Tech

report:-apple-is-testing-foldable-iphones,-having-the-same-problems-as-everyone-else

Report: Apple is testing foldable iPhones, having the same problems as everyone else

the story unfolds —

Don’t expect these clamshell-style foldables in 2024 or 2025 or maybe ever.

Report: Apple is testing foldable iPhones, having the same problems as everyone else

Samuel Axon

Apple is purportedly working on a foldable iPhone internally, according to “a person with direct knowledge of the situation” speaking to The Information. They’re said to be clamshell-style devices that fold like Samsung’s Galaxy Z Flip series rather than phones that become tablets like the Galaxy Z Fold or Google’s Pixel Fold.

The phones are also said to be “in early development” or “could be canceled.” If they do make it to market, it likely wouldn’t be until after 2025.

The report has a long list of design challenges that Apple has faced in developing foldable phones: they’re too thick when folded up; they’re easily broken; they would cost more than non-foldable versions; the seam in the middle of the display tends to be both visible and feel-able; and the hinge on an iPad-sized device would prevent the device from sitting flat on a table (though this concern hasn’t stopped Apple from introducing substantial camera bumps on many of its tablets and all of its phones).

If many of those challenges sound familiar, it’s because it’s a detailed list of virtually every bad thing you could say about current foldable Android phones, even after multiple hardware generations. Our first Pixel Fold didn’t even survive the pre-release review period, and those well-earned durability concerns plus the relatively high cost have limited foldable phones to roughly 1.6 percent of all smartphone sales, according to recent analyst estimates.

It makes sense that Apple would be testing some big swings as it thinks about the next era of iPhone design; our iPhone 15 review called them the iPhone’s “final form,” insofar as it feels like there’s not much room to continue to improve on the iPhone X-style full-screen design that Apple has been iterating on since 2017. It sounds like foldable phones will only be in Apple’s future if the company can manage to overcome the same issues that have tripped up other foldables—though to be fair, the company does have a pretty good decadeslong track record on that front.

Report: Apple is testing foldable iPhones, having the same problems as everyone else Read More »

youtube-tv-is-the-us’s-4th-biggest-cable-tv-provider,-with-8-million-subs

YouTube TV is the US’s 4th-biggest cable TV provider, with 8 million subs

Still not covering that $2 billion-a-year Sunday Ticket deal, though —

Google’s $73-a-month service is going toe-to-toe with the cable companies.

YouTube TV is the US’s 4th-biggest cable TV provider, with 8 million subs

YouTube is still slowly dripping out stats about its subscriber base. After the announcement last week that YouTube Premium had hit 100 million subscribers, the company now says YouTube TV, its cable subscription plan, has 8 million subscribers.

Eight million subscribers might sound paltry compared to the 100 million people on Premium, but Premium is only $12. YouTube TV is one of the most expensive streaming subscriptions at $73 a month. The cable-like prices are because this is a cable-like service: a huge bundle of 100-plus channels featuring cable TV stalwarts like CNN, ESPN, and your local NBC, CBS, and ABC channels. $73 is also the base price. Like cable TV, there are additional add-on packages for premium movie channels like HBO and Showtime, 4K packages, and other sports and language add-ons. Let’s also not forget NFL Sunday Ticket, which this year became a YouTube TV exclusive, as a $350-a-year add-on to the $73-a-month service (there’s also a $ 450-a-year standalone package).

The subscriber numbers come from a “Letter from the YouTube CEO” blog post for 2024 from YouTube CEO Neal Mohan. With YouTube basically unable to get any bigger as the Internet’s defacto video host, Mohan says the “next frontier” for YouTube is “the living room and subscriptions.” Mohan wants users “watching YouTube the way we used to sit down together for traditional TV shows—on the biggest screen in the home with friends and family,” and says that “viewers globally now watch more than 1 billion hours on average of YouTube content on their TVs every day.”

YouTube TV’s 8 million subscribers make it one of the biggest cable TV providers. Leichtman Research Group‘s subscriber numbers for “Major Pay-TV Providers” (that means cable companies and their competitors) in Q3 2023 had No. 1 Comcast and No. 2 Charter both in the 14 million user range, with DirectTV in third with 11.9 million, and Dish in fourth at 6.7 million customers. Leichtman had YouTube TV in fifth, with 6.5 million users. With No. 4 Dish losing customers every quarter, YouTube TV is in fourth place now. It might be No. 3 soon. Leichtman’s numbers had YouTube TV as the fastest grower of the bunch, adding 600,000 customers in Q3, while DirecTV was the biggest loser, with half a million customers dumping their satellite dishes. Q3 marked the start of NFL Sunday Ticket moving from DirecTV to YouTube TV.

