The moment Vjeran and I got to the demo room at Framework’s San Francisco event, we knew right away — the mint and pink version of the Framework Laptop 12 was the one I wanted to touch, the one he wanted to photograph and film.

One of the women standing next to the demo stations thanked us for saying that out loud, because she wasn’t initially sure mint and pink would make the cut! She introduced herself as Esther Yen, the senior industrial designer of this entire laptop — and says they were the colors she had asked for.

Yesterday, Framework announced the 12-inch convertible notebook as its first attempt to fix budget laptops, by making them modular and repairable. (It’s kind of Framework’s thing, but this is its …

Read the full story at The Verge.

Framework’s mission is to “fix consumer electronics, one category at a time” by making them modular, repairable, and upgradable. It’s the only laptop maker to ever truly succeed at that “upgradable” part. But desktop PCs are already modular, so why is Framework making one?

At first, I thought it saw a unique opportunity to make a cute yet badass tiny gaming PC with AMD’s unusual Strix Halo processor and decided to shoot its shot. As you’ll read below, I’m excited by the result. But I also have another idea I’ll share with you afterward.

Let’s start with the gadget part: yes, the tiny 4.5-liter gaming desktop that Framework announced yesterday is just as cool in person as it was in Framework’s photos, and yes, it can game.

At first, I wasn’t sure we’d be able to meaningfully try that last part. Almost all the Framework Desktops at the company’s live event in San Francisco yesterday were either running games that don’t offer a great sense of performance (Counter-Strike and Street Fighter) or were unplugged so we could take photos from every angle. But partway through the event, someone had fired up Cyberpunk 2077, and I saw my chance.

The first b …

Read the full story at The Verge.

Two years ago, the last time Framework had an event in San Francisco, California, the highlight was the Framework Laptop 16 — a laptop promising the “holy grail” of upgradable graphics cards, and easily one of the most ambitious laptops ever made.

But today, the Framework Laptop 16 got little mention at its new event, which focused on the new, similarly gamer-oriented Framework Desktop instead. While the Desktop and Framework’s original 13-inch laptop both got the latest AMD processors today, we don’t yet know if or when the Framework Laptop 16 might leap forward too.

The only mention of Framework’s flagship laptop was a new “One Key Module” that will experimentally be available for the Framework community to build their own electromechanical keyboards, should they choose to, that would be thin enough to fit in the Laptop 16’s extremely thin Input Module bay. (You can see how the Input Modules work in my video here.)

While Framework did finally ship its promised M.2 adapter bay in December, which will let you stick extra SSDs or other peripherals into a Framework Laptop 16 instead of a discrete graphics card, my big question is: is the promising but somewhat problematic laptop a dead end, or will it get new mainboards and new chips in the future?

I tracked down Framework CEO Nirav Patel at the event today, and he wouldn’t say much, but he was clear on one thing: “We’re nowhere near done with Framework Laptop 16.”

I pushed my luck, asking: Is today the day he can assure us that the Laptop 16 will eventually see at least one GPU upgrade or snap-on secondary battery?

“Today is not that day,” he told me.

I want to see Framework succeed, and perhaps it’s too early to begin to wonder otherwise — it was still shipping preordered batches of the Laptop 16 to buyers through the middle of last year.

But we’ve pushed the company pretty hard on the GPU in the past specifically because it’s a thing rivals have tried and failed at before — Dell/Alienware even got sued over the failed promise of the Alienware Area-51m, which never bothered to ship a second generation of its supposedly upgradable GPUs.

Framework has resisted our pushes so far, stopping short of confirming it in our 2023 story: here’s his exact language at the time. It’d be nice if Framework could assure buyers that the upgrades are absolutely coming. But personally, there’s more than a few things about that laptop I’d like to change, and I wouldn’t be surprised if Framework was doing a bit of a rethink even if it does deliver.

Framework, the modular computer company, has just announced its first desktop PC, which is something it absolutely did not need to do — but I’m glad it did.

Partly because the world needs more tiny 4.5-liter mini-ITX PCs, partly because it uses AMD’s most powerful APU ever (Strix Halo) with some actual gaming chops and up to 128GB of unified LPDDR5x memory… and partly because it looks like this.

Yes, you can create your own front panel out of 21 interchangeable (and freely 3D printable) tiles, pick your own two front I/O ports, and yes, that’s a standard-size mini-ITX motherboard below, along with a custom thermal system co-developed by Cooler Master and Noctua with standard 120mm fans, a semi-custom 400W Flex ATX power supply co-developed by FSP, a whole lot of seemingly perfect-length cables for a relatively cable-light system, your choice of black or translucent side panels… and, at left, an optional LAN party carry handle!

Just don’t look for any memory slots — it’s soldered. “We spent months working with AMD to explore ways around this but ultimately determined that it wasn’t technically feasible to land modular memory at high throughput with the 256-bit memory bus,” writes Framework.

