gaming – Page 3

Nintendo Switch is the second-bestselling game console ever, behind only the PS2

gaming, Nintendo, Nintendo DS, Nintendo Switch, nintendo switch 2, ps2, Sony, sony playstation 2, Tech / Kelly Newman / February 3, 2026

Although it was finally replaced last year by the new Switch 2, the orginal switch isn’t done just yet. Many recent Switch games (and a handful of major updates, like the one for Animal Crossing) have been released in both Switch and Switch 2 editions, and Nintendo continues to sell all editions of the original console as entry-level systems for those who can’t pay $450 for a Switch 2.

The 9-year-old Switch’s continued availability has helped it clear a milestone, according to the company’s third-quarter financial results (PDF). As of December 31, 2025, Nintendo says the Switch “has reached the highest sales volume of any Nintendo hardware” with a total of 155.37 million units sold, surpassing the original DS’s lifetime total of 154.02 million units. The console has sold 3.25 million units in Nintendo’s fiscal 2026 so far, including 1.36 million units over the holidays. Those consoles have sold despite price hikes that Nintendo introduced in August of 2025, citing “market conditions.”

That makes the Switch the second-bestselling game console of all time, just three years after it became the third-bestselling game console of all time. The only frontier left for the Switch to conquer is Sony’s PlayStation 2, which Sony says sold “over 160 million units” over its long life. At its current sales rate (Nintendo predicts it will sell roughly 750,000 Switches in the next quarter), it would take the Switch another couple of years to cross that line, but those numbers are likely to taper off as we get deeper into the Switch 2 era.

Nintendo Switch is the second-bestselling game console ever, behind only the PS2 Read More »

Wing Commander III: “Isn’t that the guy from Star Wars?”

C:ArsGames, Chris Roberts, christopher roberts, flight sim, FMV, gaming, gog, Good Old Games, Origin, origin systems, retrogaming, space combat, space combat sim, Wing Commander, wing commander 3 / Shannon Garcia / February 3, 2026

C:ArsGames looks at a vanguard of the multimedia FMV future that never quite came to pass.

It’s Christmas of 1994, and I am 16 years old. Sitting on the table in our family room next to a pile of cow-spotted boxes is the most incredible thing in the world: a brand-new Gateway 66MHz Pentium tower, with a 540MB hard disk drive, 8MB of RAM, and, most importantly, a CD-ROM drive. I am agog, practically trembling with barely suppressed joy, my bored Gen-X teenager mask threatening to slip and let actual feelings out. My life was about to change—at least where games were concerned.

I’d been working for several months at Babbage’s store No. 9, near Baybrook Mall in southeast suburban Houston. Although the Gateway PC’s arrival on Christmas morning was utterly unexpected, the choice of what game to buy required no planning at all. I’d already decided a few weeks earlier, when Chris Roberts’ latest opus had been drop-shipped to our shelves, just in time for the holiday season. The choice made itself, really.

Screenshot of John Rhys-Davies and Mark Hamill in the WC3 intro — Gimli and Luke, together at last! Credit: Origin Systems / Electronic Arts

The moment Babbage’s opened its doors on December 26—a day I had off, fortunately—I was there, checkbook in hand. One entire paycheck’s worth of capitalism later, I was sprinting out to my creaky 280-Z, sweatily clutching two boxes—one an impulse buy, The Star Trek: The Next Generation Interactive Technical Manual, and the other a game I felt sure would be the best thing I’d ever played or ever would play: Origin’s Wing Commander III: The Heart of the Tiger. On the backs of Wing Commander I and Wing Commander II, how could it not be?!

The movie is on my computer!

It’s easy to pooh-pooh full-motion video games here in 2026; from our vantage point, we know the much-anticipated “Siliwood” revolution that was supposed to transform entertainment and usher interactivity into all media by the end of the millennium utterly failed to materialize, leaving in its wake a series of often spectacularly expensive titles filled with grainy interlaced video and ersatz gameplay. Even the standout titles—smash hits like Roberta Williams’ Phantasmagoria or Cyan’s Riven—were, on the whole, kinda mediocre.

But we hadn’t learned any of those lessons yet in 1994, and Wing Commander III went hard. The game’s production was absurdly expensive, with a budget that eventually reached an unheard-of $4 million. The shooting script runs to 324 printed pages (a typical feature film script is less than half that long—Coppola’s working script for The Godfather was 136 pages). Even the game itself was enormous—in an era where a single CD-ROM was already considered ludicrously large, WC3 sprawled ostentatiously across four of the 600MB-or-so discs.

Photograph of WC3's original CD-ROMs. — Still got these damn things in my closet after all these years. Credit: Lee Hutchinson

Why so big? Because this was the future, and the future—or so we thought at the time—belonged to full-motion video.

The Wing Commander III opening cinematic in all its pixelated glory.

That’s Wing Commander III’s epic opening cinematic, upscaled for YouTube. Even without the upscaling and watching it on a 15-inch CRT, I was entranced. I was blown away. Before the credits were done rolling, I was already on the phone with my buddies Steve and Matt, telling them to stop what they were doing and get over here immediately to see this thing—it’s like a whole movie! A movie, on the computer! Surely only Chris Roberts could conceive and execute such audacity!

And what a movie it was, with an actual-for-real Hollywood cast. Malcolm McDowell! John Rhys-Davies! Jason Bernard! Tom Wilson! Ginger Lynn Allen, whom 16-year-old me definitely did not want his parents to know that I recognized! And, of course, the biggest face on the box: Luke Skywalker himself, Mark Hamill, representing you. You, the decorated hero of the Vega campaign, the formerly disgraced “Coward of K’Tithrak Mang,” the recently redeemed savior of humanity, now sporting an actual name: Colonel Christopher Blair. (“Blair” is an evolution of the internal codename used by Origin to refer to the main character in the previous two Wing Commander titles—”Bluehair.”)

Screenshot of Malcolm McDowell as Admiral Tolywn — I’d watch Malcolm McDowell in anything. Malcolm McDowell is my cinematic love language. Credit: Origin Systems / Electronic Arts

Once the jaw-dropping intro finishes, the player finds Colonel Blair as the newly invested squadron commander aboard the aging carrier TCS Victory, wandering the corridors and having FMV conversations with a few other members of the carrier’s crew. From there, it’s a short hop to the first mission—because beneath all the FMV glitz, Origin still had to provide an actual, you know, game for folks to play.

