DarkSwitch Review – Barking Up The Wrong Tree

With the incredible deluge of city-builders available on Steam, each new entry in this subgenre must do something to stand out in the crowd. Airborne Kingdom has players building a flying city and accounting for physics and gravity; Against the Storm is a roguelike RPG where each combat encounter is building an entire city; The Wandering Village lets you build an entire civilization on the back of a giant turtle; Frostpunk perfectly melds strategic city-building with hardcore survival; etc. These are the titles that stick with us long after playing them, because they have both flair to bring in the players and the substance to wow them. Sadly, I have not once been wowed by DarkSwitch.

The “gimmick” for DarkSwitch is that it is a primarily vertical builder, taking place on a circular map around an enormous Avatar-like tree and the surrounding lands. This part of the concept actually works great – one of my favorite features is the ability to switch between a free camera (controllable with WASD as with other city-builders) and a rotational camera that pivots on an axis around a tree with the A and D keys. I’d love a hotkey to swap between cameras instead of having to click the icon, because exploration on the ground works much better with the free camera while building on the tree trunk feels better with the rotational camera. DarkSwitch‘s promise of building both up and out in sync works very well.

The game begins with fully rendered cutscenes establishing a post-apocalyptic fantasy story, introducing dozens of words I don’t understand at the same time as a wealth of characters I cannot begin to care about. I tried to pay attention to both the cutscenes and dialogue over the first two hours of DarkSwitch, but frankly it’s very poorly written. In the Frostpunk games, which DarkSwitch has taken more than a few inspirations from, I quickly became both attached to the characters and invested in the civil wars, parliament factions, and social uprisings that transpired. In DarkSwitch, all these same kind of things happen, complete with key decision points and personal stories, but all of it is so incredibly boring.

Take this for example on the poor writing quality. As your fantasy civilians begin to colonize the tree in the Prologue, you’ll encounter the Luminarcs, the faction that serves as DarkSwitch‘s “bad guys”. This incredibly powerful cult shows up to the horror of our civilians and begins taking them for tests to see if they are heretics. Nothing about what formed the Luminarcs, why they are so powerful, or the Empire’s relationship with them is explained; but it is clearly information all the characters have while I as the player do not. At the end of the Prologue, there is a gotcha moment where it is revealed these Luminarcs are not real Luminarcs, and upon being found out they flee. I am never told what any of this means or how it relates to my story of survival, or even what differentiates a real Luminarc from a fake one.

Beyond the story, which is admittedly not the primary reason for playing a strategy game like this, is a pretty traditional city-builder. Civilians need housing, can be assigned to jobs with day and night shifts, and need to be fed. You’ll set up stations to harvest food, harvest wood, harvest metal, cook food, etc. Probably the only interesting thing is the fog that creeps in every night. At around 11 PM, the fog overtakes the ground up to the base of the tree and drives anyone caught in it mad. Madness takes a few days to set in, so it’s never that big a deal, because you’ll set up beacon towers that cast light around a radius at night to keep the fog out and slowly heal madness. If driven totally mad, civilians will either die or leave the settlement.

The main issue plaguing DarkSwitch will make itself apparent very soon. There are three types of civilians (toilers, craftsmen, and rangers), but I really couldn’t tell the difference between them on sight because they’re all just people. Against The Storm takes advantage of its fantasy setting by having the different classes of civilians be humans, beaver-people, fox-people, harpies, etc. Despite this fantasy setting, all civilians are human here, just dressed slightly differently, and all have different things they’re good and bad at. The game balance problems begin here and persist through the rest of the game.

For the first two hours during the Prologue, the rangers are entirely useless outside of the five you’re allowed to use to scout. Toilers are needed for all things related to harvesting, which, early game, is everything you’re doing. The craftsmen don’t become useful until you start reaching higher levels of automation, yet you’re given a starting spread of an equal amount of the three classes making the Prologue infuriating. The rangers also seem to be bad at everything that isn’t scouting or fighting, so you’re literally better off having a fisherman’s hut at half capacity than adding rangers in to fill the gaps.

While scouting points of interest on the ground, you’ll find deposits of materials or groups of people, or perhaps something to advance a side quest. The problem is that these finds are entirely random, so there’s no strategy involved in scouting besides getting lucky. If I’m running low on metal I need to hope I happened to click a POI that has some metal lying around. Other city-builders I’ve played will offer choices at exploration points that create a new level of strategy instead – for instance, if I came to a location and met new characters and could either give up some food to have them join or loot their materials and leave them to die, greatly increasing Fear, DarkSwitch would be a whole lot more interesting.