Naturally, these are all US numbers, and being nationwide puts YouTube TV on the same playing field as satellite companies, a big advantage compared to regional cable TV providers. YouTube TV has bigger ambitions than just the US, though. During the January earnings call, Google said it was “looking closely at” expanding the service to more countries. YouTube TV would need to clear an expansion with every single channel partner on the service, though, so it has a lot of negotiations to work through.

YouTube TV is the US’s 4th-biggest cable TV provider, with 8 million subs Read More »

those-free-usb-sticks-in-your-drawer-are-somehow-crappier-than-you-thought

Those free USB sticks in your drawer are somehow crappier than you thought

Race to the bottom of the serial bus —

Rejected chips, hidden microSD cards plague the USB stick market.

Textless microSD card fused ont a USB controller

Enlarge / A microSD card of “unknown origin” is soldered onto a USB interface board to serve as makeshift NAND storage.

CBL Data recovery

When a German data recovery firm recently made a study of the failed flash storage drives it had been sent, it noticed some interesting, and bad, trends.

Most of them were cheap sticks, the kind given away by companies as promotional gifts, but not all of them. What surprised CBL Data Recovery was the number of NAND chips from reputable firms, such as Samsung, Sandisk, or Hynix, found inside cheaper devices. The chips, which showed obvious reduced capacity and reliability on testing, had their manufacturers’ logo either removed by abrasion or sometimes just written over with random text.

Sometimes there wasn’t a NAND chip at all, but a microSD card—possibly also binned during quality control—scrubbed of identifiers and fused onto a USB interface board. On “no-name” products, there is “less and less reliability,” CBL wrote (in German, roughly web-translated). CBL did find branded products with similar rubbed-off chips and soldered cards but did not name any specific brands in its report.

  • While most chips had their manufacturer’s name scratched off their seals, one cheap USB stick simply stamped enough capital-letter text over the name to make it unintelligible.

  • Detail on a NAND chip that has its make and original name removed by abrasion (look for the circular pattern in a pre-defined area on the chip cover).

Beyond obvious physical corner-cutting, a general trend in NAND storage cells has contributed to a lower overall reliability, according to CBL. SLC, or single-level cell storage, has one bit per cell, 1 or 0, which are two different voltage levels. A QLC (quadruple-level) chip uses four bits per cell, which means 16 voltage levels that must be correct. QLC allows for denser storage, but, as we noted previously: “As the data density of NAND cells goes up, their speed and write endurance decreases—it takes more time and effort to read or write one of eight discrete voltage levels to a cell than it does to get or set a simple, unambiguous on/off value.”

With high-quality chips, there’s a lot of work put in to correct errors and control temperatures. With chips that are not actually chips or were grabbed from the quality-control discard bin and scrubbed of their logo, “data loss is not surprising,” CBL writes.

All told, CBL’s report makes the case for never putting anything you really need to keep stored long-term on a USB stick. This might not be a revelation for those who have read up on proper storage practices, but CBL has further recommendations for those keeping anything at all on USB sticks:

  • Keep them stored somewhere cool
  • Don’t use promotional sticks for anything of any importance
  • Write and read to a USB stick once or twice a year, to engage error correction (at least in higher-quality sticks)
  • Don’t stuff the disk full, if you can avoid it, to give data maintenance and error correction a fighting chance.

The market for affordable, pocket-sized storage has proven itself to be a messy one over the last few years. High-capacity storage is, in fact, getting cheaper, but not in every corner—at least, not when you look closely. In mid-2022, a “30TB” external SSD was listed on Walmart and AliExpress for just over $30. Inside were two microSD cards, hot-glued to a USB 2.0 board and loaded with firmware that both misrepresents itself to Windows and simply rewrites its limited space over and over as you copy to it.

Similarly, a “16TB” SSD, listed for a relatively reasonable $70 and sporting dozens of five-star reviews, seemed to be actually 64GB worth of microSD cards, as Review Geek discovered. We noted a plethora of similar cons when we wrote about it, along with the problem of Amazon sellers’ ability to disappear as soon as the jig is up, only to reappear soon after with a new batch of microSD cards upsold with exponentially more faux-capacity.

Those free USB sticks in your drawer are somehow crappier than you thought Read More »

your-current-pc-probably-doesn’t-have-an-ai-processor,-but-your-next-one-might

Your current PC probably doesn’t have an AI processor, but your next one might

Intel's Core Ultra chips are some of the first x86 PC processors to include built-in NPUs. Software support will slowly follow.

Enlarge / Intel’s Core Ultra chips are some of the first x86 PC processors to include built-in NPUs. Software support will slowly follow.

Intel

When it announced the new Copilot key for PC keyboards last month, Microsoft declared 2024 “the year of the AI PC.” On one level, this is just an aspirational PR-friendly proclamation, meant to show investors that Microsoft intends to keep pushing the AI hype cycle that has put it in competition with Apple for the title of most valuable publicly traded company.

