Working with Level Design (2/2)

Part two with Graham Hill gets practical. On manual versus automated placement, I put buildings down by hand: "I like being able to create interesting elevations and shapes within the world itself, like nice slopes and views of a larger area." Graham's case for the same discipline: "It's a real benefit to be this meticulous if you want to create a world with a believable history."

Modular kits are how a team survives a long development cycle: "I think we need them in terms of sustainability and the ability to maintain a long development cycle." My approach is a hybrid, with unique treatment for landmarks and layered variation everywhere else. "It's like a puzzle, and you need to work out how all the different pieces fit together."

We each name our favorite pitfall. Graham's is scale: things built in an editor viewport run too big in-game, so he builds one correctly-scaled room first and stacks character models as a ruler. Mine is jumping into 3D too quickly, before the design intent is nailed down. And on environmental storytelling: design a block around the people who would live there. No room is ever just a room.

My advice for aspiring level designers, in full: build a fundamental understanding of level design before chasing tools; knowing a little programming and art helps, but you don't have to be an expert; practice in engines with active modding communities so real players react to your work; adapt mechanics from games you love instead of copying a studio's house style, because a fresh pair of eyes is exactly what studios want; and mark up your portfolio images to show your intention. A pretty screenshot with no reasoning on it is just a pretty screenshot.