The World Design of Metroid 1 and Zero Mission | Boss Keys

There aren’t many video game franchises that have an entire genre of games named after them. But the sci-fi adventure series Metroid does indeed hold that honour. The term Metroidvania is used to describe games that have a similar structure to the mainline Metroid games. Which means the games takes place in a single, contiguous, interconnected map that spill off into all different directions: up, down, left, right. They double back on themselves, and connect up through shortcuts.

Until: you hit a dead end, blocked by an insurmountable obstacle. Perhaps, an armour-plated baddy who shrugs off your weak attack. At which point you must explore somewhere else, and hunt down a new, permanent power-up, that gives your character an additional ability. Like being able to fire a powerful projectile. Now you can backtrack to those obstacles from earlier, and use your newfound powers to make further progress.

It’s a simple set-up, but it creates a potent game about navigation and exploration. It tasks you with creating a map of the space - either mental or physical - and rewards you when you remember where those dead ends were, with further progress… or at least some handy upgrades. Plus, there’s the thrill of slowly unlocking the area, bit by bit, until you can go anywhere, giving you full control over the domain. I compared this sensation to picking apart a knot, in the first season of Boss Keys, which looked at the dungeons in the Legend of Zelda series.

And so, as more and more developers attempt to make their own Metroid-style adventure games, I wanted to take a look at the genre’s most important titles and figure out how they’re put together. I’ll start by looking at some of the major Metroid titles, including Super Metroid and Metroid Prime, before branching out to look at the games that were inspired by this franchise. And with each title, I want to find out how developers tackle the biggest challenge of the genre: making a game that gives players freedom and choice, but doesn’t leave them completely lost and frustrated.

And to do so, I’ll be looking to answer some key questions - inspired by what I learnt during Boss Keys season one. I want to know how the world is laid out. How the player knows where to go at each turn. How much of the world can be explored at each step. How much choice the player has in their path through the game. And how much backtracking is needed to finish the game. By the end of the project, we should have a pretty good idea of what makes an effective Metroidvania title.

But for now, we need to go back to the beginning and look at the game that started it all: the very first Metroid. The world of Metroid 1 takes place in the bowels of the planet Zebes, which is split into five chunks. There’s Brinstar, which holds most of the power-ups and acts as a central hub. This connects to Kraid’s lair - home of the game’s first mini boss. And also to the fiery depths of Norfair, which then links up to Ridley’s lair - where you’ll find the second mini boss.

If you kill both Kraid and Ridley, you can head on to Tourian, where you’ll finally face the titular Metroid monsters, fight the end boss, and then escape. Each area has a unique design… well, as much was possible with NES limitations. Which is useful, as Metroid does not have a map screen of any kind, so you’ll need all the orientation help you can get when exploring Zebes’s same-y looking corridors. Of course, you won’t get far in this world without picking up some power-ups.

And Metroid establishes this rule in the first few moments, with a very elegant introduction to its overarching structure. When you enter Zebes, you’ll probably run right. It’s an NES platformer, after all, and that’s what you do. But you’ll soon hit a dead end: a wall that’s too low for you to enter. So you’ll turn around, go left this time, and find the first power-up: the morph ball. Now, you can go back to that dead end and proceed.

So in the first three rooms, Metroid has told you how to progress through its world: explore as much of the space as possible, and take note of obstacles. Then go find power-ups that let you overcome those obstacles. Come back to that point, and continue onwards. And this continues for much of the game. There’s an area that’s too high to jump in Kraid’s lair - so come back with the high jump boots. And the entrance to Tourain is clearly marked with two statues of the two mini bosses - so remember to come back here once Kraid and Ridley are toast.

However, this simple structure starts to fall apart, as Metroid’s designers mix up clear, memorable obstacles, with obscure, secret walls. So to get to Kraid’s lair, you have to bomb through here: a place with clearly crumbling blocks, and an obvious route below. Nice. But to get the ice beam, you need to bomb the floor in this random tunnel, and fall into lava. In other rooms it hurts you, but here you just fall through it.

And to get the hi jump boots or find Ridley, you need to bomb through a completely random block in the corner of Norfair. At this point, making progress is not about exploration or memory or map making: its about poking at brick walls to find obscure hidden paths. Sure, bombing through walls would become a staple of Metroid map design, but only for secrets and non-essential upgrades: hardly ever in the way of the critical path.

