The Zelda campaign ended during August. Normally I’d do a postmortem of sorts to explain what the campaign was, what went right, what went wrong, and where we go from here. But I don’t think I can compress the lessons from this campaign into a single post, and there’s a lot to talk about besides what worked and what didn’t. Instead I’m splitting the normal postmortem into a few posts, each of which covers one of the things that made this campaign “Zelda-y”.
No matter how much Nintendo tries to break things up with world maps, quests, NPCs, and unskippable cutscenes, Zelda is a game of dungeons. The whole series is based on a structure of explore -> learn -> find dungeon -> get item -> fight boss -> acquire jewelry/girl/plot coupon. It’s only right that we start by talking about the dungeons in the campaign. A while has passed since I posted the first round of maps. Here’s the rest:
Some of those fonts were hard to draw.
Recall that the first half of the campaign was cave -> open-air forest -> maze of rotating rooms, all alike -> town -> giant pool of poison -> jail. The second half was grid of catacombs -> abandoned mineshaft -> beanstalk -> an ordinary house, nothing to see here, move along -> looping ice caverns -> legit palace. I didn’t escalate every time, but I did get the upward trend in complexity I wanted.
Dungeon eight has two maps because I wanted to subvert Zelda’s “the map is always right” trope. The players got the first map early in the dungeon, and they were so happy getting it because it meant they wouldn’t get lost in the twisty passageways. They immediately set out to explore the whole place, figuring out where the boss was and such. Only after a few dead ends did they conclude that the map represented an outdated layout of the mine, and the real map expanded the dungeon and explained what those purple veins of ore actually were.
By this point in the campaign, each dungeon was one overarching puzzle that often leveraged the dungeon item. The catacombs had a single beam of light the players needed to reflect through lenses to focus it until it could destroy walls of necrotic energy. The mineshaft’s history and current status slowly become clear over time. The beanstalks had leaves that could rise and lower based on proper gardening care. The manor only allowed travel through one color of door at a time. The ice maze forced the players to spell out the names of NPCs before they could get anywhere. The palace had a hard 24-hour time limit and patrolling guards so the players had to balance speed, stealth, exploration, puzzles, combat, and rest. Interestingly the difficulty of each puzzle was inversely proportionate to how complex the zone around the dungeon was. I’m going to pretend that was intentional.
I do think I didn’t design these quite as well as I could have, and I don’t know how much of that I can blame on the tabletop setting. I erred on the side of giving the players freedom to explore, so I lost some control over my pacing. I wanted the players to discover an unsolvable problem, leave it to explore elsewhere, discover a solution to that problem, then go back and solve it. For example, Link might see a cliff too high to scale, find the hookshot, then go back and grapple to the top of the cliff. But in D&D players have a half-dozen other ways to get up a twenty-foot wall, and there’s more than one player. The Deku Leaf is a great way to let Link glide from one place to another, but only after the players got it did I realize only one character could use it at a time. They ended up hanging from each other’s knees while the strongest character (who was also the worst at flying) took control. If I did the campaign over, I’d either scrap any items that don’t work for a group or take the Four Swords route and distribute identical, less-powerful items to each character. But I didn’t, so the process of exploring a dungeon was less satisfying that it could have been.
I did like how each monster and boss (and most of the puzzles) made sense narratively within a dungeon’s story. There wasn’t much in the way of “this conveyor-belt moth is the boss of an underground dungeon made of skulls where you got the fire rod, because shut up”. It was more “this lightning elemental is the boss of this beanstalk because you’re very high up and the campaign boss who controls weather has it in for you”. My players were willing to accept any hand-waving I did when I let a fun combat override a sensible world history, but I like to think I did pretty well.
I don’t know how much more level design like this I’m going to do. These days we don’t do many campaigns where we have proper dungeon explorations, so it’s a muscle I don’t flex as much as I did here. I think if I was to get deeper into this I’d break out software and try to give the maps a sleeker look with more character. Maybe I’ll do some rough dungeons with blank keys and post them here for readers to steal at will. Most of what I find online are black and white maps with an old-school, hand-drawn aesthetic. That’s great, but sometimes I want something that doesn’t look like its heyday was twenty years ago.
…the Zelda campaign itself notwithstanding.