Geoffrey Bunting | The Verge

Oeuf is a punishing platformer in a cozy shell

2026-03-19T13:01:58-04:00

The funny shape of eggs is the curious lifeblood of Oeuf, the new physics platformer by prolific developer Increpare Games. In a gaming landscape saturated with complex systems dropped into simple games, that grapples with metaphor within straightforward narratives, and that is desperate to bring cinematic sensibilities into gaming, Oeuf only asks that you briefly consider how an egg might move as you roll, slide, and hop across its world.

That world is realized in crunchy, ’90s-era 3D that brings to mind Ultima and Might and Magic. Like this archaic-seeming style — that Oeuf was released within a month of Resident Evil Requiem is a fun graphical comparison — Oeuf is refreshingly simple. Blown from the nest atop a church steeple, you, a brown-speckled egg, must navigate church grounds, climb up trees, cross sloping roofs, and clamber over jutting bricks to get home.

Your nameless egg is unpredictable. Smooth but prone to being overzealous in its haste when on its side; less controllable but more balanced on its end. Much of the challenge in Oeuf comes from positioning yourself, through tiny movements on similarly tiny platforms, to build momentum for a jump and arresting that motion before you tumble to the ground on the other side. When you fall — and you will fall — the game greets you with an “oof.” Being an egg, however, this is rendered as a pleasing “oeuf!”

If this is all sounding too Getting Over It with Bennett Foddy, Oeuf is generous with checkpoints that remove much of the grinding irritation of similar games while maintaining the surmountable frustration that makes them satisfying. Those checkpoints create a comfortable rhythm.

There is a temptation in games built on repetition and accompanying frustration to keep mounting the difficulty until challenge turns to drudgery. In mixing up the length of sequences with a clever weaving of shorter sections between longer ones, Oeuf creates a rewarding sense of forward motion with a pulsing mix of punishing and more relaxed sequences. Players breaking through a tense run of incremental platforming are often greeted by several shorter areas based on different mechanics like completing impossible jumps through building up speed on slopes.

Mileage on that may vary, likely based on how you relate to Oeuf’s 3D world. I preferred navigating lateral crossings and challenging climbs; others may gravitate toward bouncing off obstacle courses and puzzles governed by player momentum. One of the most challenging sections I encountered came early, as I navigated my egg over a series of descending ramps, while I had little trouble with climbing a tree soon after. For others, it may well be the inverse.

Oeuf understands that players can only get so frustrated before they stop playing. In fact, its choreographed rhythm appears to encourage that. Checkpoints already delineate clear opportunities to start and stop, but a generous attitude to maintaining player progress means stepping back in response to a frustrating section is often the recipe for coming back soon after and blitzing a sequence on which you’d previously been stuck. The joy of overcoming is only increased as you race through the following sections — sometimes literally, particularly when tasked with navigating slopes. It’s a level of technical self-awareness that eludes so many similar games.

That said, Oeuf is at its best when it’s asking you to jump. Some of its momentum-based puzzles can be finicky, particularly when the way the egg moves and bounces renders some jump inputs unresponsive. This appears to be a side effect of how the egg lifts off the ground when bouncing on its end, rather than a programming error.

Not every section enjoys the same generous checkpoints. One particularly provoking area that sees you traverse a series of sloping surfaces left me stumped across multiple sessions. Not for a lack of solutions — the only option was to balance on the corners of blocks — but for how unforgiving it was. When I finally made it through, I was surprised not to find a checkpoint waiting in line with the rhythm that both preceded and followed the section. Instead, the eventual checkpoint came after a lengthy series of jumps. This was the only discordant moment I perceived in Oeuf’s otherwise comfortable rhythm.

You can, however, skip areas using the built-in map editor — in which you can also build your own courses — or you can edit frustrating sections into something more forgiving to avoid actual bottlenecks. You may even, in stepping away from a frustrating section, unwind on one of the custom maps already included. Intended or not, the function offers an interesting accessibility solution to break out of any loop in which you find yourself.

Not that those are going to be common for most players. Oeuf is, paradoxically, a chill game. You may grind your teeth in places, squeeze the controller too hard, but Oeuf is trying — not always successfully — to get you to decompress. In the mellow soundtrack, the natural tones and sound of the wind, Oeuf gives you every chance of entering a relaxed state of pure concentration. Whether you can or not likely depends on how you react to being perched precariously on the edge and desperately twitching the joystick to keep your bottom-heavy egg from tipping into the abyss.

Still, it’s hard to get too frustrated with Oeuf. With quick restarts, surmountable obstacles, and generous checkpoints, the relationship with failure it fosters is a comfortable one. A popular framework in video game reviews is to find meaning in a game outside the obvious. Games secretly represent moments in time, capture bygone feelings, or help you discover something about yourself. Oeuf really is just a game about being an egg and hopping up a series of platforms. It’s just fun — really, really fun.

_{Oeuf is available now on Steam.}

How Last Samurai Standing adds kinetic action to the Battle Royale formula

2025-12-22T08:21:24-05:00

Last Samurai Standing begins with a familiar premise. Desperate samurai dispossessed by the restoration of the emperor enter into a deadly game for a life-changing cash prize — all for the entertainment of anonymous elites. Unlike its inspirations Battle Royale and Squid Game, however, Last Samurai Standing’s violence is chaotic, fast-paced, and kinetic, though it hides a careful choreography that makes the series a more electric proposition than its predecessors.

Viewers have Junichi Okada to thank for that. As well as starring in and producing Last Samurai Standing, he serves as the series’ action planner. Many will be familiar with the results of an action planner’s work — sometimes called an action director, elsewhere a “coordinator,” and even “choreographer” — though perhaps not what the role entails. In the case of Last Samurai Standing, it’s a role that touches on nearly every aspect of the production, from the story to the action itself.

“I was involved from the script stage, thinking about what kind of action we wanted and how we would present it in the context of this story,” Okada tells The Verge. “If the director [Michihito Fujii] said, ‘I want to shoot this kind of battle scene,’ I would then think through the content and concept, design the scene, and ultimately translate that into script pages.”

The close relationship between the writer and director extends to other departments, too. Though an action planner’s role starts with managing fight scenes and stunt performers, they also liaise with camera, wardrobe, makeup, and even editorial departments to ensure fight scenes cohere with the rest of the production.

It’s a role which might appear a natural progression for Okada, who is certified to teach Kali and Jeet Kune Do — a martial art conceived by Bruce Lee — and holds multiple black belts in jiujitsu. Though the roots of his progression into action planning can be traced back further, to 1995 when he became the youngest member of J-pop group V6.

“Dance experience connects directly to creating action,” he says. “[In both] rhythm and control of the body are extremely important.” Joining V6 at the age of 15, that experience has made Okada conscious of how he moves in relation to a camera during choreography, how he is seen within the structure of a shot, and, critical to action planning, how to navigate all of that safely from a young age.

That J-pop stardom also offered avenues into acting, initially in roles you might expect for a young pop star: comic heartthrobs and sitcom sons. But he was steadily able to broaden his output. A starring turn in Hirokazu Kore-eda’s Hana followed, as did voice acting in Studio Ghibli’s Tales From Earthsea and From Up on Poppy Hill. A more telling departure was a starring role in 2007’s SP, in which he played a rookie in a police bodyguard unit, for which he trained for several years under shootfighting instructor Yorinaga Nakamura.

“What I care about is whether audiences feel that ‘this man really lives here as a samurai.’”

