London startup Electronic Materials Office (EMO) has set out to change that with Altar II, a fully mechanical keyboard that’s just 4.75mm thick. Think about it: that's less than half a centimetre. Super-thin or what?

Machined from a single block of aluminium, the Altar II packs in haptic feedback, a built-in speaker, red backlighting, up to 30 days of battery life and a magnetically detachable dial for volume and media. It’s designed first and foremost for Mac users, though it works with Windows and Linux too.

Altar II launches on Kickstarter on 10 November, and will retail at $349. But if you reserve one on EMO’s website now for a refundable $1, you can buy it for $249. Delivery is expected in early summer 2027. I asked founder Andrew Emerson how it all came together.

How did it all start?

Mechanical keyboards were a lockdown hobby for me. We all had our weird things to keep us busy I suppose. Most people baked sourdough but I became fixated on mechanical keyboards. I was obsessed with how they felt, and how far people would go to get their keyboards just the way they liked. A keyboard is also something you use every day, so I figured it was a worthwhile investment.

But after trawling the internet and Reddit, watching all the reviews, and buying a few keyboards, I just couldn’t find anything I liked. They all looked horrendous. Creaky plastic, glossy, gigantic, obnoxiously loud, and RGB (I hate RGB). It was really dreadful. So I decided that I would try and make my own.

The learning curve was huge. CAD, 3D visualisation, DFM, CNC milling, electrical design, firmware. This was all very much outside my wheelhouse as a software designer at the time. But it was a lot of fun to learn everything and pour myself into the project on nights and weekends. Eventually I came out the other side with a functional prototype of what would ultimately become Altar I.

I stood up a small website for Altar I with an email sign-up field, posted it online and... crickets. Annoyed, I went for a quick lunchtime run. When I came back the website had blown up on Hacker News (a popular tech forum), had hundreds of comments, and by the end of the week there were 10,000 people on the waitlist. That month I told my boss I was quitting to make keyboards full time.

Why did you make the keyboard so thin?

Firstly I will say I am absolutely not chasing the title of “thinnest keyboard ever”. The slim design of Altar II is the result of a few things.

Typically, keyboards come in two flavours. The first is thin keyboards, like the Apple Magic Keyboard, Logitech MX Keys, or the keyboard in your laptop. These keyboards usually look really nice on a desk. The switch mechanism they use is a rubber dome that collapses a membrane, which feels squishy, indistinct, and generally pretty bad.

The other kind of keyboard is the mechanical keyboard, which is typically tremendously thick because it houses a large mechanical switch. These switches have springs and electrical contacts, and when you press them, they feel really nice and tactile.

Altar II takes the best aspects of both these worlds and puts them together: an amazing-feeling mechanical switch, housed in a very slim chassis. That is the true innovation of Altar II. You can have your cake and eat it too. It also helps that a thin keyboard is more ergonomic and easier to transport.

There is a philosophical aspect to making Altar II as slim as possible too. I believe that technological advancement comes in two modes: miniaturisation and higher performance. I think we tapped out on increasing performance on computer keyboards in the mid 1980s, which leaves us with miniaturisation.

Making a very thin mechanical keyboard wasn't enough of a challenge though, I also wanted to pack in as much technology as possible: wireless Bluetooth, haptic feedback, a speaker, a magnetic dial. To me that really represented a technological leap and a challenge.

Are you still doing all this by yourself?

Altar I was very much a one-person passion project, with a little help from a few people along the way. But with Altar II the ambitions have been raised significantly. The device is an order of magnitude more complex and it really swings for the fences. To do this the team has expanded from one person to a small team of very talented people.

I look after the industrial design, graphic design, the website, conceptualising, marketing, and day to day running of the company. Steven and Jessica Nersesian, of Metric Designworks, are the people who make the designs reality, looking after CAD, DFM, mechanical engineering, and being the link between us and our manufacturers.

They have worked on products like the Playdate and various Teenage Engineering products. Kurtis Lew is our embedded engineer and is responsible for firmware, and Moritz Plattner is our electrical engineer and looks after things like our circuit board design.

Where exactly one person’s role ends and another’s begins is not clear in reality. Everyone has opinions and thoughts on everyone else’s domain. Altar II is a single product, and all these domains live within it. People using Altar II don't care whose idea a detail was, they just care that it feels right.

As for my background, before EMO I spent about ten years as a software designer across automotive, music, healthcare, and fashion. I’m also a traditionally trained graphic designer. At design school I focused on editorial design and typography. I think you can see some of that in EMO’s products.

Overall, Altar II is a much more professionalised operation than Altar I. Rather than being a pure passion project, we are now building Electronic Materials Office to be a long-term sustainable business.

The M-dial is genius: how did you come up with it?

M-Dial is a magnetically detachable, swappable, clickable dial for Altar II. It sticks on to the keyboard in a very satisfying way. There are two M-Dials included in the Altar II box: a standard issue size and a low-profile variant for travel. We’ll also be releasing the specs for M-Dial publicly, so you’ll be able to 3D print your own in whatever shape and size you like.

Why a dial at all? When you use a dial to change the volume on your computer, it just makes sense. It’s so much more intuitive than using buttons. There are a few extra niceties built into M-Dial too: you can click the dial once to play or pause media, double click to skip, triple click for previous, and long press to invoke Siri on macOS. These are all customisable in our companion app as well.

When we launched Altar I we got some feedback that although the overall shape of the device was easy to transport, the dial would sometimes get caught on things like people’s bags. So now it’s detachable. It also lets us slim down the Altar II packaging to make it more efficient, which is better for the environment.

What are the main changes from Altar I?

There’s a question to answer here first: Why make a successor to Altar I at all? When I sat down at my desk after shipping Altar I, I put it next to my Apple Magic Keyboard, and I thought: 'Hmm, I think I have unfinished business here.' Don’t get me wrong, I adore Altar I, and I’m incredibly proud of it. I thought with all the learnings from Altar I, it'd be possible to go even further and make an even more impressive product that could supplant the likes of the Magic Keyboard. But I would only make a successor product if it could be a true generational leap and an improvement in every way over Altar I.

And that’s what Altar II is. It really is a generational leap. Almost nothing carries over between the two devices, we’ve taken the opportunity to build everything from the ground up.

The most noticeable difference is the thickness. Altar II is almost 70% thinner than Altar I. Which is especially impressive when you consider that Altar I is itself a “low-profile” keyboard!

The other notable differences are that the switch platform is completely different, so the keycaps and stabilisers are totally redesigned. Of course the Altar II dial is magnetically detachable, and Altar II has a backlight, haptic feedback and a built-in speaker. We really threw everything we had at Altar II, and took so many lessons from Altar I as well.

I was going to be mainly writing and editing, rather than designing or making videos. So although it was basic, it did everything I need. A stopgap, I told myself, until I could afford something “proper”.

I said all this to myself, I have to add, through gritted teeth. This machine was a huge step-down, power-wise, from the top-of-the-range desktop Mac I'd been able to access as associate editor of Creative Bloq. So what happened next was a total shock. My little Chromebook turned out to be the most enjoyable computer I’ve ever owned.

Why? Partly because it was so simple. Pretty much all there was on my screen was a browser, and it turned out, that was pretty much all I needed. But also, because it had a touchscreen. And oh my gosh, did I love using that.

Suddenly, scrolling through a feature, tapping a link or nudging an image into place felt natural in a way a trackpad never had. It felt like walking on air. I was getting more done, more quickly, but it all seemed so effortless.

Goodbye to my Chromebook

It didn’t last, though. As time went on, I started to do different types of work, some of which needed software that wouldn't run in the browser. So over time, I moved to Windows and eventually back to a Mac.

Today, macOS is the best home for the creative tools I rely on. But because there are no MacBook touchscreens, I’ve never experienced that sense of flow I had on that little Chromebook. Which means I'm super-excited that Apple is about to make my dreams come true.

Bloomberg’s Mark Gurman, who's rarely wrong about these things, has said that Apple is preparing to unveil OLED touchscreen versions of the MacBook Pro at an event on 28 October. And the actual laptops could theoretically be in shops as soon as 6 November.

That’s a shock, because suspicions of touchscreens has been in Apple's DNA ever since Steve Jobs called touchscreen laptops ergonomically terrible back in 2010, stating that stating that "touch surfaces want to be horizontal". Similar remarks have been made since, by software chief Craig Federighi in 2018 and John Ternus, now Apple’s CEO, in 2021.

