By the time it left early access in 2018, Subnautica had a lot going for it. Despite being in the already tired genre of early access survival-craft-em-up, it was committed to a design philosophy that kept it feeling fresh, emphasizing mystery in all its components. I wrote a pretty glowing piece on it shortly after launch, and in it, I mentioned being excited for the potentially Lovecraftian direction of the sequel. While I’m a little disappointed that they didn’t go in that direction, the game we did get – which is currently near the end of its own early access life cycle – ended up being fascinating nonetheless. On the surface, it feels like Subnautica 2.0, an upgraded version of the systems and ideas of the original, but with more polish. It absolutely does do that, brilliantly at times, and I would be more than happy to write another piece in the style of my first one, looking at each component and how it contributes to the intended experience. But I found the most interesting parts of uncovering Below Zero’s many mysteries to be in its comparisons to its predecessor, both in how it improves, and what was sacrificed to attain that improvement. Before I dive in deeper, I do want to clarify that this is not the case of a sequel that sacrificed vision for polish and mass appeal; Below Zero is still confidently its own game. However, there is a slight difference in tone that I think is telling for the future of this design philosophy. Because, while I love Subnautica and Below Zero, its exploration and mystery-focused philosophy is really what I am interested in. In the time since Subnautica, we’ve seen the release of The Outer Wilds, another phenomenal exploration-based game in a similar vein. This gives me some hope that this design philosophy will be adapted beyond just Subnautica, and makes Below Zero’s higher budget attempt feel like a trial run for pumping more money and time into this genre. Fortunately, this first attempt is a very successful one. Below Zero upgrades the toolkit of the original Subnautica substantially, and uses those upgraded tools to create levels more complex and polished than anything in the first game. However, along the way, it sacrifices some of the mystery that made the first game’s systems so compelling to explore. Nonetheless, it stands as an example of how to apply Subnautica’s design philosophy to an experience more in line with contemporary, higher-budget design philosophies, without sacrificing its soul.
The Upgraded Toolkit
Below Zero’s first chunk of upgrades to its predecessor deal with tightening and polishing the core gameplay loop, to the point where it’s probably the best it’s ever been. Minerals in general feel more carefully balanced and useful, with fewer single-use items or poorly explained resources. On my first playthrough, even in early access, I never once had to go to the wiki to figure out where mineral X or blueprint fragment Y was, something I often had to do for some of the many scattered fragments of the cyclops in the base game. Navigation is generally made more interesting as well, strangely through its weather condition system. The first game always had the Aroura on the horizon, a crashed space ship that helped players orient themselves in the mapless game world. Below Zero has multiple above-water landmarks, which do help at times, but during harsh weather conditions, the player’s visibility can be so limited that landmarks become impossible to see. This adds an extra layer of tension, but also probably contributed to the decision to place physical maps in Below Zero’s world. These aren’t comprehensive and don’t show the player’s location, but they do help the player orient themselves, and more interestingly, point to locations the player knows nothing about. A mysterious marker for “Omega Labs” can entice the player to go exploring, as well as help them find out-of-the-way locations that they might otherwise miss. So, improvements are less interesting in patching old problems than they are in adding mystery and tension to even more aspects of the game.
Unfortunately, they didn’t do much to address perhaps the most tedious aspect of the series, inventory management. Nearly all survival games have this problem, but it’s especially difficult for Subnautica, where the player has piles and piles of varied and important resources. In my piece on the previous game, I cited my difficulties moving bases as the apex of that problem, but something I found interesting in Below Zero was that I never tried to move bases or build forward outposts, because the first game had trained me that this was too much of a hassle to be worth it. It would take a decent amount of inertia to get me to overcome those trained instincts, and that will be a recurring theme in this piece. It does seem like the team was aware of the issue, as exemplified by the Quantum Locker item. All quantum lockers, ostensibly, share the same inventory, so you can deploy one near your base, toss some emergency food in it, and carry another locker with you if you ever run out. Due to some early access bugs, I was never able to get this to work, but the idea is solid. However, the storage space of the quantum locker is so limited, it’s closer to an emergency backup than a core feature. So, unfortunately, I ended up installing inventory mods to increase the storage size of the wall-mounted lockers in the game. Once I did this, I felt the pacing of the game dramatically improve. No longer was I shuffling items between different, awkwardly-placed lockers with weird and ever-changing organizational systems. Instead, I just opened the locker, and grabbed what I wanted. I feel like many survival games could benefit from a “one giant locker” item, but I hope future titles try to address this issue in other ways. Because, ultimately, inventory management has very little to do with the core of the Subnautica experience. I could see the game working well with a Resident Evil-style inventory system, where the player’s personal inventory is incredibly limited, but they can find boxes that share the same inventory with infinite space. Maybe the seatruck could be emphasized as a primary storage mechanism. There are no doubt dozens of strategies to make the inventory system thematically interesting, or at least less intrusive, and it’s easy to propose ideas without implementing and testing them. However, this is my biggest sore spot with the series so far, because I have to spend so much of my time on an activity that is ultimately meaningless.