In the years since, Okada has cemented himself as one of Japan’s most recognizable actors, hopping between action starring roles in The Fable to sweeping period epics like Sekigahara. Those two genres converge in his Last Samurai Standing role of Shujiro, a former Shogunate samurai now reduced to poverty, working through his PTSD and reckoning with his bloodthirsty past in the game. These days, it’s less of a concern that the character butts up against his past idol image, he suggests. “What I care about is whether audiences feel that ‘this man really lives here as a samurai.’”

For Okada’s work on Last Samurai Standing, as both producer and action planner, that involved lacing high-octane but believable action with the respect for history and character studies of the period dramas he loves. “Rather than being 100 percent faithful to historical accuracy,” he adds, “my goal was to focus on entertainment and story, while letting the ‘DNA’ and beauty of Japanese period drama gently float up in the background.”

A focus on what he defines as “‘dō’ — movement,” pure entertainment that “never lets the audience get bored” punctuated — with “‘ma,’” the active emptiness that connects those frenetic moments. Both can be conversations, even if one uses words and another communicates dialogue through sword blows. This is most apparent when Shujiro faces his former comrade Sakura (Yasushi Fuchikami) inside a claustrophobic bank vault that serves as a charnel house for the game’s less fortunate contestants.

“The whole battle is divided into three sequences,” Okada says. The first starts with a moment of almost perfect stillness, a deep breath, before the two launch into battle. “A fight where pride and mutual respect collide,” he says, “and where the speed of the techniques reaches a level that really surprises the audience.” It’s all captured in one, zooming take with fast, tightly choreographed action reminiscent of Donnie Yen and Wu Jing in Kill Zone.

So intense is their duel that both shatter multiple swords. The next phase sees them lash out in a more desperate and brutal manner with whatever weapons they find. Finally, having fought to a weary stalemate, the fight becomes, Okada concludes, “a kind of duel where their stubbornness and will are fully exposed” as they hack at each other with shattered blades and spear fragments.

It’s a rhythm that many fights in Last Samurai Standing follow, driven by a string of physical and emotional considerations that form the basis of an action planner’s tool kit: how and why someone fights based on who they are and their environment. Here it is two former samurai in an elegant and terrifyingly fast-paced duel. Elsewhere we see skill matched against brutality, or inexperience against expertise.

“I define a clear concept for each sequence,” Okada says, before he opens those concepts up to the broader team. From there, he might add notes, but in Last Samurai Standing, action is a collaborative affair. “We keep refining,” he says. “It’s a back-and-forth process of shaping the sequence using both the ideas the team brings and the choreography I create myself.”

There is a third factor which Okada believes is the series’ most defining. “If we get to continue the story,” he says, “I’d love to explore how much more we can lean into ‘sei’ — stillness, and bring in even more of a classical period drama feel.”

As much of a triumph of action as Last Samurai Standing is, its quietest moments are the ones that stay with you. The charged looks between Shujiro and Iroha (Kaya Kiyohara) or their shuddering fright when confronted with specters of their past. Most of all, Shujiro watching his young ward, Futaba Katsuki (Yumia Fujisaki), dance before a waterlogged torii as mist hovers. These pauses are what elevate and invigorate the breathless action above spectacle.

The pauses are also emblematic of the balance that Last Samurai Standing strikes between its period setting and pushing the boundaries of action, all to inject new excitement into the genre. “Japan is a country that values tradition and everything it has built up over time. That’s why moments where you try to update things are always difficult,” Okada says. “But right now, we’re in the middle of that transformation.”

That is an evolution that Okada hopes to support through his work, both in front of and behind the camera. If he can create avenues for new generations of talent to carry Japanese media to a broader audience and his team to achieve greater success on a global stage, “that would make me very happy,” he says. “I want to keep doing whatever I can to help make that possible.”

_{The first season of Last Samurai Standing is streaming on Netflix now, and a second season was just confirmed.}

Demonschool channels Buffy and Persona for delightfully demonic RPG action

2025-11-19T08:30:11-05:00

In one particularly memorable sequence in Demonschool — the new tactical RPG from Necrosoft that, after a few hiccups, releases on November 19th — I throw a doll at a zombie’s head, dispatch it with a roundhouse kick to the jowls as I watch my friends sweep demons off the floor in vortices of blood, and celebrate by returning to town and petting a dog. All in the name of passing a college assignment that I’m juggling alongside deciphering a millennia-old apocalyptic prophecy and making new friends.

For Demonschool’s demon-hunting protagonist, Faye, that would all preferably be achieved by kicking the apocalypse, and coursework, into submission across 10 in-game weeks of reference-laden gore.

Necrosoft isn’t shy about those inspirations, either. Launching from a setup that screams Buffy the Vampire Slayer, a color palette ripped from Dario Argento’s Suspiria, and a smörgåsbord of horror — your first in-game quest is to track down a videotape that kills you within three days of viewing — Demonschool injects its own philosophy into what it draws from what’s come before. If “take your time” in Persona 5, another of Demonschool’s many influences, felt more like a threat in a 100-plus-hour game, Demonschool’s deployment of the phrase is more sincere. There’s no time pressure to its main quest and, but for a few rhythmic missteps, the time you spend in Demonschool is yours to do with as you will.

That time will be split between two things. The first: exploring an island and engaging with its increasingly eerie residents — including a child whose ambition is to grow up to be a bench — and the comic interludes you and your growing band of new pals consistently wind up in. Upon arriving on the island, for instance, Faye drags reluctant new best friend Namako to explore, only to end up in a fight with local thugs (a hobby of Faye’s we’ll revisit shortly). “We’re hitting the town,” she tells Namako. “Pretend those guys are the town!”

The rest will be spent in Demonschool’s rendition of grid- and turn-based combat, with a focus on positional gameplay, where you’ll be maneuvering characters behind enemies. Faye’s basic attack pushes enemies back a square — enemy attacks can similarly influence your position — while Namako can phase through demons and drag them toward her original position. Drive enemies between characters and you can unleash a powerful combo; do enough damage and you can access characters’ special moves. Tactical combat becomes a string of considerations of how to maneuver enemies into your line of attack and stay out of range of their (often devastating given characters’ limited health) rebuttals.

That’s just the “planning phase.” The “action phase” follows, in which everything you’ve made your characters do is played out and, in the right combinations, sees your team weave around one another in cinematic — and cool-as-hell — ballets of punching, kicking, and arcane magic.

It lends a directorial bent to already tactically deep gameplay, as you decide not just what will finish a fight but how to look good while doing it. Which is fortunate, because you will be getting into a lot of fights. Getting to know your friends? Time for a fight. Completing an incremental step of a weekly assignment? Fight. Taking a break with a side quest? Fight. Going to the restroom? Fight… with 50 ghosts.

This isn’t unusual for an RPG, though these are unavoidable scripted events rather than random encounters, nor are these battles especially hard. But not being able to move to a new screen without running into a fight can get old fast. It only took me until the second week to toggle on the “almost invincible” accessibility option — Demonschool lets you rewind your movements and restart battles without penalty to clean up mistakes, too — just for a cognitive break while encountering fights every few minutes.

There’s a hint of self-awareness about the frequency of fights, which are occasionally refreshed as you develop new skills and find new companions, as Namako early on bemoans not being able to walk into Temsk’s cemetery without being ambushed by gangsters and the undead (yeah, Namako, I know how you feel).

The weight of that specific frustration is lessened, however, by everything else Demonschool offers. Let’s be honest: main quests are fine, but I’d rather decorate my clubhouse with friends, explore Temsk’s nightlife, and pull demon fish from the depths in a now-obligatory-in-RPGs fishing minigame.