But on a technical level, it is true that PCs made and sold in 2024 and beyond will generally include AI and machine-learning processing capabilities that older PCs don’t. The main thing is the neural processing unit (NPU), a specialized block on recent high-end Intel and AMD CPUs that can accelerate some kinds of generative AI and machine-learning workloads more quickly (or while using less power) than the CPU or GPU could.

Qualcomm’s Windows PCs were some of the first to include an NPU, since the Arm processors used in most smartphones have included some kind of machine-learning acceleration for a few years now (Apple’s M-series chips for Macs all have them, too, going all the way back to 2020’s M1). But the Arm version of Windows is a insignificantly tiny sliver of the entire PC market; x86 PCs with Intel’s Core Ultra chips, AMD’s Ryzen 7040/8040-series laptop CPUs, or the Ryzen 8000G desktop CPUs will be many mainstream PC users’ first exposure to this kind of hardware.

Right now, even if your PC has an NPU in it, Windows can’t use it for much, aside from webcam background blurring and a handful of other video effects. But that’s slowly going to change, and part of that will be making it relatively easy for developers to create NPU-agnostic apps in the same way that PC game developers currently make GPU-agnostic games.

The gaming example is instructive, because that’s basically how Microsoft is approaching DirectML, its API for machine-learning operations. Though up until now it has mostly been used to run these AI workloads on GPUs, Microsoft announced last week that it was adding DirectML support for Intel’s Meteor Lake NPUs in a developer preview, starting in DirectML 1.13.1 and ONNX Runtime 1.17.

Though it will only run an unspecified “subset of machine learning models that have been targeted for support” and that some “may not run at all or may have high latency or low accuracy,” it opens the door to more third-party apps to start taking advantage of built-in NPUs. Intel says that Samsung is using Intel’s NPU and DirectML for facial recognition features in its photo gallery app, something that Apple also uses its Neural Engine for in macOS and iOS.

The benefits can be substantial, compared to running those workloads on a GPU or CPU.

“The NPU, at least in Intel land, will largely be used for power efficiency reasons,” Intel Senior Director of Technical Marketing Robert Hallock told Ars in an interview about Meteor Lake’s capabilities. “Camera segmentation, this whole background blurring thing… moving that to the NPU saves about 30 to 50 percent power versus running it elsewhere.”

Intel and Microsoft are both working toward a model where NPUs are treated pretty much like GPUs are today: developers generally target DirectX rather than a specific graphics card manufacturer or GPU architecture, and new features, one-off bug fixes, and performance improvements can all be addressed via GPU driver updates. Some GPUs run specific games better than others, and developers can choose to spend more time optimizing for Nvidia cards or AMD cards, but generally the model is hardware agnostic.

Similarly, Intel is already offering GPU-style driver updates for its NPUs. And Hallock says that Windows already essentially recognizes the NPU as “a graphics card with no rendering capability.”

Your current PC probably doesn’t have an AI processor, but your next one might Read More »

boston-dynamics’-atlas-tries-out-inventory-work,-gets-better-at-lifting

Boston Dynamics’ Atlas tries out inventory work, gets better at lifting

also better at stumbling —

Atlas learns to pick up a 30-lb car strut and carefully manipulate it.

  • Boston Dynamics’ Atlas research robot.

    Boston Dynamics

  • Atlas’ new spindly, double-jointed fingers are capable but a bit creepy.

    Boston Dynamics

  • Atlas’ old hands were rudimentary clamps, and look at all the damage they did to this plank of wood. It was just crushing things.

    Boston Dynamics

  • More finger movement.

    Boston Dynamics

  • From the robot’s point of view. The video overlays the real world with 3D models of Atlas’ hands and the object.

    Boston Dynamics

  • Atlas has to first balance the strut on a shelf, then it can slide it into place.

    Boston Dynamics

  • You can see all the work that goes into this lift. Recognize the object, wrap your one hand around it, pull it out enough to balance it on the edge of the container, wrap your other hand around it, then torque your upper body to rotate the strut into position.

    Boston Dynamics

The world’s most advanced humanoid robot, Boston Dynamics’ Atlas, is back, and it’s moving some medium-weight car parts. While the robot has mastered a lot of bipedal tricks like walking, running, jumping, and even backflips, it’s still in the early days of picking stuff up. When we last saw the robot, it had sprouted a set of rudimentary hand clamps and was using those to carry heavy objects like a toolbox, barbells, and a plank of wood. The new focus seems to be on “kinetically challenging” work—these things are heavy enough to mess with the robot’s balance, so picking them up, carrying them, and putting them down requires all sorts of additional calculations and planning so the robot doesn’t fall over.