Below, find a closer look at Framework’s desktop motherboard and cooling; you only get a PCIe x4 port, not PCIe x8 or x16, and no legacy connectors like SATA, but it’s more loaded than the image lets on: Framework says it’s got onboard 5Gbps Ethernet, two USB4, two DisplayPort, one HDMI, not one but two M.2 2280 NVMe SSD slots for up to 16TB of storage, and a Wi-Fi 7 module, plus what appear to be two full-size USB-A ports and a headphone jack (at the rear).

Here’s a little bit of the build process, as told by the images Framework sent us:

And here are some better images of the Framework Desktop mainboard’s I/O, in a rack-mounted, daisy-chained configuration that the company expects some AI enthusiasts might snap up — thanks to the local AI chops and 128GB of memory on the highest-end Ryzen AI Max Plus 395 Plus config.

“With Framework Desktop, you can run giant, capable models like Llama 3.3 70B Q6 at real-time conversational speed right on your desk,” the company claims, adding, “With USB4 and 5Gbit Ethernet networking, you can connect multiple systems or Mainboards to run even larger models like the full DeepSeek R1 671B.”

Framework CEO Nirav Patel says it was also designed with LANs in mind — though with a bit less bulk than the massive CRT moniotrs and desktops of old.

What kind of gaming chops might it really have? My colleague Antonio has seen in his time with the Asus Z13 gaming tablet that Strix Halo is roughly around the performance of an Nvidia RTX 4060 mobile chip. AMD gaming architect boss Frank Azor was also here with some 1080p benchmarks at high settings; it apparently can’t quite play the most demanding games (like Black Myth Wukong and Starfield) at over 60fps at native resolution with everything turned up, but the claim is that even 1440p at 60fps is possible with AMD’s FSR upscaling.

Even if you’re not getting your money’s worth out of AI, though, the prices on these desktops don’t seem all that outlandish. While a desktop with 16 CPU cores, 40 graphics cores, 80MB of cache of the 395 Plus chip, and 128GB of memory will cost $1,999 — not including bring-your-own storage and OS — you can pick one with 8 CPU cores, 32 graphics cores, 40MB of cache and 32GB of memory for $1,099. There’s also a 395 Plus with just 64GB of RAM for $1,599. Or, you can buy a mainboard alone for $799 if you provide your own mini-ITX case and power supply, too.

If you are looking for the most powerful mini-PC, Patel suggested the $1,999 model compares favorably to an Apple Mac Studio, which can cost over twice as much for the same 128GB of RAM.

Framework says these desktops and mainboard should be available to preorder today, with plans to ship in Q3. If you’re reading these words shortly after they were published, I’m currently at a Framework event in San Francisco, where the company also just introduced a new AMD-powered version of its Framework Laptop 13 and the new Framework Laptop 12: its first budget laptop, its first touchscreen, and its first convertible.

Framework, the modular repairable computer company, says budget laptops generally suck — and it plans to fix that this year. Today, it’s previewing a computer called the Framework Laptop 12 that’ll be its smallest laptop yet, its first with a touchscreen, its first with a 2-in-1 convertible hinge, and its first with a rubber-esque shock absorbing TPU outer edge to make it more suitable for kids and students, too.

“We build products to fix what we see as a broken industry, and few categories are as emblematic of the problems with consumer electronics as entry-level laptops,” writes Framework CEO Nirav Patel, continuing:

They tend to be janky, locked-down, disposable, underpowered, and frankly, boring machines.  Shamefully, these are the products that PC brands market for use by students and young people around the world.  Instead, we believe these are the people who most need thoughtfully designed, long-lasting computers.

Patel says the Framework Laptop 12 “is in many ways the product I started the company to create,” and that it will be “our easiest product ever to repair.” It will come with a 13th Gen Intel Core i3 or i5 chip (circa 2023), a 1920×1200 screen at over 400 nits of brightness with both touch and stylus support, up to 48GB of RAM and up to 2TB of NVMe storage, and Wi-Fi 6E.

Based on the image below, it would appear that the Framework Laptop 12 will also have four of Framework’s Expansion Card slots, which is how Framework lets you pick your own ports.

Despite the claim that the Framework Laptop 12 is designed to fix entry-level laptops, Framework isn’t sharing any idea of pricing today, beyond that it will be “lower cost” compared to the Framework Laptop 13, which typically starts at around $750 for a DIY model with previous-gen chips or $1,100 for a prebuilt with the latest ones.

But while Framework products often cost more than you’d pay for the same specs with Dell, HP, or Lenovo, it is the first and only company that has repeatedly delivered on the promise of modular upgrades, letting you easily swap out the entire motherboard and processor as a single module for a new one down the road. For example, Framework is also announcing a new AMD mainboard for its Framework Laptop 13 today, one that can breathe new life into any previous iteration of that laptop, even the original Intel version it released in 2021.