A mixed experience

I get their point, in theory, at least. Holding your arm up to poke a screen all day isn’t ideal, and nobody wants a display covered in fingerprints. But what I think they're missing is that no one uses a touchscreen as the sole input on a laptop. On my old Chromebook, I didn’t abandon the keyboard and trackpad. I just used whichever one was easiest and quickest in the moment.

And I think for a broad range of creative work, the potential is huge. Think about scrubbing through a timeline in your video editing software, cropping a photo in Photoshop, pinching to zoom on a layout, or flicking through a client’s PDF proof. These are the kind of tasks that a trackpad is just not designed for. And exactly the kind of tasks that a touchscreen is.

So I'm excited. Although as ever with Apple, I'm not sure I'll be buying one straight away, because the price is likely to be super-high, especially with memory costs soaring right now. I'm also happy for other people to discover the bugs that will inevitably come with a new release, and for Apple to fix those over time at their expense.

Over time, though, assuming prices come down, I reckon it's going to be a winner. Because in my view, Macbooks can't be beaten in terms of specs. But using a laptop for creative work isn't just about specs, it's about user experience. And if Apple can recapture that sense of joy I had from that first Chromebook, I’ll be a happy man indeed.

A spokesperson for Goth has called the viral ruse "disconcerting on a fundamental level," pointing the blame at Instagram for allowing the false image to circulate. But could pointing the finger at Meta just be scratching the surface of the issue?

The AI images of Goth appeared on the verified @miagoth_ Instagram account, which, despite the name, has no affiliation with the actress. Believing the fan account was real, numerous news sources reported on the AI images, further spreading the false information.

“The fact that this fan account meets Instagram’s criteria to qualify for a ‘blue check’ is disconcerting on a fundamental level,” says Christine Tripicchio, spokesperson for Goth. “Beyond that, it is incomprehensible that Instagram evidently prioritizes the dissemination of forged and manipulated images over the protection and integrity of a public figure’s likeness.”

While Meta certainly has its own issues with AI slop, it's worth bringing into question how audiences and 'news' sources alike are quick to be trusting of AI imagery. AI labelling is simply not keeping up with the quantity of slop content. And when fan accounts can easily charade under the status of a blue checkmark, is it surprising that Goth's AI likeness went viral?

There's an undeniable level of parasociality between celebs and the public when it comes to social media, making stars appear more accessible and relatable. The internet loves a bit of drama, and the refinement of AI technology only makes it easier to stir the pot; Mia Goth's 'selfie' was too good to resist.

Whether the AI image of Goth was created as a playful (albeit inappropriate) gesture to her divorce news or with the more malicious intent of creating fake drama, the rise of AI celebrity hoaxes is concerning. With recent deepfake drama like the Taylor Swift TikTok scam, it's unsurprising that stars are increasingly taking a stand against AI deepfakes by trademarking their voice and likeness.

The production car won't arrive until at least 2028, may not be road legal and can be reserved by invite only, reportedly requiring a deposit of $500,000. For anyone not able to part with that amount of money on a reservation agreement that explicitly states that there's no guarantee of delivery, Ikea thinks it has a safer, more affordable option that's available in store today.

Ikea's marketing departments already have a reputation for ingenious quick thinking. We recently saw it jump on 2026's Wicker trend faster than you can say "Alexander Skarsgård". But Ikea Taiwan's response to the Tesla Roadster concept's flying abilities is as impressive as the car's own 0.9-second 0-60mph record.

Its answer for a Tesla Roadster Rocket Edition setup suitable for any home is to combine a 4-piece $12.99 Ikea Lillabo toy car set with a Brokglim bath towel. "Real life, baby," it says in the Instagram post above, echoing Tesla's caption in its own post below, before adding "Safe, affordable."

I'm not sure how you're supposed to fix the towel in the air like that, and I don't think the Lillabo has a fan-based suction system to keep it on the track, but kids have even more imagination than Tesla; I'm sure they can work something out. You could also try a slot car race track.

The good news for non celebs is that the Sennheiser 480 Pro is reduced by £40 from £339 to £229 at Wex Photo and video. We've tested and reviewed dozens of headphones and this is our favourite closed-back option so far, as reviewed in our guide to the best wired headphones.

Why the Sennheiser 480 Pro are our favourite closed-back wired headphones

The Sennheiser HD480 Pro only came out in the summer of 2026, but they immediately stood out in the world of closed-back wired headphones. They're very well tuned, incredibly comfortable and they look cool too. While designed for professional studio use, they're also a treat for audiophiles or anyone who wants to discover what makes a neutral wired earphone so special.

The clear, largely uncoloured sound profile evades the issue of overly boomy bass and unrealistic timbre that haunts the closed-back headphone space. Passive noise isolation is well executed and the mature tuning fits all kinds of music at all volume levels.

While the music video itself is a masterpiece of storytelling, set and character design, a tiny animated section has been accused of using AI. Having previously denounced the use of AI in her music, the allegations come as a shock for many of Rodrigo's fans.

The controversial scene in question features cartoon Rodrigo and Roan flying past on a wheel-of-fortune-inspired spaceship. Observant fans were quick to notice that one shot appears to remove one of Chappell's hands, while Olivia appears to have inherited her missing extremity.

Is it the work of AI, or simply a bit of good ol' human error? Either way, fans were dissapointed by the blunder. "Not the AI..." one fan wrote, while another added, "No way, and I just liked a tweet cheering her for not using AI."

Recently telling The Hollywood Reporter, "I would never use AI in any of my music. Absolutely no,” the AI allegations came as a surprise to many. Some fans came to Olivia's defence, claiming, "This mistake happens in 2d animation all the time," while another wrote, "Honestly, I think the artist might’ve just messed up the colors because honestly it’s the only mistake I saw in those drawings and it’s a minor mistake."

For more design news, check out why Tyla was hit with heated AI allegations over her new animated music video or take a look at Apple Music's new logo that's giving major mixed messages.

It might seem sensible at this for anyone else to stay out of it and let them sort it out between themselves, but not Angry Birds 3. The upcoming movie has jumped on the controversy to cheekily remind the world that it's due for release at cinemas in a couple of months.

If you somehow missed the online scuffle of the week, the controversy is basically a “he said, she said” row over a $15,000 mechanical bird, and whether Kelly Clarkson was or wasn't treated like she couldn't afford it. The singer says that when she asked about the bird at Art of Play, a member of staff offered to find her a cheaper alternative. Writing on X, she compared the experience to a scene from Pretty Woman in a post that's received over 11,000 comments.

The shop then responded to complain that it was being hounded by Kelly Clarkson fans and claimed that the bird, called Molly, hadn't been for sale in the first place.

It's not an obvious branding opportunity, but Angry Birds 3 quickly saw the potential. In the animation above posted on the same platform, Chuck, AKA Yellow Bird, is cast in the role of Molly. He doesn't take sides, instead sitting back to watch others being angry for a change, but the animated poster design pokes fun at Molly's price tag, pointing out that we can see Chuck for the price of a movie ticket.

In a pre-programmed sequence with no driver, the two-seater concept vehicle performed a kickflip off a half-pipe before gliding to the ground. It was later shown to lift off from stationary and 'hop' over a row of parked vehicles.

"Flying cars are real," Elon Musk said, perhaps referring to rival company Alef's Model A?

Tesla's next-gen Roadster electric sports car has been in development for around a decade. Musk wrote on Twitter back in 2018 that there would be a "SpaceX option package" featuring "10 small rocket thrusters" to improve acceleration, top speed, braking and cornering. "Maybe they will even allow a Tesla to fly," he wrote then.

Delivery was initially slated for 2020. The company now says 2028.

The concept car at yesterday's event was fitted with rechargeable "ultra high-pressure air tanks" to achieve vertical propulsion and to cushion landings. It was able to cling to the steel ramp in the demo thanks to a fan suction system.

The thrusters can also fire rearwards to increase acceleration. This allowed the car to reach 60 miles per hour in 0.81 seconds, Tesla said.

Tesla chief designer Franz von Holzhausen has said that the Roadster's angular design and scissor doors were inspired by an "aerospace aesthetic", while Musk described the concept vehicle as "what happens when you have a collaboration with the world's best car company and the world's best rocket company".

"Success is not guaranteed" but "it will probably work" said the CEO, who in 2022 claimed that the Cybertruck would be "waterproof enough to serve briefly as a boat, so it can cross rivers, lakes & even seas that aren’t too choppy".