In the latest video, we’re on to what looks like “phase 2” of picking stuff up—being more precise about it. The old clamp hands had a single pivot at the palm and seemed to just apply the maximum grip strength to anything the robot picked up. The most delicate thing Atlas picked up in the last video was a wooden plank, and it was absolutely destroying the wood. Atlas’ new hands look a lot more gentle than The Clamps, with each sporting a set of three fingers with two joints. All the fingers share one big pivot point at the palm of the hand, and there’s a knuckle joint halfway up the finger. The fingers are all very long and have 360 degrees of motion, so they can flex in both directions, which is probably effective but very creepy. Put two fingers on one side of an item and the “thumb” on the other, and Atlas can wrap its hands around objects instead of just crushing them.

Sadly all we’re getting is this blurry 1 minute video with no explanation as to what’s going on.

Atlas is picking up a set of car struts—an object with extremely complicated topography that weighs around 30 pounds—so there’s a lot to calculate. Atlas does a heavy two-handed lift of a strut from a vertical position on a pallet, walks it over to a shelf, and carefully slides it into place. This is all in Boston Dynamics’ lab, but it’s close to repetitive factory or shipping work. Everything here seems designed to give the robot a manipulation challenge. The complicated shape of the strut means there are a million ways you could grip it incorrectly. The strut box has tall metal poles around it, so the robot needs to not bang the strut into the obstacle. The shelf is a tight fit, so the strut has to be placed on the edge of the shelf and slid into place, all while making sure the strut’s many protrusions won’t crash into the shelf.

One limitation here is that at least some of the smarts in the video are pre-calculated—at one point, we see what looks like Atlas’ vision processing, and it has a perfect 3D scan of the car strut ready to go. So this is either attempt number 5,000, and it has already seen the strut from all angles, or Atlas was pre-programmed with topographical data for this exact model car strut. Either way, for all the lifts in the video, Atlas is saved from trying to figure out the shape of the object in real time. Atlas has a lidar sensor on its face and can generate a point cloud of what it’s looking at, so it just needs to line up the pre-baked model with the point cloud, and it has perfect knowledge of the strut topography. A harder level of difficulty would be picking up an object Atlas has never seen before, but you’ve got to break down the challenges into smaller parts and start somewhere.

When Atlas picks up a strut, it has to walk around a pallet, and as always, the robot shines when it comes to bipedal movement. The simpler way to move around the pallet would be a set of straight-line walking paths with pivots in between. Atlas’ path-planning is way more complicated, though, and involves more advanced side-step moves, leaning into turns, and just dynamically stumbling around the pallet any way it can. This version of Atlas moves less like a robot and more like a drunk person, which is a big compliment. At one point, it even stumbles and recovers, drawing an excited reaction from onlookers in the background.

Boston Dynamics’ Atlas tries out inventory work, gets better at lifting Read More »

google-and-mozilla-don’t-like-apple’s-new-ios-browser-rules

Google and Mozilla don’t like Apple’s new iOS browser rules

Surely US regulators will help us… —

Google and Mozilla want iOS’s new EU browser rules to apply worldwide.

Extreme close-up photograph of finger above Chrome icon on smartphone.

Apple is being forced to make major changes to iOS in Europe, thanks to the European Union’s “Digital Markets Act.” The act cracks down on Big Tech “gatekeepers” with various interoperability, fairness, and privacy demands, and part of the changes demanded of Apple is to allow competing browser engines on iOS. The change, due in iOS 17.4, will mean rival browsers like Chrome and Firefox get to finally bring their own web rendering code to iPhones and iPads. Despite what sounds like a big improvement to the iOS browser situation, Google and Mozilla aren’t happy with Apple’s proposed changes.

Earlier, Mozilla spokesperson Damiano DeMonte gave a comment to The Verge on Apple’s policy changes and took issue with the decision to limit the browser changes to the EU. “We are still reviewing the technical details but are extremely disappointed with Apple’s proposed plan to restrict the newly-announced BrowserEngineKit to EU-specific apps,” DeMonte said. “The effect of this would be to force an independent browser like Firefox to build and maintain two separate browser implementations—a burden Apple themselves will not have to bear.” DeMonte added: “Apple’s proposals fail to give consumers viable choices by making it as painful as possible for others to provide competitive alternatives to Safari. This is another example of Apple creating barriers to prevent true browser competition on iOS.”

Apple’s framework that allows for alternative browser engines is called “BrowserEngineKit” and already has public documentation as part of the iOS 17.4 beta. Browser vendors will need to earn Apple’s approval to use the framework in a production app, and like all iOS apps, that approval will come with several requirements. None of the requirements jump out as egregious: Apple wants browser vendors to have a certain level of web standards support, pledge to fix security vulnerabilities quickly and protect the user’s privacy by showing the standard consent prompts for access to things like location. You’re not allowed to “sync cookies and state between the browser and any other apps, even other apps of the developer,” which seems aimed directly at Google and its preference to have all its iOS apps talk to each other. The big negative is that your BrowserEngineKit app is limited to the EU, because—surprise—the EU rules only apply to the EU.