Framework says it’ll open preorders in April and ship in “mid-2025.”

If you’re reading this soon after it was published, I’m probably still at Framework’s event in San Francisco, looking for both this laptop and the just-announced Framework Desktop.

Framework has done it again — it’s built a new AMD-powered modular computer that fits into any previous version of its 13-inch laptop for $449, or that you can buy as a brand-new laptop starting at $899 for DIY or $1,099 entirely prebuilt for you. If you’re reading these words, I’m currently at Framework’s event, where I’m checking it out alongside the just-announced tiny Framework Desktop and possibly the new budget touchscreen convertible Framework Laptop 12.

This 13-inch machine is the company’s second with AMD processors, this time featuring the new-ish Ryzen AI 300 “Strix Point” chips in Ryzen 5 and Ryzen 7, all the way up to the Ryzen 9 HX 370 with 12 CPU cores, half-decent portable gaming chops and 50 TOPS of AI performance. And yes, that makes this Framework’s first Copilot Plus PC.

The changes don’t stop there: Framework says its second AMD laptop gets all the big upgrades of last year’s Intel Core Ultra model, including the optional 2.8K 120Hz screen and presumably the better webcam.

While you don’t get four fully functional Thunderbolt 4 / USB4 ports like you do on Intel models, you can now power four external displays from each of its four ports, up from three on the last AMD, in addition to new Wi-Fi 7, a new cooling system with a single larger 10mm heatpipe like last year’s Core Ultra model (up from twin 5mm heatpipes) with Honeywell PTM7958 phase change thermal interface material, and “a new key structure on the wide keys (e.g. spacebar and shift) that reduces buzzing when your speakers are cranked up.”

They’ve got a plastic structure that spans the full width of the key instead of metal linkages, and says they’ve been thoroughly tested, joking that there’d be no Apple Butterfly Keyboard issues in Framework’s future.

Speaking of the keyboard, it does have Microsoft’s mandatory Copilot key if you buy it as a prebuilt Windows 11 laptop — but you can ditch that if you buy DIY! (Which I’d recommend unless you really need a preinstalled copy of Windows or are quite squeamish, as even Framework’s DIY machines are mostly prebuilt anyhow.)

Lastly, Framework’s now offering translucent bezels, including new colors purple, green, and black, and matching translucent USB-C expansion cards too.

The new Framework Laptop 13 boards and laptops should be available to order today, shipping in April, and Framework says it’s reducing the price of its previous-gen AMD notebooks, as usual, as these new machines roll out.

At the beginning of today’s Framework event in San Francisco, CEO Nirav Patel says he believes Framework is now the fastest growing laptop brand, and “probably the only company that can announce a product that looks the same four years in a row and get cheers.”

“We’re proving it’s possible to build a company around product longevity by actually doing it,” he says.

Today is the third anniversary of Valve’s Steam Deck, the handheld gaming PC that all but created the market for handheld gaming PCs. It was a mess to start! But three years later, The Verge has data showing how it has dominated the nascent market. While Valve told us in November 2023 that it had sold “multiple millions” of the AMD-powered handheld, we’ve never had a good glimpse at how big it is or how Windows competitors stack up… till now. It seems the Steam Deck, so far, has been bigger than all its competitors combined.

Market research firm IDC uses supply chains to estimate just how many handheld gaming systems have shipped around the world, and creates spending forecasts. When I asked IDC market research analyst Lewis Ward if he’d be willing to isolate SteamOS and Windows gaming handhelds from that data, he said yes.

So here are the estimated combined shipments of the Steam Deck, and the Windows-based Asus ROG Ally, Lenovo Legion Go, and MSI Claw from 2022 through 2024, and an estimate for 2025:

Add it up, and that’s just under 6 million shipments in three years. One way to view that: it’s …

Read the full story at The Verge.

I’ve never reviewed a perfect product, but Logitech’s Powerplay Wireless Charging System comes close. For over three years and counting, I’ve never even had to think about charging my wireless mouse. It’s so dead simple, it feels like magic, and it’s a shame that most people probably can’t afford it at $120.

The good news: Logitech is releasing a new $100 version in March, called the Powerplay 2, and it’s just as easy to set up. Plug in mousepad, snap a magnetic “Charging Coin” into the base of your mouse, then put mouse on mousepad to continuously charge.

The bad news: It’s only $20 cheaper, yet it feels like Logitech made its mousepad more than $20 cheaper to hit that goal.

The mousepad does come with improvements. Logitech boasts it has a 15 percent wider charging area and is thinner at just 3.5mm, and that’s what I see with my review unit. Now, as long as the entirety of my G502 Lightspeed mouse is resting within any corner of the mousepad, the charging indicator lights up, which wasn’t quite true of the original. My calipers do read 3.5mm when I’m measuring the charging base and its thin fabric mousing surface together.