There is a refreshing lack of extravagance to this side content, especially compared to where Necrosoft draws its inspiration from. Like all good high school horror, Demonschool is less about fighting the big bad than bringing together a found family of complex and relatable misfits to do so. In its straightforward approach to this aspect of Demonschool, Necrosoft not only lets its writing — particularly of its characters, who genuinely develop and grow — shine through, but also relieves potential for frustration. But for the odd misplaced karaoke lyric, if you spend time with your friends you become closer to them.

Demonschool has rough edges, particularly in how it balances its excellent writing with slightly-too-frequent (though still interesting) combat. But in its open approach to its influences — while avoiding feeling like a lesser copy — its wit, and its charm, it builds a unique identity from its melting pot of inspirations to leave its deficiencies far outweighed by how much there is to love.

_{Demonschool launches on November 19th on the Switch, PC, PlayStation, and Xbox.}

How a bunch of hackers freed the Kinect from the Xbox

2025-11-03T16:10:14-05:00

In 2010, when Microsoft unveiled the Kinect, it pitched the camera as a revolutionary new gaming device. Swing an imaginary lightsaber and that would be translated onscreen. Throw a football and it would be caught on your TV. Fifteen years later, we know the Kinect as an expensive failure. Microsoft overestimated the demand for playing games with your body. But the Kinect did still turn out to be revolutionary — just not for gaming.

Now, we understand the Kinect is anything but a gaming device. It became a robotics game changer, enjoyed a brief dalliance with pornography, and is now upsold as a ghost hunting toy. None of which would have been possible had a community of hackers not come together to fashion open source drivers for the Kinect, freeing it from the limitations of being locked to the Xbox 360 and opening new frontiers of experimentation, creative expression, and commercial advancement.

“Technically, nothing the Kinect did was entirely new,” says Memo Akten, an artist working with code, data, and AI and an assistant professor at the University of California, San Diego. The small camera projected a grid of infrared dots and read deformities in that pattern to discern depth. In an early example of machine learning, it recognized human limbs and gestures. “Those capabilities existed in research and industrial systems for many years,” he adds. Those systems cost in the region of $5,000 to $12,000. Here was Microsoft selling a variation of the technology for $150.

“What had previously required very expensive equipment and/or complex multi-camera setups with manual alignments, calibration, and correspondence was now available off the shelf,” Akten continues.

Kyle Machulis, CEO of Nonpolynomial and founder of buttplug.io — an open source project for controlling sex toys — was working on $250,000 mapping systems not dissimilar to the Kinect in 2010. He quickly recognized the peripheral as an opportunity to “democratize that technology.”

He recalls heading out on November 4th to pick up a Kinect to reverse engineer. An hour later, New York-based DIY electronics producer Adafruit announced OpenKinect: a bounty of $1,000 — a prize that it would raise to $3,000 — for whomever offered evidence of the Kinect working on any operating system.

“Imagine being able to use this off the shelf camera for Xbox for Mac, Linux, Win, embedded systems, robotics, etc.,” Adafruit wrote in its announcement. “We know Microsoft isn’t developing this device for FIRST Robotics, but we could! Let’s reverse engineer this together, get the RGB and distance out of it and make cool stuff!”

Doing so was not a simple case of taking the Kinect apart or plugging it in. Though it could connect to a PC via USB, the way they communicated was unknown and the only way to get at it was to watch the Kinect and Xbox 360 speaking to one another. “Since the Kinect didn’t have PC drivers, we needed this piece of hardware called a USB sniffer,” Machulis tells The Verge. A colloquial term for a protocol analyzer, a USB sniffer is a tool that could record the data passed between the Kinect and Xbox 360. In 2010, that cost $1,200 and, Machulis says, “I really didn’t want to buy it.”

Some information could be gleaned by simply connecting the Kinect to the PC, but it was mostly unhelpful — power consumption, packet sizes, and confirming the Kinect is, in fact, a camera. Hackers could start sending random packets and possibly work something out, but it was just as liable to brick the Kinect completely.

Hackers and reverse engineers around the world were raring to go. But it appeared that whoever got their hands on a sniffer would win the bounty almost by default. That race wasn’t just for the money, however, but also the cachet of being the first to hack such a high-profile device. With the community stalled over the massive expense — almost half the bounty — it opened the door for someone outside the community to potentially snatch the glory away.

To keep the contest equitable — and, perhaps, to try and maintain the bounty’s and the company’s momentum in the press, Adafruit took on that expense, ordering a sniffer to then release the logs to the community. But while everyone waited for the device to ship to Brooklyn, it appeared the worst had already happened and someone had beaten them to the punch.

“AlexP” released a video the next day demonstrating control of the Kinect’s motor on PC. It prompted a panicked response from Microsoft denying the Kinect could be hacked while threatening to explore legal options. Microsoft quickly U-turned when it became clear no one was trying to hack consumers’ cameras. But as the community reeled from potential litigation, AlexP returned with a second video. This time, he showed off depth and RGB images on PC. The contest was over before it had begun.

AlexP didn’t claim the bounty, however. Instead, his company, Code Laboratories, opened a $10,000 “fund,” upon payment of which Code Laboratories would release the source code to the open source community. For some in the OpenKinect community, this wasn’t so far removed from what Adafruit — which was already benefitting from significant press — was doing. Even if Adafruit was looking to open source the Kinect drivers and Code Laboratories to sell the drivers as it had with the PlayStation Eye before. For others, it was tantamount to a ransom, withholding code that could make the Kinect more accessible and unleashing its potential easier. “But that was great motivation for the community to just be like: Let’s take $10,000 away from you, actually,” Machulis says.

A competitive edge shouldn’t be surprising in a contest for a bounty. According to Machulis, however, there was more to it. “That’s the thing about reverse engineering; It’s who gets their name on it first and loudest.”

On the evening of November 9th, Adafruit finally uploaded the logs collected by its sniffer, and the community began to pore over them. They were searching for the protocols that controlled the Kinect, exploring packets that might turn on a light, enable a camera, or operate the motor. It was incremental, tedious, and exhausting work.

As US hackers dropped off in the early hours, the clock ticked over to the Kinect’s European release date. Like his American peers, 20-year-old Hector “marcan” Martin purchased a Kinect and, armed with Adafruit’s logs, went through packet by packet to divine the Kinect’s protocols.

When the US woke, it did so to the fruits of his dogged examination: a video of Martin demonstrating RGB and depth on Linux. It had taken six days for OpenKinect to hack the Kinect from its release — really, once the logs became available, it had taken Martin less than 24 hours.

That was far from the end, however. “Hector certainly did most of the hard, technical work in terms of getting the initial packets set up,” Machulis says. “Then everyone realized this shit is gonna get big.” The bounty claimed, the community set its sights on more drivers.

Theo Watson spent three weeks at his computer working on OpenKinect — every day, 10 hours a day. The Kinect revolutionized how Design I/O, which Watson co-founded, developed interactive installations, and he still uses it today. In 2010, however, he was 30 years old, recently transplanted to the US, isolating limb data from infrared cameras.

“I saw that time off as an R&D investment,” he says, a way to open more efficient avenues to bigger and better interactive experiences. “I really wanted to be the first person to get the Kinect running on a Mac.”

The OpenKinect community had coalesced around a remote nucleus of reverse engineers from different countries, generations, and demographic groups. There was drama, frayed nerves, but also a common goal, as Watson discovered while trawling through the process of running the Kinect on OSX.

“I think it really helped because it felt like you were part of a team effort,” he says. “If people were running into problems, it’s like having a collective brain. The Kinect basically needs everything to be perfect. If you’re off by one little thing, you don’t get anything. Then, suddenly, someone notices something and it works.”

