There’s a moment from Breath of the Wild that I think about a lot. I was gliding across Hyrule, and I spotted a small island off in the distance — remote, mysterious, clearly put there for a reason. So I flew over. Landed. Found a Guardian. Fought it, realized my weapons were nowhere near good enough, and after a few failed attempts, opened my map, dropped a pin, and left. “I’ll come back when I’m stronger.”
That tiny, forgettable interaction — mark it and leave — is the implicit contract most action-RPGs make with you. The world is full of things you can’t do yet. Come back later. Check a guide. Make sure you’re on the right path.
I accepted that contract for years, because I didn’t know there was an alternative.
Then I played Hollow Knight.
A game that teaches you to stop worrying
Hollow Knight rarely tells you to come back later — and when it does, it never punishes you for being there. There is no “you’re not ready for this area.” You wander into Deepnest five hours too early, terrified, under-equipped, and the game just… lets you. And the remarkable thing is: it works. You might die a lot. You might scrape through by the skin of your teeth. But there’s a grub in a jar behind that horrible corridor, a pale ore past that gauntlet, a new spell after that brutal boss. The challenge is real, but so is the reward. Every single time.
After enough hours, something shifts in you. You stop opening the wiki. You stop wondering if you’ve taken a wrong turn. You just go — down whatever path looks interesting, into whatever boss arena you stumble upon — because the game has taught you, through dozens of consistent experiences, that it will never waste your time. If you can reach it, you’re ready. If it’s hard, it’s worth it. If you’re stuck, the answer is practice, not grinding.
That’s what I call the trust principle. Not difficulty, not fairness exactly — trust. The quiet confidence that the game is on your side, that curiosity will be rewarded, that you and the designers are working toward the same goal. Hollow Knight builds this trust so gradually and so completely that you don’t even notice it’s happening until it’s already the reason you can’t put the controller down.
Like many others, I started Silksong right after it launched (and Steam finally recovered from the launch-day meltdown) and finished it in a few days. Let me be clear: it’s a great game. Hornet moves like a dream, the art is staggering, and some of the boss fights are the best Team Cherry has ever designed. I don’t regret a minute of my playthrough. But something nagged at me the whole way through — a low-grade unease I couldn’t name. It took weeks of turning it over before I connected the dots: Silksong is great in all the ways you’d expect, but it forgot the one thing that made Hollow Knight more than great. It broke the trust.
Three things in particular.
The tools that solved the wrong problem
In Hollow Knight, you fight with your nail. Charms modify your playstyle — longer reach, faster attacks, a protective shell of flying insects — but they’re enhancements, not requirements. You can beat almost every boss in the game with a base nail and good fundamentals. The Charm system lets you express a preference. It never makes you dependent.
Silksong introduces tools: consumable items that deal damage, inflict status effects, create traps, and generally add a tactical layer that didn’t exist before. Used well, they can drastically simplify a fight. And that’s exactly the problem — because if tools can trivialize a fight, the bosses have to be designed with that power budget in mind.
Now the old approach — walk into the arena, learn the patterns, die, learn more, eventually win through pure mastery — no longer reliably works. Some bosses in Silksong are tuned around the assumption that you’ll use the right tools at the right moments. If you don’t realize this — if you’re still operating under the Hollow Knight contract of “just go in and practice” — you’ll hit walls that feel not hard but unfair. You’re not failing because you lack skill. You’re failing because you’re playing by the wrong rules, and the game never clearly told you the rules changed.
This is a trust violation. In Hollow Knight, the answer to “how do I beat this boss?” was always the same: practice. In Silksong, the answer is sometimes “practice,” sometimes “change your loadout,” and sometimes “did you bring enough shell shards?” The consistency is gone, and with it, the confidence that got you through every door in Hallownest without hesitation.
The rewards that stopped coming
Here’s a simple piece of math that explains a lot of Silksong‘s emotional texture.
Hollow Knight has a certain number of meaningful upgrades: mask shards, soul vessel fragments, nail upgrades, new spells, movement abilities. It also has a certain number of significant challenges: bosses, difficult platforming gauntlets, locked rooms, hidden areas. The ratio between those two numbers is remarkably close to 1:1. Almost every hard thing leads to a good thing.
Silksong is a much bigger game. More bosses, more locked rooms, more platforming gauntlets, more areas to explore. But the pool of critical upgrades — masks, silk, nail levels, core abilities — is roughly the same size as the original’s. The challenges multiplied. The rewards didn’t.