The concept Roadster's gymnastics were impressive, but whether there's any real-world use for them is another matter. Requiring heavy tanks packed with compressed air, along with the valves and thrusters needed to contain and release it, the 'flying' can only last a matter of seconds before the tanks need to be refilled. That inefficiency makes the flying Roadster feel like a very elaborate gimmick.

On the Tesla website reserving a basic version with no boosters requires a £50,000 deposit while a “Thrust” version with rearward boosters for acceleration requires a $250,000 deposit. A single-seater “Rocket” variant with a “full flight system” is labelled as being available only by invitation. It's not clear whether that variant would be road legal at all, but buyers certainly won't be jumping over traffic or driving up buildings along public roads.

The bigger question is whether yesterday's theatre matters for regular customers and whether it can do anything to help repair confidence in the Tesla brand. Since the announcement of the Roadster in 2017, competitors like Rimac, Porsche, Xiaomi and BYD have already delivered high-performance electric cars, nudging Tesla out of the spotlight in a category it once hoped to dominate. There's also the upcoming Jaguar Type-01, Bentley Torcal and Ferrari Luce.

It seems the markets may have become skeptical of the CEO who cried wolf. Tesla shares remain down by 4.5% on last week and over 16% for the year to date.

The Roadster isn't the first design for a car that can hover a bit. Alef Aeronautics has begun limited hand-built production on its Model A for controlled testing.

“Welcome, Tesla, to the flying car race; we’ve been here alone for eleven years,” CEO Jim Dukhovny commented after Tesla's event. The California-based company posted a video on YouTube showing its Model A 'flying' over a Tesla Roadster. "This flight marks 1000th recorded flight by Alef flying car," it claimed.

While Tesla's Roadster uses compressed air thrusters for short hops and glides, Alef's much lighted Model A uses distributed electric propulsion via eight internal, enclosed propellers integrated into a mesh-like shell.

The banner recreates Rockstar’s Grand Theft Auto VI cover art, with Lucia and Jason front and centre and joins the growing marketing campaign that started in earnest with yesterday's GTA 6 launch trailer. Look closely, though, and two weapons have vanished: the pistol Lucia holds in the original, and an assault rifle in another panel. Kotaku spotted the changes after fans started comparing the finished banner with the official image (X user Rockstar Mags had been following the ad's creation and took photos).

On paper, removing a gun from an image sounds like a five-minute job. In practice, it leaves Lucia with no pistol and no hand to hold it, and the character in the lower image with an arm thrust into thin air, as if they’ve just spotted a particularly interesting cloud, so a canny zoom has solved the problem.

But these guns aren’t background clutter for GTA 6; Rockstar’s art centres on its characters, their poses, and what they’re holding, and the weapons give those silhouettes attitude and tell us a little about who they are and what we can expect from the game. Take them out and the balance tips, leaving gestures that no longer quite make sense. It's a similar, though perhaps less dramatic, edit that made Amazon's shockingly bad James Bond posters so appalling by removing Bond's famous pistol.

No guns, no problem

It’s a handy reminder of how tricky it is to adapt key art for different spaces. An image designed for game boxes, websites, and trailers won’t always hold up when blown up across the side of a building, especially when bits have to be cut out because locals don't want a 30ft gun staring down at them as they pop out for a croissant.

Other GTA 6 murals in the US reportedly keep their weapons, so this looks like a Paris-only tweak for the local area, which also shows how aware Rockstar is of the impact its campaign can have. I get the feeling Rockstar doesn't want to start culture-war fires this time and wants people to focus on the game.

That said, it’s a strange compromise for a game defined by crime, chaos and bad behaviour. A gun-free GTA is a bit like advertising a racing game without the cars, or James Bond without a Walther PPK. And once you’ve spotted those empty hands, you can’t stop looking for them.

However, Powerade's response, the 'Never Just Sports' campaign, adds one word and turns the whole furore into a different debate about women in sport. The Ogilvy-made campaign has been running on digital screens in New York's Times Square, directly opposite the Novig billboard, with images of women swimming and clearing the high jump, and it never mentions Sweeney or Novig; it just nicks the phrase, changes it just enough to make a difference and trusts you to remember the original – but it's also right there next to Sweeney, so it's not hard.

What interests me more than who has the better line or how clever Coca-Cola feels right now is why an ad with a Hollywood actor in a cheeky pose ended up as a row about how women are shown in sport at all, whether we really are returning to 'ads' old dirty tricks', and whether Millennials and Gen Z read the same pictures differently, because I suspect the answer is messier than one lot being offended and the other shrugging.

One word, a very different picture

Novig's idea is commercially clear: it wants to stand out as a sports-only prediction market app, unlike platforms that let you trade on politics and world affairs. Its CEO says Sweeney helped develop the campaign's creative, and that it's an eye-catching way into something few people care about (I mean, really?). The gag is that she's wearing "just sports", footballs and other kit doing the job of clothes, a play on words that's as subtle as "good jeans".

Just like American Eagle's slogan, it worked for what was needed at the time, with plenty of attention both good and bad, and Novig reporting a rise in activity on its platform: it claims first-time depositors rose more than 218% in the 20 days around the 9 September launch, and the Instagram reel passed 37 million views in four days. (Honestly, if you, like me, never felt clear on what a naked Sweeney and Novig were selling, I'd argue that alone could have driven the search.)

By contrast, Powerade's athletes skip the Austin Powers prancing-around-his-love-pad routine and use their bodies for sport. The physicality remains, shown as the result of training and competition, with no '90s lads mag revival.

Yet the bad-ad-versus-good-ad framing is too easy, because women can be glamorous without undermining what they've achieved; Serena Williams and Naomi Osaka have done just that in their careers. So what are these ads actually saying, beyond getting people to look? And who comes out on top?

Gen Z versus Millennials (again)

The easy answer is Millennials raised on lad mags and tabloids against a Gen Z fluent in social media and the language of representation, but plenty of Millennials hate sexualised advertising, plenty of Gen Z see Sweeney confidently making her own choices (she's Gen Z herself), and a fair few probably think it's ordinary attention-seeking marketing, so you can't pin a view on a whole generation.

Millennials watched brands move from traditional ads to blogs, Facebook, and YouTube, and saw influencers take over, while Gen Z grew up with influencer celebrities selling them everything from games to makeup, with brand deals and staged images all muddled together. Gen Z has its own cues for knowing when they’re being sold to (the #ad tag, the suspiciously candid photo), plus the vocabulary to call it out on the spot. Which is how the same billboard can be a knowing joke with one lot and the oldest trick with another.

So we're in a place where one person sees Sweeney owning her image and getting paid on her own terms, while another sees the old shortcut of using an attractive woman to grab attention, and both can be right, since she can have agency over her image while a brand still uses it to sell something. Powerade is selling too, using a live row to position itself alongside female sports achievement, and it's playing the same attention-grabbing game.

We've been having the same debate since tobacco ads became a thing in the 1870s, and I'm old enough to remember Madonna making a career out of it, from Material Girl to the Sex book. So what's changed?

We’ve been here with Sweeney before

Sweeney's American Eagle 'Great Jeans' campaign had already been through this, with arguments about beauty standards, race, genetics and political identity, because the "genes" pun plus a blonde, blue-eyed actor led some people to read it as being about far more than denim, and it showed how an ad can pick up a meaning its makers never intended and can't control once the internet gets hold of it. Adweek's Weissbrot called it a return of “male gaze” advertising, and many people read it as the end of the brand-purpose decade.

In many ways, that's what American Eagle can teach us about branding in the 2020s, because the Novig row is really about how women are presented in ads, here in sports, and whether sexualised imagery undermines female athletes, and whether criticising the ad is an attack on Sweeney's right to make her own choices. When Madonna did it, the provocation at least started a conversation; now polarisation turns everything into a shouting match.

Female athletes have criticised it, from Ariarne Titmus to Felicity Palmateer and Gracie Kramer; Novig has defended it and kept promoting it, and Powerade elbowed its way in with a polar opposite view, one that, let's be honest, is happy to play off the furore too, only from a different angle: women powering through the water and lifting the weight.

Reactive ads are easy

Powerade's timing has an obvious commercial angle: it chose to wade into a mess, make a point without building awareness from scratch, and the one-word swap hands journalists like me a simple way to describe the idea and write about how clever it all is.