None of this is helped by the fact that Metroid 1 lets you explore a lot of the world at once. After you collect the morph ball at the beginning of the game, you can now explore most of Brinstar, and even venture into Norfair for a bit. And once you have the bombs, you can explore most of Kraid’s lair, check out a whole bunch of Norfair, and even head off and fight Kraid or Ridley. This certainly gives the player a feeling of grand exploration - but it’s almost too much.

It’s overwhelming, and gives the player so many areas to check in one go. Finding those hidden paths wouldn’t be so hard, if you didn’t have access to such a large portion of the world. There are some caveats to this grumbling, though. One is that a lot of items are in fact optional. The Varia Suit, Long Beam, Screw Attack, Hi Jump Boots, and Wave Beam will prove handy, but are all completely skippable. So while it’s a pain that the Varia Suit is found by shooting through a random ceiling, at least you can get through the game without it.

And then there’s the fact that there are often multiple paths to get to places in Zebes. Kraid, for example, can be reached by hopping up here or by finding a secret path down here. There are also areas where you can use either the high jump or the ice beam - a rarity in Metroidvania design, where it’s often strictly one type of upgrade for one type of obstacle. And there are two places to find the ice beam, meaning that obscure location is a little less grating.

Plus, you can fight Ridley or Kraid in either order: Metroid 1 is massively non-linear in its approach. To borrow my dependency graph designs from Boss Keys season 1, the overall structure of Metroid looks like this: short and wide, which shows there’s tonnes of branching paths and loads you can do at once. It will be rare to find a game that looks like this again. A Metroidvania is not just about finding the path forward, of course.

It’s also about finding non-essential upgrades, in this case: tanks for your missiles and energy. Which are extremely important in this game because bosses need a lot of missiles to go down, and also it’s just bloody hard. But Metroid 1 is a bit weird in this regard. Many power-ups are just sitting out in the open, including these five easy-to-find missile tanks in Norfair. And others are hidden behind obscure walls, that require random bombing or a subscription to Nintendo Power to find.

We’ll see secret design improve in future titles. Ultimately, Metroid was a bold, inventive new approach to level design. It set a bunch of standards we still see today, and effectively set players loose in a free-roaming, non-linear, choice-driven world. But it was not without mistakes. The extremely hidden items, the indistinguishable locations, and a world that opens up a tad too fast, will inevitably lead to players getting lost and frustrated.

Which means Metroid is important and memorable, but difficult to enjoy today. So what would happen if Nintendo got another shot at it? A chance to take the original Metroid, and make it again for a more modern audience? Well that’s exactly what happened in 2004, when it released Metroid Zero Mission on Game Boy Advance. It’s got new areas, new bosses, new power-ups, a whole new epilogue, and a snappy control scheme. But how is the world designed?

Let’s find out. If Metroid was too hands-off. If it let you explore too much. If it let you make too many decisions. If it hid the path forward too well. Then Zero Mission is the opposite, with Nintendo being almost aggressively helpful in leading you through the game world. So now, the player’s path through Zebes is limited at every step, with the world only opening up a small amount with each new power-up. Here’s how much you can explore after you’ve collected the morph ball.

And the long beam. And the missiles. And the bombs. And so on. And at each stage, there’s maybe only one or two new locations to check out with your new power-up. So, in Metroid 1, when you get the missiles you can go up here and find the long beam, or go here and find the entrance to Tourian, or go here and find the bombs. But in Zero Mission, there’s just one door to check: the path towards the bombs. And, you know, the cool thing about getting a new power-up in a Metroidvania game, or a Zelda dungeon, is that it recontextualises everything you’ve seen before.

Suddenly, all those dead ends are paths you can now explore. Maybe you’ll find the next bit of the game, maybe you’ll find a handy upgrade, maybe you’ll find another dead end. Who knows? It’s exciting! But when it’s just one path to check, it’s obviously less interesting. It’s just rote backtracking to get to the next bit of the game. Also, there’s hardly any choices to be made in Zero Mission. Unless you’re sequence breaking - more on that in a bit - you’ll have to fight Kraid before Ridley, you’ll have to get the power-ups in the predetermined order, and there are hardly any optional power-ups.