You feel this constantly. You push through a brutal gauntlet, expecting the dopamine hit of a mask shard or a new ability, and you get… rosary beads. Or shell shards. Or nothing. You beat a boss that took you forty minutes to learn and the reward is access to a room with some lore and a bench. It’s not that these things are worthless. It’s that the implicit promise — suffer, then be rewarded proportionally — has been broken. And once it’s been broken a few times, your relationship with the game changes. You stop approaching every new challenge with excitement and start approaching it with calculation: “Is this going to be worth it?”
That shift — from exploration-as-joy to exploration-as-calculation — is the death of the trust principle.
The shards that broke the economy
Every resource in Hollow Knight fits neatly into one of two categories. There are renewable resources — HP (masks) and Soul — that reset every time you sit on a bench. (These are essentially the same resource, since Soul converts into masks.) And there are permanent resources — pale ore, mask shards, story items — that you collect once and keep forever. This binary is clean, intuitive, and perfectly suited to a game built around retry loops. Bosses can be attempted infinitely because every attempt starts from the same baseline. You never come back weaker than you were before.
Shell shards in Silksong break this framework. They’re consumable but not fully renewable. They deplete when you use tools, and they don’t magically refill at a bench. If you burn through shards during repeated boss attempts, you can find yourself in a situation the original game would never have allowed: you are now worse off than when you started.
Think about what that means for the retry loop. In Hollow Knight, death costs you nothing but geo, which is recoverable. The economy of failure is generous. Every death makes you better without making you poorer. In Silksong, death can cost you shell shards, which means repeated failure depletes the very resources that give you an edge, which makes the next attempt harder, which leads to more failure. It’s a punishment spiral instead of a learning loop.
The worst part is the meta-choice it forces on you. Do you use your tools against the boss and risk running out? Do you hoard them and fight naked, without the tactical options the game was balanced around? Or do you leave the boss arena and go farm shell shards from regular enemies — a tedious, flow-destroying activity that has no equivalent in Hollow Knight?
A friend of mine had a phrase for this: 缺乏文化自信 — roughly, “a lack of cultural confidence.” It’s a term usually aimed at countries that imitate foreign trends instead of trusting their own traditions, and it’s funnier in Chinese, but it’s also exactly right. Consumable crafting resources are a staple of traditional RPGs, where the entire loop is built around gathering and crafting. But Hollow Knight‘s loop was never about that. It was about flow: bench → exploration → boss → reward → bench. Shards import a mechanic from a completely different design tradition and graft it onto a system that doesn’t need it and actively suffers from it. It feels like a solution to a problem nobody had.
Why “what makes games fun” deserves to be taken seriously
Here’s the question that haunts me: if the trust principle was so central to Hollow Knight‘s brilliance, why didn’t Team Cherry protect it in Silksong?
My answer — just a guess, but an honest one: because they never knew it was there.
The trust principle wasn’t a design document or a mission statement. It was emergent — the product of thousands of small, intuitively correct decisions made by a tiny team following their instincts. “Should this room have a reward? Yeah, feels right.” “Should this boss be beatable with just the nail? Yeah, that’s the spirit of the thing.” Those instincts were extraordinary, and they produced one of the most internally coherent games ever made. But because the principle was felt rather than articulated, it couldn’t survive the expansion.
Silksong is a much bigger, more ambitious game. More systems, more content, more mechanics. Every addition was an opportunity for the unspoken principle to get violated — and without an explicit rule to check against, nobody caught it. The shell shard system wasn’t flagged because there was no document that said “every resource must be either fully renewable or permanently acquired.” The reward dilution wasn’t caught because nobody had written down “every significant challenge must have a proportionate reward.” The tool-dependent boss design wasn’t questioned because the principle of “every fight should be winnable through pure skill mastery” existed only as a vibe.
That ineffable spark — call it creative intuition, call it lightning in a bottle — is real, and it produces extraordinary art. But it’s fragile. It can’t survive a change in scope, a new collaborator who doesn’t share the same gut feelings, or simply the volume of decisions that a larger project demands.
This is what’s at stake when people mock the attempt to intellectualize fun. “You’re overthinking it.” “Just let artists create.” “Not everything needs to be analyzed.” I understand the impulse — nobody wants great art reduced to a checklist. But what I’ve described above is the cost of not analyzing. Theory isn’t a replacement for intuition. It’s a safety net — a way to preserve what intuition discovers so it isn’t lost the moment the project outgrows the team that felt it into existence. Formalizing “what makes this fun” isn’t the death of art. It’s how art learns to survive its own success.
Hollow Knight didn’t need a theory. It had lightning in a bottle. But lightning doesn’t strike the same place twice — unless you build a rod.