Through a rose-tinted helmet visor

The game itself is…fine. It’s fine. The polygonal graphics are a welcome step up from the previous two Wing Commander titles’ bitmapped sprites, and the missions themselves manage to avoid many of the “space is gigantic and things take forever to happen” design missteps that plagued LucasArts’ X-Wing (but not, fortunately, TIE Fighter). You fly from point to point and shoot bad guys until they’re dead. Sometimes there are escort missions, sometimes you’re hitting capital ships, and there’s even a (very clunky) planetary bombing mission at the very end that feels like it directly apes the Death Star trench run while doing everything it can to shout “NO THIS IS NOT STAR WARS THIS IS VERY DIFFERENT!”

Screenshot of Mark Hamill saluting badly — That salute is… definitely a choice. Credit: Origin Systems / Electronic Arts

The space combat is serviceable, but the game also very clearly knows why we’re here: to watch a dead-eyed Mark Hamill with five days of beard stubble fulfill our “I am flying a spaceship” hero fantasies while trading banter with Tom Wilson’s Maniac (“How many people here know about the Maniac? …what, nobody?!”) and receiving fatherly advice from Jason Bernard’s Captain Eisen. And maybe, just maybe, we’d also save the universe and get the girl—either Ginger Lynn’s chief technician Rachel Coriolis or fellow pilot “Flint” Peters, played by Jennifer MacDonald.

(Perhaps unsurprisingly, given the primary purchasing demographic, players tended to overwhelmingly choose Rachel—though this might also have something to do with the fact that if you don’t choose Rachel, you can’t customize your missile loadout for some important missions near the end of the game).

Screenshot of the player picking a love interest — Who doesn’t enjoy a good old-fashioned space love triangle? Credit: Origin Systems / Electronic Arts

Worth a revisit? Definitely!

I will let others more qualified than me opine on whether or not Wing Commander III succeeded at the game-y things it set out to do—folks looking to read an educated opinion should consult Jimmy Maher’s thoughts on the matter over at his site, The Digital Antiquarian.

But regardless of whether or not it was a good game in its time, and regardless of whether or not it’s an effective space combat sim, it is absolutely undeniable that it’s a fascinating historical curiosity—one well worth dropping three bucks on at the GOG store (it’s on sale!).

There are cheats built into the game to help you skip past the actual space missions, which range in difficulty from “cream puff” to “obviously untested and totally broken” because just like in 1994, what we’re really here for is the beautiful failed experiment in interactive entertainment that is the movie portion of the game, especially when Malcolm McDowell shows up as Admiral Tolwyn and, in typical Malcom McDowell fashion, totally commits to the role far beyond what would have been required to pull it off and turns in his scenery-chewing best. (He’s even better in Wing Commander IV, though we’ll save that for another day.)

You could find a worse way today to spend those three bucks. Slap on that flight suit, colonel—the galaxy isn’t going to save itself!

Lee is the Senior Technology Editor, and oversees story development for the gadget, culture, IT, and video sections of Ars Technica. A long-time member of the Ars OpenForum with an extensive background in enterprise storage and security, he lives in Houston.

Wing Commander III: “Isn’t that the guy from Star Wars?” Read More »

Looking back at Catacomb 3D, the game that led to Wolfenstein 3D

carmack, Doom, gaming, history, id Software, john, PC Gaming, romero, wolfenstein 3d / Beth Washington / February 2, 2026

No longer keen on more Commander Keen

While id’s decision to lean into fast, action-oriented first-person games might seem obvious in retrospect, the video reveals that it was far from an easy decision. Catacomb 3D earned the team just $5,000 (about $11,750 in December 2025 dollars) through a contract to deliver bi-monthly games for Softdisk’s Gamer’s Edge magazine-on-a-disk. Each episode of the Commander Keen series of run-and-gun 2D games, on the other hand, was still earning “10 times that amount” at the time, Romero said.

That made sticking with Commander Keen seem like the “obvious business decision,” Romero says in the video. The team even started work on a seventh Commander Keen game—with parallax scrolling and full VGA color support—right after Catacomb 3D‘s release. At the time, it felt like Catacomb 3D might be “just like a weird gimmick thing that we did for a little bit because we wanted to play with a different technology,” as John Carmack put it.

A tech demo shows early work on Commander Keen 7 that was abandoned in favor of Wolfenstein 3D.

That feeling started to fade away, Carmack said, after his brother Adrian had an “almost falling out of his seat” moment while pivoting toward an in-game troll in Catacomb 3D. “It automatically sucked you in,” Adrian Carmack said of the feeling. “You’re trying to look behind walls, doors, whatever… you get a pop-out like that, and it was just one of the craziest things in a video game I had ever seen.”

That kind of reaction from one of their own eventually convinced the team to abandon two weeks of work on Keen 7 to focus on what would become Wolfenstein 3D. “It kind of felt that’s where the future was going,” Carmack said in the video. “[We wanted to] “take it to some place that it wouldn’t happen staying in the existing conservative [lane].”

“Within two weeks, [I was up] at one in the morning and I’m just like, ‘Guys, we need to not make this game [Keen],’” Romero told Ars in 2024. “‘This is not the future. The future is getting better at what we just did with Catacomb.’ … And everyone was immediately was like, ‘Yeah, you know, you’re right. That is the new thing, and we haven’t seen it, and we can do it, so why aren’t we doing it?’”

Looking back at Catacomb 3D, the game that led to Wolfenstein 3D Read More »

Why reviving the shuttered Anthem is turning out tougher than expected

anthem, Bioware, EA, gaming, revival / Kelly Newman / January 28, 2026

Despite proof-of-concept video, EA’s Frostbite Engine servers are difficult to pick apart.

Anthem may be down, but it’s not quite out yet. Credit: Bioware

On January 12, EA shut down the official servers for Anthem, making Bioware’s multiplayer sci-fi adventure completely unplayable for the first time since its troubled 2019 launch. Last week, though, the Anthem community woke up to a new video showing the game at least partially loading on what appears to be a simulated background server.

The people behind that video—and the Anthem revival project that made it possible—told Ars they were optimistic about their efforts to coerce EA’s temperamental Frostbite engine into running the game without access to EA’s servers. That said, the team also wants to temper expectations that may have risen a bit too high in the wake of what is just a proof-of-concept video.

Andersson799’s early proof-of-concept video showing Anthem partially loading on emulated local servers.

“People are getting excited [about the video], and naturally people are going to get their hopes up,” project administrator Laurie told Ars. “I don’t want to be the person that’s going to have to deal with the aftermath if it turns out that we can’t actually get anywhere.”

Keep an eye on those packets

The Anthem revival effort currently centers around The Fort’s Forge, a Discord server where a handful of volunteer engineers and developers have gathered to pick apart the game and its unique architecture. Laurie said they initially set up the group “out of little more than spite for EA and Bioware around the time the shutdown got announced” back in July.