But Powerade has a history of campaigns focused on mental and physical well-being, hard work, and training, which gives this context beyond a swipe at a rival brand. The response has landed with at least some of the people it's aimed at, too: a female athlete working in marketing wrote in PRWeek that many women in sport, herself included, greatly appreciated it.

A reactive ad like Powerade's still has to be judged as advertising, so it's fair to ask whether it gives female athletes a real platform, changes how people see their achievements, or just gets remembered as a clever troll to another saucy Sydney Sweeney campaign. The same goes for any brand taking a stand, because people are wary of companies that adopt a cause only when it becomes useful.

Powerade could have issued a statement or run a montage of sporting achievements, but instead it took a phrase with a public meaning and flipped it, short enough for a bold ad and with a second life for anyone who's seen Sweeney's butt on a billboard and Novig's pun. For brands, an empowerment slogan, a provocative image or a well-timed reply only gets you so far; what lasts is whatever holds up once the joke or shock wears off.

Will Gen Z and Millennials agree on what that is? Probably not. As a Gen X, I've seen it all before too. Move on.

Featuring one of the most iconic typefaces of all time, the design has surprised creatives thanks to its unexpected appearance on the single's branding. But while some are clutching their metaphorical pearls, others see the controversial font choice as a return to form for the pop music design world.

So what typographical sin has ol' Tate committed? She's used an ultra light Helvetica font. I know, the horror.

"She will pay for her crimes," one X user wrote, while another joked, "It's grotesque tbh." While one critic claimed "Tate McRae has NO REASON to be using an ultra light Helvetica font," others gestured to album covers of the early 2000s, like Britney's I Love Rock 'N' Roll, and Madonna's Bedtime Stories. Remember those? Exactly... that's why we don't use ultra light Helvetica, folks.

For more music news check out why Kpop group IVE's new album cover felt so unsettling or take a look at Phoebe Bridgers' 'internet-free' album campaign.

Below, he talks us through four examples of his work.

Wrath Unleashed

“This creature contest entry was a specific challenge for me to switch to realism, leveraging my professional skills in texturing and dramatic, mood-driven lighting.”

Pastel & Bones

“I aimed for a natural flow within a highly stylised look, paying special attention to the hair shapes while maintaining a rapid, decisive sculpting process.”

Project Zero

“This piece focuses on complex metal panelling, precise armour modeling, and glowing emissives. I wanted to balance futuristic cybernetics with an ornate, regal silhouette.”

Fractured Personas

“Inspired by Ev Ganin’s concept, this piece explores character identity. I focused on telling a deep story while keeping my process streamlined through sculpting and hand-painting in ZBrush.”

You can see more of Skander’s work on his ArtStation profile

This article originally appeared in ImagineFX. Subscribe to ImagineFX to never miss an issue. Print and digital subscriptions are available.

Today's drawing tablet deals

See our guide to the best drawing tablets for more options.

Creative Bloq is now easier to access than ever before with our on-the-go app, which brings you all the content you know and love from our website, but in a super-streamlined design.
Download the Creative Bloq app for iOS
Download the Creative Bloq app for Android

With the new iOS 27.2 update, Apple debuted a new icon for Apple Music, which caused quite a stir among users. Monochromatic and confusingly basic, the new logo design feels like a regression to Apple's minimalist roots, but not all fans are on board.

The new design is essentially a black-and-white version of the previous music note logo, with slightly bolder lines. Completely flat and admittedly a little characterless, the new look confused fans who believed Apple was on the road to embracing a quirkier look. With the recent leak of the new Apple Music wordmark, it appeared the brand was waving goodbye to minimalism, but now I'm not so sure. Talk about mixed messages...

"This ain’t it chief," one X user wrote. "The new Apple Music branding is miserable how do you sap a thing of all of its colour like this and think it's okay?" another added. "My best guess is that they want to turn Apple Music into more of a brand to compete with Spotify, and this new mark will allow artists to provide the color and emphasis in billboard ads," one X user theorised.

For more Apple news, check out the chaotic Apple HomeView advert that's proving to be stressful for viewers or take a look at Apple's perfect iPhone Duo TikToks that nail foldable phone marketing.

Speaking to journalists, including The Nerds of Colour, he explained that the film’s storyboards and constantly moving camera meant the sets couldn’t always make physical sense. One example was a battlefield shot where a character on one side sees someone on the other, who in real life would be smaller than a pixel. His team mocked up options, tested them in camera, and got many of them wrong because the problems were so unconventional. They kept going until it worked. He’s done this a long time, he said, but with challenges that raw, “you just make mistakes, and you learn from them.”

What can we learn from Lowry? Most of us only see the finished version of a painting or project: the polished portfolio, the behind-the-scenes reel where everything looks smooth. Lowry’s story is the part those leave out. If your current illustration, animation, or design isn’t working, it may be at the stage where the first attempt failed and the second hasn’t happened yet.

If LAIKA's artists can spend seven years making characters move millimetres at a time, making mistakes, learning, moving on and improving, then maybe the rest of us can cut ourselves a bit of slack, too. You can see every inch of that work at the Design Museum exhibition, too.

Source: Mike Manalo, The Nerds of Color, 21 September 2026. Visit LAIKA for behind the scenes insights for Wildwood and more.

Best LAIKA art books:

After the great Hexed vertical orientation conspiracy theory and the character design disappointment, the latest issue concerns an unexpected addition to the cast. It's reminding some people of where they think Disney took the movie's concept from.

Throughout the build up to the release of Hexed, some fans of The Owl House, Dana Terraces' Disney Channel animation that was cut short, have insisted that they think the cancelled show inspired the movie. Disney chief creative officer Jared Bush denied the idea at D23, but it's come up again now that Disney has named Sarah-Nicole Robles as part of the voice cast. The actor previously voiced Patience in Knights of Guinevere and Luz Noceda in The Owl House.

"They're not even trying to hide it any more," one person complains.

I'll point out that Sarah-Nicole is a fine actor and previously voiced Señora Ozma in Encanto. I don't think Disney is trying offend anyone here. But it shows the sense of grievance that remains among TOH fans.

Meanwhile, people continue to dissect the Hexed character designs and to diagnose an acute case of Disney same-face syndrome. Attention has turned to the question of why mother and daughter Alice and Billie Doe look the same age.

Disney has a 'rule' that makes it almost impossible for its animations to show a mother who looks a believable age, one person believes. It all goes back to a 1943 paper by Konrad Lorenz "about why babies look cute", apparently: large eyes, a high forehead, small nose and chin and round cheeks.

"Disney will even draw an old woman with real wrinkles when the role calls for it, like Gramma Tala in Moana. The mom in her forties is the one who falls through the cracks, and Disney Animation hasn't had to solve that much, because it almost never puts a mom at the center of the story," they suggest, writing on X.

The movie's directors, Josie Trinidad, who was head of story on Ralph Breaks the Internet and Zootopia, and Jason Hand, co-director of Moana 2 head of story on Encanto, have said that Billie is very different from Disney heroines like Moana or Mirabel, with a sharp and more dangerous personality.

The ensemble voice cast also includes Henry Winkler, Dan Stevens, Rachel Dratch and Ariana Madix.

Hexed will be released on 25 November.

There are a lot of big talking points, including the depiction of Aslan and the way it seems the movie may take the novel's creation myth as a means to explore the act of creation and the nature of creativity. But among Narnia fans, there's also a lot of debate around a small background detail that appears at around the 1.48 mark in the trailer below.

As shown in the post on X below, there's scene towards the end of the trailer where Jadis appears to be using magic in London, something that wasn't possible in the novel. As a possible allusion to this deviation from the literature, she also appears to destroy a reference to one of CS Lewis's favourite hangouts.

The pub in the background is named the Eagle & Child, the same as a pub in St Giles', Oxford, where Lewis and JRR Tolkien's literary discussion group The Inklings would meet. There are lots of pubs in England with the same name, and the scene is presumably London, not Oxford, but it seems unlikely to be coincidental given the attention to detail in other scenes, such as when removal van logo offers a premonition of Fledge.

The movie also moves the setting around 50 years forward in time, from 1900 to the 1950s when CS Lewis was writing the novel, increasing the possibility for meta references.

Some have suggested that the inclusion of the pub is a symbolic gesture, or a "confession", intended to show that movie will break with the novel. It's been compared to how Stanley Kubrick used a yellow VW Beetle in his film of The Shining and briefly showed a damaged red Beetle as if to clarify that the movie is a different story to that of Stephen King's novel.