Fortunately, the seatruck had more success in its improvements to the original game. I’ll be blunt: piloting the cyclops in the first game…sucked. Navigating it never felt natural, the camera system was janky and awkward, I kept getting stuck on terrain, and the monster attacks felt more like annoyances than actual threats. And the seatruck is not without its own jank. This is most apparent in the interactions with the Moonpool, a building that docks and charges the player’s vehicles. When the seatruck docks to it, its attached modules detach and float idly nearby. When undocking, the player has to back the seatruck into those modules – a process that has cost me more than one newly-constructed piece – or get out of the ship and manually drag the modules over. Neither system feels natural, and this is the most “early access feeling” part of the game. It felt like the game needed an expanded Moonpool to let you swap out modules on the fly, which would have improved utility in general. Still, once you’re in the seatruck, the experience gets much better. With modules to allow for storage and fabrication, a dock for the prawn suit, and even an aquarium, it does function as a portable mini-base. It even includes a teleportation module, which allows instant teleportation back to the seatruck from any distance. Unfortunately, I didn’t find this module until reading the wiki after completing the game, so I can’t attest to its usefulness personally. Overall, the seatruck feels like a much more customizable and useful version of the cyclops and seamoth. While playing the first game, I regularly wanted to try a cyclops-only playthrough, eschewing a more established base, but it never worked reliably. The seatruck makes that feel actually possible, and is unlocked much earlier in the game. There is still some jank to be fixed, but the seatruck makes the experience of navigating Below Zero, especially the endgame, much more engaging.
Finally, Below Zero features an updated enemy roster, which is a general improvement on the first game’s. The most noticeable difference in early-game enemies is the greater variety of enemies. Below Zero has around the same number of creatures as the first game, but those enemies feel much more unique. In the base game, stalkers, sandfish, and other enemies felt more or less the same, with some slight differences in attack pattern. They were rarely a threat once the player got their seaglide, and were pretty easy to ignore. Fortunately, Below Zero spends most of its time developing these mid-sized enemies, and making them interesting. Brutesharks and Squidsharks attack the player directly, while Brinewings fire freezing projectiles at them and other fauna. Spike traps will grab the player from a distance, with the faster and larger Cryptosuchus charging head on in pairs. And independent of combat, Sea Monkeys will grab the player’s equipped tools, and even trade items with the player later on. This makes the early and mid game much more interesting for the player, because these enemies require different strategies to work around. Unfortunately, the primary leviathan creature does not fare as well. The first game had the terrifying reaper leviathan, a creature with a distinctive movement pattern and roar that still makes me sweat even after multiple playthroughs. But Below Zero’s Chelicerate…doesn’t really stack up. It’s just as mechanically dangerous, but it’s goofier design and more common presence in highly trafficked areas make it feel both more mundane and more unsettling. In the first game, the roar of a reaper leviathan meant it had seen you and was closing in. It was a giant flashing sign in 10-foot-tall letters saying, “GET OUT OF HERE NOW.” But because the Chelicerate hunts nearby fauna and other creatures have a similar roar, the player is almost constantly hearing creature roars in the Thermal Spires, Purple Vents, and Lilly zones. The end result is a sense of constant, mid-level stress, at least in my case, and that makes exploring those zones exhausting. If the Chelicerate spawns were slightly reduced, and the roar frequency reduced, I think they would be much more engaging creatures. Ultimately, while the new enemy roster has its problems, the enemies become a much more active part of the game, all the way through the endgame.