Speaking of Google, Google’s VP of engineering for Chrome, Parisa Tabriz, commented on DeMonte’s statement on X, saying, “Strong agree with @mozilla. @Apple isn’t serious about supporting web browser or engine choice on iOS. Their strategy is overly restrictive, and won’t meaningfully lead to real choice for browser developers.”

Today, you can download what look like “alternative” browsers on iOS, like Chrome and Firefox, but these browsers are mostly just skins overtop of Apple’s Safari engine. iOS app developers aren’t actually allowed to include their own browser engines, so everything uses Safari’s WebKit engine, with a new UI and settings and sync features layered on top. That means all of WebKit’s bugs and feature support decisions apply to every browser.

Being stuck with Safari isn’t great for users. Over the years, Safari has earned a reputation as “the new IE” from some web developers, due to lagging behind the competition in its support for advanced web features. Safari has gotten notably better lately, though. For instance, in 2023, it finally shipped support for push notifications, allowing web apps to better compete with native apps downloaded from Apple’s cash-cow App Store. Apple’s support of push notifications came seven years after Google and Mozilla rolled out the feature.

More competition would be great for the iOS browser space, but the reality is that competition will mostly be from the other big “gatekeeper” in the room: Google. Chrome is the project with the resources and reach to better compete with Safari, and working its way into iOS will bring the web close to a Chrome monoculture. Google’s browser may have better support for certain web features, but it will also come with a built-in tracking system that spies on users and serves up their interests to advertisers. Safari has a much better privacy story.

Even though only EU users will get to choose from several actually different browsers, everyone still has to compete in the EU, and that includes Safari. For the rest of the world, even they don’t get a real browser choice; competing in the EU browser wars should make the only iOS browser better for everyone. The EU rules have a compliance deadline of March 2024, so iOS 17.4 needs to be out by then. Google and Mozilla have been working on full versions of their browsers for iOS for at least a year now. Maybe they’ll be ready for launch?

Google and Mozilla don’t like Apple’s new iOS browser rules Read More »

windows-version-of-the-venerable-linux-“sudo”-command-shows-up-in-preview-build

Windows version of the venerable Linux “sudo” command shows up in preview build

sudo start your photocopiers —

Feature is experimental and, at least currently, not actually functional.

Not now, but maybe soon?

Enlarge / Not now, but maybe soon?

Andrew Cunningham

Microsoft opened its arms to Linux during the Windows 10 era, inventing an entire virtualized subsystem to allow users and developers to access a real-deal Linux command line without leaving the Windows environment. Now, it looks like Microsoft may embrace yet another Linux feature: the sudo command.

Short for “superuser do” or “substitute user do” and immortalized in nerd-leaning pop culture by an early xkcd comic, sudo is most commonly used at the command line when the user needs administrator access to the system—usually to install or update software, or to make changes to system files. Users who aren’t in the sudo user group on a given system can’t run the command, protecting the rest of the files on the system from being accessed or changed.

In a post on X, formerly Twitter, user @thebookisclosed found settings for a Sudo command in a preview version of Windows 11 that was posted to the experimental Canary channel in late January. WindowsLatest experimented with the setting in a build of Windows Server 2025, which currently requires Developer Mode to be enabled in the Settings app. There’s a toggle to turn the sudo command on and off and a separate drop-down to tweak how the command behaves when you use it, though as of this writing the command itself doesn’t actually work yet.

The sudo command is also part of the Windows Subsystem for Linux (WSL), but that version of the sudo command only covers Linux software. This one seems likely to run native Windows commands, though obviously we won’t know exactly how it works before it’s enabled and fully functional. Currently, users who want a sudo-like command in Windows need to rely on third-party software like gsudo to accomplish the task.

The benefit of the sudo command for Windows users—whether they’re using Windows Server or otherwise—would be the ability to elevate the privilege level without having to open an entirely separate command prompt or Windows Terminal window. According to the options available in the preview build, commands run with sudo could be opened up in a new window automatically, or they could happen inline, but you’d never need to do the “right-click, run-as-administrator” dance again if you didn’t want to.

Microsoft regularly tests new Windows features that don’t make it into the generally released public versions of the operating system. This feature could also remain exclusive to Windows Server without making it into the consumer version of Windows. But given the command’s presence in Linux and macOS, it will be a nice quality-of-life improvement for Windows users who spend lots of time staring at the command prompt.

Microsoft is borrowing a longstanding Linux feature here, but that road goes both ways—a recent update to the Linux systemd software added a Windows-inspired “blue screen of death” designed to give users more information about crashes when they happen.