But my calipers also show the charging base is exactly the same 2.7mm thickness as before, and the old mousepad wasn’t all that much thicker: just 4.3mm in total before vs. 3.5mm in total now, a difference I do not feel. And do you leave your mouse all the way at the corners of your mousepad? Again, I’ve spent over three years charging this mouse on the old mousepad without even thinking about it. I never bother to reposition my mouse on my old Powerplay to make sure it’s charging; I just drop it when I’m done using it, and I’ve never once run out of charge.

What do we lose with the Powerplay 2? First, while Logitech has ditched the old micro-USB cable, we’re not getting USB-C. Instead, Logitech’s opted for a fixed cable, so I can’t as easily take the mousepad off my desk on days I need more space there.

The big one: there’s no more wireless mouse receiver built into the Powerplay 2, a feature I found handy with the original. Now, my mouse requires two full-size USB ports instead of one because I still have to leave the mouse’s dongle plugged into my PC, too. I can’t leave the dongle stowed in the mouse for grab-and-go travel, and I can’t leave it in my laptop and switch between laptop and desktop anymore by turning the mouse on and off and yanking the Powerplay’s plug.

There’s also no more programmable RGB light inside the Logitech G logo. I don’t terribly miss that, since I don’t sync up gamer lights. But the dull black Logitech G feels cheaper; before, the RGB was at least a nice reassuring reminder that my mousepad was properly receiving USB power and ready for action.

And, while I do like the new thinner mousepad that comes with the Powerplay 2, which looks like it might not delaminate from its backing as easily as the original (it’s the one piece of my Powerplay that has deteriorated over the past three years), the original Powerplay came with two mousepads (one hard, one cloth) in the box. Now, you get the one.

(Also, just as FYI, the new Powerplay 2 charging coin doesn’t seem to work with the original pad and vice versa. You can’t mix and match those parts.)

I tried hard to get Logitech to show me more benefits, because the original’s one of my favorite products. Perhaps this one’s so much easier for Logitech to produce that it’ll offer some great discounts, or sell amazing bundles after launch? Or perhaps that 15 percent larger charging area will come in handy for possible additional supercapacitor mice that’d react worse than battery mice if they aren’t getting reliably fed, though the current supercap G309 seems to work fine with the original Powerplay in my early tests.

Logitech wouldn’t comment on future supercap products, wouldn’t hint at sales, and wouldn’t promise its own bundles — though Logitech does “anticipate retailers will be offering bundles shortly after launch,” according to Logitech senior global marketing manager Andrew Siminoff.

The original Powerplay is no longer in stock at major retailers, so I expect it will soon fetch a premium price on eBay. But the Powerplay 2 still seems like a good product that achieves the core goal. Fingers crossed that come Black Friday, we’ll be able to buy a combo pack with it and Logitech’s cheapest compatible mouse — that G309 — for under $100 in total.

The Powerplay 2 should be available on Amazon and Logitech’s site on March 11th.

When Nvidia originally confirmed that some of its new RTX 50-series graphics cards had a “rare” manufacturing issue that left them missing some promised render units and a slight amount of performance as a result, it only named three affected cards: the RTX 5090, RTX 5090D, and RTX 5070 Ti. But now, Nvidia has confirmed to us that RTX 5080 production was affected by the same issue as well.

“Upon further investigation, we’ve identified that an early production build of GeForce RTX 5080 GPUs were also affected by the same issue. Affected consumers can contact the board manufacturer for a replacement,” Nvidia GeForce global PR director Ben Berraondo tells The Verge.

In response to The Verge’s questions, Berraondo adds that “no other Nvidia GPUs have been affected” — we specifically asked about the upcoming RTX 5070, and he says it’s not affected either. Nor should any cards be affected that were produced more recently: “The production anomaly has been corrected,” he says. In case you’re wondering, he also told us that Nvidia was not aware of these issues before it launched these GPUs.

Here’s the company’s full amended statement:

We have identified a rare issue affecting less than 0.5% (half a percent) of GeForce RTX 5090 / 5090D, RTX 5080, and 5070 Ti GPUs which have one fewer ROP than specified. The average graphical performance impact is 4%, with no impact on AI and Compute workloads. Affected consumers can contact the board manufacturer for a replacement. The production anomaly has been corrected.

One specific Redditor was the one to discover that their RTX 5080 also demonstrated the issue; he’s since said he’s worked out a deal to hand that card to GamersNexus, which is investigating the RTX 50-series issues, for more study.

While it doesn’t seem like a lot of GPUs were affected, given how few of these GPUs have shipped so far, and Nvidia is also promising replacements, it’s the latest in a line of annoyances with Nvidia’s new cards.

The Lenovo Legion Go S was supposed to change things. It was poised to show Valve isn’t the only one that can build an affordable, portable, potent handheld gaming PC — you just need the right design and the right OS. 