Driven by the community, with a bit of help from Martin, Watson had the Kinect talking to Mac by November 12th. More drivers and code filled the community’s GitHub, and hacking efforts wound down. “There’s only so much to extract and then you have to be able to do something with that data,” Machulis says. “Those are two fairly different skillsets.”

Instead, people like Akten stepped in. The first open source drivers, libfreenect, didn’t include body tracking — Microsoft released its own skeletal SDK in 2011, its hand forced by OpenKinect — only granting access to raw depth data. “We could still do a lot with that,” Akten says. “For one thing, we had the 3D data, which allowed all kinds of creative, playful interpretations.”

The community chat flooded with experiments, many Akten’s. He explored drawing in 3D space, later moving on to develop machine learning algorithms to detect poses, and even controlling drones. The Kinect was open, bringing with it a host of explorations of how to exploit it creatively. Suddenly, an affordable way for robots to detect obstacles and map environments in real time became available, surgeons explored examining scans contactless, rapid 3D models of rooms and objects became a real possibility, teachers used the Kinect as an interactive learning device, and, if you really wanted, someone could now control a sex toy over a video call.

“This thing on the front camera,” Watson says, pointing to the black bar at the top of his iPhone’s screen, “that, I think, is a miniature Kinect.”

He’s almost wistful. Apple purchased PrimeSense, the Israeli company behind the Kinect’s sensor technology, in 2013. “I was so disappointed,” he says, “because I just knew that was the end of the Kinect technology.”

The sale prompted Microsoft to explore a new system for its next Kinect — OpenKinect went and hacked that one too — discontinuing Kinect for Windows shortly after its release in 2014 and shutting down manufacturing for the Kinect in 2017 as sales diminished and it focused on the Kinect 2 and development of the third-generation Kinect Azure. Yet, the technology has lived on, incorporated into countless Apple devices as part of its facial recognition and 3D mapping to the point of being ubiquitous.

That sense of loss extends, in part, to the internet from which OpenKinect emerged. “It was way more punk rock!” Watson laughs. “No one had really established the rules.”

In 2010, the internet was unruly; it had yet to coalesce around the hubs it has today. Piracy was in its heyday, pre-AlexNet — a major neural network architecture that paved the way for modern AI models like Stable Diffusion — with GitHub, now an online staple, having released only three years before (the same year as Tumblr and the iPhone’s reveal). “We were only four or five years into the maker movement,” Machulis says. “The idea of a product like this that has taken a massive amount of R&D cost to be put out and hacked this quickly — it was basically unheard of.”

Now, with better tools, it’s far more common. Which is part of why we don’t hear about it as much as before — that, and not being attached to, as Machulis puts it, a “shining sun” of a product. “It is in general easier to make some of this stuff,” Machulis continues. “There’s way more communities online, there’s more content creators talking about this stuff.” The kind of effort surrounding opening the Kinect has now lost some of its buccaneering flavor, some of its sense of counterculture, simply by virtue of becoming more mainstream and, in many ways, more frequent. “I don’t think anything fizzled out,” Machulis adds. “I think it just got quieter and spread out.”

Still, there is a sense that how we approach technology has changed irreparably. “I think technology has become more of a product now and less something that you get involved with. That’s kind of sad,” Watson adds. “I kind of fear that the current generation is growing up just thinking the internet is inflexible. It is the way it is and nothing will ever change. We were constantly surrounded by that change. And it really made things feel more free and more open.”

Rather, similar communities to OpenKinect may feel invisible without a subject as high-profile as the Kinect. As the economic bubble inflating around AI grows more opaque as corporate interests scramble to make the technology a profit-turning industry, hackers have turned their attention to open sourcing its models. Aligned to a sprawling technology constantly in the public eye, it may well be that AI grants us our next big communal reverse engineering effort to echo OpenKinect. This is especially curious given the Kinect gave many of us our first interactions with AI, and it is AI that is finally interrupting the Kinect’s spirited afterlife.

“When the original Kinect came out, it took what might have been 100 hours of me writing computer vision code with a standard black-and-white infrared camera and gave me something that would shave that time off our development for a project and give better-quality results,” Watson says. “AI with code is doing a similar thing; they just take away the painful aspects of the work and let us focus on the creative part.”

Now, 15 years after hackers opened the Kinect to computer vision creatives, AI can do everything it did better, faster, and using standard RGB cameras. Watson shows The Verge a video of AI’s real-time tracking, its superior occlusion of limbs and digits blitzing across the screen as members of a K-pop group weave around one another, each marked by a colored skeleton on the screen — all pulled from an ordinary camera. “AI is made to make decisions about many things very quickly, and we need a decision about every pixel in an image,” Machulis says. “Since we can tell so much just from images now we may not need all the extra hardware, with methods like gaussian splatting we’re already seeing that ability to, what looks like, create information from thin air.”

“Next time we chat, we might have gone back to infrared cameras,” Watson says, before adding: “AI might kill the Kinect.”

Kirby Air Riders’ impressive accessibility features are a rarity for Nintendo

2025-10-24T12:59:50-04:00

The latest Nintendo Direct on October 23rd brought us an hour-long dive into Kirby Air Riders, as well as a surprising accessibility announcement. Hosted by game director Masahiro Sakurai, the Direct showed off more features, gameplay, and modes for the racing game releasing for the Switch 2 on November 20th. Tucked toward the end of the presentation, however, Sakurai devoted a few minutes to Kirby Air Riders’ accessibility options. It was a welcome spotlight on an aspect of game design Nintendo regularly forgets, though not necessarily a sign of broader change for the company.

An impressive suite of options includes button remapping, changeable text size — something missing from many games, but vital on the Switch 2’s smaller undocked screen — colorblind filters, preset options for changing the size and opacity of the HUD, and a toggle for onscreen outlines (though this was the only menu option not demonstrated during the Direct). Unlike with previous first-party Switch 2 games, inputs can be remapped in-game as opposed to solely through the Switch 2’s system-level options. You can also save named profiles for quick switching and set up the controls so the game is playable one-handed.

After presenting this list, Sakurai focused in on something he wanted to implement “no matter what” — motion sickness mitigations. In Kirby Air Riders you can add onscreen markers, ostensibly of varying thicknesses and colors, and change the field of view to reduce the perception of motion. Alternatively, you can turn camera shake and tilt off altogether, all of which can be visualized within the menu. For greater simplicity, there are also preset options for visual accessibility from “none” to “medium” and “strong.”

It’s a boon for players, especially disabled players — not just that these features are included but that they’re being communicated ahead of release. Accessibility menus beyond the bare-bones remain rare from Nintendo’s first-party titles, and it’s significant that a tentpole title from the Switch 2’s release year takes such a major step toward a more accessible and customizable experience. That said, we should temper our expectations of what this means going forward.

To be clear, this should be a precedent for Nintendo to follow — not just in the maintenance of these features in future titles, but also further accessibility development. But it should be noted that Sakurai has taken a greater personal interest in accessibility over the past few years, particularly in the impact of motion sickness. The evidence of the past six months suggests that the impetus behind Kirby Air Riders’ accessibility considerations likely comes from Sakurai and his team, rather than Nintendo as a whole.

The system-level accessibility improvements in the Switch 2 were a similar cause for excitement that they might represent a meaningful change for Nintendo. Instead, the first-party games that followed released with few, if any, accessibility considerations. This culminated in Drag x Drive, where that apathy was more keenly felt in a game depicting wheelchair basketball and yet utilizing inaccessible mouse controls. Nintendo’s suggested players seek mitigations in system-level menus.

