"The Silent War - 2086" pre-visualizations

I started officially working on the art for "The Silent War - 2086" in a high school class called Media Art. I had an amazing teacher who taught me how to use the latest digital art technology, but it took a whole year just to draw a few pages of the graphic novel. My friends joked that by the time I finished it, it would be take place in the present day, and at my rate, they weren't wrong! 

I looked for ways to use technology to increase my productivity. For example, I found a helpful way to manage perspective in complicated buildings (e.g. ones with curved surfaces) is creating a model of the building in Tinkercad, a program that allows users to easily form shapes together to make 3D models in the computer. The viewpoint can be set from within the building at any height, and this can be screenshotted and put into Photoshop. Using tools like these, and sometimes even posing and photographing physical models, I could work around some of my shortcomings as an artist. I always enjoy looking for creative solutions to "impossible" problems!

In college, I spent a summer taking zoom classes from a fantastic art teacher who taught me a ton about things like page design and pacing and encouraged me to continue using technology to find hacks to improve the quality of my work. Unfortunately, I realized that the time it would take before I could start publishing chapters was an insurmountable obstacle, so as much as I needed to share my story and characters, I eventually put the project on the back burner. 

When A.I. chatbots were invented, I couldn't help but see what they could do with my panels. At first, I'll admit that I was very impressed. Compare this image to my hand-drawn panel in the prior section. I thought I could clean up all the mistakes, inconsistencies, and facial expressions in Photoshop (more on that later) and create the panels exactly as I imagined. I saw a video recently from a comic book artist who said that the #1 commandment of making comics is to "finish the comic." From that point of view, given the time it took me to create my panels by hand (which didn't even live up to my imagination), the choice seemed pretty clear.

However, there were also clear downsides of continuing down this path. Artists did not consent to having generative AI train on their artwork and many are now having a more difficult time finding work due to the technology. There is also the small issue that I'd probably be giving up any chances of actually making money from my story (book publishers want copyright clarity). There had to be a way to get the story out there AND have 100% human-drawn work, with the additional benefits of character consistency, style consistency, and copyrightability.

I contacted several artists (without luck) to see if I could hire them to redraw my panels the way I visualized them and eventually hired three artists on Fiverr to draw test panels based on my pre-visualizations. I got to experience firsthand why artists are angry about A.I. job displacement when it turned out that two of the people I paid (who charged me up to $85 per panel) were actually exclusively using generative A.I. (I knew from experience the types of changes that are probably impossible through prompting alone). Human artists trying to find work are hard to find as they're being crowded out by chatbot users. But I had another problem: I estimated that my graphic novel will have about 900 panels. If an artist charges anything close to $80 per panel, it doesn't take a lot of math to see that the cost would be quite unreasonable for a recent college graduate like myself.

I eventually decided that the first commandment of making comics is sacred, and wanted everyone to read and enjoy my story the way I imagined it, even if I wouldn't make a penny out of it.  However, even that was easier said than done...

The inconsistencies ("slop") that come from A.I. visualizations are not always easily fixable in Photoshop. Trust me, I've been working on this full-time for months and still haven't gotten it to where I'm satisfied with it. The "six-finger problem" is just the beginning. I spent endless hours trying to keep consistency with the characters, their clothing, their hair color and styles... And on top of that, sometimes objects and buildings would be photo-realistic, which looks cool but has a downside. We are accustomed to seeing the work of human artists and all of these artifacts draw attention and distract readers from what I want them to focus on: the story.

One night, a promising idea popped into my mind: instead of fighting with the fully rendered images, what if I created pencil-sketch versions of the panels? Not only would that be much easier to Photoshop to my liking, I could hire a colorist to give the panels consistent coloring and finalize the look in a way that actually matches what we expect comic books to look like!