I was intrigued when Valve’s own Steam Deck designers told me this Windows handheld would double as the first authorized third-party SteamOS handheld this May. When I heard Lenovo had procured an exclusive AMD chip that would help that SteamOS version hit $499, I got excited for a true Steam Deck competitor. 

But I’m afraid that chip ain’t it. 

I’ve spent weeks living with a Legion Go S powered by AMD’s Z2 Go, the same chip slated to appear in that $499 handheld. I’ve used it with both Windows and Bazzite, a SteamOS-like Linux distro that eliminates many of Windows’ most annoying quirks. I tested both directly against a Steam Deck OLED and the original Legion Go, expecting to find it between the two in terms of performance and battery life. But that’s not what I found.

Watt for watt, its Z2 Go chip simply can’t compete with the Steam Deck, and it’s far weaker than the Z1 Extreme in last year’s handhelds. That’s inexcusable at the $730 …

Read the full story at The Verge.

It’s true: Nvidia has just confirmed it shipped some RTX 5090, RTX 5090D, and even some RTX 5070 Ti graphics chips that were missing render units, as TechPowerUp originally reported — and that you’ll be able to get a replacement if your card was affected.

Nvidia GeForce global PR director Ben Berraondo tells The Verge:

We have identified a rare issue affecting less than 0.5% (half a percent) of GeForce RTX 5090 / 5090D and 5070 Ti GPUs which have one fewer ROP than specified. The average graphical performance impact is 4%, with no impact on AI and Compute workloads. Affected consumers can contact the board manufacturer for a replacement. The production anomaly has been corrected.

In the grand scheme of things, that doesn’t sound like a lot of affected GPUs, particularly because there weren’t a lot of 5090s on shelves to begin with, nor was it a huge hit to performance — as those who discovered the missing render units can already attest. But it is the latest in a line of annoyances with Nvidia’s latest pricy cards, including launch driver issues (including some ongoing black screen issues that Nvidia is still investigating) and some melting power connectors.

While limited, the manufacturing issue affected multiple Nvidia graphics card partners: reports came in of Zotac, MSI, Gigabyte, Manli, and even an Nvidia Founders Edition card with missing ROPs. You can use GPU-Z to check your card and see if it’s showing the proper number of 176 ROPs; if fewer, you should probably get it replaced.

Nvidia has yet to explain why it launched its GeForce RTX 5090 and 5080 GPUs with barely any inventory, some major launch driver issues, and the occasional melting power connector, but it has apparently reconsidered its stance when it comes to scalpers. The company’s just announced a way for Nvidia fans to sign up for “Verified Priority Access” to buy the elusive two-slot SFF-friendly RTX 5090 and 5080 Founders Edition graphics cards.

Like a similar Verified Priority Access program for the RTX 4090, the new program is invite-only, but this time you’ll apply for access by filling out this form rather than being pre-selected. The site will check that you’ve already had an Nvidia account (accounts created after January 30th need not apply) and ask you whether you’d prefer a 5090 or a 5080. Then, it’ll apparently use an algorithm to figure out if you’re a real gamer (analyzing your Nvidia app / GeForce Experience use) before offering a card. Limit one per person.

“Invites will begin rolling out next week,” writes Nvidia. The company doesn’t say how many cards have been allocated to this program, so it’s difficult to tell if this is a meaningful way to get cards to gamers rather than scalpers.

Remember PhysX, the GPU-accelerated technology that let games realistically simulate destructible cloth, shattering glass, moving liquids, smoke, fog, and other particle effects? It only ever got deployed in a few dozen games — but with 32-bit PhysX turned on, those games reportedly now run faster on Nvidia’s last-gen cards than they do on a new RTX 5090, 5080, 5070 Ti or beyond.

That’s because Nvidia has quietly removed support for PhysX in its latest graphics chips, the company confirmed this week, after buyers noticed PhysX games like Borderlands 2 were mysteriously taxing their CPU instead of their GPU and either chugging or failing to work. Nvidia points to a support page from January where it did say that the RTX 50 series would not support 32-bit CUDA applications, but that page doesn’t explicitly mention PhysX, and the company’s other PhysX support pages are several years old.

Again, we’re talking about mere dozens of games here, all over a decade old, but a good number of them were marketed by Nvidia as showcase PC games thanks to their physics simulations — and a handful of those were at least cult hits.

I fondly remember PhysX cloth effects in Mirror’s Edge (above) and the Batman: Arkham games. The best Assassin’s Creed also used it (Black Flag, the pirate one, don’t @ me) and so did Metro 2033 and Metro: Last Light.

And as you can see in the video just above, PhysX just doesn’t run terribly well without a GPU’s assistance, tanking performance when its effects are most vividly felt on screen.

Over on the ResetEra forums, RandomlyRandom67 is maintaining an updated list of 32-bit PhysX games and how they perform on a CPU instead of a GPU — users are reporting that Mirror’s Edge, Batman: Arkham Asylum and Batman: Arkham City, and Cryostasis all have significantly lower lows.