Windows version of the venerable Linux “sudo” command shows up in preview build Read More »

google-will-no-longer-back-up-the-internet:-cached-webpages-are-dead

Google will no longer back up the Internet: Cached webpages are dead

We need a cached version of Google, the company —

Google Search will no longer make site backups while crawling the web.

A large Google logo is displayed amidst foliage.

Google will no longer be keeping a backup of the entire Internet. Google Search’s “cached” links have long been an alternative way to load a website that was down or had changed, but now the company is killing them off. Google “Search Liaison” Danny Sullivan confirmed the feature removal in an X post, saying the feature “was meant for helping people access pages when way back, you often couldn’t depend on a page loading. These days, things have greatly improved. So, it was decided to retire it.”

The feature has been appearing and disappearing for some people since December, and currently, we don’t see any cache links in Google Search. For now, you can still build your own cache links even without the button, just by going to “https://webcache.googleusercontent.com/search?q=cache:” plus a website URL, or by typing “cache:” plus a URL into Google Search. For now, the cached version of Ars Technica seems to still work. All of Google’s support pages about cached sites have been taken down.

Cached links used to live under the drop-down menu next to every search result on Google’s page. As the Google web crawler scoured the Internet for new and updated webpages, it would also save a copy of whatever it was seeing. That quickly led to Google having a backup of basically the entire Internet, using what was probably an uncountable number of petabytes of data. Google is in the era of cost savings now, so assuming Google can just start deleting cache data, it can probably free up a lot of resources.

Cached links were great if the website was down or quickly changed, but they also gave some insight over the years about how the “Google Bot” web crawler views the web. The pages aren’t necessarily rendered like how you would expect. In the past, pages were text-only, but slowly the Google Bot learned about media and other rich data like javascript (there are a ton of specialized Google Bots now). A lot of Google Bot details are shrouded in secrecy to hide from SEO spammers, but you could learn a lot by investigating what cached pages look like. In 2020, Google switched to mobile-by-default, so for instance, if you visit that cached Ars link from earlier, you get the mobile site. If you run a website and want to learn more about what a site looks like to a Google Bot, you can still do that, though only for your own site, from the Search Console.

The death of cached sites will mean the Internet Archive has a larger burden of archiving and tracking changes on the world’s webpages.

Google will no longer back up the Internet: Cached webpages are dead Read More »

“rasti-computer”-is-a-detailed-grid-compass-tribute-made-from-framework-innards

“Rasti Computer” is a detailed GRiD Compass tribute made from Framework innards

But can it play Pitfall!? —

It’s a custom keyboard, an artfully dinged-up case, and a wonderful throwback.

Penk Chen's Rasti Computer

Enlarge / Penk Chen’s Rasti Computer, built with 3D printing, Framework laptop internals, and a deep love for the first laptop that went to space.

If I had to figure out what to do with the insides of a Framework 13 laptop I had lying around after today, I might not turn it into a strange but compelling “Slabtop” this time.

No, I think that, having seen Penk Chen’s remarkable project to fit Framework parts into a kind of modern restyling of the Grid Compass laptop, I would have to wait until Chen posts detailed build instructions for this project… and until I had a 3D printer… and could gather the custom mechanical keyboard parts. Sure, that’s a lot harder, but it’s hard to put a price on drawing unnecessary attention to yourself while you chonk away on your faux-used future laptop.

The Rasti Computer, which Chen writes is “derived from the German compound word ‘Rasterrahmen’ (grid + framework),” has at its core the mainboard, battery, and antennae from the highly modular and repairable first-generation Framework laptop. It takes input from the custom keyboard Chen designed for the chassis, with custom PCB and 3D-printed keycaps and case. It sends images to a 10.4-inch QLED 1600×720 display, and it all fits inside a bevy of 3D-printed pieces with some fairly standard hex-head bolts. Oh, and the hinges from a 2012 13-inch MacBook pro, though that’s possibly negotiable.

  • Rear view of the Rasti Computer, with “a touch of silver dry brushing [that] added the beat-up metal look.”

  • Semi-exploded view of the Rasti Computer.

  • You can, of course, run Windows on this device, if you like. But it might feel dissonant to put so much custom work in to run a stock OS.

Chen’s project derives from, and pays tribute to, the Grid Compass (styled “GRiD” by its maker, GRiD Systems Corp.). The Compass was probably (again, probably) the first clamshell-style laptop made. It saw use by NASA’s Space Shuttle program, as well as by military and other entities needing a laptop that was both compact and throw-it-at-a-wall durable. It had 256KB of memory by default (less than half the amount Bill Gates didn’t say you should ever need), a 320×240 pixel screen, and an Intel 8086 processor. Some models contained a 1,200bps modem. It cost more than $8,000 in 1982, or almost $25,000 today.