"They deliberately chose to destroy Lewis and Tolkien's favourite pub," one person complains on X. "She’s destroying his story, so why wouldn’t she destroy the pub that the Inklings met in?" someone else asks.

Well, every act of creation is first of all an act of destruction, Picasso supposedly said.

The Dell UltraSharp 27in 4K reduced from $799.99 to $649.99 at B&H Photo was a little on the expensive side at its original price, but a $150 discount brings it into sweet-spot territory given how many boxes it ticks. The IPS Black panel provides deeper contrast on a bright display with 4K resolution, solid colour coverage and a higher refresh-rate than most cheaper monitors can offer. It also has top-tier docking/Thunderbolt 4 connectivity.

If you're not bothered about the fast refresh rate and brightness, a solid budget 4K monitor option is the LG UltraFine 27BU650B reduced from $399.99 to $329.99.

Also see our pick of the best monitors for graphics artists to compare where these sit against other options.

• Browse monitor deals at Amazon

See more monitor deals below.

The environmental detail, lighting, character animation is of a quality that in many games would impress even as a pre-rendered cinematic, but Rockstar says the trailer was captured entirely in-game on a PlayStation 5. Yet, stunning graphics are only part of what makes the new trailer so powerful. Photorealism alone isn't enough for a game this big; the real magic is in the art direction.

More than just the end of the video game look, the new GTA VI trailer blurs the line between game design and cinematography. The saturated Florida-inspired colour palette, neon-soaked nightlife, sun-bleached suburbs and social-media-inspired imagery all work to build a world that's as cohesive and convincingly realised as a well-directed movie.

Rockstar has clearly thought about how it builds the story through environment design: the colour scripting in turquoise, pink, orange and gold, the scenes backlit by sunsets, the contrast between glamour and decay. The lighting is expressive rather than realistic, with humid atmospheric haze, neon spill lighting and strong silhouettes.

Meanwhile, Jason and Lucia's relationship is finally taking centre stage, pushing the Bonnie-and-Clyde narrative harder and giving the game a more character-driven identity. They look more natural and expressive than in earlier marketing: small gestures, posture changes and eye movements convey convincing relationship dynamics. References to Jason's military background and Lucia's time in jail build deeper motivation and a backstory with the power to draw in newcomers as much as lifelong GTA fans.

It's a story as much as a game. The impact is made complete by the introduction of Flower in a Hurricane, an original track from the French music producer Justice featuring Tame Impala, SZA, and King Gizzard & The Lizard Wizard. Included on Grand Theft Auto VI: The Album, this adds another rich cinematic element, and a blend of electro house and retro vibes that seems the perfect soundtrack for Leonida and Vice City, the game's fictionalised versions of Florida and Miami.

Highlights include the introduction of new ways to populate and simulate complex 3D worlds in C4D, more complete production workflows in ZBrush and faster rendering feedback in Redshift.

• Get Cinema 4D

Following the recent introduction of support for third-party chatbots in Cinema 4D, 2027.0 introduces a new Scatter System, delivering one of the most-requested capabilities. Artists can now procedurally populate environments with vegetation, rocks, debris, crowds and other assets while retaining precise artistic control.

Environment scattering has long been an area where Cinema 4D lagged behind dedicated ecosystem-building tools and some competing DCC applications. It makes creating forests, landscapes and large-scale scenes dramatically easier.

The new Scatter Generator and Scatter Tag support distribution, filtering, clustering and transform controls, while integration with Cinema 4D’s MoGraph toolset lets artists use familiar Fields and Effectors to shape, animate and influence scattered objects. The updates give artists a quicker and more approachable path to building dense environments without having to turn to complex workarounds or separate third-party scattering tools.

The other feature generating buzz is Liquid FX, which adds automatic foam, spray and bubble generation to Cinema 4D's simulation system. The secondary effects respond to the underlying liquid behavior, adding the visual detail that makes pours, splashes, waterfalls and other fluid effects feel more convincing without requiring artists to construct separate complex simulations.

Cinema 4D for iPad remains in beta, with a full release expected by the end of the year.

What else is new in Maxon's 2027 releases?

• ZBrush gets an integrated UV Editor with UDIM and texel-density controls, helping artists move more quickly from sculpting and painting into texturing and downstream production. 
• ZBrush for iPad gets non-destructive 3D Layers so artists can experiment more freely with sculpting, polypaint and masking on a layer, giving more control over sculpting changes, detail blending and creative iteration. 
• Redshift gets faster interactive rendering and new procedural fur, fiber and textile generation at render time.
• Red Giant gets new art-directable VFX and relighting capabilities.
• Autograph now has support for Model Context Protocol (MCP) as well as its first direct Cinema 4D connection: artists can copy and paste 3D elements directly from C4D, including animated elements, cameras, lights, nulls and compatible primitives.

That desire to expand from 2D to 3D isn't uncommon, as with the developers of Moonlighter 2: The Endless Vault, but instead of expanding the team, Deconstructeam partnered with Selkie Harbour, a two-person studio with 3D expertise. But before they even started making their sledgehammer-based narrative game, they also had time to experiment together, which has already been part of the studio's process between projects.

"Oftentimes if you get too experimental the result is that you fail, so we tried to accelerate that process by making many small things to eventually find out what's next," De Paco explains. "We took that same approach with 3D games, just making small games and releasing them for free on itch.io, so we could just fail freely. The most important part of developing a successful creative endeavour is developing a context that allows you to fail. So we created that space so we could fail several times and find what's interesting for ourselves to do in 3D."

From those projects, including the short interactive essay Many Nights a Whisper, which is a low-cost $2.99 / £2.49 to download on Steam, the two studios arrived at Virtue and a Sledgehammer, which has the player character returning to their hometown where "you are so angry, the only way through is just smashing everything" with a sledgehammer.

Although Unreal 5 has Chaos physics that allows for destructible physics, creating the destructible gameplay in Virtue and a Sledgehammer, which allows you to take your sledgehammer to any building or even the android inhabitants, required heavily modifying how its physics work.

"The logical system is totally different because the model is using a Voronoi noise on top of a mesh," Selkie Harbour's Guillermo Ferrando explains. "You have a static mesh, and you can subdivide it. When you start, you see a building that hasn't been broken; you see the static mesh is a proxy, but the moment you break it, it detects what's the closest Voronoi chunk that you've hit, generates a sphere around it, and then it breaks around that space in that moment. Instead of drawing a single mesh, you're drawing as many chunks as that mesh is subdivided into."

That kind of complexity can affect shadows and dynamic lighting, throttling the GPU, though fortunately there was an easy way to handle it. "The moment you break something, if you are playing in low or medium graphics, we kill the shadows," says Ferrando. "It's not noticeable when you play it unless you look at it, but otherwise the Steam Deck wouldn't be able to run it."

Even as a 3D game, Deconstructeam's 2D roots are still present in other ways. Take the dramatic framing of its cutscenes, which De Paco admits was directly influenced by indie narrative game 1000xResist's cutscenes, but even more so by comic books.

"To me, it's like an illustration, and I need to have in my mind where the text is going to come up and where the characters are," he explains. "It would have been way easier just to have subtitles, but to lay out the exact point, like graphic design for a magazine, for every shot was really worth it because it really draws you into where we want you to be looking at with the text."

Meanwhile, the lo-fi 3D aesthetic gives off a PS2 vibe, the console that first gave us Red Faction and is also famous for its destructible-environment gameplay. However, De Paco says it neither specifically emulates lower-poly PS2 graphics nor is it a performance choice in Unreal Engine 5.

"It's a result so that our audience from previous titles who were expecting a pixel art style that has been Deconstructeam's signature style until now can still connect with this 3D game," he explains. "We actually work with pixel art concept art, and we develop mockups for how this would look if it was one of our 2D games. Then we think about how we can port this into a 3D space."

Ferrando adds: "It's also faster to develop the pixel art models as we can get away with just displaying the shadows because the fidelity is not super high. So it makes the whole development easier, even though that's not the main reason we chose this aesthetic."

Despite the low-poly textures, the environments still had distinct details, from the architecture to small appliances, that made it clear this hometown was set in a mountain forest area on Spain's east coast rather than, say, the Pacific Northwest.