While Laurie has some experience with the community behind Gundam Evolution revival project Side 7, they knew they’d need help from people with direct experience working on EA’s Frostbite engine games. Luckily, Laurie said they were “able to catch the eyes of people who are familiar with this line of work [without] searching too much.”

One of those people was Ness199X, an experienced Frostbite tinkerer who told Ars he “never really played much Anthem” before the game’s shutdown was announced. When a friend pointed out the impending death of the title, though, Ness said he was motivated to preserve the game for posterity.

Initial efforts to examine what made Anthem tick “came up empty,” Ness said, largely because the game uses EA’s bespoke Frostbite engine differently than other EA titles. To begin mapping out those differences, Ness released a packet logger tool in September that let contributors record their own network traffic between the client and EA’s official servers. In addition to helping with reverse-engineering work, Ness writes on the Fort’s Forge Discord that players who logged their packets should be able to fully recover their characters if and when Anthem comes back in playable form.

Catching Frostbite

By analyzing that crowdsourced packet data, Ness said the Fort’s Forge team has been able to break Anthem down into three essential services:

EA’s Blaze server: Used for basic player authentication.
Bioware Online Services (aka BIGS): A JSON web server used to track player information like inventory and quest progression.
The Frostbite multiplayer engine: Loads level data and tracks the real-time positions of players and non-player characters in those levels.

Early efforts to emulate the Blaze and BIGS portions of that architecture helped lead directly to last week’s proof-of-concept video. Andersson799—who says he’s been tinkering with Battlefield and other Frostbite games since 2015—said he was quickly able to use his own logged Anthem packets to create a “barebones anthem private server” that served as a “quick and dirty” sample that he decided to share via YouTube.

“I basically made the tool to just simply reply with the packet captures that I got,” Andersson told me. That was enough to “get in to the game with player profiles loaded and everything.” And while Ness says there’s still some effort needed “to [make Blaze and BIGS] work well and smoothly in terms of quest progression, etc.,” the path forward on those portions is relatively straightforward.

It’s the Frostbite engine and its odd client-server architecture that forms the biggest barrier to getting Anthem up and running again without EA’s servers. “Due to how Frostbite is designed, all gameplay in a Frostbite game runs in a ‘server’ context,” Ness explained. Even in a single-player game like Mass Effect: Andromeda, he said, “the client just creates a separate server thread and pipes all the traffic internally.”

“I feel like with Anthem, it heavily relies on online data that was stored in Bioware’s server,” Andersson added. “In my initial testing, the game couldn’t load into the level without that data.”

Anthem‘s Fort Tarsis area loads its data from local files, rather than EA’s servers.

There’s some hope that this crucial level data is still available and recoverable, though. Ness points out that Fort Tarsis, the game’s lobby area, already runs using offline data piped through a local “server” thread, meaning the rest of the game could theoretically be coerced to run similarly.

Just as important, he says, “as far as we have been able to discern, all the logic for the other levels, which when the game was live ran on a remote server, also exists in the client,” Ness said. “By patching the game we can most likely enable the ability to host these in process as well. That’s what we’re exploring.”

“To be honest we’re not entirely sure…”

While all that local level data should be usable in theory, seemingly random differences between Anthem and other Frostbite games are getting in the way of loading the data in practice. Anthem acts like a standard Frostbite game “for the most part,” Ness said, but at times will show unusual behaviors that are hard to pin down.

“For example, when we try to load most maps, no NPCs spawn, but in some maps they do,” he said. “And we have yet to determine why. Ness has some suspicion that the odd behavior is connected to the “fairly extensive amount of player data the game keeps as part of its online RPG nature,” but adds that “to be honest we’re not entirely sure how deep the differences go, other than that the engine didn’t behave how we expected it to.”

Ness said he’s about 75 percent confident that the team will be able to figure out how to fully leverage the Frostbite engine to power a version of the game that runs without EA’s centralized servers. If that effort succeeds, he says a playable version of Anthem could be back up and running in “months, or less even, depending on motivation.” But if the efforts to pick apart Anthem’s take on Frostbite hits a brick wall, Ness says “the amount of work increases fairly exponentially and I’m a lot less confident that we have the motivation for that.”

“I’m fairly confident that we can get this game to be playable again, like how it is supposed to be,” Andersson said. “It’ll just take time as most of us have our own life to manage besides this.”

Engaging in some expectations management on the Fort’s Forge Discord. Credit: Laurie / The Fort’s Forge

In the meantime, Laurie is still trying to manage expectations set by the somewhat premature posting of Andersson’s proof-of-concept video. “Please, do not expect frequent updates,” Laurie wrote in the Fort’s Forge Discord. “We had not anticipated releasing anything this early, nor should the expedience of this video’s release serve as any kind of benchmark for how fast we make progress.”

Laurie also took to Reddit to publicly call the video “a really hacky thing so I want to ask people to manage their expectations just a bit. A lot of stuff clearly doesn’t work as ‘intended,’ and definitely needs at minimum, more polish.”

At one point last week, Laurie says they had to stop accepting new members to the Fort’s Forge Discord, “mostly to prevent an influx of people in response to… news coverage.” And while people with Frostbite engine modding experience are encouraged to reach out, the small team is being cautious about growing too large, too fast.

“We’re a little reluctant to add developers right now as we have no real code base to work from,” Ness said, describing their current efforts as “scratch work” maintained in separate forms by multiple people. “But once we firm that up (hopefully in the next weeks), we will look to add more [coders].”

Kyle Orland has been the Senior Gaming Editor at Ars Technica since 2012, writing primarily about the business, tech, and culture behind video games. He has journalism and computer science degrees from University of Maryland. He once wrote a whole book about Minesweeper.

Why reviving the shuttered Anthem is turning out tougher than expected Read More »

How to get Doom running on a pair of earbuds

can it run doom, coding, Doom, gaming, hack, id, port / 9u50fv / January 26, 2026

Hard to believe the gameplay on this website is powered by a set of earbuds. Credit: DoomBuds

Squeezing the entirety of Doom onto modern earbuds wasn’t an easy task, either. The 4.2MB of game data won’t quite fit on the PineBuds’ 4MB of flash memory, for instance. That means the project needed to use a 1.7MB “squashware” build of Doom, which eliminates some animation frames and shortens some music tracks to make the game even more portable.

The earbuds also have just under 1MB of RAM, requiring the coding of a new version of the game that optimizes away many of the bits that usually fill up a full 4MB of RAM in the standard game. “Pre-generating lookup tables, making variables const, reading const variables from flash, disabling DOOM’s caching system, removing unneeded variables… it all adds up,” Sarkisan writes.

