The Daily Inference
Culture & Games · Column

Microsoft Closed Hi-Fi Rush's Studio, Calling the Game a Breakout Hit

The mid-sized, finite, risk-taking game is not extinct, but the money that kept it safe is going elsewhere. Generative AI could cut costs, and it will not help unless publishers choose to fund the middle.

Every few months someone declares the mid-budget game dead, and every few months someone answers with a list of mid-sized games released last week. Both are right, which is why the argument never goes anywhere.

So take "dead" as a provocation. Games of that size are still made. What is dying is the safety of making them, and my position is simple: the middle of the games industry is being priced out, and it will come back only if the people who pay for games decide to pay for it.

By the middle I mean no budget figure, because nobody I can cite offers a clean one. I mean a shape: a game with a clear voice, a finite scope and a cost that does not demand franchise-scale returns to justify itself. Ambitious but bounded. A game you can finish, and that its makers could finish too.

The grave worth visiting

The best-known case is Hi-Fi Rush, the game from Tango Gameworks. Microsoft itself called it a "break out hit" that met its key measurements and expectations, according to PC Gamer's account of the company's statement [1].

In May 2024 Microsoft closed Tango Gameworks, the studio that made it. Tango was one of four Microsoft-owned studios that PC Gamer said were being shut in that round [1].

I am not going to tell you the game's budget or sales sealed the studio's fate. Nothing I can cite says so.

Microsoft's stated reason, in an email from Xbox Game Studios head Matt Booty that PC Gamer said it obtained, was that the company was "prioritizing high-impact titles." Booty also said the closures were not a reflection of the teams' creativity or skill [1].

Set those two facts side by side. A game meets its expectations and is praised by its own publisher as a breakout, and the team behind it is still among those closed.

PC Gamer's headline for the industry's reaction carried the words "will no longer keep you safe" [1]. Whoever chose them was describing the same contradiction.

To me the lesson is that meeting expectations is no longer the bar. The bar is "high-impact," and that phrase is a trap for the middle. A bounded game, by design, does not try to be the biggest thing of its year.

What the other side gets right

The strongest case for the companies is real. Matthew Ball's January 2024 essay, an analyst's argument and not an industry dataset, lays out the pressures: rising costs, escalating budgets, tougher competition, changing business models and weaker growth. He ties them to the layoffs and to cancelled or underperforming games [2].

In that climate, a publisher with a finite pile of money has every reason to be careful. Nobody gets fired for backing the franchise.

But look at where the carefulness points. When every project must justify itself as high-impact, the projects that clear that bar are the ones with the biggest budgets and the longest tails, the games built to be played for years. The money gathers at the top. The middle, whose whole virtue is that it asks for a modest return, never gets a hearing, because a modest return is a dull pitch when capital is scarce.

That is how a category can be squeezed without anyone deciding to kill it. Ball's analysis does not prove a defined mid-budget class has vanished. It does explain why the pressure on it is so heavy.

The crowd below

At the other end sits the cheap game, made by a handful of people on laptops, which can reach players without any publisher's blessing. Many are lovely.

There is also a loud complaint that Steam, the biggest PC storefront, is drowning in sludge. I cannot put a number on how many games it releases, how many are junk, or what AI-generated submissions have done to the pile, and I will not pretend to. The complaint may be right.

What I can say is that nothing in it changes the argument.

If the top is closed to anything that is not a giant bet, and the bottom is a crowd, a game that wants a few million dollars and two years has nowhere to stand. It is too small for the publishers' priorities and too large for a laptop.

The AI question

This is where the industry says help is coming. It is a fair thing to hope for, so here is what the evidence says.

In a survey by the Computer Entertainment Supplier's Association, the Japanese industry body known as CESA, 1,349 developers gave valid responses. Of them, 63.0 percent said they used generative AI daily and 22.8 percent occasionally [3]. That counts tool use at work. It does not say anyone put AI-generated assets into a released game [3].

In a separate survey of member companies, 48 answered a question on what they expect from the technology. Thirty-eight chose better efficiency and productivity, 30 chose shorter development periods and 29 chose lower development and operating costs. They could pick more than one answer [3].

Those are hopes, not savings.

CESA's executive director, Tsutomu Masuda, said generative AI was being adopted "widely and steadily" in development. He also raised copyright and other rights concerns and said human checking and supervision remained important [3]. The full report is due in early December [3].

Suppose the hopes come true and games get cheaper to make. Who gets the saving?

There are two ways to spend it. A publisher can use cheaper production to make the giant bets cheaper, or to make them bigger. Ball's essay is, after all, about budget escalation [2], and a tool that lowers the cost of each feature has a way of raising the number of features. That is my inference, not a finding.

Or a publisher can use the same saving to make a bounded game survivable: a smaller team, a shorter schedule, a result that pays its way without becoming a phenomenon.

The technology is indifferent between the two. The choice belongs to whoever signs the cheques, and so far the stated priority is the high-impact title.

What to think, and what to ask for

I have lately thought a good deal about what gets kept and what gets lost, and I have come to value the finished thing. A game with a beginning, a middle and an end is a complete object. It can be handed to someone, played in a week and remembered afterward. Not every idea deserves to become a forever-world, and very few deserve to be a miracle hit.

So if you play games, spend on the middle. Buy the strange, compact, well-made thing in its first month, because a studio's survival is decided by numbers like that. If you work at a publisher, ask what the saving from your new tools is for.

The middle tier is worth rebuilding because it lets a good idea become a whole, playable game, and lets its makers survive long enough to make another.

The studio that made the breakout hit was told, in effect, that it was not high-impact enough. Think about what that means for everyone else.

Topics: Games industry · Microsoft · Generative AI · Steam

Every edition in brief, three times a day, on our Telegram channel, on Bluesky and on Threads.

Sources
  1. Making good, profitable games 'will no longer keep you safe': industry expresses fury and heartbreak over closure of Hi-Fi Rush and Prey studios | PC Gamer pcgamer.com
  2. The Tremendous Yet Troubled State of Gaming in 2024 - MatthewBall.co matthewball.co
  3. CESA Survey Finds 85.8% of Game Developers Use Generative AI godisageek.com