Similarly, previous ostensible precedents have proven otherwise. In 2020, The Last of Us Part II appeared to be a turning point for accessibility in gaming going forward. Notably, Sakurai praised The Last of Us Part I’s accessibility and cited it as a title from which he learned a lot about accessibility not long before starting development on Kirby Air Riders — only for the broader industry to deprioritize accessibility in years since.

None of which is to say we shouldn’t celebrate this surprising and sudden increase in accessibility in a first-party Nintendo release, nor that players shouldn’t be excited. Looking at the care Sakurai and his development team have taken to make Kirby Air Riders more accessible is a rare bright spot in this industry. It behooves us, however, to not get carried away in regarding this as a turning point at Nintendo given how strongly history suggests otherwise.

Easy Delivery Co. is a cozy, Lynchian dream

2025-10-20T14:49:52-04:00

I have been a fan of David Lynch ever since a friend’s older brother interrupted a 13th birthday celebration to insist we all watch Eraserhead. In the realm of horror movies, it’s a common way to be introduced to a seminal film: be it in a friend’s basement, illicit underage viewings, or a stray recommendation from a pot-smoking sibling. It feels apt, then, that my first introduction to Easy Delivery Co. was a friend insisting I drop everything because a game exists that lets you play as a cat driving a kei truck.

I admit, those two things don’t immediately correlate. But it’s clear that Lynch, particularly Twin Peaks, was a major influence for developer Sam Cameron, in what the Steam description explains is “a relaxing driving game, with definitely no secrets.”

Easy Delivery Co. starts you off gently. You are a nameless black cat, rendered in the style of Animal Crossing on the GameCube, working as a delivery driver in a sleepy mountain community. Day and night, you shuttle supplies between shops and businesses snowed in by a perpetual blizzard. Like most real-world delivery drivers, resting is discouraged. Instead, you neck energy drinks and coffee to keep at bay the biting cold that threatens to overcome you the moment you set foot outside the adorably small cabin of your kei truck.

That cold will likely be your first sign that all is not what it seems. Succumbing to the blizzard as you shuffle to a vending machine or take too long shutting your truck’s tailgate briefly transports you to a dark and ostensibly endless maze, before you wake to resume your deliveries. Yet, as your world expands to other towns, you’ll also find that, curiously, all the shopkeepers remain the same. Then there’s MK, the only character you’ll meet outside — and the only dog in town — who leads you away from the repeating web of deliveries to set you on the path to unraveling the mountain’s mysteries.

Easy Delivery Co. is not a complex game. You drive, pick stuff up, and spend your earnings on the means to do more of that. This low-poly mix of Lake and Silent Hill makes for an interesting concept, even if, like many indie games that hit upon an intriguing hook, the story is a little clumsy — especially as the game shifts from a meandering string of ostensibly unrelated tasks toward its narrative climax.

There are rough edges, too. Among a few minor bugs I encountered, dropping cargo — which you’ll do often on these icy roads — can sometimes leave you unable to retrieve it even though it’s right in front of you. Those rough edges, however, are not terminal. That specific bug is easily remedied by aborting the job and picking another, as Easy Delivery Co. is pleasingly forgiving, levying few penalties for leaving a job or having a rough drive.

Driving, the thing you’ll do the most in this game, is relaxing without being passive, a hard balance to strike. Setting Easy Delivery Co. in a snowy tangle of mountain roads, delivering a variety of items that impact how your truck handles, is a smart turn. You can zone out somewhat, but not completely, as you have to adjust for the conditions and loads. This is something you’ll become accustomed to after you careen off the road and roll down the mountain a few times — something I did a little more often for playing with the perfectly manageable keyboard controls, rather than a controller. None of that is frustrating. In fact, more than once a mistake led to an accidental shortcut, achieved mostly by somehow surviving a lengthy cartwheel down an incline with my cargo intact.

If I have one glaring issue with Easy Delivery Co., it’s the balance between direction and freedom. Many mechanics in-game are found in readable tutorials but also able to be puzzled out. For instance, I appreciated being able to ignore guidance and work out how to make coffee myself or that my lighter will let me stay out in the cold for longer. Being ushered through the game’s main quest by MK, I found myself — oddly for someone usually allergic to obfuscation — wishing there was more of the central mystery to discover independently, more secrets beyond the odd collectible to solve myself. Though that, I think, comes from a desire to spend more time in Easy Delivery Co.‘s eerier side. Either that or I’m just a cat person.

It’s a sign of how versatile Easy Delivery Co. can be, however. You can, if you’re inclined, race through the game’s main story with minimal deliveries in the space of a few hours. You can also completely ignore it — and MK — and stay frozen in a sequence of deliveries, basking in the low-poly, nostalgic environment and listening to catchy jingles through the radio.

My favorite way to play was to make Easy Delivery Co. small in the corner of my screen, where it remained perfectly visible and playable, while watching YouTube or TV… which may have contributed to how much time I spent off-road. I might be biased, if only because Easy Delivery Co. feels like it’s been developed from a Venn diagram of my specific interests (especially adorably proportioned motor vehicles). But one of my gaming holy grails is something I can play and enjoy while watching something else to pass tired hours — in Easy Delivery Co. I’ve finally found that.

_{Easy Delivery Co. is available now on Steam.}

Drag x Drive is more drag than drive

2025-08-13T07:33:40-04:00

When Nintendo announced Drag x Drive, a Joy-Con mouse-controlled wheelchair sports game, for the Switch 2 I was tentatively excited. I have a lot of time for developers trying new things, and sports video games are hardly replete with disability representation. Having been hands-on with the game, however, Drag x Drive has left me baffled and in significant pain.

As a “driver” — wheeled robots that come in three mostly indistinct flavors: guard, center, and forward — Drag x Drive has you play pickup games of three-on-three wheelchair basketball. You navigate the game by pushing and pulling your Joy-Cons across a surface in a motion vaguely similar to moving a wheelchair. This with a combination of shoulder button inputs and motion controls lets you move, brake, pass, and shoot.

Those basic controls are initially intuitive. This includes a satisfying controller rumble if you’re moving, and that’s a big if. My Drag x Drive experience revolved around desperately trying to get the Joy-Cons to register consistently across tables, floors, my bed, and my body. Yet, more often than not, I’d be spinning out and losing control while taking umpteen unintended captures — the home button blessedly appears to be disabled while moving the mouse. My driver regularly took both hands off its wheels unbidden in a comical, yet appropriate, shrug.

The game’s 20-minute tutorial tells you to move and then lift the Joy-Cons and to do so in long motions on a flat surface or on your legs. Yet, these actions on their suggested surfaces only contribute to an unresponsiveness that makes controlling Drag x Drive laborious. Any imperfection, be it an uneven table or folds in your clothing, has a significant impact on whether your movements translate appropriately onscreen.

I found some consistency by sitting cross-legged on my bed, hunched over, using small movements on my bedsheets (and weeping for my 34-year-old back). My legs, too, though error-strewn, were more responsive and quieter than any table I tried. This was especially true — before anyone questions my dedication in finding some way to enjoy this game — when I shaved my upper-legs and used the Joy-Cons on my bare thighs. I can’t tell whether the mouse sensors are too sensitive or not sensitive enough, but it does appear that a soft, malleable, and smooth surface yields the best, if still often frustrating, results.