For those without their own PineBuds to test this wild idea, Sarkisan has set up an interactive Twitch stream that players can queue up to control for 45-second sessions via doombuds.com. It’s a great little break-time diversion, especially for people ready to marvel that a set of $70 earbuds can now run a game that required a $1,000-plus computer tower a few decades ago.

How to get Doom running on a pair of earbuds Read More »

TR-49 is interactive fiction for fans of deep research rabbit holes

computer, gaming, inkle, interactive fiction, narrative, narrative deduction, research, steampunk / Mike M. / January 23, 2026

If you’re not comfortable staring at a screen like this for hours, you’d better stop reading right now. Credit: Inkle

While the catalog contains short excerpts from each of these discovered works, it’s the additional notes added to each entry by subsequent researchers that place each title in its full context. You’ll end up poring through these research notes for clues about the existence and chronology of other authors and works. Picking out specific names and years points to the codes and titles needed to unlock even more reference pages in the computer, pushing you further down the rabbit hole. Picture something like Her Story, but replace the cinéma vérité surveillance video clips with a library card catalog.

You’ll slowly start to unravel and understand how the game world’s myriad authors are influencing each other with their cross-pollinating writings. The treatises, novels, pamphlets, and journals discussed in this database are full of academic sniping, intellectual intrigue, and interpersonal co-mingling across multiple generations of work. It all ends up circling a long-running search for a metaphysical key to life itself, which most of the authors manage to approach but never quite reach a full understanding.

Matching titles to reference codes form the most “gamey” part of the game. Credit: Inkle

As you explore, you also start to learn more about the personal affairs of the researchers who collected and cataloged all this reference material and the vaguely defined temporal capabilities of the information-synthesis engine in the computer you’ve all worked on. Eventually, you’ll stumble on the existence of core commands that can unlock hidden parts of the computer or alter the massive research database itself, which becomes key to your eventual final goal.

Through it all, there’s a slowly unfolding parallel narrative involving Liam, the unseen voice guiding you through the research process itself. Through occasional voice clips, Liam eventually hints at the existence of a powerful and quickly encroaching threat that wants to stop your progress by any means necessary, adding a bit of dramatic tension to your academic pursuits.

TR-49 is interactive fiction for fans of deep research rabbit holes Read More »

Finally, a new controller that solves the Switch 2’s “flat Joy-Con” problem

controller, drift, gaming, hall effect, joy con, Nintendo, nyxi, Switch, switch 2 / Shannon Garcia / January 22, 2026

When I reviewed the Switch 2 back in June, I noted that the lack of any sort of extended grip on the extremely thin Joy-Con 2 controllers made them relatively awkward to hold, both when connected to the system and when cradled in separate hands. At the time, I said that “my Switch 2 will probably need something like the Nyxi Hyperion Pro, which I’ve come to rely on to make portable play on the original Switch much more comfortable.”

Over half a year later, Nyxi is once again addressing my Switch controller-related comfort concerns with the Hyperion 3, which was made available for preorder earlier this week ahead of planned March 1 shipments. Unfortunately, it looks like players will have to pay a relatively high price for a potentially more ergonomic Switch 2 experience.

While there are plenty of third-party controllers for the Switch 2, none of the current options mimic the official Joy-Cons’ ability to connect magnetically to the console tablet itself (controllers designed to slide into the grooves on the original Switch tablet also can’t hook to the successor console). The Hyperion 3 is the first Switch 2 controller to offer this magnetic connection, making it uniquely suited for handheld play on the system.

And although I haven’t held the Hyperion 3 in my hands yet, my experience with the similar Hyperion 2 on the original Switch suggests that the ergonomic design here will be a welcome upgrade from the relatively small buttons and cramp-inducing flat back of the official Switch 2 Joy-Cons (“Say Goodbye to Tendonitis,” as Nyxi claims in its marketing materials). The controller can also connect wirelessly via Bluetooth 5.0 for when you want to switch to docked play, unlike some Switch Joy-Con replacements that only work in portable mode.

Finally, a new controller that solves the Switch 2’s “flat Joy-Con” problem Read More »

Why adding modern controls to 1996’s Tomb Raider simply doesn’t work

C:ArsGames, dos gaming, gaming, gog, PC Gaming, platformer, shopping, tank controls, Tomb Raider / 9u50fv / January 21, 2026

For our C:ArsGames series, we look at the controls conundrum of early 3D.

The graphical updates to Tomb Raider are modest but effective. Credit: Aspyr

For a lot of the games I’ve written about in the C:ArsGames series, I’ve come to the conclusion that the games hold up pretty well, despite their age—Master of Orion II, Jill of the Jungle, and Wing Commander Privateer, for example. Each of those have flaws that show now more than ever, but I still had a blast revisiting each of them.

This time I’d like to write about one that I think doesn’t hold up quite as well for me: For the first time in almost 30 years, I revisited the original Tomb Raider via 2024’s Tomb Raider I-III Remastered collection.

You might be thinking this is going to be a dunk on the work done on the remaster, but that’s not the case, because the core issue with playing 1996’s Tomb Raider in 2026 is actually unsolvable, no matter how much care is put into a remaster.

The age of tank controls

Tomb Raider was part of the first wave of multiplatform games with fully 3D gameplay, releasing the same year as similarly groundbreaking 3D titles Super Mario 64 and Quake. I think you could make a pretty compelling case that most of the modern AAA games industry can trace its lineage in some way back to those three titles.

Because it was the beginning of mass-market 3D games (yes, I know other, more niche 3D games existed before), there were no established best practices for things like the controls or the camera.

Tomb Raider opted for a modality that was common for a few years before it was replaced by clearly better solutions: what we now call “tank controls,” where forward or back moves the character forward or back, but hitting left or right turns the character on its axis in place without moving.

The way it works is naturally intuitive enough, which is part of why it was so popular early on. But the industry has moved on because it’s frustratingly sluggish and clunky. I loved Tomb Raider‘s level design and atmosphere, and the designers did about as good a job as they could designing around the limitations of the controls for most of the combat sequences. But ultimately, there was enough combat that the sluggishness of this input method significantly detracted from my enjoyment.

In 1996, I had little to compare it to, and the novelty of these vertically stacked 3D levels played from a third-person perspective was powerful enough that I had no complaints. But after 30 years of new ideas and iteration, the industry’s designers have solved all the problems this game has with controls.

That’s why the studio behind the remaster tried including an alternative modern control scheme. Unfortunately, that doesn’t work for Tomb Raider at all.

Prince of Persia and grids

When work started on the original Tomb Raider, its developers are said to have had a specific cocktail of influences in mind: They wanted to combine the truly 3D navigable environments they had seen in the groundbreaking Ultima Underworld and the polygonal characters from Virtua Fighter, with gameplay inspired by the 1989 Jordan Mechner classic Prince of Persia.