Speaking of Fear, it’s probably my biggest imbalance gripe. Every time something bad happens, such as civilian death, madness, Luminarc abductions, hunger, etc, the Fear meter goes up. It seems to rise disproportionately fast, and when everything is going well and all civilians are happy it drops extremely slowly. In other similar games, these meters are often there to create a sense of urgency, and can be decreased rapidly by making smart, quick decisions. In DarkSwitch, no matter how efficient I got, I failed to see any meaningful effect on the Fear meter. Higher Fear also decreases worker efficiency, so it traps you into this death-loop of increasingly less efficient work shifts until you get a game over. In addition, every time there was a story point that asked me to make a decision (like whether to allow mad settlers in), every single decision I made massively increased Fear. It’s very, very poorly implemented.

As you might surmise, my first run did not end well. As I approached the end of the two hour Prologue, I found my Fear rapidly increasing with literally no way out. I had tried everything, and since the Fear had decreased my worker efficiency so badly I was unable to harvest enough wood to build watchtowers, which meant I got all my resources stolen by monsters, so all my civilians starved, and I was unable to staff more than one or two workers at each building. It was, and had been for the past hour, genuinely miserable. At the end of the Prologue, the fog overtook the tree and I had to choose whether to activate an ancient device or not. I decided to try it, and then the whole civilization was wiped out. I was stunned, and then it turned out that was part of the story, and no matter how you play the Prologue you lose everything and start fresh in Chapter 1. This was very frustrating.

I continued on with restarting the civilization in Chapter 1, harvested wood from the ruins with which I could start over, but not 20 minutes later I was attacked by demons and got a Game Over. I did not understand why or how this happened, or what it was a consequence of. If I had not been reviewing DarkSwitch I would have uninstalled it right then and there, permanently. I restarted a new campaign fresh, this time knowing the rules, and optimizing during the prologue to build upwards and in safe zones while lighting areas before harvesting from them. My second run at the Prologue was going well, and this time I had built sky rails early on to increase efficiency, this time ending up with hundreds of each resource, more than I knew what to do with.

I also decided that since every decision I made in the last run had gone badly, I’d make the opposite decision at all the story points. Lo and behold, it did not help at all. Things still went badly every time and even at the end when I chose not to activate the device, the whole civilization and all my resources were wiped out. Everything was exactly how it was last run. So I can very definitively say that literally nothing you do or build in the prologue matters in any way, shape or form, so that’s two wasted hours guaranteed with DarkSwitch.

As I continued on with this second campaign, everything suddenly became way too easy – I was blowing past all the challenges without effort, had more civilians and resources than I could staff, and flame towers and watchtowers set up 24/7 all around the tree. I even got into an extremely boring “boss fight” with a giant centipede lady that involved me sitting and watching my rangers hack at her for five minutes with no input from me until she died. So here you can see the problem; DarkSwitch is very bad at explaining how it works, so the Prologue is incredibly difficult and dissatisfying. Once you understand how it works through trial and error, it becomes way too easy to manipulate all the systems. One thing I did notice is that my sick civilians kept dying 100% of the time, even with the hospital staffed to max capacity with the right class of workers. I have no idea why. All told, at no point did I have fun playing DarkSwitch.

The UI design pairs nicely with the game balancing in that it is abysmal. Everything on screen is so, so small, I had to squint to see all the little circles and icons, much less figure out what they mean. Thankfully the text for the objectives is the right size, but all the HUD elements on the top and bottom of the screen are conservatively half the size they need to be. Also, mousing over icons does not show text of what they are. This is massively disappointing and unhelpful, as well as the fact that there is no in game setting to adjust any of this nor see previous tutorials. There’s a metric ton of information to absorb every few seconds, so not being able to click on something and review the tutorial is enough to make me not recommend DarkSwitch all on its own.

I have a fairly high end PC rig: 12th Gen i9-12900K, GeForce RTX 4070, 32 GB RAM, 12 GB VRAM. You can imagine my surprise when I saw that at high graphics settings, it took up to 10 seconds for all textures and shadows to pop in every single time I moved the camera. Which, in city-builders, is happening continuously every second. DarkSwitch was totally unplayable like this, so I lowered the graphics settings to medium. This reduced the pop-in load times to about one second, which was still jarring but not so bad that I couldn’t play it at all. This pop in included all foliage, trees, shadows, and effects like steam and lights. DarkSwitch is very poorly optimized with its textures, although that at least seems like it can be patched and fixed.

Despite how harsh I’ve been on it, DarkSwitch is not a disaster by any means. While it doesn’t offer anything that other popular titles in this genre don’t do better, it’s at least competently made and the art looks good. Some ideas, like the light vs fog mechanic or warring factions, could be very fun if fine-tuned. The music from Silent Hill composer Akira Yamaoka is excellent, but it sets a mood that the game can’t match. I do not recommend DarkSwitch to anyone in its current state, but if the texture issues are fixed, balancing is adjusted, and UI scaling and accessibility options are added, I could see this being worth your time. If you’re interested, keep it on your wishlist for now. In the meantime, check out Frostpunk 2 if you are craving politics and diplomacy or Against the Storm for high fantasy city-building.

Nirav reviewed DarkSwitch on PC with a provided code. This review is based on the version of the game available at the time of writing and our score will not be changed.