We have it on good word from some resident vintage computer collectors that the Compass remains a rare and expensive item to get. Rebuilding a Framework mainboard into a modern-day Grid-like doesn’t seem particularly cheap, depending on your 3D printer setup, or lack thereof. Nor is it likely to be easy, given a glimpse at how it goes together. But it will give you a unique portable and conversation piece, one that runs programs beyond Grid-OS.

You can read more about the Grid Compass at Cooper Hewitt, the firm where Compass designer Bill Moggridge worked as design director from 2010 until his 2011 passing. If you remember bubble memory, it’s a dip back into that genial trauma. Hackaday, where we first saw the Rasti project, wrote up a similarly Compass-inspired laptop, the GRIZ Sextant, with a Raspberry Pi at its core.

“Rasti Computer” is a detailed GRiD Compass tribute made from Framework innards Read More »

new-6gb-version-of-the-rtx-3050-may-be-nvidia’s-first-sub-$200-gpu-in-over-4-years

New 6GB version of the RTX 3050 may be Nvidia’s first sub-$200 GPU in over 4 years

clearing a low bar —

Exciting? No. New technology? Also no. But it ought to be better than a 1650.

New 6GB version of the RTX 3050 may be Nvidia’s first sub-$200 GPU in over 4 years

Gigabyte

Nvidia launched three new GPUs last month, part of a Super overhaul of the RTX 40-series designed to improve the value of the company’s $600-and-up graphics cards.

But today, the company is quietly doing something that it hasn’t done in over four years: launching a sub-$200 graphics card. As spotted by TechPowerUp, Nvidia partners like Gigabyte have begun officially announcing a 6GB version of the old RTX 3050 graphics card, albeit with less memory and memory bandwidth, fewer CUDA cores, and lower power requirements.

The announcement follows a few days of leaked retail listings, which generally point to an MSRP of roughly $179 for the new-old card. This would make it Nvidia’s first sub-$200 graphics card launch since the GeForce GTX 1650 Super came out in late 2019, a four-year gap caused partially by a cryptocurrency- and pandemic-fueled GPU shortage that lasted from late 2020 into mid-to-late 2022.

RTX 3050 6GB RTX 3050 8GB GTX 1650 Super GTX 1650 RTX 2060 6GB RTX 3060 12GB RTX 4060
CUDA Cores 2,048 or 2,304? 2,560 1,280 896 1,920 3,584 3,072
Boost Clock 1,470 MHz 1,777 MHz 1,725 MHz 1,665 MHz 1,680 MHz 1,777 MHz 2,460 MHz
Memory Bus Width 96-bit 128-bit 128-bit 128-bit 128-bit 192-bit 128-bit
Memory Clock 1,750 MHz 1,750 MHz 1,500 MHz 2,000 MHz 1,750 MHz 1,875 MHz 2,125MHz
Memory size 6GB GDDR6 8GB GDDR6 4GB GDDR6 4GB GDDR5 6GB GDDR6 12GB GDDR6 8GB GDDR6
TGP 70 W  130 W 100 W 75 W 160 W 170 W 115 W

We weren’t in love with the original 8GB version of the 3050—”weird,” “overpriced,” and “could’ve been worse,” we wrote of it during the depths of the GPU shortage in early 2022—and the 6GB version is cut down even further. Its 6GB of memory is attached to a narrow 96-bit memory bus. And while reports differ on exactly how many CUDA cores we can expect—TechPowerUp and Tom’s Hardware say 2,048, while Gigabyte’s 6GB 3050 product pages list 2,304—it’s clear that there are fewer than in the original RTX 3050, and their clock speed is lower to boot.

These specs should make it an OK card for 1080p gaming at medium to high settings, depending on the game. It should also be a decent fit for small-form-factor systems—one of Gigabyte’s versions is a low-profile card suitable for many office desktops—and its 70 W power requirement should mean that the cards can draw all the power they need from the PCI Express slot, without the need for an external power connector.

As a cut-down version of a 2-year-old card based on a 3.5-year-old last-generation GPU architecture, the RTX 3050 isn’t very exciting. But it gets more interesting once you consider that Nvidia’s partners are currently selling the old non-Super version of the GTX 1650 for around the same price. It’s difficult to predict how much the narrower memory bus will impact the 6GB 3050’s performance, but compared to the 1650, it has a significantly higher number of newer CUDA cores, ray-tracing support, another 2GB of RAM, GDDR6 instead of GDDR5, and support for Nvidia’s DLSS upscaling technology (albeit not DLSS Frame Generation, which remains exclusive to the 40-series).