If you’ve played Prince of Persia, you know the platforming in that game is both precise and challenging. To make jumps, you had to carefully position yourself before launching—one step forward, one step back, until you reached the perfect starting point.

The same goes for Tomb Raider. In fact, the entire game—all the puzzles, layouts, and platforming challenges—adheres to a strict grid system. Players can predict exactly how far protagonist Lara Croft will jump based on where they are on that grid. They can count steps to position themselves, and it’s basically required if you want to consistently navigate the game’s complex and precise jumping sequences without frustration.

Using the game’s original tank controls, you could step forward or backward in predictable ways, or side step, jump to the side, jump forward, jump backward, and so on, with specific numbers of presses on the arrow keys. The entire game was built around this principle.

As frustrating as tank controls are to a modern player, there was an exquisite elegance to this.

The remaster’s modern controls option works more like Tomb Raider Legends from the 2000s, and it’s that general approach that has become standard in almost all modern third-person 3D games.

They feel so much nicer and more responsive to a modern player who has been trained on that for the past two decades, even if that player is someone like me who did play the original games with tank controls back in the day. That short window of three to five years of muscle memory and comfort based on tank controls has been completely overwritten by more than 20 years with what the modern control scheme offers.

Unfortunately, the flexible modern controls lose almost all connection to that elegant grid system. What used to be a precise process—for example, “X steps forward, X steps to the left, then a backflip from exactly this spot”—is now a guessing game of feeling things out. And the platforming sequences aren’t designed with that in mind. As a result, the combat feels a lot better with modern controls, but just about everything else is much more frustrating than before.

Embracing Tomb Raider

I’m not the first to observe this about the remaster; reviewers and Reddit dwellers debated this at length when this release happened two years ago. But I hadn’t gotten to playing the remasters—or revisiting Tomb Raider at all since the ’90s—until I decided to try it out for C:ArsGames.

Tomb Raider is still worth revisiting, but it is frustrating to leave behind 20 years of muscle memory to return to a previous paradigm that ended up being an evolutionary dead end.

The more time you put into it, the more natural the tank controls feel, but without the wow factor of groundbreaking new 3D gameplay, it’s harder to put up with.

Tellingly, Tomb Raider has already gotten a complete remake (distinct from this remaster) once, and another one is coming. Both radically reinvent the gameplay and seem to turn away from the grid system that made the original what it was. Many modern players won’t put up with the tank controls, but not being willing to embrace those means you simply can’t experience Tomb Raider as it was originally intended.

And again, I’m not knocking the work done on this remaster. Fittingly, it was made by Aspyr, the same studio that ported the original games to the Mac in the ’90s. (For a few years, they absolutely dominated the Mac game market with their Windows-to-Mac ports.) They’re still porting games to Mac, Linux, iOS, and Android today—notably, they did all the Civilization VI ports—as well as remasters of classics for modern platforms.

There’s no version of the modern controls that would truly work from this game, so it’s not an execution issue, and I actually think that Tomb Raider I-III Remastered is possibly Aspyr’s most well-crafted work.

The remaster includes the ability to flip between classic graphics and a more contemporary look that I think does a great job of walking the line between honoring the ’90s original and looking nice to 2020s eyes. They even hired Timur “XProger” Gagiev, a developer known for work on Tomb Raider open source engine OpenLara, to be the remaster’s technical director.

The Tomb Raider franchise is about to enter a new era (controversially) under Embracer Group and Amazon Games; it remains to be seen whether it will be a good one. But if you want to go back to where it all started, I recommend grabbing this remaster (available on GOG and other storefronts, as well as on consoles) instead of playing the original release. Just stick with the tank controls, and I hope you adapt back to them more easily than I did!

Ars Technica may earn compensation for sales from links on this post through affiliate programs.

Samuel Axon is the editorial lead for tech and gaming coverage at Ars Technica. He covers AI, software development, gaming, entertainment, and mixed reality. He has been writing about gaming and technology for nearly two decades at Engadget, PC World, Mashable, Vice, Polygon, Wired, and others. He previously ran a marketing and PR agency in the gaming industry, led editorial for the TV network CBS, and worked on social media marketing strategy for Samsung Mobile at the creative agency SPCSHP. He also is an independent software and game developer for iOS, Windows, and other platforms, and he is a graduate of DePaul University, where he studied interactive media and software development.

Why adding modern controls to 1996’s Tomb Raider simply doesn’t work Read More »

The first new Marathon game in decades will launch on March 5

Bungie, delay, destiny, Development, extraction shooter, gaming, Halo, marathon, ps5, release date, steam, xbox series / Shannon Garcia / January 19, 2026

It’s been nearly three years now since Destiny maker (and Sony subsidiary) Bungie formally announced a revival of the storied Marathon FPS franchise. And it has been about seven months since the game’s original announced release date of September 23, 2025 was pushed back indefinitely after a reportedly poor response to the game’s first Alpha test.

But today, in a post on the PlayStation Blog, Bungie revealed that the new Marathon would finally be hitting PS5, Windows, and Xbox Series X|S on March 5, narrowing down the month-long March release window announced back in December.

Today’s pre-rder trailer revealing the Marathon release date.

Unlike Destiny 2, which transitioned to a free-to-play model in 2019, the new Marathon sells for $40 in a Standard Edition or a $60 Deluxe Edition that includes some digital rewards and cosmetics. That mirrors the pricing of the somewhat similar Arc Raiders, which recently hit 12 million sales in less than 12 weeks.

A new kind of Marathon

Unlike the original Marathon trilogy on the ’90s Macintosh—which closely followed on the single-player campaign corridors and deathmatch multiplayer of the original Doom—the new Marathon is described as a “PvPvE survival extraction shooter.” That means gameplay based around exploring distinct zones and scavenging for cosmetics and gear upgrades in exploratory missions alone or with up to two friends, then seeing those missions “break into fast-paced PvP combat” at a moment’s notice, according to the game’s official description.

The first new Marathon game in decades will launch on March 5 Read More »

Reports of ad-supported Xbox game streams show Microsoft’s lack of imagination

cloud gaming, game streaming, gaming, streaming, xboxmicrosoft / 9u50fv / January 19, 2026

You can do better than that

That’s a moderately useful option for cloud-curious Xbox players that might not be willing to take the plunge on a monthly subscription, we suppose. But it also feels like Microsoft could come up with some more imaginative ways to use Cloud Gaming to reach occasional players in new ways.

