• ReaderGrev
  • Posts
  • Is 'Starfield' Actually Good? Uhhhhhhhh.

Is 'Starfield' Actually Good? Uhhhhhhhh.

It’s a book but not a story. It’s a canvas and paint but not art.

Foreground illustration by Sonny Ross

Hi, I’m Mikhail Klimentov. You may know me for my past video game coverage at The Washington Post, such as my investigation into the “culture of fear” at TSM, or this feature about my three days with Frost Giant, which is building a successor to StarCraft and Warcraft.

If you value reporting and criticism, consider subscribing on Patreon.

Invariably, whenever someone learns that I’m playing “Starfield,” they ask: Is it good?

“Starfield,” the new space adventure role-playing game from the makers of “Skyrim,” has a lot to offer. A vast array of planets to settle and plunder. Dozens of guns and even more unique attachments to wield against several galaxies’ worth of generic bad guys. Good quests! Vastly more just-OK quests. Ships. Loot. Collectible resources. Dialogue trees. The list goes on.

But is it good? Uhhhhhhhhhhhhhhhhhhhh.

The first few times1 I was asked this question, I said yes. The character creator is delightfully robust2 , and early on, I cleared out an abandoned zero-gravity casino (by “cleared out,” I mean both the pirates that had taken over and the casino vault) over the planet Nesoi — the type of story a space-faring barfly could dine out on for decades.

But as I’ve played more of the game (my save file says I’m at just under 30 hours) my opinion has changed from a straightforward “yes” as the precious quality of exploration has given way to — and in some cases revealed — the limits of “Starfield’s” world.

Writing a good song is hard. It requires a certain kind of creative exertion — a person or people making artistic decisions toward a positive end.

It’s not hard, by contrast, to use audio editing software. The software I use, Ableton, is very complicated. But in theory, there are discrete steps one can take to achieve whatever they want in a piece of software. The difference between an Ableton power user and somebody following instructions on how to achieve their desired outcome is speed.

I love to make music. I wouldn’t say I love using Ableton.3

Unfortunately, when I play “Starfield,” I don’t feel as though I’m writing a song — good or bad. I’m not exercising artistic discretion, or making expressive choices. Instead, I’m mucking around in a user interface.

“Starfield,” like Ableton, is not difficult. There’s a lot to learn, and plenty of stuff in the vastness of space that a beginner would struggle with: planets that are just out of grav-jump range; enemy ships with too-thick hulls, and too-powerful lasers; bounty hunters, space pirates, soldiers, security guards, robots and dangerous aliens with insurmountable health bars. But the challenge therein derives not from difficulty, but progression.

Crassly, so much of “Starfield” is about reaching an arbitrary level that allows you to achieve your objective, and then doing that thing in the exact way you did it before — with the exact same mechanics — except now your machine or gun or spacecraft can do the thing, and before, it couldn’t.

This kind of system is not unique to “Starfield,” nor is it necessarily problematic in and of itself. But the bad feeling it contributes to, I think, stems from the game’s overarching design.

If you were to catalog a typical “Starfield” session on a whiteboard — stripping away a layer or two of abstraction — you’d see that the game is, crudely, a series of bubbles split up by loading screens. In each bubble, there’s usually some immediate thing to do (a resource to harvest, a fight, a person to talk to, a city, a landmark) and then nothing else that would be so uniquely compelling as to encourage me to linger in that bubble.

Start in the city bubble. Play the dialogue-and-maybe-combat minigame. Open a menu. Click on space to go to space. Loading screen. Now you’re in a space bubble, playing the flying-and-maybe-combat minigame. Open a menu. Click the planet you want to fly to. Loading screen. You’ve landed in a planet bubble, which has several smaller bubbles inside it. Run 500 meters to the settlement bubble to play the shoot-at-space-enemies minigame. Collect loot. Leave the settlement bubble. Run 600 meters to the structure bubble. Play the shoot-at-space-enemies minigame again. Fast-travel to your spaceship. Loading screen. Now you’re in the ship bubble. From there, you can go up to the space bubble (via loading screen) or go back to the planet bubble (via loading screen). Your choice.

In this kind of regimented sequence, the game’s systems and mechanics can start to feel like unrelated tabs in a browser that the user is merely asked to toggle between.