"We try to be really Spanish and convey our culture," De Paco concludes. "In cinema or other works, we are used to seeing other cultures, but for some reason with video games, we worry about making things more palatable to bigger audiences. So we need to try to actively change and be able to have different settings and cultures in our games."

Virtue and a Sledgehammer is releasing on PC on 27 October, and a free demo is available to download on Steam.

Now, the first hands-on previews are out, and the conversation has moved on. Journalists got an hour or two wandering Kokiri Forest, the Great Deku Tree, Castle Town and the Forest Temple, and the general feeling is that this is a lot more than a fresh coat of paint on the 1998 classic, with new controls, richer environments, extra dialogue, voiced cutscenes and fixes for some of the creakier bits of the original, even if not everybody is sold on every change.

01. It looks much better than the reveal suggested

Pretty much every preview agrees. The new Hyrule keeps the shapes and colours of the N64 original but rebuilds all of it with far more detail, and Nintendo Life’s Gavin Lane loved the little environmental touches, from the grain in the wood and foliage to the stitching on Link’s tunic, while GamesRadar+ reckons the softer, warmer art direction works a lot better in motion than it did in that first reveal.

The lighting gets praise, and the Forest Temple gets pools of sunlight, deep shadow, drifting mist, and wooden corridors you could stare at for ages, so places feel different rather than just sharper. Some people have spotted shimmering in close-up footage of hair and foliage, which is worth keeping an eye on, but plenty of doubters seem to have come round.

02. It’s a proper remake, not an HD remaster

Nintendo could have upscaled the 3DS version and called it a day, and clearly didn’t. The environments have been rebuilt, the camera updated, and familiar areas given new touches, with Castle Town now feeling connected rather than like the string of fixed-camera spaces we remember.

The broad layout is still what you’d expect, so you’ll know exactly where you found the Kokiri Sword and where the Deku Tree and Forest Temple sit, and VGC’s Chris Scullion came away feeling the appeal of the original had survived while the clunkiness that lingered in the 3DS version had been smoothed off. Nobody’s turning this into Breath of the Wild, which is the point.

03. Link finally has a jump button

It sounds minor, but in the original, Link hopped automatically off certain ledges, which was clever for its time and a pain when you were lining up something fiddly. Now there’s a dedicated button, and you decide when to leap, but Nintendo Life says it takes some getting used to, especially mid-fight, when years of muscle memory will have you reaching for the old way. Don’t expect Link to start scaling every wall in sight, mind you, because exploration is still the structured, Ocarina kind.

04. Combat is smoother, and your thumbs will need retraining

Z-targeting is still there, so locking on and waiting for an opening is still how fights work, which should put anyone who feared Nintendo would swap the paced encounters for something more frantic at ease. The jump button changes jump attacks: you now jump and then swing instead of doing it in one go, and veterans will fumble that for a while. The bigger gain is the move to 60fps, and VGC points out the original ran at 20, so dodging and reacting to enemies feels far more responsive than it ever did on N64.

05. Items and the map have caught up with the modern era

Anyone who remembers pausing the game repeatedly to swap boots in the Water Temple will understand why this matters. You can now switch equipment quickly without sitting through the old pause-menu animation, so the bow, Hookshot, and potions are far less of a faff, and the on-screen map makes it easier to track where you are and which rooms you’ve done.

You also get different HUD options, including a minimal one and one that turns the HUD off completely, which suits both the modern-conveniences crowd and the cinematic-purist crowd. The Water Temple will still test your patience, but at least the boots are less of a chore.

06. Castle Town might be the biggest surprise

You’d expect the dungeons to get the biggest changes, but it’s the town that seems to have had the most thoughtful overhaul. In the original, the market and its backstreets felt like separate spaces, with fixed camera angles hiding what the hardware couldn’t do; now the centre, shops, and alleys all connect, and you can rotate the camera freely.

It’s livelier too, with more detailed surroundings and extra dialogue, and characters will chat to Link as he walks past instead of waiting for him to start every conversation. GamesRadar+ also spotted new backstreets and a fleshed-out layout, plus the famously bearded man in the market, who can now point Link to another character by using his beard as a sort of tracking device, daft and brilliant in equal measure.

07. The cutscenes are voiced, and people are split

This is the one that’ll keep getting argued about until the full game lands. Cutscenes feature voice acting, including Navi and the Great Deku Tree; Ganondorf gets his own lines, and Link stays silent.

Some people love hearing the characters speak because it brings the game closer to modern Zelda, while others find the performances a bit cartoonish compared with the atmosphere they remember, and early Reddit threads are split between both camps, with a few comparing the English voices to an anime dub. It isn’t a fully voiced RPG, mind you; ordinary conversations still rely on text, and the voices are reserved for selected scenes.

08. The Forest Temple is properly creepy now

The overhaul isn’t just about prettier scenery, and the Forest Temple is the best example so far, with dark wooden corridors, heavy shadow, stained-glass windows, and shafts of sunlight that make it feel more unsettling than the stark original. GamesRadar+ was especially taken with the lighting, and The Verge’s preview said much the same about how the visuals make the temple scarier.

Ocarina was never a horror game, but its dungeons were always a bit eerie, and the music and ghostly encounters here are some of the most memorable in the series, so leaning into that makes sense. All that detail can backfire, too, because Kotaku pointed to a preview account of a player missing an important key chest simply because the room was too busy to look at.

09. The puzzles and pacing are largely untouched

For longtime fans, the most reassuring thing is that the dungeon layouts, puzzle solutions and progression all seem faithful to the original. Don’t go in expecting Breath of the Wild or Tears of the Kingdom, because Ocarina’s dungeons focus on keys, locked doors, backtracking, and figuring out where a newly found item fits on the bigger map.

The game tweaks enemies and encounters, and the new movement options will change how you approach a few situations, but nothing suggests a radical redesign. Nobody can say yet whether that slower, more deliberate structure will land with players raised on modern Zelda, and a couple of hours isn't enough to judge every dungeon.

10. Fans are warming up, with a few reservations

The mood has shifted a lot since the reveal. On Reddit, some people who were unsure about the art style say the new footage has won them over, while others are excited to play with modern controls and a few extra details. The voice acting is divisive; some aren’t convinced by the new dialogue, veterans will have to adjust to the jump button, and Navi’s extra commentary has people wondering whether the remake sometimes explains things the visuals already make obvious.

These are previews, not reviews, so we still don’t know how the changes hold up over the whole adventure, how it runs in handheld mode, or what else Nintendo is keeping back. Either way, Nintendo clearly took this seriously and revisited how Ocarina of Time feels to play instead of just updating the graphics, which could make it a lot more than a nostalgia trip for anyone who played it in 1998.

The Legend of Zelda: Ocarina of Time launches on Nintendo Switch 2 on 5 November, 2026.

As a key figure in the franchise, Aslan's character design is not something you want to mess up. Taking a more abstract approach to the lion's look, Narnia purists are already kicking up a fuss, proving that, to some, character design can make or break a movie.

Our glimpse of Aslan in the new trailer is fairly brief, but by no means forgettable. Prowling out of the darkness, the lion emerges as a cluster of stars whose mighty roar births the universe, but the more spiritually abstract take on the character has divided audiences.

"Looks like an LG TV commercial," one critic wrote. "Tell me this doesn’t look like AI I dare you," another added. "Why are people calling this magic? It looks like every cheap sci-fi flick. The average fan just eats up any blurry colors and calls it genius," one X user critiqued.

Many critics have brought up the book's supposed representation of Aslan as Jesus, stating that Gerwig's explosive universe-building is a take on the Big Bang, and thus a snub to creationism. Others have said the scene isn't book-accurate, as Aslan sings Narnia into existence, contrasting with his ferocious roar in the trailer.

I find myself torn on Aslan's representation, and until I see the full film, I'm anxious to pass judgement. To me, Gerwig's adaptation captures the magic and whimsy of the original book series, even if it isn't strictly accurate. Having a less literal depiction of Aslan feels more enriching than the 2005 film, giving him a transcendental quality that's open to interpretation. At its core, the Narnia franchise is for children, and the director's modernised take is an immersive experience set to ignite the imaginations of a new generation of fans.

For more design inspiration, take a look at the Narnia posters or check out the response to the Narnia logo.