The Performance Impact: How The Tools Are Used
While the upgraded toolkit might have been a mixed bag, the improvements to level design have been almost universally positive, and it all starts with the performance. That’s a strange place to start on level design, but the technical limitations of the first game severely hampered what the designers could actually do with the world. In Below Zero…those issues are gone. Completely. While the first game was full of egregious framerate hitches, stuttering, and pop-in, Below Zero has none of that, and it feels like the level designers have been set free. Without looking at the code, I can’t tell what tricks they pulled to fix this; the only noticeable one is a slightly lower draw distance and more aggressive fog. But, from a player’s perspective, there doesn’t seem to be any tradeoffs, just improvement. Because the increased performance lets them significantly increase the density of assets and create more vertical maps that the series feels like it was made for. The opening zone, the Twisty Bridges, is perhaps the single best example of this new design. Starting as a shallow, cozy area for the player to explore, but with crevices that go down hundreds of meters. In fact, some great mid-late-game upgrades are found by diving down to the lower area of the biome. This introduces the player to one of the core appeals of Subnautica right away: finding a safe area to get established in, but being teased with further depths. And, it introduces the player to this loop even more effectively than the first game did. Without crafting the Rebreather, the player can only descend a hundred meters before severely hampering their oxygen supply. And, before getting the seatruck as a portable source of oxygen,
even having enough oxygen to get to 100 meters is unlikely. This means the player spends more time pushing the limits of their oxygen in the early game, diving just deep enough to grab this new upgrade or that new material. The game rewards this with expanded environmental interaction systems; for example, the oxygen plants that replace the role of brain coral in the first game. Brain coral required the player to sit and wait for it to produce oxygen, while O2 plants just require a simple click, with a longer recharge time. This lets the player keep their momentum while diving deeper and deeper, maybe deeper than they should be.
These levels are not just deep crevices, however. They’re fully fledged cave networks. Exploring around the Lilypad Islands, for example, might have the player start in an area of open water, then duck into a crevice to grab some materials, then entering a small cave network, all within the same area. This makes discoveries and dramatic changes in the environment much more common. In the base game, the player spent the majority of their time in open water, with the sea above them. This made for some great moments of terror when the player was hundreds of meters deep, looked up to the surface of the ocean, and saw nothing but blackness. Below Zero keeps those moments, but the player spends much more time in cave networks or mostly covered areas that provide cover from the more threatening predators. And this is where the performance impact becomes so noticeable, because these zones simply could not have existed in the first game. The closest Subnautica came was the Mushroom Forest zone, which was an infamously performance-heavy area with the worst pop-in in the game causing collisions as items appeared right in front of the player. So, I feel like I can’t overstate the value of these optimizations, because they allowed for the complex and winding levels that really made me fall in love with Below Zero. For example, the previously mentioned Lilypad Islands is Below Zero’s take on the Underwater Islands zone from the first game, a visually gorgeous but otherwise mostly forgettable zone. The Lilypad Islands, however, are much more tense, filled with cave networks, and giant, decaying lily pads that would have brought the previous game to its knees. The end result is a game world that is much more in line with the themes of the game, where mystery and discovery are an even greater part of the core game loop.
Interestingly, however, Below Zero also contains some of the most traditionally linear level design in the series. Many of the experiences feel more tightly controlled and designed, especially on land and near the endgame. This begins with the structure of the critical path, which is a lot more clearly defined. In the first game, you weren’t even aware that there *was* a critical path; you had just crash landed on this planet and needed to survive. The fact that Subnautica had a story at all was sort of a surprise to the player. Below Zero opens with a very clear narrative goal: find out what happened to your character’s sister. This gives the player a more specific focus to follow, and while it does follow the first game’s rule of multiple, concurrent narratives, it’s communicated a lot more clearly to the player. This has its advantages and disadvantages, and I personally prefer the first game’s approach because of how naturalistic your gradual immersion in the story feels. However, from a purely gameplay approach, a clearer critical path does help players get over the early game learning curve. This does not mean that the story is all directed; the player still has multiple leads to follow up on, new areas to explore, a whole on-land segment to tackle, and just new zones to find. If anything, Below Zero’s structure feels more open-ended, especially towards the endgame. This chart showing the depth of the first game’s zone highlights that, while the early and mid game had many zones spanning the available depth ranges, the last few zones all followed a strict, linear progression, basically
preventing exploration. Below Zero’s endgame does contain two linear zones, and I’ll expand on that shortly, but it otherwise feels like a much more explorable world. There is something interesting to be found in nearly every zone at nearly every stage of the game, even the starting zone. Further cementing this is the continuing usefulness of the seatruck. In the first game, you had a clear progression path from seaglide, to seamoth, to cyclops, with each vehicle being useful in specific situations or at specific depths. This meant that, functionally, your seamoth was useless in the endgame, and the cyclops was always the best choice. The seamoth couldn’t even be upgraded to go as deep as the cyclops. However, in Below Zero, the seatruck is the first vehicle the player builds, and it is used until the last moments of the endgame. Which, unfortunately, are where the level design kind of falls apart.