One Redditor claims that when they forced PhysX on in Borderlands 2, they “got drops to below 60 FPS by just standing and shooting a shock gun at a wall,” despite having a system with today’s top PC components (an RTX 5090 and an AMD Ryzen 7 9800X3D). They claim their RTX 4090 never dipped below 120fps in the same game.

I won’t terribly miss PhysX, because modern games have plenty of other ways to do physics built into their various engines, but Nvidia could probably have communicated this better. It’s also just the latest evidence that its RTX 50-series cards aren’t the upgrades we’d hoped. My colleague Tom reviewed the new RTX 5070 Ti today, and found it’s largely a cheaper RTX 4080, except maybe not cheaper because you probably won’t be able to find one anywhere near its $750 sticker price.

Nvidia’s older cards don’t seem to be losing value as a result of this year’s upgrades, either: the street price of an RTX 4080 has actually jumped over the past month, from an average eBay selling price of around $1,000 in January to over $1,200 now.

Acer CEO and chairman Jason Chen says your laptop will cost an extra 10 percent in the United States next month — and that his rivals might attempt price gouging if they think you’ll pay even more.

“We think 10pc probably will be the default price increase because of the import tax. It’s very straightforward,” he told The Telegraph, referring to President Trump’s 10 percent tariff on incoming goods from China. While big tech companies have generally been quiet on how they’ll respond to Trump tariffs, Acer says it just made the decision to increase prices last week, and it’ll take effect a few weeks from now.

It sounds like the price hike may not affect desktops, as Chen told The Telegraph that Acer had moved its desktop computer manufacturing outside China during Trump’s earlier tariffs during his first term. He said Acer might now consider moving some of its laptop manufacturing outside China too, with US manufacturing as “one of the options.”

But as of today, the vast majority of the world’s laptops are assembled in China — even US-based companies like Apple, Dell, or HP all look to contract manufacturers there.

The Telegraph doesn’t have a quote from Chen about possible price gouging, but rather puts it like this:

He said that some companies were likely to use the tariffs as an excuse to raise prices by more than 10pc.

We’ve reached out to other PC makers to see if they’ll follow suit. Apple, Dell, HP, and Lenovo didn’t immediately respond to a request for comment and didn’t comment for our earlier story; Asus and Razer are getting back to us.

Framework tells The Verge that its modular laptops themselves probably won’t be affected, but some of the modules themselves might. “Because we manufacture Framework Laptops and Mainboards in Taiwan, we have limited impact from the additional recently introduced tariffs. Some of our modules are manufactured in China, so we are taking this into account for future module pricing for US customers in the Framework Marketplace as we also continue to diversify our supply base,” says CEO Nirav Patel.

Personally, I can’t imagine using my smartphone as a baby steadicam anymore now that I have DJI’s Osmo Pocket 3 — we saw it everywhere at CES 2025, and its one-inch-type sensor offers image quality beyond what today’s smartphones allow.

But if you like the idea of sticking your phone into a set of stabilized motors on a stick, DJI’s just-announced $149 Osmo Mobile 7 Pro is the most feature-packed spin on the idea I’ve ever touched.

Now you can plop your selfie-stick down onto a surface with a new built-in tripod, pull eight inches of telescoping arm out of the top to give it reach, then snap on a new “Multifunctional Module” with a built-in computer vision camera to automatically track you, start and stop filming, and adjust framing, just by gesturing with your hands.

The Multifunctional Module unlocks other neat tricks too: it’s got a built-in fill light with adjustable color temperature that you can turn on by holding down the OM7 Pro’s side dial, and it’s a receiver for DJI’s excellent wireless microphones, including the DJI Mic Mini. It gets wirelessly powered right through DJI’s smartphone clamp thanks to a set of spring-loaded pogo pins, and you can attach it facing either direction if you want the assist for your phone’s selfie camera.

Perhaps most importantly, the module could dramatically reduce your need to use DJI’s app for filming. Previous gimbals required it for tracking, but historically it’s not always played nice with the newest camera features that smartphone makers offer. Now you can use your phone’s built-in camera app and still have it follow you autonomously. (DJI says its own Mimo app does now play nicer with the multi-lens switching, 4K60 recording and electronic image stabilization on more Android phones, but it’s more futureproof this way.)

The module’s tracking definitely works in my early tests, though the Multifunctional Module does get a bit hot — as it warns right on the module’s side. It’s also a bit of a battery drain. While DJI has upsized the battery to 12.06 watt-hours and quotes an upgraded 10 hours of battery life, the module cuts it in half to five hours with tracking, or four hours with tracking and the fill light on. (The Osmo Mobile 6’s 7.74Wh pack promised up to 6.5 hours, for comparison.)