The idea is that local AI agents running on your machine can act as “teammates” rather than “tools”. RTX Spark lets you use them as productivity assistants, which could be a real help for anyone who struggles to get their admin done. But the bigger news is that the built-in agents (Hermes, Perplexity and OpenClaw) are designed to interact directly with a whole host of apps, including creative ones. You can read more in my RTX Spark explainer, which goes into the benefits of local AI as opposed to running it on the cloud and exactly what the technology can do.

The premise is that an agent saves you time on busy work, freeing you up to do more creating. Nvidia also hopes it will help specialist creatives become more multidisciplinary, a skill increasingly in demand. That raises a bigger question about whether this is a good thing. The more technology democratises skills, the more creatives are expected to do, and the implications for the creative industries are enormous. But for the individual creative who can use Blender but not DaVinci Resolve (for example), there is clear benefit, and it could help small teams punch above their weight.

I’m not a professional creative by any means: Photoshop is about the limit of my skills. But I wanted to see how these agents work with creative apps, so I picked Blender, which I don’t know how to use and which gave me the highest possible barrier to entry. A creative pro will explore multiple apps in more depth soon, but for now I tried some tasks in Blender to see how easy it was to dive in as a complete beginner, whether the agent could do the work of a teammate, how quickly it worked, and what the experience was like.

How I tested

• Laptop: ASUS ProArt P16 with RTX Spark, 32GB
• Agent: Hermes
• Model: Qwen3.8 27B running locally
• App: Blender.

The set up

Setup was straightforward, with instructions for installing the Hermes agent. Hermes then guided me through which software versions I needed, how to download them and how to grant the relevant permissions.

The Blender test

As a beginner, I started by asking the agent what it could actually do in Blender. It came back with an extensive list of abilities, from editing and scripting to rendering stills and setting up cameras and lights. It also said it could do “scene surgery”, such as moving an object from one scene to another, and batch file operations.

The barbershop edit

I ran a few tasks through Hermes, and it handled the simple ones really well. I started by asking it to open a stock file of a barbershop scene and make the lighting much warmer. It did this with a script instructing Blender, then offered to confirm visually that the scene looked warmer. I said yes, and it checked and confirmed. I then asked it to put the before and after images side by side and export, which it did easily.

Editing a head

Hermes struggled with more complex, multi-step edits on this first time round. I asked it to change the hair colour of a stock model with curly hair and add clown make-up to its face. It changed the hair and eye colour quickly and easily.

The clown make-up was harder. Hermes spent a long time trying to render the additions before bringing back an approximation of the design. From then on it worked one step at a time, asking for approval almost every time, even for small iterations like bringing the mouth more in line with, well, the mouth. I stopped after about half an hour, and this is where we got to (above). If I had the skills and inclination, I could refine the model by hand, so it could work as a starting point for a more in-depth project. Alternately if I had more experience in Blender, I could probably have been clearer with my instructions, which may or may not have helped.

My mouse

I then started a project from scratch and asked the agent to build a mouse with a long tail and big ears (see above). The result was a little rough looking (that mouth!), but it was built quickly and simply. The first render didn’t show the long tail because of the camera position, so Hermes asked if I wanted to change the camera angle so it could build a visible tail, which was helpful. I confirmed and it made the image quickly.

Moving one object to another scene

Next I asked it to remove a model from one file and put it in the barbershop chair in my barbershop scene. After about 25 minutes of prep work, it estimated the job would take 5-10 minutes, and promised me it could do it after I repeatedly checked it was sure. By my deadline it had been working for about an hour and a half and still hadn’t finished.

So the agent is slow when you first ask it to complete a new task. But the technology works by learning the skills you train it in, so when you return to a task it should know what to do, including logging any mistakes and feedback. If you understand the software well, you can probably speed this up with better prompts. With time invested, it’s easy to see how this could become more useful, whether for beginners or for repetitive tasks you’d rather avoid.

The experience of talking to the agent

The agent talks to itself constantly while working through more complex tasks. It’s unclear whether that is a genuine thought process or a simulation designed to make you feel in control. It is reassuring to see what it’s doing, but frustrating to watch it tie itself in knots over something that could be searched online in a couple of seconds. At times I wanted it to just say “I can’t do that” rather than go around in circles – but maybe it would have got there in the end if I'd had left it for long enough. That may be part of its learning process, and as it logs its steps into skill files the behaviour may lessen over time, but I can’t say for sure.

Final thoughts

As a beginner, I appreciated how Hermes offered a gateway into a creative app I had never used before. It lowered the barrier to entry in a practical way, which could help creatives who are branching out or new hobbyists. Human direction will still be needed to steer, choose and refine the final outputs, but for many people this could be a significant productivity boost. That said, it was slow and needed a lot of hand-holding on complex tasks, and the usual AI concern remains that the technology could replace human teammates. We’re a long way from anywhere near that skill that at the moment – currently it has potential as a useful assistant, which is exactly how Nvidia is pitching it.

It’s impressive how the agent communicates with Blender through Python scripts and keeps the user informed throughout. Creative software apps are reportedly being rewritten for Spark, which suggests the creative tech companies are behind it. If the agent keeps improving with training, it could become more useful for the tasks creatives would rather not spend time on, and for speeding up workflows. It's one to keep an eye on.

Stand by for our full testing overview from a creative professional across multiple apps, coming soon.

Nvidia Spark is here to usher 'a new paradigm of personal computing' that puts agentic AI front and centre on a Windows PC. The RTX Spark technology puts a Blackwell RTX GPU and an Arm CPU on one chip, sharing up to 128GB of memory, in Windows laptop built for local AI agents. This could change the workflow of creatives in a host of different ways – including opening up how you interact with creative software and accelerating a spectrum of creative workflows.

I went to the launch day last week, and saw a room full of demos showcasing the tech's potential for creative workflows (see my Blender testing here) and gaming – including video editing, CAD, community game modding, productivity overhauls and more. It's clear Nvidia and its partners are committed to the future of agentic computing: but what does this mean for creatives?

Firstly, what is RTX Spark?

NVIDIA announced RTX Spark on 31 May at GTC Taipei, calling it a superchip for Windows PCs built around personal AI agents. It joins a Blackwell RTX GPU, with 6,144 CUDA cores and fifth-generation Tensor Cores, to a 20-core Arm-based Grace CPU over NVIDIA's NVLink-C2C link. MediaTek collaborated on the CPU design.

The CPU and GPU share up to 128GB of unified memory, and NVIDIA quotes up to 1 petaflop of AI compute. Laptops from ASUS, Acer, Dell, HP, Lenovo, Gigabyte, Microsoft Surface and MSI are all up for release, in 14 to 16-inch designs as thin as 14mm.

It runs Windows 11 on Arm. Apps with a native Arm version run directly, and everything else goes through Microsoft's Prism emulation layer. NVIDIA also says the laptops deliver equivalent performance on battery and plugged in – which is one of the biggest improvements compared to its older laptops.

Why this matters for creatives

A normal laptop GPU has its own small pool of video memory. On RTX Spark the GPU shares the CPU’s pool of up to 128GB, so big scenes and models aren’t capped. NVIDIA says that means rendering 90GB+ 3D scenes, editing 12K 4:2:2 video and running 120-billion-parameter language models locally. The GPU’s share is adjustable in the NVIDIA App, and NVIDIA’s largest local model, Qwen3.8-Flash-Next, needs a 128GB machine with at least 78GB given to the GPU.

Important parts for creators are privacy and cost. Local models mean footage and client assets don't have to leave the machine, and there are no per-generation cloud fees. The data example I saw at the launch is that the 61m tokens it would take to complete a heavy creative workflow will cost $20 in electricity, as opposed to the $1200 token cost it would take in API fees.

As this is hybrid intelligence, and the local AI can't yet do everything, there's the NeMo Switchyard Verifier, which kicks in when the task is sent to the local LLM. This decides if the task can stay local, or needs to be sent up to the cloud.

The software

One of the most impressive parts of the launch is how many apps Nvidia has worked with to ensure optimisation with the Spark technology. More than 100 Windows software providers, including Adobe, Blackmagic Design, Blender, CapCut, ComfyUI and OTOY, are backing the platform. With Adobe actually rewriting its Premiere Pro and Photoshop apps from the ground up to suit Spark, it's clear many software companies see the potential here.

I saw a CAD workflow move from proposal to fully rendered 3D design in 26 minutes using AI. This is a process that usually takes weeks. For solo creatives, this kind of tech is potentially a good thing – it will allow you to create whole studio's worth of work individually. But as usual with AI advancements, the impact on team structure and the need for jobs will be immense. When we're talking about some creative workflows, the old Adobe-style argument is that that just means creatives can do more – but in this instance, how many buildings really need to be built?