Zero Mission adds in a bunch more stuff like the speed booster, power grip, and super missile, but most of them are needed to get through the game. And most of the optional items from Metroid 1, like the wave beam and high jump, have been paired up with obstacles, so they’re now necessary for making progress. There’s only a few optional items now, like the screw attack and charge beam. A dependency graph for Metroid Zero Mission is more like this: a tall, thin chart, which represents a very linear and restricted path through the game.

But the most obvious way that Nintendo guides you through Zebes comes in the form of these Chozo statues. At the start of the game, you’ll roll into this statue, and it will add a blinking beacon to your map, telling you where to go next. Namely, the room where the long beam is held. And this sets the tone for the adventure. There are seven Chozo statues in the game, and they will lead you from one power-up to the next, with only a few exceptions.

Now at some points, you will have to make a detour from the statue’s waypoint. Like, after getting the bombs it points you to the ice beam: but you can’t actually get there until you wander off in the opposite direction, head into Crateria, and get the power grip. So at least you have to pay attention there. And other statues are skippable. Like, after you defeat Kraid and return to Brinstar, a path will open up in the elevator room and there’s a statue there which points you towards Norfair - but you can run past it if you like.

But I think it really changes the way most players will approach the game. Instead of venturing out into the unknown, players simply follow a path, laid out for them like breadcrumbs. It’s hard to feel like an explorer when the path forward is lit up by a flashing beacon. And this affects backtracking, too. Because backtracking isn’t necessarily a dirty word. When you’re deciding to revisit an area you saw earlier, because you remember seeing an obstacle there that you now have the tools to overcome - it actually feels great.

Like, you make a mental note of this super heated room in the top of Norfair, and then once you get the Varia suit you say “aha! I can go back to that room I saw earlier!”. And inside is a missile tank. You made this decision off of your own back, and you’re rewarded with an upgrade. That’s when backtracking works. But when you find a Chozo statue that says “okay, now just travel all the way back to Norfair to continue”, it’s a chore.

You’re told where to go and you’re simply having to make the trek to get back there. It’s just a task, and that’s when backtracking feels like meaningless busy work. Do you see the difference? Hopefully. Luckily, many of the power-ups in Zero Mission make travelling faster and more enjoyable, so it is quite fun to speed boost back through rooms that gave you trouble earlier. Plus, having Brinstar as a hub area makes it quite quick to get between areas.

And there are clever shortcuts, too. The trip back from Ridley to Tourian in Metroid 1 is a slog: but a tunnel in Zero Mission makes it effortless to return to Brinstar. The Chozo statues also affect secrets. Because you know where the actual path is (the Chozo statue told you as much) you have to make the decision to specifically go off the critical path to get secrets. It’s not just part of exploring, it’s this specific task you perform.

Again, it’s a very subtle distinction, but I think it changes the way the game feels quite dramatically. Thankfully, the secret design in Metroid Zero Mission is really good. They have interesting puzzles to solve, or require smart use of Samus’s advanced movement techniques like shine-sparking, or ask you to remember and backtrack to obstacles. And exploring the entirety of Zebes at the end of the game, with all the power-ups, as you soak up secrets and upgrades, is pretty fun.

Now many true blue Metroid fans will be watching all of this and be saying that actually, if you know what you’re doing, you can really open Zero Mission up. If you find hidden blocks, or use speed running and sequence breaking tricks like wall jumps, bomb jumps, and shinesparking, you can bypass certain areas, skip certain items, and even fight Ridley before Kraid. And that’s totally true! And awesome! But it’s also not what 99% of players will be experiencing on their first go through.

Instead, the vast majority of players will be given a guided tour of Zebes. A limited path, that’s constrained at every point. A linear progression, through a bunch of must-have items. So, ultimately, Zero Mission feels like an overcorrection from Metroid 1 - an attempt to fix a confusing and overwhelming game, but one that went too far and lost that sense of exploration. That sense of venturing into the unknown. That sense being a lone bounty hunter in a scary, alien world.

What we really need is a game that learnt from the mistakes of Metroid 1, but is also very much its own beast. And that’s what we’ll see next time on Boss Keys, when we look at the world design of the incredible, Super Metroid. Hey thanks for watching! A reminder that we’re running a Game Jam at the end of August. That I stream every Wednesday over on Twitch. That GMTK and Boss Keys is paid for by viewers like you over on Patreon.

And that you’re awesome. Keep it up.