I also like that the Osmo Mobile 7P still has a 1/4-inch tripod screw hole in addition to its built-in tripod, so you can mount it to other things, and you can still top up your phone from its USB-C port. I’m especially pleased to say that DJI will start selling a magnetic quick-release adapter so you can plop phones directly onto the gimbal instead of having to stretch apart a clamp — though that adapter won’t be compatible with the Multifunctional Module, and unlike that module, it won’t generally come in the box.

As far as I’m concerned, DJI’s own Osmo Pocket 3 sucked the air out of the room for smartphone gimbals, except in those cases where it’s easier to justify the $150 purchase for an existing phone instead of dropping $520 for better quality or because your iPhone workflow demands it. But I have to admit I’m jealous of the Osmo Mobile’s new tricks. Maybe DJI could make an Osmo Pocket 4 with a telescoping gimbal, pop-out tripod, and hand gestures?

The Pocket 3 ditched most of the Pocket and Pocket 2’s modular parts, proving that modularity wasn’t that product line’s real superpower, and I don’t think pocketability is necessarily it either. I think it’s having an ultra-steady camera that’s better and more versatile than today’s phones while still fitting into a shoulder bag, and I’d welcome even more of that versatility.

In addition to the $149 Osmo Mobile 7P, the company’s also announcing a new budget $89 Osmo Mobile 7 that comes in white rather than black. It also has the pull-out tripod, but ditches the telescoping selfie-stick, doesn’t come with the Multifunctional Module in the box, and doesn’t have the side dial to easily turn on the fill light and adjust focus or zoom.

Both should be shipping today from DJI’s website.

Two weeks ago, we exclusively reported Meta CEO Mark Zuckerberg’s remarks on how many pairs of Ray-Ban Meta smart glasses the company had recently sold and might theoretically sell — 1 million pairs in 2024, with the possibility of reaching 2 million or even 5 million by the end of 2025.

But glasses giant EssilorLuxottica, which produces those glasses for Meta, has now publicly revealed 2 million pairs of Meta Ray-Bans have sold since their October 2023 debut, and that it’s aiming to produce 10 million Meta glasses each year by the end of 2026.

“A pair of eyeglasses will be the main digital platform addressing our daily needs,” EssilorLuxottica CEO and chairman Francesco Milleri said on the company’s FY2024 financial results call (via UploadVR).

Milleri says his company is “planning for the long term with Meta” and thinking about Meta’s Ray-Bans as a “shared platform” that’s “ready to embark on third-party brands,” not just a single product. Bloomberg reported last month that Meta is planning Oakley-branded glasses too. (Oakley is one of EssilorLuxottica’s many other brands of glasses.)

“In the light of such evolution, and in line with our ambitious plan, we are currently expanding our production capacity for Ray-Ban Meta, set to reach 10 million annual unit by the end of next year,” Milleri said on the call. In September, the two companies announced a long-term partnership through at least 2030.

Milleri suggested that all of Meta’s glasses would be “supported by AI,” hinted that subscription services are coming as well, and said EssilorLuxottica is looking forward to Meta’s multimodal AI features expanding worldwide.

My colleague Victoria Song recently tested Meta’s new Live AI and live translation features for the Meta Ray-Bans, and found both features intriguing but not quite ready for primetime — but we’ve called the glasses themselves a “turning point”, and clones were everywhere at CES.

Meta is spending big to make its glasses a hit, paying for not one but two Super Bowl ads and a limited edition Super Bowl set of the shades, and it just hired the former CEO of luxury goods site The RealReal as its new VP of retail for wearables.

To become “the main digital platform addressing our daily needs,” Meta and EssilorLuxottica would need to grow glasses by two orders of magnitude; Samsung and Apple each sell over 200 million phones per year. Ten million per year would barely scratch that surface.

Meta is reportedly working on a pair of glasses with a display for later in 2025; my colleague Alex Heath has written it will ship alongside a neural wristband you can use to control it.

As always, the most important Nvidia graphics card is the one you can actually buy, and Nvidia’s talked a big game for its RTX 5070, making the dubious but nuanced claim it can deliver RTX 4090 performance for just $549. On February 28th, AMD will get its chance to intercept with the Radeon RX 9070 and 9070 XT, in a streaming event it just announced today. But Nvidia has now made its own wiggle room, delaying the launch of the RTX 5070 from February to March 5th, its product page reveals today.

Nvidia will ship its $749 RTX 5070 Ti ahead of AMD’s event, though, on February 20th, a week from today.

AMD has telegraphed that it won’t be competing with Nvidia’s latest and greatest cards, so price is the one big lever that AMD can potentially pull in order to compete. (The AMD Radeon 9070 cards appeared to be targeting Nvidia 4070 Ti and 4070 Super levels of performance, not necessarily higher.) But Nvidia, a company that can now make $20 billion in pure profit in a single quarter, could theoretically counter that if it feels it needs to, and now has more room to do so.

