Creating a Character
Posted on April 5, 2013 by Steven
When we joined Execution Labs, Lightning Rod Games had no dedicated artist and while we tried to play it cool, deep down, we were a bit worried. Our intention was to iterate on design and technology before addressing art. However, we were not sure what kind of talent was out there. Clearly, we found no shortage of it in our artist, Kurtis.
This post will discuss the process of how we went from having no art pipeline the week before the Game Developers Conference to creating the Henchmen! poster in time for GDC 2013.
When we were interviewing artists, we looked for someone with a style that would fit well with our game. We wanted to play to the artist’s strength so that production would be as smooth as possible. The cartoony villains that Kurtis submitted after his interview (above) won us over and we were especially impressed by the magician character – who took the villain idea in a different direction from what we had originally envisioned. So, on his first day, we talked to Kurtis about what we liked from his submissions, and asked him to create a few concepts.
We liked them, but didn’t feel they were right for the game – even though, they were exactly what we asked for.
It dawned on us that Kurtis was an artist and not a psychic. With him working remotely, and with a three hour time difference, we would need to make a special effort to ensure that the team stayed synchronized.
So, like good designers, we made a spreadsheet. It contained a name; a one sentence description of the character’s background; one or two sentences about the physical appearance; a few henchmen or trap ideas; and a link to any references that inspired us. We thought about our favourite villains and what made them so memorable and decided that we should write stories for each character which – as Kurtis confirmed for us later – is how they conceptualize characters in television: from an existing script.
After roughing out about a dozen villains, we decided to focus on one in order to establish the style – and ultimately – the tone of the game.
Villainman was our challenge. If we could tell his story in a single picture, then we’d have a much easier time when we created more elaborate characters later on.
These are the exact scribbles from the spreadsheet:
Villainman: A former sidekick to “Heroman”, the most legendary superhero to ever exist. Witnessed the death of Heroman and took up his mantle. But he could never live up to the legend. Beaten down and humiliated by the villains, he gave up and quit. Eventually, the guilt of his mentor’s death overwhelmed him, driving him to desperation. Became Villainman.
We weren’t sure if we wanted a spoiled brat, an anti-hero, or a buffoon. However, we did know that he had baggage and gravitated towards the third character on the bottom row (above). We also knew that we wanted more exaggerated features. “More Disney” is what we told Kurtis. The spreadsheet wasn’t enough, so we also created Google documents for each character that would contain our notes for each revision.
#2 (above) was our guy. The slouch and eyes made him feel heavy. This was the start of something good. However, this was a player character, so he needed to feel more powerful – a powerful character who had been mentally crushed. Mark felt he also needed a mask.
Mark was right – he did need that mask.
From the three characters (above) we cobbled together what we hoped would be the Villainman…
We were pretty stoked but something was missing. Maybe the background could help tell the story…
Even though his past lies in ruins, it still looms over him (sweet scrap paper sketch!).
Kurtis created several tone sketches. They all looked great but didn’t feel right. Mark and I simply could not articulate what it was that we wanted. Thankfully, we knew someone that could.
We had spoken to Sam previously, and she had emphasized that the character design should also tell us who the character is on the inside – unless we’re trying to deliberately mislead the audience.
And we thought we were following her advice with the slumped shoulders, downward gaze, etc. However, during our second meeting, we realized that we weren’t taking it far enough. For the story we were trying to tell, everything about the character needed to be heavy. The camera position, as well as the upward lines created by the statue, made Villainman feel like a hero.
His hair, posture, and even his ears needed to drag him down. We needed to see that the character had baggage and let the background show us what that baggage was.
He needed to feel big, but small at the same time; strong, but unstable.
The camera could not look down at him because we did not want the viewer to pity him. We could not use pillars to create the lines we wanted because they would clutter the image. The monument would need to do this for us. The first changes we made were to the horizon (flattening out the camera viewing angle) and at the same time flipping the edges of the monument to point down instead of up.
Mark felt that we could extend the legs and use them as pillars (which led to the third above). We also decided the monument was too close and decided to push it back so that it could really LOOOOOM (which – for me – was the word of the project).
We were onto something.
I was getting pretty excited but the composition wasn’t feeling right, so I decided to vandalize the picture.
I first noticed that the background drew my eyes up to the knees. Once this was addressed, there were still strong diagonals pulling my eyes to the right and away from the subject. The background was an easy fix, but I was worried about the statue (I felt so attached to the tone concept above). When I expressed my concerns to Andrew from Miscellaneum Studios – who sits at the desk next to mine – he stared at me blankly and said something along the lines of “…then why don’t you just move the legs?”. I was pretty groggy and needed more convincing, so I had Elin from Imaginary Games take a look. She said that the legs framed him and made the composition feel too stable for a villain. I called Kurtis, and he wanted to move the legs too.
So we moved the legs.
It felt much better. However, when we added the title, the legs faded into the top of the image and it wasn’t clear that they belonged to a broken statue. So, they were broken off lower. However, that break was too well lit and drew the eye too much… and we went on like that for a while, fixing the little things (including the emo hair which Mark disliked so much).
The reaction from people at GDC was phenomenal. It’s amazing how a bit of art can engender interest in something.
When the final poster was produced we decided that there were three interpretations that we would be happy with:
- Guessing the entire back-story – which was a bit of a stretch
- That Villainman was a fallen hero
- That he just tore down the statue (which he may or may not have actually done)
- Artists are not psychics.
- Creating good lines of communication between the local team and people working remotely does not happen organically. It takes effort.
- Sweat the small things. It makes a huge difference. If you feel something is off, it probably is.
- If you’re not an expert, befriend one. Sam’s help probably saved us from driving Kurtis crazy!
- Collaboration is important. Create a constant feedback loop and don’t be afraid to bring people in from outside.
- Take a nap. Walk away from something and return to it later with fresh eyes.
Wisdom from KurtisWhen we asked Kurtis to lend us some insight into his process for this post, he gave us this: “I don’t do anything special. I just try to concentrate.” So there you have it. Stay tuned for more character art in the future.
- On April 05, 2013