of the picture, where another framing tree (5) brings us back up above ground. Finally the denuded canopy brings us back to where we began.

The other aspect of THD’s design I’ll touch on is the golden section grid. I think most artists use the golden section concept unconsciously. Some, like Terry Moore use it explicitly. I could see an argument either way for THD. Several elements line up perfectly with the grid, but others don’t.


We can see that the horizon lies exactly on the upper golden section. Then there’s the left side of the tree above ground – that lines up with the left golden section. The big guy underground is ambiguous, though. The face itself doesn’t lie on any of the grid lines, though it’s close to the lower line. Neither is there anything in particular at the grid’s intersection points. My guess, then, is that Miller’s years of design experience led to him creating this naturally. That is, without actually overlaying the grid or figuring out the exact measurements needed. Still, I’d only bet the price of a cup of coffee on that.

The Lines

Let’s shift gears. We’ve been looking at THD as a whole. Now let’s zoom way in to see how Miller is getting these values and defining his forms. I’ll go ahead and get this out of the way now: yes, there’s an insane amount of detail work. How can he possibly do that? Won’t he go blind?

For example, let’s look at the roots/tentacles around the creature’s face:


Yes, he drew each tentacle. Then, he hatched each one with simple C’s. The spacing and thickness of those lines establish value, while their curve defines the form and turns them from flat wiggles into tubes. Or look at the rock formation to our left of the critter. Miller uses cross hatching and line direction to establish form and shading there.

While I’m not completely sure, I think that Miller probably did most of this drawing in ink first and only then went back to add color. To get an idea of what that might have looked like, I tried to remove the color and leave the black parts of the drawing in this next detail. It shows the claw that probably belongs to that face. Here is the color version, followed by the black and white simulation of the ink-only image.



The second detail here is a reasonable approximation of what Miller’s inkwork looked like to me in person. I’ll come back to color more later. Here, I think it’s interesting that Miller uses a contour line around the claws but doesn’t use them around the roots and rocks right next to it. That helps the claws stand out – we should notice them before we notice some random roots. Not even the toes behind the claws have them. Miller is sparing with contour lines.

Why, then, can we clearly see the change in planes on the rock surface next to the claws? He gives each facet its own hatching direction. Sometimes we see a slight curve to the lines that define form. Other times the lines are straight. For darker values he cross hatches in 2 or 3 directions. Line thickness also comes into play – the lines on the left side of the claws are thicker than the same lines on the top side.

Here is an example of a face on the right-most tree.


This is another example of lines following the form of an object. The lines around the mouth, for example, remind me of Doré’s linework. The forehead has an interesting mix of curved lines that suggest the form and straight horizontal lines that don’t. Something else to notice here is that Miller didn’t do all of the hatching work in black. The grooves on the right side have lines that are green, as does the face’s cheek. These lines are lower contrast than black lines would be, so they help define form while reducing the impression of shade and darkness.

Now let’s look at something I didn’t notice until I saw the larger version of the picture.


These people stand at the foot of the upper tree. It dwarfs them. Completely. Suddenly the tree and creature beneath it take on a completely different scale. Miller just uses silhouette to suggest them, but they still have a lifelike gesture (posture). Like the last detail image, we can see green hatching here as well. It also shows a good example of how Miller handles the finer parts of the tree branches.

Finally, here is a close up of the creature in the lower left corner of the picture. See if you can pick out the lines that suggest the creatures form. How about the straight hatching lines that contribute to shading but not form?


The Colors

We’ve looked at the colored hatching lines above, but what about the larger role of color? A quick look shows that greens and blues dominate. What’s your impression of how Miller uses brightness and saturation? Did you notice the small bits of red that show up in the creatures’ eyes?

Those trace amounts of red as an accent color complete a triad on the color wheel. I sampled the picture down to 32 colors and plotted them on the color wheel below. You can get a sense of the brightness by each dot. The dots closer in to the center are less saturated, tending toward grey. The ones toward the outside are more saturated, vibrant.


It’s easy to see the trend on the wheel from blue to green. For the most part the saturation is at the 50% mark or below. I find that interesting because I look at the picture and find it to be pretty saturated. More of the colors are dark than light, also. By putting lines around the dots we can see the triad that Miller used. While I doubt that he sat down and thought, “I’m going to use a triad today!” that’s the color scheme that he ended up with. My guess is that he instinctively knew that a deep red was the right accent color to use with the green and blue scheme of the picture. Yellow, while bright, wouldn’t have balanced the blue and green nearly as well. It would also produce a much narrower triangle on the color wheel.

The Elements

To wrap up we’ll go through Lee Moyer’s Elements of a Successful Illustration.

Focus: It’s a landscape, and the focus is on the tree above ground and the creature with the root/tentacle system underground.

Composition and Design: There is a clear path for our eyes through the painting. Some elements of it line up exactly with the golden sections, while others don’t.

Palette: Blues and greens, with the accent color (red) balancing the two of them on the color wheel. It’s got moderate saturation and wide range of light and dark values.

Value: Miller controls value mostly through his hatching and lines.

Mass: Often the hatching lines indicate the volume of the objects. Miller uses thickness, spacing, direction, and color of the lines to get subtle effects.

Texture: The detailed linework gives us a consistent texture throughout the work. Even when reduced to the point where we can’t see individual lines they combine to produce a distinctive look.

Symbolism: I’m not really sure. Since I don’t know the original context I can only guess. It’s got enigmatic symbols, how’s that?

Micro/Macro: Miller is certainly one for the detail work. This micro/macro concept is about knowing which details to leave out and which ones to include. For now I’ll just mention the pair of people at the base of the tree. I mentioned that above. There are also two more people in the lower section, who I didn’t show. This one small thing makes us re-evaluate the scale of the entire picture.

Ornament: It’s easy to get lost in the finely rendered plants and rocks and trees.

Narrative: I don’t see a specific story in this one picture, but there is certainly a mood of foreboding. Dark things move beneath the bright surface world.

Juxtaposition: Speaking of dark things and bright surfaces, that juxtaposition of light and dark parts is what this picture is all about.

Stylization: Miller’s style is distinctive for both his fine and precise linework as well as a certain viewpoint. See his artwork page for more examples of what I mean, but it should hardly surprise you to know that he has done many covers and illustrations for Lovecraft stories.

Character: The creatures we see in this painting seem more like forces than specific individuals. It’s the landscape itself that has the character.

Tension: When and how will the darkness underneath come into the open? How will it affect the tree above and the people’s world?

Line: Miller’s lines provide texture, values, and form. The one thing Miller doesn’t use them for extensively is outline.

Research/Reference: I don’t know, but my guess is that Miller did not use much reference for this. He knows his plants (real and imagined) as well as his geology.

Vignette: The real silhouette for us to see is the central tree. As a landscape, vignette is not a major part of the painting.

Perspective: There is no need for vanishing points and formal perspective here, but Miller does use atmospheric perspective. There is a structure of some sort in the far distance. Also, Miller uses fewer black lines on the central tree. It’s further from our viewpoint than the creatures in the foreground. He still hatches the tree to define form, it’s just that more of the lines are green rather than black. The reduced contrast pushes the tree further into the distance.

Next week, something by Aly Fell.

1 Comment

    Social comments and analytics for this post…

    This post was mentioned on Reddit by SubaruKayak: A better appreciation for the detail could be achieved if a larger image was provided- something in the ballpark of 2000 pixels wide would suffice… as it is, I am only mostly in awe at the magnificenc…