All that said, there is a rumor that AMD may be working on a more potent card than its already-announced RX 9070 and 9070 XT, one with 32GB of RAM. AMD gaming marketing boss Frank Azor says the 9070 XT won’t be a 32GB card, but did not address the larger rumor.

I like my phone, so I put a ring on it back in 2022 — the year my colleagues Jen and Allison introduced me to the Anker ring we now call “The best MagSafe phone grip”. It’s a simple magnetic ring attached to a non-magnetic finger ring that doubles as a kickstand.

But what if both rings were magnetic? What if both of them were the same size? What if each of them could stick to a phone — or a fridge, or a filing cabinet, or anywhere else you can stick one of their included adhesive-backed steel discs?

This thing exists. There are at least a dozen different varieties on the market now, each promising a folding quick-release variable angle phone mount that you can pocket and take a …

Read the full story at The Verge.

Nearly two years ago, Framework announced one of the most ambitious laptops ever live on a San Francisco stage, plus an AMD upgrade for its modular Intel machines. On February 25th, it’s promising “an even bigger set of announcements,” and that it’ll show “everything that we’ve been working on for the last two years.”

What could that mean? Well, the modular repairable gadget company did raise $18 million last April to expand beyond the laptop, but it’s not at all clear that’s what we’ll see, and a spokesperson wouldn’t comment when I asked. Instead, the company’s choosing to tease us with some pictograms! Look:

So we’re getting a “beach ball downward facing dog,” something Flash-y, and something worthy of a gamepad-only LAN party that serves knockoff Cheetos, yes?

(In all seriousness, my money’s on that first one being a convertible Chromebook with a backflipping 360-degree hinge — after all, Lenovo’s “Yoga” laptops famously cemented the convertible category.)

What are your guesses?

Three more things that caught my eye:

• “We will be opening pre-orders on at least one new item that day, so if you’re in the market for repairable, upgradeable, long-lasting consumer electronics products, you may want to create an account ahead of time to be ready.”
• “[W]e’re ready to bring this mission and product philosophy to even more of the world, one category at a time.”
• Framework fans willing to travel in the SF Bay Area can apply to attend in person.

Here’s the event page, with a live countdown and notify me button. Lastly, here’s the pre-prepped livestream which should go live at 10:30am PT / 1:30pm ET. Oh, and I’ll be there in person.

In 2023, when I realized small phones were truly dead, I decided I couldn’t wait for the Small Android Phone Project to revive them. That project, from the founder of the Pebble smartwatch, just wasn’t far enough along. Unfortunately, things aren’t any better in 2025 for small phone lovers — now that founder Eric Migicovsky is bringing back the Pebble smartwatch (yay!), he tells me he’s only “tangentially” working on small phones, and that they’re no longer “the top priority.”

“I really do hope someone else makes one so I don’t have to 😂,” writes Migicovsky.

I’m now one of a contingent of Samsung Galaxy Z Flip owners who is seriously considering trading in my Flip for a Galaxy S25. It’s not small, but it’s the closest thing we’ve got and my Flip’s battery doesn’t remotely last until bedtime anymore. I ask Migicovsky whether I can safely upgrade without feeling I might miss out on a possible Small Android Phone by, say, mid-2026. He says I can, without commenting on my proposed timeline.

At one point the phone project was targeting a 2024 release date and had gotten as far as creating an entire brand, Beep, to market it with, according to the portfolio of Small Android Phone Project industrial designer Alex De Stasio. He’d even dreamt up a billboard:

I’m still pretty excited for Migicovsky’s new project to revive the Pebble smartwatch — I loved my Pebble Time Steel and even owned two of them for a bit. You shouldn’t get your expectations up too high for that project, either, though: Migicovsky told the Android Faithful podcast that “this will not be a watch for everyone,” and expanded on that in a February 6th blog post:

Please don’t get your hopes up that the new watch will have X/Y/Z new feature. It’s going to be a Pebble and almost exactly as you remember it, except now with open source software that can you can modify and improve yourself. More hardware details will be shared in the future.

For the uninitiated, the original Pebbles were low-power devices whose best features were dead simplicity and battery life — no touchscreen, no digital crown, just a few buttons to help you read notifications and run a catalog of charming basic apps. That’s fine with me as long as the buttons are good!

A few other things he’s confirmed about the new Pebble so far:

• It will be new hardware, not an existing watch
• It’s targeting this specific chip as processor, which is marketed primarily as a Bluetooth SoC!
• You’ll be able to load your own firmware if you want to develop your own features

He told Android Faithful there are two software features he’d like to add. His “big thing” is to someday have a chat client, one that could address a limitation of early Pebbles by letting you see a whole conversation history on your wrist. The other is a basic AI handoff: “Pebble has a microphone, it has a screen… why can’t you talk to ChatGPT?”

Migicovsky should be in Shenzhen this week to meet with suppliers and factories for the new smartwatch.