Rain World is one of the most gorgeous 2D platformers. It gained a reputation for being unfairly difficult, and became an indie darling of the “git gud” club.
My first time playing Rain World with Survivor (the default playable slugcat) I bounced between wrath and apathy. One of my main complaints back then was the unfairness, which isn’t a direct result of developer intent, but a consequence of the virtual world behaving like an “ecosystem.” This buzzword gets thrown around a lot as a justification for why the game is the way it is, and I’m not well-versed enough in programming to have a coherent opinion on the “system” part, but I have thoughts on the world’s biology.
It seems that, for the slugcat to function as a protagonist, it needs to be somewhere in the middle of the food chain; not powerful enough to take on the larger predators like Vultures and whatever the fuck’s that enormous thing in Shoreline, but also not so weak that it couldn’t accomplish all the narrow escapes that constitute the only interesting mechanical aspect. I’m pointing this out because, for as much as people online like to throw the word “ecosystem” around, most “peaceful” creatures exist solely to get hamburgered by my character. The appeal of this virtual world is that, unlike Elder Scrolls or Fallout, hostile creatures have their own directives, independent of the player character. And yet, this seems like a thin facade that breaks once you die a couple too many times in the same location. Spawn points for most lizards seem static, and for large predators, there is a certain randomness that makes the game feel strange, as in you can’t point out if you (the player) did something correctly, or you got the right number of berries out of the RNG slot machine.
The latter randomization is what creates the unfair situations people are (rightly) furious at. Because your character isn’t John Wick, there is no reliable way of dealing with the unfairness, and in turn, it doesn’t create interesting design dilemmas due to the save point system.
Assuming you’ve hamburgered enough batflies and worms, your bunker will provide you with a save spot you’ll mostly stumble upon by accident while exploring. But depending on the placement of your bunker, which was decided by the developer, if you die, you’ll likely have to repeat the same actions until you reach your next spot. The RNG creates a sort of dull tedium that saps the energy to look for creative ways of solving problems.
But then, what problems are there to solve, besides engineering narrow escapes using platforming skills? You can technically fight, but there’s no mechanical incentive to do so. You can also try to pit critters against each other, but the largest threats are often alone or working somewhat together. You can try and plan ahead your exact steps, but the aforementioned randomness is a monkey wrench well-versed in wasting my time. And the map feature is so unfriendly to users that using external maps pulled from the wiki is my prefered way of playing Rain World.
Because make no mistake; as fantastic as the environments are, and as well-programmed the physics-based movement of the slugcat and its foes is, there is little that’s well designed in Rain World. Take, for example, the most controversial of the game’s mechanics: the karma gates. Karma gates are, according to the developer, a way of checking player skill before entering a tough zone. The zones are interconnected, and following the “standard” route will yield you karma gates with lesser requirements, but will not protect you from farming karma (cycles of hibernating with a full stomach). What’s baffling is the vocal minority that defends this poor design choice. As I stated somewhere else, if something is so bad that indie devs patch it up or offer a setting to turn it off, it means it’s really bad. In this case, karma gates are an inelegant solution to a problem created by the developer: a desire to do two things at once. On one hand, the game wants to offer me the chance of reaching a high skill ceiling and be rewarded with a flexible progression through the levels. On the other, the game needs a firm grasp over my balls as I can’t really go wherever I want unless I farmed enough cycles. There’s no reason to assume player deaths happen because of skill issues here; unfair deaths due to the “ecosystem” behaving in hostile ways are probably the most likely cause of player deaths. There’s no reliable way to measure “skill” in Rain World other than memorizing the map.
Karma gates don’t solve this. They punish people who die with farming. More elegant solutions were available, but not implemented:
The problem with the second proposal is that levels aren’t “designed” as much as they exist to hold space for critters and evasive maneuvers. To implement the second proposal you’d need level design that takes into account the possibility of players returning because they stumbled upon a challenge that’s too big.
And as much as I loathe to admit it, I like Rain World, I would’ve refunded it last year if I didn’t like it. It’s right up my alley; challenging, beautiful, purely based on gameplay, without trashy cutscenes and mandatory bad combat. But it just does so little from a game design perspective that the only solutions to its woes, through the years, was implementing an insane mess of a “remix” settings screen that makes the game lesser, but more palatable.
So honestly, just play with Monk. If you don’t intend to lose yourself in speedrunning tactics, there’s no reason not to choose the easier option.