Now the game is technically still a few months out from its full 1.0 release, so it is possible that these zones will be reworked, but the final zones of Below Zero – the Crystal Caves and Fabricator Caverns – are once again the weakest. I previously criticized the linearity of the first game’s ending levels, how they were essentially tunnels to the next story beat. Below Zero takes that to its extreme. While I can’t find any maps of the caves to prove this, these zones felt incredibly linear, with a single point of interest in each zone. Even the Lost River, Inactive Lava Zone, and Lava Lakes from the first game still had points of interest to find, hidden side areas and caches to explore, and more unique assets such as the lava castle and two alien bases. Below Zero’s zones are much shorter, and lack these. The only unique discovery in these zones, other than the two plot-required points of interest, is Kyanite, a resource required for advanced fabrication that only exists in these caves, and a small alien cache. Strangely, however…I still prefer Below Zero’s endgame, because it is so much shorter. I think these kinds of levels are a necessary endpoint for the fantasy Subnautica is presenting, about Diving Too Deep into the mysterious oceans. Both games may have struggled in the execution, but the existence of these lower levels makes the game more imposing, knowing that some eldritch horror lurks further down. And while, systemically, these might not click as well, emotionally, the experience of the caves is much better.
“Proceed with caution. A leviathan class creature is near.”
That is the message that plays when entering the crystal caves. And it nearly got me to turn around and leave. But I kept going deeper, at least partially because I wanted to get the jump scare over with, and see whatever new horror the developer had cooked up. I tried to keep close to the walls, keeping the openings to other caves in my field of vision. Then, a few dozen meters below, I saw it. Slithering through the water with dozens of tiny arms clicking back and forth, and a pulsing, bioluminescent underbody. I panicked, and tried to turn around, but it was too late, the creature had already grabbed my seatruck, and was crushing it. I Alt-F4ed out of the game.
Subnautica’s leviathan creatures inspire a level of pure terror in me that not many games, even dedicated horror games, can match. I don’t know if I was all that afraid of the ocean before I played Subnautica, but after playing it, I most certainly am. And while I take issue with the design of the game’s last two areas, I’d be lying if I said it wasn’t one of the most effective horror experiences in the medium. Part of what makes the endgame zones more tolerable is that it doesn’t feel like the game even wants you to explore them. There are so few easy hiding places or options to retreat, that the player basically has to be on alert every second, always watching and listening for these shadow leviathans. Mechanically, at least, these experiences can be smoothed over. Once you realize that the leviathan’s grab attack takes two full hits to kill your ship, they become a lot less of a threat. Mechanically. But emotionally…I cannot be at ease when they’re nearby, even if I know they’re not a threat. So, the majority of the endgame wasn’t a challenge in system mastery, but a challenge of courage. I knew that the optional strategy was to let the leviathan grab you, then trigger your ship’s electric defense system, letting you escape unscathed. Evaluating the threat from the numerical perspective of DPS and health pools, it wouldn’t have even been a consideration. But the terror of a giant sea creature attack made it just as intense as if I were playing a permadeath run. Still, this terror can turn into exhaustion when repeated too much. Because of some unclear signposting, I thought I needed to get to the end of the final zone much earlier than I actually did, and made the treacherous, terror-inducing trip to the bottom only to be greeted with a message from the game that “you should come back here later”. And then, faced with the requirement to swim all the way back to the surface, I felt more exhaustion than dread and terror. Fortunately, if you are playing optimally, you only need to make that trip once, whereas the base game required numerous trips. This continues the trend of these endgame zones being better because they’re shorter, and for future takes on this idea, I do think that’s the approach they should take, if they’re not willing to completely overhaul it. A short, flashy conclusion to an entire game’s worth of building dread. So, when given a choice between shorter or more complicated, it’s fortunate that the on land sections of Below Zero essentially do both.