What’s stopping Microsoft from offer streaming players a 30-minute timed demo stream of any available Xbox Cloud Gaming title—perhaps in exchange for watching a short ad, or perhaps simply as an Xbox Live Arcade-style sales juicing tactic? Or why not offer discounted access to a streaming-only Game Pass subscription for players willing to watch occasional ads, like Netflix? Microsoft could even let players spend a couple of bucks to rent a digital copy of the title for a few days, much as services like iTunes do for newer films.

Those are just a few ideas off the top of our heads. And they all feel potentially more impactful than using ads as a way to let Xbox players stream copies of games they already purchased.

Back in 2019, we noted how Stadia’s strictly buy-before-you-play streaming business model limited the appeal of what ended up as a doomed cloud-gaming experiment. Microsoft should take some lessons from Google’s failure and experiment with new ways to use streaming to reach players that might not have access to the latest high-end hardware for their gaming experiences.

Reports of ad-supported Xbox game streams show Microsoft’s lack of imagination Read More »

10 things I learned from burning myself out with AI coding agents

AI, AI coding, AI coding tools, Biz & IT, Claude Code, Claude Opus, coding agent, Features, gaming, large language models, Opus 4.5, vibe coding / 9u50fv / January 19, 2026

Opinion: As software power tools, AI agents may make people busier than ever before.

Credit: Aurich Lawson | Getty Images

If you’ve ever used a 3D printer, you may recall the wondrous feeling when you first printed something you could have never sculpted or built yourself. Download a model file, load some plastic filament, push a button, and almost like magic, a three-dimensional object appears. But the result isn’t polished and ready for mass production, and creating a novel shape requires more skills than just pushing a button. Interestingly, today’s AI coding agents feel much the same way.

Since November, I have used Claude Code and Claude Opus 4.5 through a personal Claude Max account to extensively experiment with AI-assisted software development (I have also used OpenAI’s Codex in a similar way, though not as frequently). Fifty projects later, I’ll be frank: I have not had this much fun with a computer since I learned BASIC on my Apple II Plus when I was 9 years old. This opinion comes not as an endorsement but as personal experience: I voluntarily undertook this project, and I paid out of pocket for both OpenAI and Anthropic’s premium AI plans.

Throughout my life, I have dabbled in programming as a utilitarian coder, writing small tools or scripts when needed. In my web development career, I wrote some small tools from scratch, but I primarily modified other people’s code for my needs. Since 1990, I’ve programmed in BASIC, C, Visual Basic, PHP, ASP, Perl, Python, Ruby, MUSHcode, and some others. I am not an expert in any of these languages—I learned just enough to get the job done. I have developed my own hobby games over the years using BASIC, Torque Game Engine, and Godot, so I have some idea of what makes a good architecture for a modular program that can be expanded over time.

In December, I used Claude Code to create a multiplayer online clone of Katamari Damacy called — In December, I used Claude Code to create a multiplayer online clone of *Katamari Damacy* called “Christmas Roll-Up.” Credit: Benj Edwards

Claude Code, Codex, and Google’s Gemini CLI, can seemingly perform software miracles on a small scale. They can spit out flashy prototypes of simple applications, user interfaces, and even games, but only as long as they borrow patterns from their training data. Much like a 3D printer, doing production-level work takes far more effort. Creating durable production code, managing a complex project, or crafting something truly novel still requires experience, patience, and skill beyond what today’s AI agents can provide on their own.

And yet these tools have opened a world of creative potential in software that was previously closed to me, and they feel personally empowering. Even with that impression, though, I know these are hobby projects, and the limitations of coding agents lead me to believe that veteran software developers probably shouldn’t fear losing their jobs to these tools any time soon. In fact, they may become busier than ever.

So far, I have created over 50 demo projects in the past two months, fueled in part by a bout of COVID that left me bedridden with a laptop and a generous 2x Claude usage cap that Anthropic put in place during the last few weeks of December. As I typed furiously all day, my wife kept asking me, “Who are you talking to?”

You can see a few of the more interesting results listed on my personal website. Here are 10 interesting things I’ve learned from the process.

1. People are still necessary

Even with the best AI coding agents available today, humans remain essential to the software development process. Experienced human software developers bring judgment, creativity, and domain knowledge that AI models lack. They know how to architect systems for long-term maintainability, how to balance technical debt against feature velocity, and when to push back when requirements don’t make sense.

For hobby projects like mine, I can get away with a lot of sloppiness. But for production work, having someone who understands version control, incremental backups, testing one feature at a time, and debugging complex interactions between systems makes all the difference. Knowing something about how good software development works helps a lot when guiding an AI coding agent—the tool amplifies your existing knowledge rather than replacing it.

As independent AI researcher Simon Willison wrote in a post distinguishing serious AI-assisted development from casual “vibe coding,” “AI tools amplify existing expertise. The more skills and experience you have as a software engineer the faster and better the results you can get from working with LLMs and coding agents.”

With AI assistance, you don’t have to remember how to do everything. You just need to know what you want to do.

Card Miner: Heart of the Earth is entirely human-designed by AI coded using Claude Code. It represents about a month of iterative work. — *Card Miner: Heart of the Earth* is entirely human-designed, but it was AI-coded using Claude Code. It represents about a month of iterative work. Credit: Benj Edwards

So I like to remind myself that coding agents are software tools best used to enact human ideas, not autonomous coding employees. They are not people (and not people replacements) no matter how the companies behind them might market them.

If you think about it, everything you do on a computer was once a manual process. Programming a computer like the ENIAC involved literally making physical bits (connections) with wire on a plugboard. The history of programming has been one of increasing automation, so even though this AI-assisted leap is somewhat startling, one could think of these tools as an advancement similar to the advent of high-level languages, automated compilers and debugger tools, or GUI-based IDEs. They can automate many tasks, but managing the overarching project scope still falls to the person telling the tool what to do.

And they can have rapidly compounding benefits. I’ve now used AI tools to write better tools—such as changing the source of an emulator so a coding agent can use it directly—and those improved tools are already having ripple effects. But a human must be in the loop for the best execution of my vision. This approach has kept me very busy, and contrary to some prevailing fears about people becoming dumber due to AI, I have learned many new things along the way.

2. AI models are brittle beyond their training data

Like all AI models based on the Transformer architecture, the large language models (LLMs) that underpin today’s coding agents have a significant limitation: They can only reliably apply knowledge gleaned from training data, and they have a limited ability to generalize that knowledge to novel domains not represented in that data.

What is training data? In this case, when building coding-flavored LLMs, AI companies download millions of examples of software code from sources like GitHub and use them to make the AI models. Companies later specialize them for coding through fine-tuning processes.