If you can get moving, matches are slow and repetitive. Playing bots, loading into the three-minute games took about two minutes, and that wait can be longer online in the hub “Parks,” where you are often forced to wait for other games to finish. The loop of matches is mostly one player wheeling the ball up the court, shooting, and then the opposing team doing the same. Tackling can only be achieved head-on and, with controls that are inconsistent in their responsiveness, intentionally stealing the ball is a major achievement. Similarly, consistent shooting requires being lined up with the basket which is hard at speed, so dunking by riding up the court’s curved edges — something you’ll be doing a lot accidentally anyway — is the best tactic. Some of this sluggishness could be allayed if the games remained chaotic, but after each basket it is impossible to interact with the ball for a few seconds, meaning a delay before play can start again.

But Drag x Drive isn’t just an onscreen experience, it’s a physical one. This was always going to be an issue for disabled players, but I remain surprised how uncomfortable Drag x Drive is to play — turtling over my bedsheets for grip notwithstanding. I found the narrow claw needed to hold the Joy-Cons in this configuration agonising and lengthy play impossible.

Controller support is absent. Though, using the Joy-Con as a traditional mouse to navigate menus isn’t so bad (you can also toggle menu movements to the joystick, thankfully). Nor does it appear any additional support, even for the Hori Flex adaptive controller, is planned. Nintendo tells The Verge, “We wanted to focus on the innovative experience of using the Joy-Con 2 mouse and motion controls for this game. However, certain accessibility features can be found and utilized at the hardware system level.”

What features Nintendo references here, I cannot tell. After Mario Kart World’s reliance on — an initially impressive — system-level accessibility, it’s a concerning trend that Nintendo directs players to system menus rather than implementing mitigations into games. Though, perhaps it’s something we might have seen coming. Despite the obvious conception of these characters, Nintendo has appeared reluctant to acknowledge the disability inspiration behind the game leading up to the its official release, reportedly instructing staff to use the term “vehicles” when attendees said “wheelchairs” at hands-on events to the confusion of many. After several inquiries, Nintendo ultimately tells The Verge, “It has always been the case that the vehicles were inspired by wheelchairs. This was miscommunicated at some events. The gameplay is inspired by a mix of wheelchair basketball, wheelchair rugby, wheelchair motocross, and skateboarding.” It’s notable, however, that the only in-game references to the player characters are the terms “driver” and, on one occasion in the lengthy tutorial, “ballers.”

That said, I could not find anything that might help how frustrating and physically uncomfortable Drag x Drive’s control setup rapidly becomes.

That’s not to say there are no accessibility considerations in the game. Drag x Drive is visually easy to parse, with opposition players highlighted prominently and indicators for what’s happening offscreen being bright and bold. Nor is it entirely joyless. The controls for shooting, a satisfying and responsive flick of the wrist, are a blast when a shot lands. Though, the crispness of the game’s motion controls do further highlight the mouse controls’ deficiencies. Similarly, Rebound Scramble, one of two minigames that breaks up match play, is fleetingly fun as a madcap dash to catch a bouncing ball, though this too soon gets old. The other minigame, an obstacle course, was nothing but infuriating for me.

I could see Drag x Drive being a somewhat diverting Rocket League clone if it wasn’t stubbornly attached to Nintendo’s gimmicky, throwback to the Wii attitude to controls or simply a frustrating part of a broader suite of sports a la boxing in Wii Sports. But with sluggish, repetitive matches, two minigames, and a few activities in parks to win mostly useless rewards, it’s hard not to notice just how little Drag x Drive has to offer.

If I can say one good thing about Drag x Drive: it does demonstrate that a game based on wheelchair sports could be great in the right hands. But if you’re hoping for a fast-paced, representative, or even fun foray into wheelchair basketball, this isn’t it.

What happens when AI comes for our fonts?

2025-06-26T10:05:43-04:00

Monotype is keen for you to know what AI might do in typography. As one of the largest type design companies in the world, Monotype owns Helvetica and distributes Futura and Gill Sans — among 250,000 other fonts. In the typography giant’s 2025 Re:Vision trends report, published in February, Monotype devotes an entire chapter to how AI will result in a reactive typography that will “leverage emotional and psychological data” to tailor itself to the reader. It might bring text into focus when you look at it and soften when your gaze drifts. It could shift typefaces depending on the time of day and light level. It could even adapt to reading speeds and emphasize the important portions of online text for greater engagement. AI, the report suggests, will make type accessible through “intelligent agents and chatbots” and let anyone generate typography regardless of training or design proficiency. How that will be deployed isn’t certain, possibly as part of proprietarily trained apps. Indeed, how any of this will work remains nebulous.

Monotype isn’t alone in this kind of speculation. Typographers are keeping a close eye on AI as designers start to adopt tools like Midjourney for ideation and Replit for coding, and explore the potential of GPTs in their workflow. All over the art and design space, creatives are joining the ongoing gold rush to find the use case of AI in type design. This search continues both speculatively and, in some places, adversarially as creatives push back against the idea that creativity itself is the bottleneck that we need to optimize out of the process.

That idea of optimization echoes where we were a hundred years ago. In the early 20th century, creatives came together to debate the implications of rapid industrialization in Europe on art and typography at the Deutscher Werkbund (German alliance of craftspeople). Some of those artists rejected the idea of mass production and what it offered artists, while others went all in, leading to the founding of the Bauhaus.

“It’s almost as if we are being gaslighted into believing our lives, or our professions, or our creative skills are ephemeral.”

The latter posed multiple vague questions on what the industrialization of typography might mean, with few real ideas of how those questions might be answered. Will typography remain on the page or will it take advantage of advances in radio to be both text and sound? Could we develop a universal typeface that is applicable to any and all contexts? In the end, those experiments amounted to little and the questions were closed, and the real advances were in the efficiency of both manufacturing and the design process. Monotype might be reopening those old questions, but it is still realistic about AI in the near future.

“Our chief focus is connecting people to the type that they need — everywhere,” says Charles Nix, senior executive creative director at Monotype, and one of Re:Vision’s authors. This is nothing new for Monotype, which has been training its similarity engine to recognize typefaces since 2015.

But the broader possibilities, Nix says, are endless, and that’s what makes being a typographer now so exciting. “I think that at either end of the parentheses of AI are human beings who are looking for novel solutions to problems to use their skills as designers,” he says. “You don’t get these opportunities many times in the course of one’s life, to see a radical shift in the way technology plays within not only your industry, but a lot of industries.”

Not everyone is sold. For Zeynep Akay, creative director at typeface design studio Dalton Maag, the results simply aren’t there to justify getting too excited. That’s not to say Dalton Maag rejects AI; the assistive potential of AI is significant. Dalton Maag is exploring using AI to mitigate the repetitive tasks of type design that slow down creativity, like building kern tables, writing OpenType features, and diagnosing font issues. But many designers remain tempered about the prospect of relinquishing creative control to generative AI.

“It’s almost as if we are being gaslighted into believing our lives, or our professions, or our creative skills are ephemeral,” Akay says. She is yet to see how its generative applications promise a better creative future. “It’s a future in which, arguably, all human intellectual undertaking is shed over time, and handed over to AI — and what we gain in return isn’t altogether clear,” she adds.

For his part, Nix agrees: the more realistic and realizable use of AI is the streamlining of what he calls the “really pedantic” work of typography. AI might flatten the barrier to entry in design and typography, he says, but “creative thinking, that state of being a creative being, that’s still there regardless of what we do with the mechanism.”

“Thirty-five years ago there was a similar sort of thought that introducing computing to design would end up replacing designers,” he continues. “But for all of us who have spent the last 35 years creating design using computers, it has not diminished our creativity at all.”

“For all of us who have spent the last 35 years creating design using computers, it has not diminished our creativity at all.”