In stark contrast to its underwater endgame zones, Below Zero’s on land sections are refreshingly terror-free. This makes them some of Below Zero’s most interesting levels because of how they adopt the Subnautica formula to an environment without that extra Z axis. Some are more straightforward replacements: oxygen is replaced by cold, with warmth-refilling plants, the seatruck is replaced by the snowfox, there are mid-sized predators that can mostly be avoided with your vehicle, and larger ones that require stealth. However, each aspect has its own flair to it. The ice worm keeps that leviathan-sized spectacle, but with a more formalized stealth system to deal with them. Noise and elevation all matter when avoiding it, and the player is even given a gadget, called the Thumper, to simulate movement and distract the ice worm. In contrast to the straightforward shadow leviathan encounters, I felt like I could actually plan around the ice worms, and use different tools if I got stuck. The problem was, I almost never got stuck. As cool of an the Thumper is, I never felt the need to use it, and I never got killed by the ice worm or any other hostile fauna. The cold system fares a bit better, and it did actually kill me a few times when I forgot to bring heat-refilling items. Harsh weather conditions can really amplify its effectiveness, and make a totally mundane environment feel as hostile as a leviathan. So, it feels like a bit of a messy first pass that needs more tuning, but for a first pass on a new system in the back half of an already pretty solid game, it’s a damn impressive one.
However, I basically ignored the surface completely until the endgame, and I think this clashes with one of Subnautica’s core approaches to design: giving the player so many different paths to take. The designers specifically cite a desire to have multiple narrative threads running concurrently, and this works wonderfully for encouraging exploration, but can make it possible to ignore some systems until the endgame. Because it surrenders so much agency to the player, the player’s actions are much less predictable, more subject to random whims and habits. For example, I personally didn’t go to the surface because the first game had trained me that on land segments were brief expeditions to grab new technology or plot elements, then return to the ocean. I felt at home under water in this game because 99% of it was already happening under water. Add to this the harshness of the cold weather conditions, and the surface can seem like a hostile but also insignificant place, a roaring blizzard to make you feel more comfortable when you return to the water. The game does have a few earlier tasks pushing you in that direction, but because of how removed it feels from the base game, the player has to overcome a lot of inertia to go there. This, absolutely, has its disadvantages, but I think it comes with the upside of making the surface feel like a sort of forbidden fruit and final arena. By the time the player actually does get there, it’s been built up enough to be intriguing as all hell. It allows the game to drop some of its strongest narrative beats, which are made all the more effective by how long you’ve waited to get there.
Speaking of which, Subnautica has a story now.
A Proper Narrative: How The Tools Are Contextualized
This is it, the part of the game that makes Below Zero a test run, and the core of the game’s structural question: can you apply Subnautica to something with a traditional narrative? Characters, 3 acts, inciting incidents, all that. Like I mentioned previously, the first game did have a story, which was actually a surprise for most players expecting an empty survival game. The existence of a greater narrative at all was one of the first mysteries the player uncovers, and as much as I *love* that trick, it’s not a trick that can work twice. If they released Below Zero with seemingly no story until hour 5, players would be expecting it. So, I understand why they decided to go a different route, and it’s clear they cared very deeply about getting it right. When their first pass on the story didn’t work, they scrapped it entirely, writing an entirely new story pretty late into the early access process at considerable expense. So it’s with that in mind that I say that Below Zero has too much narrative for a Subnautica game. I really enjoy the game’s core story, I even like its core characters, and compared to just about any other game, the narrative to other content ratio is pretty low. But it’s not low enough to keep that lonely feeling that the first game delivered on so well. In Subnautica, you were *incredibly* isolated, perhaps the only sentient being on the planet, and no one knew
you were there. In Below Zero, there are Altera settlements, satellites in orbit, old, abandoned structures, years of history, and even at least two sentient beings active in the same sector as you. The first game had a few old settlements, but they were decaying, from dead explorers, implying that you might meet the same fate. Here, one of those characters from the first game is alive and well, holed up in a base you can visit at any time. It makes the exploration feel just a bit less satisfying, because you often feel like you’re not really discovering anything, just seeing other people’s work. This carries over into the approach to text logs as well. In the first game, I scanned every object and read every scrap of flavor text, just wanting to learn more about this world and its ecosystem. In Below Zero, it feels like far too much narrative content. I’m listening to audio logs of the workplace drama for some dystopian future space company, not learning hidden secrets about alien races. Sifting through pages and pages of contrite fathers who miss their daughters, little workplace romances, and complaining about equipment regulation. This is coupled with an AI companion that starts conversations about history, philosophy, and personal experiences, all while the player is swimming around the overworld. Again, none of these are bad – I actually really like the conversations between AL-AN and the player – it just feels misplaced. If there were fewer audio and text logs, maybe one or two fewer bases, and less frequent dialog, then I think Below Zero could have captured the same feeling of isolation that the first game had, while keeping its narrative ambitions. As it stands, though, the first game just delivers on this narrative mystery better than its predecessor.