The ability of AI agents to use trial and error—attempting something and then trying again—helps mitigate the brittleness of LLMs somewhat. But it’s not perfect, and it can be frustrating to see a coding agent spin its wheels trying and failing at a task repeatedly, either because it doesn’t know how to do it or because it previously learned how to solve a problem but then forgot because the context window got compacted (more on that here).

Violent Checkers is a physics-based corruption of the classic board game, coded using Claude Code. — *Violent Checkers* is a physics-based corruption of the classic board game, coded using Claude Code. Credit: Benj Edwards

To get around this, it helps to have the AI model take copious notes as it goes along about how it solved certain problems so that future instances of the agent can learn from them again. You also want to set ground rules in the claude.md file that the agent reads when it begins its session.

This brittleness means that coding agents are almost frighteningly good at what they’ve been trained and fine-tuned on—modern programming languages, JavaScript, HTML, and similar well-represented technologies—and generally terrible at tasks on which they have not been deeply trained, such as 6502 Assembly or programming an Atari 800 game with authentic-looking character graphics.

It took me five minutes to make a nice HTML5 demo with Claude but a week of torturous trial and error, plus actual systematic design on my part, to make a similar demo of an Atari 800 game. To do so, I had to use Claude Code to invent several tools, like command-line emulators and MCP servers, that allow it to peek into the operation of the Atari 800’s memory and chipset to even begin to make it happen.

3. True novelty can be an uphill battle

Due to what might poetically be called “preconceived notions” baked into a coding model’s neural network (more technically, statistical semantic associations), it can be difficult to get AI agents to create truly novel things, even if you carefully spell out what you want.

For example, I spent four days trying to get Claude Code to create an Atari 800 version of my HTML game Violent Checkers, but it had trouble because in the game’s design, the squares on the checkerboard don’t matter beyond their starting positions. No matter how many times I told the agent (and made notes in my Claude project files), it would come back to trying to center the pieces to the squares, snap them within squares, or use the squares as a logical basis of the game’s calculations when they should really just form a background image.

To get around this in the Atari 800 version, I started over and told Claude that I was creating a game with a UFO (instead of a circular checker piece) flying over a field of adjacent squares—never once mentioning the words “checker,” “checkerboard,” or “checkers.” With that approach, I got the results I wanted.

A screenshot of Benj's Mac while working on a Violent Checkers port for the Atari 800 home computer, amid other projects. — A screenshot of Benj’s Mac while working on a *Violent Checkers* port for the Atari 800 home computer, amid other projects. Credit: Benj Edwards

Why does this matter? Because with LLMs, context is everything, and in language, context changes meaning. Take the word “bank” and add the words “river” or “central” in front of it, and see how the meaning changes. In a way, words act as addresses that unlock the semantic relationships encoded in a neural network. So if you put “checkerboard” and “game” in the context, the model’s self-attention process links up a massive web of semantic associations about how checkers games should work, and that semantic baggage throws things off.

A couple of tricks can help AI coders navigate around these limitations. First, avoid contaminating the context with irrelevant information. Second, when the agent gets stuck, try this prompt: “What information do you need that would let you implement this perfectly right now? What tools are available to you that you could use to discover that information systematically without guessing?” This forces the agent to identify (semantically link up) its own knowledge gaps, spelled out in the context window and subject to future action, instead of flailing around blindly.

4. The 90 percent problem

The first 90 percent of an AI coding project comes in fast and amazes you. The last 10 percent involves tediously filling in the details through back-and-forth trial-and-error conversation with the agent. Tasks that require deeper insight or understanding than what the agent can provide still require humans to make the connections and guide it in the right direction. The limitations we discussed above can also cause your project to hit a brick wall.

From what I have observed over the years, larger LLMs can potentially make deeper contextual connections than smaller ones. They have more parameters (encoded data points), and those parameters are linked in more multidimensional ways, so they tend to have a deeper map of semantic relationships. As deep as those go, it seems that human brains still have an even deeper grasp of semantic connections and can make wild semantic jumps that LLMs tend not to.

Creativity, in this sense, may be when you jump from, say, basketball to how bubbles form in soap film and somehow make a useful connection that leads to a breakthrough. Instead, LLMs tend to follow conventional semantic paths that are more conservative and entirely guided by mapped-out relationships from the training data. That limits their creative potential unless the prompter unlocks it by guiding the LLM to make novel semantic connections. That takes skill and creativity on the part of the operator, which once again shows the role of LLMs as tools used by humans rather than independent thinking machines.

5. Feature creep becomes irresistible

While creating software with AI coding tools, the joy of experiencing novelty makes you want to keep adding interesting new features rather than fixing bugs or perfecting existing systems. And Claude (or Codex) is happy to oblige, churning away at new ideas that are easy to sketch out in a quick and pleasing demo (the 90 percent problem again) rather than polishing the code.

Flip-Lash started as a “Tetris but you can flip the board,” but feature creep made me throw in the kitchen sink, losing focus. Credit: Benj Edwards

Fixing bugs can also create bugs elsewhere. This is not new to coding agents—it’s a time-honored problem in software development. But agents supercharge this phenomenon because they can barrel through your code and make sweeping changes in pursuit of narrow-minded goals that affect lots of working systems. We’ve already talked about the importance of having a good architecture guided by the human mind behind the wheel above, and that comes into play here.

6. AGI is not here yet

Given the limitations I’ve described above, it’s very clear that an AI model with general intelligence—what people usually call artificial general intelligence (AGI)—is still not here. AGI would hypothetically be able to navigate around baked-in stereotype associations and not have to rely on explicit training or fine-tuning on many examples to get things right. AI companies will probably need a different architecture in the future.

I’m speculating, but AGI would likely need to learn permanently on the fly—as in modify its own neural network weights—instead of relying on what is called “in-context learning,” which only persists until the context fills up and gets compacted or wiped out.

Grapheeti is a “drawing MMO” where people around the world share a canvas. Credit: Benj Edwards

In other words, you could teach a true AGI system how to do something by explanation or let it learn by doing, noting successes, and having those lessons permanently stick, no matter what is in the context window. Today’s coding agents can’t do that—they forget lessons from earlier in a long session or between sessions unless you manually document everything for them. My favorite trick is instructing them to write a long, detailed report on what happened when a bug is fixed. That way, you can point to the hard-earned solution the next time the amnestic AI model makes the same mistake.

7. Even fast isn’t fast enough

While using Claude Code for a while, it’s easy to take for granted that you suddenly have the power to create software without knowing certain programming languages. This is amazing at first, but you can quickly become frustrated that what is conventionally a very fast development process isn’t fast enough. Impatience at the coding machine sets in, and you start wanting more.

