This is the third and final entry into my Trilogy of articles on Intentional Game Design, where I argue that Audiences can tell when a game has been made through intention and care, versus a design by committee mess. If you’re still not convinced, I would suggest reading onward.

The game that proves my point is Clair Obscur: Expedition 33, a turn-based RPG made by a French studio called Sandfall Interactive. It launched in April of 2025, and within its first year it sold more than eight million copies, won Game of the Year at every major awards ceremony that gives one out, and cleared a path through the gaming conversation that nobody saw coming. It was directed by Guillaume Broche, who had been a developer at Ubisoft until he left to make the game he wanted to make. He couldn't have made it inside Ubisoft. Nobody in that building was going to greenlight a turn-based RPG inspired by Belle Époque painting at a publisher betting its survival on open-world Assassin's Creed sequels. So he left, and the game exists, and we're talking about it.

I want to tell you what it was like to play.

The first thing that hit me was the music. I had booted up the game, watched the opening logos, and noticed that the visuals looked a bit rough around the edges before the soundtrack came in and stopped me from caring. Lorien Testard's score moves between acoustic guitar, full choral arrangements, vocal solos, and stretches of music that should not work next to each other on paper. They work. The game was telling me that someone in the room was unwilling to compromise on the parts that mattered most.

By the time my girlfriend got home, I was right at the scene where Sophie is gommaged. There was not a dry eye in the apartment. The opening sequence had already torn through me. Everyone you meet in those first minutes is erased in the gommage, and Gustave, the man you have just been getting to know, is reduced in front of your eyes to someone ready to end his own life. Then the game pulls him forward, sword in hand, into the expedition that will either stop the next gommage or die in the attempt. In just over an hour my heart was in pieces, and I felt more invested in this protagonist than I had in any other in a decade.

That's the thing about playing games like this after a long stretch of not feeling much. You forget how good it is until something gets through.

Some of what Clair Obscur does is recognizable from the games we talked about in Part One.

The combat is turn-based, but it isn't passive. Every action has a real-time component, parries and dodges that demand actual reflex from you in the middle of a turn-based exchange, and a freeform aiming system for breaking enemy weak points. The Pictos system, which I did not understand for my first six hours of play, turns out to be a layered build-crafting system of remarkable depth, and you only see it when you start experimenting.

The same is true of the parry windows; I'm still not great at them, but I'm better than I was, and getting better is part of what the game wants from you. Mechanics build on themselves the way Soul Reaver taught me they should. You start simple, you stack, you compound. Its complex but never dumped on you wholesale.

Then there is what the game does narratively, which is closer to Ueda than to anything else.

SPOILER WARNING

Gustave dies. There is no plot armor. The character you have spent the early hours of the game with, the one whose loss in the opening I told you tore through me, is taken from the story and it can't be undone. The game asks you to keep going, echoing a line from the early hours of the adventure. "When one falls. WHEN one falls. Not if. When. We knew not all of us would make it. But "WE. CONTINUE." As long as even one of us stands, our fight is not over." This is austere storytelling in the style of Shadow of the Colossus. The game does not insulate you from consequence, it does not provide a workaround, and it does not soften the blow with a comeback arc. The cost of going on is heavy, it asks you to pay it, and the people who built the game trust you as the player to push onward.

The art direction commits the same way. Belle Époque France, painters' palettes, a coherent aesthetic vision. The score commits, the combat commits, the story commits. Every system in this game has been weighed against the question of what it adds to the experience, and the answers are visible in every minute of play. That is design by subtraction, the same discipline Ueda put on the page, the same one Crystal Dynamics put into Nosgoth, applied here by a small French studio that didn't have the budget to make any other kind of game.

The ending is where the game makes its boldest move, and I want to talk about it a little, because it ties everything together.

You're given a choice between two endings, neither of which is clean. Maelle's path keeps the painted world alive, which means the sentient inhabitants of that world continue to exist, including a child whose role is to keep painting forever, long past the point where he wanted any part of it. Verso's path ends the painting, which functions as a kind of genocide against everyone living inside it. There is no good answer. The Darkness and the Light meet in the middle, which is, of course, exactly what Clair Obscur means in French. Chiaroscuro. The play of light and dark across a single image.

I expect the youth to side with Maelle, because Maelle's path is the one that lets you stay. The mature adult in me, the one who has actually lost people, sides with Verso, and with Renoir to a degree. The path through grief is acceptance of the loss. You honor the people who came before you by carrying on in their stead, by living the life they don't get to live, by paving the road for those who come after. Renoir had the right intention and the wrong process; he wanted to control everyone else's grief for their sake and his own, rather than letting them work through it. His wife and Maelle process grief in an arguably unhealthy manner that would see them waste away, throwing their own futures away in service of an attachment to what they have lost. The game refuses to tell you which reading is correct. It hands you the brush.

This is the move a game can make when no one in the room is consulting a focus group. It is the ending a publisher would soften, because soft endings test better. Sandfall didn't have a publisher in that sense. They had Kepler Interactive, which seems to have understood that the point of funding a small French studio's painting-and-grief turn-based RPG was to let them make it.

Sandfall did this with a team of around thirty. Ubisoft, the company Broche left to make this game, employs more than eighteen thousand. The thirty-person studio out-sold and out-acclaimed everything Ubisoft put out in the same window. Eight million copies. Game of the Year at the Game Awards, the Golden Joysticks, and the D.I.C.E. Awards. The Sandfall founders were knighted under the Ordre des Arts et des Lettres by the French Ministry of Culture. This is truly a recognition from France that a small group of its citizens did something the rest of the world is paying attention to.

The industry's working thesis for the last decade has been that you need scale, you need a focus group, you need a feature checklist, you need a hundred-hour campaign, you need broad appeal, you need to keep something happening every fifteen feet. Expedition 33 sets a turn-based RPG with a hard ending and a French art-school sensibility against that thesis, and the thesis loses.

These are the things games could, should, and can do when no one is beholden to producers, spreadsheets, or boardrooms trying to make a game for a hypothetical modern gamer who exists nowhere outside their data.

Twenty-six years ago, Crystal Dynamics handed the world Raziel and trusted us to learn how to navigate Nosgoth one mechanic at a time. Twenty-one years ago, Fumito Ueda handed the world Wander and trusted us to understand what we were doing to him without being told. A year ago, Sandfall handed the world a brush and a gommaged world and trusted us to sit with what that meant.

The willingness to trust their principles is what determines whether a game is worth your time. Three games across twenty-six years, made by people who chose to make decisions instead of avoiding them, and every one of them found me when I needed it.

The light and the dark meet in the middle. The discipline is the same. The proof has been there the whole time. We just had to remember how to look for it.