RPG Level Design (Or Musings on 5 and Expensive Graph Paper)
I am presently working with some junior designers on RPG level design. What follows is a collection of thoughts from our conversations.
I wish things were more formalized, but they are not. I still believe the only way to show you how to do something is to have you do it again and again with solid corrective critique along the way. The apprentice method for game design still works best (and yes, if you’re curious, it is precisely how I structure all my studio classes). There is much we can read and take in, but ultimately, until the brush goes to canvas, all of this knowledge remains largely untested.
Here’s how it starts for me – once my story arc is created (typically in 9 acts as opposed to three), I get to work figuring out what my world is going to look like. This generally means completing something I’ll call the five-level breakout. There is no reason for me choosing five distinct areas other than habit and a knowledge that it works well as a stage. I may end up with more (or many more as I did in Wizardry 8), but five is my minimum. Naturally, these levels usually are inextricably tied to my story arc. However, I have designed complete game areas with only a desire to explore that location myself, unsure of who or what was going to populate it and what was going to happen there when I finished. Some I still love (the Trynnie area of Wizardry 8 ) and others never saw the light of day (the underwater world of Bayjin, not shipped with Wizardry 8).
Levels in RPGs are different than those in FPSs. The whole aesthetic changes. In one, you’re often going for constant tension with spikes of the same concentrated in particular areas. In RPGs, we’re most often going for exploration and discovery. There is crossover, of course, but the primary aesthetic differs, and that makes for significant change in game style. In RPGs, I don’t typically include long pause elevator/lift things which will deliver you to a horde of enemies waiting up top (saying that makes me want to do it next time, though). All this is to say that in RPGs, I didn’t create what Romero created in his FPSs. It is amusing to me to think that we were considered to be doing the same job at the time – level and game design. Ultimately, when I think of the term “level design,” I think FPS. When I think “world design” or “area design”, I think RPG, MMO, RTS or TBS.
I still draw all my levels out by hand on graph paper. It was how I learned to do it back in 1982 before I had other options. I recall one director recently saying that he had to work with film, actual film, not digital cameras. This isn’t me. I don’t have to work with paper, but there is something very relaxing and zen-like about creating worlds on graph paper with pencil and then with ink that makes it too decadent and delicious for me to consider anything else. I purchase stupidly expensive graph paper and pens. Even typing this makes me want to go do this now. I suspect every long-term game developer has a similar story about their process.
Think intrigue and secrets: In a given location, there are several things I want to do, particularly if the aesthetic is about discovery. I go in – players go in – wanting to explore every last piece of it. You reward them not by opening the door to a giant and revealed room or even a long stretch of rooms, but by making them work for that and rewarding their work. There are plenty of ways to “make them work for it” from the old secret door to a locked door to a series of lifts that must be raised before they can get to the other side. Platformers provide an endless series of suggestions about how discovery within a level can become a puzzle in itself. Beware the location that just welcomes the player in and shows him or her everything it has. It’s not a game location. It’s a gallery. You open the door, look around, and you’re done. That is not a good tradeoff in terms of asset development $ and time to play.
People take sides: We are conditioned to do it. I force people to take sides in games constantly. Part of it is for practical reasons – if you don’t have conflict, you don’t have a game. However, I like to give players a variety of options – like aligning with one group or another or even multiple groups simultaneously. I like to give the player rope to hang themselves. Take a risk and work both sides of the fence. If you get found out, the mess is yours to clean up. When you create these choices, make them meaningful. Helping a poor peasant overcome a demeaning and punishing overlord doesn’t really give me a side to choose (unless I happen to be working for the overload at the time). In choosing sides, the player should always have something to gain and something to lose, and it’s even better if there is some kind of prisoner’s dilemma going on (short term gain vs. long term gain). I don’t want them to just press A) Yes, I’ll Take Your Quest to Kill the Rats if the other choice is B) No, Thanks. I’ll Be Back In A Minute When I Realize There Is Nothing Else To Do.
Next up… we love to be stealthy.