But even if you do know the programming languages being used, you don’t get a free pass. You still need to make key decisions about how the project will unfold. And when the agent gets stuck or makes a mess of things, your programming knowledge becomes essential for diagnosing what went wrong and steering it back on course.

8. People may become busier than ever

After guiding way too many hobby projects through Claude Code over the past two months, I’m starting to think that most people won’t become unemployed due to AI—they will become busier than ever. Power tools allow more work to be done in less time, and the economy will demand more productivity to match.

It’s almost too easy to make new software, in fact, and that can be exhausting. One project idea would lead to another, and I was soon spending eight hours a day during my winter vacation shepherding about 15 Claude Code projects at once. That’s too much split attention for good results, but the novelty of seeing my ideas come to life was addictive. In addition to the game ideas I’ve mentioned here, I made tools that scrape and search my past articles, a graphical MUD based on ZZT, a new type of MUSH (text game) that uses AI-generated rooms, a new type of Telnet display proxy, and a Claude Code client for the Apple II (more on that soon). I also put two AI-enabled emulators for Apple II and Atari 800 on GitHub. Phew.

Consider the advent of the steam shovel, which allowed humans to dig holes faster than a team using hand shovels. It made existing projects faster and new projects possible. But think about the human operator of the steam shovel. Suddenly, we had a tireless tool that could work 24 hours a day if fueled up and maintained properly, while the human piloting it would need to eat, sleep, and rest.

I used Claude Code to create a windowing GUI simulation of the Mac that works over Telnet. Credit: Benj Edwards

In fact, we may end up needing new protections for human knowledge workers using these tireless information engines to implement their ideas, much as unions rose as a response to industrial production lines over 100 years ago. Humans need rest, even when machines don’t.

Will an AI system ever replace the human role here? Even if AI coding agents could eventually work fully autonomously, I don’t think they’ll replace humans entirely because there will still be people who want to get things done, and new AI power tools will emerge to help them do it.

9. Fast is scary to people

AI coding tools can turn what was once a year-long personal project into a five-minute session. I fed Claude Code a photo of a two-player Tetris game I sketched in a notebook back in 2008, and it produced a working prototype in minutes (prompt: “create a fully-featured web game with sound effects based on this diagram”). That’s wild, and even though the results are imperfect, it’s a bit frightening to comprehend what kind of sea change in software development this might entail.

Since early December, I’ve been posting some of my more amusing experimental AI-coded projects to Bluesky for people to try out, but I discovered I needed to deliberately slow down with updates because they came too fast for people to absorb (and too fast for me to fully test). I’ve also received comments like “I’m worried you’re using AI, you’re making games too fast” and so on.

Benj's handwritten game design note about a two-player Tetris concept from 2007. — Benj’s handwritten game design note about a two-player *Tetris* concept from 2007. Credit: Benj Edwards

Regardless of my own habits, the flow of new software will not slow down. There will soon be a seemingly endless supply of AI-augmented media (games, movies, images, books), and that’s a problem we’ll have to figure out how to deal with. These products won’t all be “AI slop,” either; some will be done very well, and the acceleration in production times due to these new power tools will balloon the quantity beyond anything we’ve seen.

Social media tends to prime people to believe that AI is all good or all bad, but that kind of black-and-white thinking may be the easy way out. You’ll have no cognitive dissonance, but you’ll miss a far richer third option: seeing these tools as imperfect and deserving of critique but also as useful and empowering when they bring your ideas to life.

AI agents should be considered tools, not entities or employees, and they should be amplifiers of human ideas. My game-in-progress Card Miner is entirely my own high-level creative design work, but the AI model handled the low-level code. I am still proud of it as an expression of my personal ideas, and it would not exist without AI coding agents.

10. These tools aren’t going away

For now, at least, coding agents remain very much tools in the hands of people who want to build things. The question is whether humans will learn to wield these new tools effectively to empower themselves. Based on two months of intensive experimentation, I’d say the answer is a qualified yes, with plenty of caveats.

We also have social issues to face: Professional developers already use these tools, and with the prevailing stigma against AI tools in some online communities, many software developers and the platforms that host their work will face difficult decisions.

Ultimately, I don’t think AI tools will make human software designers obsolete. Instead, they may well help those designers become more capable. This isn’t new, of course; tools of every kind have been serving this role since long before the dawn of recorded history. The best tools amplify human capability while keeping a person behind the wheel. The 3D printer analogy holds: amazing fast results are possible, but mastery still takes time, skill, and a lot of patience with the machine.

Benj Edwards is Ars Technica’s Senior AI Reporter and founder of the site’s dedicated AI beat in 2022. He’s also a tech historian with almost two decades of experience. In his free time, he writes and records music, collects vintage computers, and enjoys nature. He lives in Raleigh, NC.

10 things I learned from burning myself out with AI coding agents Read More »

Meta’s layoffs leave Supernatural fitness users in mourning

gaming, Meta, meta vr, syndication, Tech, VR headsets / DJ Henderson / January 17, 2026

There is a split in the community about who will stay and continue to pay the subscription fee and who will leave. Supernatural has more than 3,000 lessons available in the service, so while new content won’t be added, some feel there is plenty of content left in the library. Other users worry about how Supernatural will continue to license music from big-name bands.

“Supernatural is amazing, but I am canceling it because of this,” Chip told me. “The library is large, so there’s enough to keep you busy, but not for the same price.”

There are other VR workout experiences like FitXR or even the VR staple Beat Saber, which Supernatural cribs a lot of design concepts from. Still, they don’t hit the same bar for many of the Supernatural faithful.

“I’m going to stick it out until they turn the lights out on us,” says Stefanie Wong, a Bay Area accountant who has used Supernatural since shortly after the pandemic and has organized and attended meetup events. “It’s not the app. It’s the community, and it’s the coaches that we really, really care about.”

Welcome to the new age

I tried out Supernatural’s Together feature on Wednesday, the day after the layoffs. It’s where I met Chip and Alisa. When we could stop to catch our breath, we talked about the changes coming to the service. They had played through previous sessions hosted by Jane Fonda or playlists with a mix of music that would change regularly. This one was an artist series featuring entirely Imagine Dragons songs.

In the session, as we punched blocks while being serenaded by this shirtless dude crooning, recorded narrations from Supernatural coach Dwana Olsen chimed in to hype us up.

“Take advantage of these moments,” Olsen said as we punched away. “Use these movements to remind you of how much awesome life you have yet to live.”

Frankly, it was downright invigorating. And bittersweet. We ended another round, sweaty, huffing and puffing. Chip, Alisa, and I high-fived like crazy and readied for another round.

“Beautiful,” Alisa said. “It’s just beautiful, isn’t it?”

Meta’s layoffs leave Supernatural fitness users in mourning Read More »