That shift to digital type was the result of a clear and discernible need to improve typographic workflow from setting type by hand to something more immediate, Akay says. In the current space, however, we’ve arrived at the paintbrush before knowing how the canvas appears. As powerful as AI could be, where in our workflow it should be deployed is yet to be understood — if it should be deployed at all, given the less-than-stellar results we’re seeing in the broader spectrum of generative AI. That lack of direction makes her wonder whether a better analog isn’t the dot-com bubble of the late 1990s.

In many ways, it mirrors our current situation with AI. As public access to the internet increased, a wave of dot-com startups emerged and with them increased venture capital, even though the internet at the time “never connected to a practical consumer need,” Akay says. Overvalued and without a problem to solve or a meaningful connection to consumers, many of those startups crashed in 2000. “But [the internet] came back at a time when there were actual problems to solve,” she adds.

Similarly, few consumers exploring AI are professional designers trying to optimize workflow; rather, AI is increasingly the playground — and product — of executives overvaluing AI as they attempt to automate jobs and try to push creativity out of creative professions.

Both Nix and Akay agree a similar crash around AI might actually be beneficial in pushing some of those venture capitalist interests out of AI. For Nix, however, just because its practical need isn’t immediately obvious doesn’t mean it’s not there or, at least, won’t become apparent soon. Nix suggests that it may well be beyond the bounds of our current field of vision.

Nix adds that in our Western-focused view of AI, we might not see the difference in our expansive selection of typefaces and how limited those choices might be for non-Latin scripts, for instance. That, and similar areas outside the Western mainstream of design, may be where the need for change is more apparent. “The periphery may end up driving the need-state [for AI].”

For all that, it remains unlikely that current models of selling typography will change, however. We’d still be licensing fonts from companies like Monotype and Dalton Maag. But in this AI-driven process, these generative apps may well be folded into existing typography subscriptions and licensing costs passed on to us through payment of those subscription fees.

Though, that remains more speculation. We are simply so early on this that the only AI tools we can actually demonstrate are font identification tools like WhatTheFont and related ideas like TypeMixer.xyz. It’s not possible to accurately comprehend what such nascent technology will do based solely on what it does now — it’s like trying to understand a four-dimensional shape. “What was defined as type in 1965 is radically different from what we define as type in 2025,” Nix adds. “We’re primed to know that those things are possible to change, and that they will change. But it’s hard at this stage to sort of see how much of our current workflows we preserve, how much of our current understanding and definition of typography we preserve.”But as we explore, it’s important not to get caught up with the spectacle of what it looks like AI can do. It may seem romantic to those who have already committed to AI at all costs, but Akay suggests this isn’t just about mechanics, that creativity is valuable “because it isn’t easy or fast, but rather because it is traditionally the result of work, consideration, and risk.” We cannot put the toothpaste back in the tube, but, she adds, in an uncertain future and workflow, “that doesn’t mean that it’s built on firm, impartial foundations, nor does it mean we have to be reckless in the present.”

Correction, June 26th: An earlier version of this article misstated which fonts Monotype owns. It owns Helvetica font software and distributes fonts like Futura and Gill Sans; it doesn’t own Futura and Gill Sans.

Answering the Nintendo Switch 2’s lingering accessibility questions

2025-06-13T19:20:24-04:00

One of the biggest surprises of the Nintendo Switch 2’s reveal was its proposed accessibility. For years, Nintendo has been known for accidentally stumbling on accessibility solutions while stubbornly refusing to engage with the broader subject. Yet, in the Switch 2, there appeared a more holistic approach to accessibility for which disabled players have been crying out. This was supported by a webpage dedicated to the Switch 2’s hardware accessibility.

However, specifics were thin and no further information emerged ahead of the Switch 2’s debut. Now, having spent the last week with the Switch 2, I’ve found that this limited information hid, aside from a few missteps, an impressive suite of system-level accessibility considerations and advances that somewhat offset the otherwise gradual update the Switch 2 represents. But as we finally answer lingering accessibility questions over the Switch 2, there’s a nagging sense that this information should have been readily available ahead of launch.

How intuitive is the setup? Very, but blind players may need assistance

I tend to find setup procedures dense and unapproachable thanks to cognitive disability. Yet I was pleasantly surprised at how breezy the Switch 2’s setup was. Aside from a few hiccups trying to decipher Virtual Game Cards (a feature I ended up opting out of), the whole process was intuitive and fast.

Accessibility features, however, are not available during setup. The console’s text-to-speech is not enabled by default, nor can you access text sizing and zoom options. This will represent a significant barrier to entry for some, and blind players may require sighted assistance during setup.

Are Switch controllers and Joy-Cons compatible with the Switch 2? Yes, including the Hori Flex

The increased size of the Joy-Con 2 feels better this time around, as does the Joy-Con 2 grip. But the controller size and the grip’s unergonomic square shape still don’t take long to incite my hand pain. Fortunately, Switch controllers and Joy-Cons are usable on the Switch 2, and the ability to default to the original Pro Controller is welcome.

It’s encouraging that you can pair other controllers with the Switch 2 using peripherals like the Magic-S Pro 2. Though, given recent trends in third-party peripheral support, I’d be reluctant to suggest that support is here to stay. We’re already seeing some connectivity issues around third-party controllers, especially 8BitDo gamepads, with 8BitDo working on updates to get those working with the Switch 2.

What is supported, however, is Nintendo’s licensed adaptive controller: the Hori Flex. This works docked and in tabletop mode for games that don’t require mouse controls (some other features, like motion controls, may also cause issues). Keep in mind, you will need a USB-A to USB-C adapter to connect in tabletop mode.

Can you connect a USB keyboard? Yes

Not everyone finds onscreen keyboards intuitive or accessible. The good news is you can connect a USB keyboard through the console’s USB-C ports and use that instead, including during setup.

What is the text-to-speech speed? It’s inconsistent, but US English is 120–130 words per minute on average

We have no official number on the speed at which the Switch 2’s text-to-speech reads, but we can estimate. Using the information the system reads when you enable text-to-speech, we’ve found that both voice options averaged around 120–130 words per minute in US English. For UK English, it was 130–140 words per minute. Different languages will see different rates. There was some inconsistency in repeated tests that could not be accounted for simply by considering when timers were started and stopped.

That’s not the only quirk in the system. Players should be aware there is a noticeable delay between landing on a menu option and the text-to-speech kicking in. Similarly, during testing, toggling text-to-speech off led to a notification with the word “disabled.” Reenabling the function without moving out of the accessibility menu, however, did not lead to anything suggesting text-to-speech was back on.

All told, however, it’s a good system, controlled by a discrete speed slider that ranges from 50 percent to 300 percent. The slowest rate for US English users reads at around 60–70 words per minute and the max speed, where many blind players operate, is in the region of 400–420 words per minute.

What screens don’t support text-to-speech? The eShop

When enabling text-to-speech, you’re told the function may not be supported on all screens. As things stand, it looks like text-to-speech works across all system-level menus, but not on the Nintendo eShop. Given this is the first place most players visit after setup, that’s an oversight that needs rectifying soon. Most games also don’t support system-level text-to-speech, including Mario Kart World.

Can you adjust the Switch 2’s audio balance? No

Outside the ability to toggle mono audio — an important feature for hard-of-hearing players — greater audio customization is not available on the Switch 2 at a system level.

Does GameChat’s speech-to-text transcribe swearing? You bet it fucking does!

You may already know that GameChat’s speech-to-text will transcribe swear words. While this was reported widely as a bit of fun, it’s also an important accessibility feature that allows users — especially deaf players — to engage fully with in-game communication without having to decipher improper transcription and censored text.

It’s not perfect. You may find speech-to-text swaps in odd words at times, but this is an error rather than censorship of specific terms.