Describing a game in this kind of way is not always useful (or even fair). Games often work because of artifice. But in a title like this, where the core illusion of being in space is communicated via loading screen, otherwise innocent features and mechanics get pulled into the orbit of “Starfield’s” robotic, software-like vibe.4

Earlier this year, I started to wonder if maybe gamers had run into a discursive dead end with the term “game.”

The feeling crept up on me during “Tears of the Kingdom.” To me, at least, defeating Ganon — the thing that ends the game — was sort of besides the point. I did it, eventually, out of obligation. But what really kept me coming back was the itch to complete chores for NPCs.

Is that play?

In “Valorant,” a game that takes up an exceedingly large amount of my time, play is creative exertion — something close to sport. Expression and decision-making reign supreme. Spontaneity and responsiveness to incomplete information is a big part of what makes the game interesting, and sometimes even fun. I consider these qualities important to my idea of what makes a game.5

Meanwhile, I’ve spent the last ~4-5 hours of my “Starfield” playtime building outposts, little factories that extract a planet’s resources for you. I later learned, via Reddit, that outposts are “useless,” and that the whole feature is barely integrated into the rest of the game.

Is that play? Is it bad play? Is it something else? I wish there were better language to discuss these things.

I am not interested in the debate over what is or isn’t a game for the purposes of excluding certain titles from game-dom. But I do find the software metaphor useful. It breaks “Starfield” out of its default context (“It’s a bigger and has more features than X Sony game!”) and casts it into another (“The things I’m doing do not feel expressive; the tech is fussy and mediated”). It’s just a shame the recasting redounds so poorly for “Starfield” — a game I otherwise feel compelled by.

My point isn’t that “Starfield” has bad user interfaces (though one could make that case). It’s that “Starfield” is user interface, epitomizing the qualities of “software” (derogatory) in some essential way. It’s a book but not a story. It’s a canvas and paint but not art. It is software as an end unto itself, foregrounding itself at the expense of the end product.

Worse still, in an ungenerous light, “Starfield” can feel like bad, extractive software, the kind that exacts a toll before you can even really enjoy it.6

Thanks for reading ReaderGrev! If you’d like to become a paid subscriber and receive exclusive access to more newsletters like this, subscribe on Patreon.

  1. Within the game’s first 10 hours.

  2. I spent about 45 minutes in “Starfield’s” character creator. I created a plausible person with a Rami-Malek-as-Freddie-Mercury type face. He was, I realized later with regret, busy enough in the face to look wrong, but not quite weird enough to look interesting.

    As I wrote the above graf, I kept returning to the idea that there’s something particularly millennial-humor-ish about going Fegan Floop in a character creator and then showing off how awful your new freak is. But maybe this feeling transcends generations.

  3. Plenty of people know how to use Ableton. Considerably fewer have made a good song.

  4. I am still playing and enjoying “Starfield.” I’ll often write as a way to figure out what I think, and repetition in service of explanation can read as certainty or even fervor. What I wrote above is true, to me, but it’s not the entirety of what I think or feel about “Starfield.”

  5. Caveat: Not every game fulfills these criteria, and many games are good because I don’t. But because “Starfield” is a role-playing space fantasy adventure, I do feel the need to hold it to some kind of expressive standard.

  6. “Starfield” basically requires players to look up certain features to learn more about how the game works. Given how much explanatory labor is outsourced, it strikes me as unreasonable and self-defeating for Bethesda to have withheld early access to the game from outlets such as Kotaku in apparent retaliation for previous coverage. “Starfield” is one of the biggest games of the year. A lot of video game outlets rely on releases like “Starfield” for meaningful bumps in traffic and ad revenue; these bumps, in turn, help them survive and grow.

    Developers and publishers benefit from coverage by reliable outlets, particularly when it comes to tips and guides. The quality of a player’s experience with a game now depends, in part, on how useful those tips and guides are. As the number of unscrupulous and AI-driven news sites grows, developers and publishers should do everything they can to support good websites staffed with thoughtful human writers. I believe Kotaku falls into that category. Withholding access, on the other hand, increases the likelihood that websites fail, writers get laid off and the quality of coverage suffers to a degree that hurts everyone.