How I do my thumbnail sketches in Blender.

Berangias sisters, meet the Wicked sisters. Wicked sisters, meet the Berangias sister.

One of my favorite book, Lies of Locke Lamora by Scott Lynch

Who doesn’t like a good fight scene? As I was trying to think of my favorite fantasy duel, this one from The Lies of Locke Lamora by Scott Lynch was up there for a couple of reasons:

• Jean is my favorite character in the series.

• It was a tense and emotional point in the story with what just happened to the gentleman bastards.

• 2 on 1. This fight was totally unfair.

• But you knew the only way Jean came out of the fight was by being Jean.

• Plus, he names his hatchets the Wicked sisters. Did I say how much I love Jean?

First I needed to look at a lot of action poses. My brain can't just come up with a cool pose. I need to steal like an artist!

Here's my PureRef. I went with the pose in the middle because I thought it could show Jean swinging both axes with the sisters surround him.

This time I built and posed the characters in Character Creator and brought them into Blender.

And here is where I spend a lot of time, it feels like I'm in a photo shoot session. I'm moving the camera around, changing focal lengths, adjusting the poses and the lighting.

A lot of artist will do thumbnail sketches. This is kind of my version of how I do that (it’s a bit slower, but I think it’s more powerful and it’s useful even in later stages if I need to make changes).

I'm not very imaginative and I need to see something to get into a feedback loop to help me make decisions. Working in Blender helps me do that. It's a composition tool for me. I can quickly go through hundreds of angles and perspectives until I feel like it’s right (looking at this now, I probably could have gone lower on the angle).

Here, I don’t worry too much about the accuracy of the models or textures, that’ll come in the next step when I render out the images. I just need good data to work with.

This is the composition I ended up with. I like the criss-crossing angles of the arms, legs, and the weapons and even the floor. The line of motion and how it all flow, I like it. I can imagine Jean swinging his Wicked Sisters in this scene, barely keeping the sisters at bay.

Now to the rendering part. I focus on one part of the image at a time (axe, leg, Jean's face, shirt, background, sister's expression). I usually crop out a 1:1 square or a 1:2 and work on it in a separate file. I use AI to render the image using image to image with a high init strength (.6-.7) in Leonardo ai or I use Fooocus with a low denoising strength (.3-.4)

Once I'm happy, I merge the layers and bring it back into my main working image and then I'll crop out another section to work on.

Here's a video as I turn on the layers from bottom to top (lot of layers already got merged):

read.and.draw.fantasy

View more on Instagram

So that's it. Sounds simple but I still find it hard a bit time consuming. What you don't see in the above video is the discarded 90% that didn't work out with AI and the time spent painting over to fix things. Hands are always hard and easier to just paint. The sister’s expression was challenging. Every image has it’s own challenged.

AI is here. It's not going away. So what do we want out of our tools/technology as an artist?

Back when I was a graphic designer, I worked publications and had strict deadlines. Often I worked with editors standing over my shoulders, just to make things fun, while the printers were calling me for the files. In that environment, a good designer was a fast designer.

Two decades later, a part of me still has that belief. I know it's not true. But I still want to do things fast (maybe from years of getting in trouble if I wasn't fast enough..) I also have only 2-3 hours early in the morning to work before the kids and our farm wakes up. Time is precious.

The reason I wanted to learn Blender three, four years ago was to look for shortcuts. I thought it could help me create images faster. Before that I brought the Apple pencil and iPad pro when it came out because I thought, that would make things faster. When AI came around, I was excited because I thought it could help me create images faster. See a pattern? They’re all tools. They didn’t replace me, but I did add them to my tool box.

With each image, I learned something and I'm grateful for the tools and technology that helps me do that. Maybe when I'm old and the kids are all out of the house, I'll take things slow...