So, Below Zero is not as good at narrative mystery as the first game. But so what? Sequels are allowed to go in radically different directions, explore different experiences. Aliens doesn’t feel terrifying in the same way Alien does, but that doesn’t make it a bad film. So, acknowledging the differences between the two, what kind of
experience *is* Below Zero trying to deliver? Well, one that follows many of the same rules as the first one. Those multiple narrative threads I mentioned earlier are still present, with threads of finding out what happened to Robin’s sister, learning about Altrea’s bases, the precursor AI, and the sub-stories of each of the ruined sites, to name a few. And some of those narrative threads are quite good in their own right. I’ve got a particular fondness for the AI storyline, because, while I think the conversations are misplaced and too frequent, they are interesting conversations. Robin is fiercely opinionated, especially for a first-person protagonist, a role usually filed by characters who can’t speak at all or speak sparingly and inoffensively. There’s even a bit of tension when Robin defiantly states sweeping philosophical beliefs that the player might disagree with. The first game’s writing was more functional than good in its own right. It was there to create texture and tone and hint at larger mysteries. Below Zero might lose some of that tone, but the actual quality of the writing itself has improved, and I have to commend how well they did integrate it, even if they often fell flat. Because there is one piece of narrative design that makes me really hope for the future of this franchise and this design philosophy.
Deep in a cave at the northern end of the ice fields, the player stumbles upon the frozen leviathan chamber. While it has an initial moment of shock – the creature is *massive*, after all – it quickly fades towards a different tone. It’s eerie, quiet, cozy, like a good Resident Evil safe room. The massive leviathan is almost fully encased in ice, with scattered lab equipment thrown around the cavern. Safe Cave, one of my favorite tracks on the soundtrack, kicks in, reinforcing the mood that, while everything is calm now, something big happened here. So the player gets to spend time playing detective, scanning every item, looking at the destruction and piecing together how it happened, and, of course, approaching and scanning the creature itself. They learn some very important information about a main story thread, and then quietly leave.
This, to me, is the promise of Below Zero. It’s a narrative moment rendered at a scale and budget that most smaller studios probably wouldn’t be able to pull off in a feature-rich open world game like this one. It has audio logs, text, and other conventional forms of narrative, but primarily entices the player with exploration and extrapolation. It’s one of my favorite moments in the medium, and I don’t think it could have happened in any other game, including Subnautica the first. It takes the strengths of the first game – narrative mystery, player-directed intrigue – and merges them with the strengths of the second – polish, character focus, and narrative clarity. So, while I have spent a lot of this essay criticizing the narrative implementation in Below Zero, moments like these remove my worries that these two styles of design are incompatible. It’s not impossible, it just takes experimentation and iteration, like everything else in this medium. And it shows that, with enough of both, this formula can create great experiences. Unknown Worlds hasn’t announced what their next project will be, but given all the ideas they had for Below Zero, a proper Subnautica 2 doesn’t seem unlikely. I’m hoping that they can take what they learned from Below Zero, and carry it forward into a proper successor, one that delivers on its narrative ambitions. But, for now, the game we have is an exciting look at what’s to come, and a compelling experience in its own right, and I can’t wait to see what they do next.