Anyway, let's look at the impacts on other creative apps:

• Adobe. NVIDIA says Adobe is rearchitecting Photoshop and Premiere around RTX Spark, promising up to 2x faster AI, editing, colour and effects. Premiere gets a new video pipeline built on unified memory, the GPU and TensorRT. Photoshop gets an engine built around GPU-accelerated compositing, for live filters, HDR and natural brushes. Substance 3D Painter and Stager will run natively, and updates are due as the laptops go on sale. Adobe's public betas already include TensorRT-accelerated Object Mask, Enhance Audio, Scene Edit Detection and Auto Reframe in Premiere, a GPU-accelerated colour mode, and on-device object removal in Photoshop.
• Blender. Version 5.3 adds DLSS 4.5 Ray Reconstruction as a denoiser for the Cycles viewport, aiming for a near-final rendered view with far less noise. It is currently in the 5.3 alpha builds.
• DaVinci Resolve. Blackmagic Design says it expects portable RTX Spark laptops to help on-the-go production. Resolve Studio 21 has a native Arm build.
• ComfyUI. New RTX Video nodes upscale resolution, interpolate frames (up to 4x frame generation) and convert SDR to HDR, so artists can generate at lower resolution and finish higher.

What can the AI agents do for creatives?

Nvidia frames the new PC tech as moving from "tool to teammate". Essentially, you can use a trio of AI agents as a team of assistants, who will prepare assets and set up workflows. Two open-source agents, Hermes Agent and OpenClaw, are getting native Windows apps that can carry out tasks in Windows applications, work across apps, generate images and video, and search local files. Adobe says it will extend Premiere and Photoshop so people can create, edit and design with Windows agents.

That means that the most powerful advancement is that these AI agents will interface with these creative apps for you. Yes, you read that right. You don't even have to go into the app to utilise its tools and features – you can describe to the agent what you want it to do, and it will open the app, individual file, and apply the edits or create the work natively within the application.

This is exciting and alarming, depending on how you look at it. In a world in which creatives are increasingly expected to be multi-disciplinary, this functionality will allow you to use apps you don't have specialist knowledge of. Alarming because, well, obviously it's another shortcut on human skill. However, in line with much of the creative AI narrative, the sell is that the human remains the director, the refiner, the ultimate creative – the technology is there to speed it up, make it simpler, enable more creativity and eliminate busy work. Where you sit on this probably fluctates, and it isn't a simple conversation.

The agents can also work as admin/productivity assistants. They can connect to your files, email account, calendar and more and stay on top of your daily workflow. This is probably appealing for the majority of creatives, as this is the true busy work we'd all like to eliminate.

Capcom's remakes of Resident Evil 2, Resident Evil 3 and Resident Evil 4 are big, cinematic games, full of moody lighting, detailed environments and things I'd rather not encounter in a dark corridor. Read our Resident Evil 4 Remake review on PS5 to see why this game is considered a modern classic.

Getting all three running on Nintendo Switch 2 is an impressive technical achievement, and the RE Engine does a fine job of bringing their grim worlds to a portable screen, something the publisher has excelled at in 2026 following successful versions of Resident Evil Requiem, Pragmata and Dragon’s Dogma 2.

After spending time with the collection, though, I can't help but think Capcom has focused on getting the games running and given less thought to how they should play on Nintendo's handheld. Now, we all know, and if you didn’t, Resident Evil 2 Remake is exceptional, Resident Evil 3 Remake is good but takes some liberties, and finally Resident Evil 4 Remake is a fantastic update of many fans' favourite game of all time (I have my own peccadillos with its controls and repetition, but that’s me). So the question isn’t whether they're good games, but whether they're good on Nintendo Switch 2.

More motion control

My biggest gripe is the inconsistent approach to motion controls. Resident Evil 2 and Resident Evil 3 lack gyro aiming on Switch 2, even though motion controls fit this kind of game naturally. I like being able to make small aiming adjustments with the controller while using the analogue stick to turn and move. When ammunition is scarce and a zombie is closing in, that extra precision can make a real difference.

Resident Evil 4 does offer gyro aiming, but the implementation is restrictive. While you’re holding the right stick, motion aiming is locked to horizontal movement only, so you can’t use the gyro for fine vertical adjustments. It’s a curious compromise, especially when the other two games don’t offer gyro aiming at all. That’s a shame, because gyro helps most with vertical precision when lining up headshots.

I'd also have liked to see support for the Switch 2's mouse mode, since too few games offer it, and these three, RE4 in particular, would have been a fun adaptation. It would have given us another way to aim and helped distinguish these versions from straightforward ports of games we've been playing for years on PlayStation and Xbox. In fairness, Capcom producer Masato Kumazawa has stated they tried to get mouse control working, but it "confused the gameplay and became too complicated". Capcom has left some useful hardware features on the table, unless it plans to add extra touches in post-release updates, which is always on the cards for a publisher like this that loves supporting fans.

Raccoon City still looks terrific

The frustration is that the games themselves make such a strong case for playing on Switch 2, especially after RE Requiem proved so successful at bringing survival horror to the handheld. Resident Evil 2 is the highlight for me (and always has been of the three), with the Raccoon City Police Department retaining its oppressive atmosphere and carefully staged lighting. The game targets 60fps, and the overall presentation makes a convincing transition to Nintendo's console.

Resident Evil 3 fares well, too. Its gloomy, rain-soaked streets, fire effects and neon-lit environments retain much of their visual character, and the faster pace suits portable sessions. My reservations are more about the game itself than the conversion. The remake's shorter, more streamlined approach still leaves me wanting more after the excellent Resident Evil 2.

Resident Evil 4 puts more pressure on the hardware with its busier combat and detailed environments that make compromises easier to spot, particularly when playing in Switch 2’s handheld mode; for example, some textures look softer, and pop-in and performance dips can take the edge off the presentation in demanding sections, such as the infamous village encounter.

These aren't problems that make these three classic Resident Evil games unplayable. Far from it; in fact, it's worth celebrating that such substantial games can run on a handheld at all, and there's real pleasure in dipping into a survival horror campaign away from the television. But visuals are only part of the job, and controls matter just as much, which is where Capcom's approach feels a little uneven.

Should you buy the RE Remakes on Switch 2?

If you haven't experienced these three remakes yet, the Switch 2 is a compelling entry point. You get three major Resident Evil adventures in a portable format, plus bonus content to round out the package, and it’s a nice narrative run-up to Requiem for completionists. Being able to explore the police station, escape Nemesis, or work through the Spanish countryside wherever you are is a genuine draw.

I'm less convinced that existing owners need to rush out and buy them again. If you already have the games on PC or PlayStation, the main reasons to double-dip are portability, easy access (Nintendo makes downloading so simple), and the option to have a complete collection on one new system. There's no substantial new content here to change the experience, and the control options don't do enough to make these versions feel purpose-built for Nintendo's console.

That Nintendo-designed intent feels like a missed opportunity, especially since RE4 gave many a reason to buy a Wii (alongside Resident Evil 2 on N64, Resident Evil 0, and other entries that made Nintendo a destination). Gyro aiming could have been a useful addition to two games that currently lack it, while a more flexible implementation in Resident Evil 4 would have made the feature worth using. These are small changes in the grand scheme of a port, but they're the kind of details that can make a familiar game feel right at home on a new platform.

A strong technical effort that needed more polish

Capcom deserves credit for bringing three demanding remakes to Switch 2 without losing what makes them work. The atmosphere remains intact, the games look good on a portable screen, and Resident Evil 2 in particular is a fine showcase for the hardware.

But I want more than a game that simply runs on a new console. I want the developer to think about what that console can add to the experience, too. Here, the lack of gyro aiming in Resident Evil 2 and Resident Evil 3, coupled with the feature's limitations in Resident Evil 4, makes the collection feel like a missed opportunity.

If you're new to these games, I'd happily point you to them, not least because each comes as the Deluxe/Complete/Gold edition, loaded with extras, new modes, and more. If you've already played them elsewhere, I'd weigh up how much you value handheld gaming before buying again. These are excellent remakes, but Capcom could have made their Switch 2 versions feel considerably more at home.

Resident Evil 2,3 & 4 Switch 2 mini-reviews