Can you remap controls? At a system level, yes

Within the Switch 2’s settings, players can remap all inputs on connected controllers, and do so for each Joy-Con independently. You can also toggle the ability to access this menu at any time from the Switch 2’s quick menu, accessed by holding Home.

Unfortunately, current evidence suggests this option will not be present in Nintendo’s first-party games. Mario Kart World does not include any remapping options. Some might ask why this is a problem if system-level input rebinding exists, but in-game options for remapping are more instructive and convenient, and they limit rebindings to specific actions in-game rather than having to continuously rebind on a system level for every game. Being able to remap on the fly through the quick menu only mitigates this so much.

This is something Nintendo appears to understand, allowing remapping in the new GameCube games available to Switch Online subscribers.

A quiet win overall

It all adds up to an impressive suite of accessibility features and customization that will, hopefully, grow in time. The strange part is that Nintendo is being so quiet about it. The Verge reached out multiple times for clarification on the Switch 2’s accessibility and for more information relating to the questions above, but Nintendo didn’t respond.

One might expect that points to an internal awareness that features aren’t as robust as they should be. Outside a few missteps, however, this is a significant accessibility win and one would think Nintendo would want it out there. More importantly, and this is a lesson to anyone releasing a device or game: players need clear accessibility information ahead of release to make informed buying decisions and secure any help that might be needed.

Nintendo’s shift to greater accessibility is welcome. But in restricting the flow of information before release, the win is tempered somewhat by Nintendo’s willingness to keep its players in the dark.

The Access-Ability Summer Showcase returns with the latest in accessible games

2025-06-04T16:23:21-04:00

Spray Paint Simulator.

Now in its third year, the Access-Ability Summer Showcase is back to redress the lack of meaningful accessibility information across the ongoing video game showcase season. As we see progress broadly slow down, it’s also a timely reminder of the good work that’s still happening in pursuit of greater accessibility in gaming.

“At a time where we are seeing a slowdown in accessibility adoption in the AAA games space,” organizer Laura Kate Dale says, “we’re showing that there are interesting accessible games being made, games with unique and interesting features, and that being accessible is something that can bring an additional audience to purchase and play your games.”

The showcase is growing, too. In 2025, it’s longer, more packed with games, and streamed concurrently on Twitch, Youtube (where it’s also available on-demand), and on Steam’s front page. That growth comes with its own challenges — mitigated this year by Many Cats Studio stepping in as sponsor — but the AA Summer Showcase provides an accessible platform in response to the eye-watering costs of showcasing elsewhere (it has previously been reported that presenting trailers across Summer Game Fest starts at $250,000), while providing disabled viewers with the information they need to know if they can actually get excited about new and upcoming releases.

It’s lesson Dale hopes other platforms might take on board. “I grow the show in the hopes that other showcases copy what we’re doing and make this the norm,” she says. “If I could quit hosting the AA Summer Showcase next year because every other show in June committed to talking about accessibility as part of their announcements, that would be wonderful news.”

To help that along (sorry, Laura, don’t quit just yet), The Verge has collated the games featured in this year’s Access-Ability Summer Showcase below.

Visual accessibility in focus

A major theme that emerged from this year’s showcase is color blind considerations. The showcase kicked off with ChromaGun2: Dye Hard by Pixel Maniacs, a first-person color-based puzzler. In its color blind mode, colors are paired with symbols for better parsing and those symbols combine when colors are mixed.

A similar spirit is echoed in Sword and Quill’s Soulblaze, a creature-collecting roguelike that’s a bit of Pokémon mixed with tabletop RPGs (dice included). It also pairs colors and icons, adding a high level of customization to color indicators, difficulty, and an extensive text-to-speech function that supports native text-to-speech systems and NVDA.

Later, Gales of Nayeli from Blindcoco Studios, a grid-based strategy RPG, showcased its own color blind considerations and an impressive array of visual customization options.

Room to breathe

A welcome trend carried over from last year, games continue to eschew time pressure and fail states. Dire Kittens Games’ Heartspell: Horizon Academy is a puzzle dating simulator that feels like Bejeweled meets Hatoful Boyfriend. Perhaps its most welcome feature is the ability to skip puzzles altogether, though it also features customization for puzzle difficulty. Sunlight from Krillbite Studio is a chill hiking adventure that tasks the player with picking flowers while walking through a serene forest. It does away with navigation as you’ll always be heading the right way, while sound cues direct you to nearby flowers.

This year’s showcase featured two titles from DarZal Games. Quest Giver is a low-stakes management visual novel which casts the player as an NPC handing quests out to RPG heroes, while 6-Sided Stories is a puzzle game involving flipping tiles to reveal an image. The games were presented by Darzington, a developer with chronic hand pain who develops with those needs in mind and, interestingly, with their voice (thanks to Talon Voice). Both games feature no time pressure, no input holds or combos, and allow for one-handed play.

Single-handed controls are also a highlight of Crayonix Games’ Rollick N’ Roll, a puzzle game in which you control the level itself to get toy cars to their goal without the burden of a ticking clock.

Highlighting highlights

Speaking of highlights, this was another interesting trend to emerge from this year’s showcase. Spray Paint Simulator by Whitethorn Games is, in essence, PowerWash Simulator in reverse. Among a suite of accessibility features that help players chill out and paint everything from walls and bridges to what looks like Iron Man’s foot, the game allows you to highlight painting tasks and grants a significant level of control over how those highlights appear and how long they last. Whitethorn Games provides accessibility information for all its games here.

Cairn, by contrast, is a challenging climbing game from The Game Bakers which looks like transplanting Octodad onto El Capitan. As it encourages players to find new routes up its mountains, the game allows players to highlight their character’s limbs, as well as skip quick reaction minigames and rewind falls completely.

Highlights are also important to Half Sunk Games’ Blow-up: Avenge Humanity, in which players can desaturate the background and customize the size and tone of enemy outlines to make its chaotic gunplay more visible. Something Qudical’s Coming Home, which debuted during the showcase, also offers in its tense horror gameplay as you evade a group of murderers. You can switch on a high-contrast mode that highlights objects to distinguish them from the environment (including said killers).

Unsighted

If this year’s been challenging for accessibility, it’s been even more disappointing for blind players when it comes to games that are playable independently. The AA Summer Showcase, however, included an interlude showing off the best titles from the recent Games for Blind Gamers 4, a game jam in which all games are designed with unsighted play in mind and judged by blind players. Four games were featured: Lacus Opportunitas by one of last year’s standouts shiftBacktick, The Unseen Awakening, Barista, and Necromancer Nonsense. This was chased by a look at Tempo Labs Games’ Bits & Bops, a collection of rhythm games with simple controls and designed to be playable in its entirety without sighted assistance.

A difficult subject

Accessible indie games often favor the cozy, but this year’s AA Summer Showcase brought a standout game that bucked that trend. Wednesdays by ARTE France is a game that deals with the aftermath of childhood abuse. That’s certainly in keeping with the host of trauma-driven indie games out there. Wednesdays, however, positions itself as a more hopeful examination of that trauma, both through its visual novel style memories and theme park manager gameplay.

Like so many of the showcase’s games this year, Wednesdays includes mitigations for color blindness — though no essential information is tied to color in-game — as well as a comprehensive text log for cognitive support, manual and automated text scrolling, and customization options for cursor speed, animations, fonts, inputs, and more. Better yet, all those options are displayed at launch and the game always opens in a windowed mode to allow for easier setup of external accessibility tools.

It’s a curious title, for its wealth of accessibility features, naturally, but also for how it handles its subject matter — because maybe we all need a little more hope this year, yeah?