The new card also complicates one of AMD’s selling points for its new Ryzen 8000G processors, which include reasonably capable 1080p integrated GPUs. AMD compared the cost of a $329 Ryzen 7 8700G favorably to the cost of a Core i5 CPU and a GTX 1650 GPU, which deliver similar performance for more money. But the RTX 3050 should decisively outrun the 8700G, and AMD’s pricing argument was already being undercut by the extra expense of socket AM5 motherboards and DDR5 memory.

Nice as it is to see at least some twitch of life in the entry-level GPU market, we’re still a long way from where we were in the mid- to late-2010s, when sub-$200 cards like the GTX 1050 Ti and the 1650 launched not long after the other cards in the same series and used the newest GPU architectures available at the time. There’s a laptop GPU in the RTX 4050 series, but a desktop version looks highly unlikely at this point—to say nothing of a desktop version at or under $200. But a belated, uninspiring improvement is an improvement nonetheless.

New 6GB version of the RTX 3050 may be Nvidia’s first sub-$200 GPU in over 4 years Read More »

youtube-premium-announces-100-million-subscribers

YouTube Premium announces 100 million subscribers

Did someone really forget about this during the earnings call? —

Ad-free videos and YouTube Music access hits a major milestone.

YouTube Premium announces 100 million subscribers

Hot on the heels of Google’s “One” subscription plan obtaining 100 million users, YouTube is also hitting that big milestone, with 100 million people paying for Premium and YouTube Music. YouTube’s subscription data didn’t make it into the earnings call three days ago.

It’s hard to know what exactly is driving YouTube’s subscriptions. Premium gets you both ad-free YouTube videos and YouTube Music, and it’s easy to imagine people sticking to one or the other. Ad-free videos have been getting the most aggressive promotion lately, with Google cracking down on ad-block users by blocking video playback and displaying interstitial pop-ups. After warning users that ad blockers violate YouTube’s terms of service, the pop-ups show a big “try YouTube Premium” button. Premium also added an exclusive “enhanced bitrate” 1080p setting, although 2K, 4K, and 8K options have always been free.

There’s not much new on the music side of things. YouTube Music is free with ads and a more limited feature set, but subscribing gets you ad-free playback, background playback on phones, and access to YouTube Music streaming on Google’s various speakers. YouTube’s blog post highlights quotes from many big music industry CEOs celebrating the service.

Google’s announcement bundles together two different subscriptions. There’s the $13.99 per month YouTube Premium subscription, which gets you ad-free YouTube and YouTube Music, and a music-only “YouTube Music Premium” subscription, which is $10.99 per month (Google increased prices last year). If you’re a Spotify customer and don’t want Google’s music offering, the company doesn’t have a plan for you. From 2021 to 2023, Google had a music-free “YouTube Premium Lite” subscription plan available only in Europe, but the company killed the plan a few months ago.

YouTube Premium announces 100 million subscribers Read More »

google’s-pixel-storage-issue-fix-requires-developer-tools-and-a-terminal

Google’s Pixel storage issue fix requires developer tools and a terminal

Stagefright’s revenge —

Automatic updates broke your phone; the fix is a highly technical manual process.

Google’s Pixel storage issue fix requires developer tools and a terminal

Google has another fix for the second major storage bug Pixel phones have seen in the last four months. Last week, reports surfaced that some Pixel owners were being locked out of their phone’s local storage, creating a nearly useless phone with all sorts of issues. Many blamed the January 2024 Google Play system update for the issue, and yesterday, Google confirmed that hypothesis. Google posted an official solution to the issue on the Pixel Community Forums, but there’s no user-friendly solution here. Google’s automatic update system broke people’s devices, but the fix is completely manual, requiring users to download the developer tools, install drivers, change settings, plug in their phones, and delete certain files via a command-line interface.

The good news is that, if you’ve left your phone sitting around in a nearly useless state for the last week or two, following the directions means you won’t actually lose any data. Having a week or two of downtime is not acceptable to a lot of people, though, and several users replied to the thread saying they had already wiped their device to get their phone working again and had to deal with the resulting data loss (despite many attempts and promises, Android does not have a comprehensive backup system that works).

The bad news is that I don’t think many normal users will be able to follow Google’s directions. First, you’ll need to perform the secret action to enable Android’s Developer Options (you tap on the build number seven times). Then, you have to download Google’s “SDK Platform-Tools” zip file, which is meant for app developers. After that, plug in your phone, switch to the correct “File transfer” connection mode, open a terminal, navigate to the platform-tools folder, and run both “./adb uninstall com.google.android.media.swcodec” and “./adb uninstall com.google.android.media.” Then reboot the phone and hope that works.

I skipped a few steps (please read Google’s instructions if you’re trying this), but that’s the basic gist of it. The tool Google is having people use is “ADB,” or the “Android Debug Bridge.” This is meant to give developers command-line access to their phones, which allows them to quickly push new app builds to the device, get a readout of system logs, and turn on special developer flags for various testing.

Google’s instructions will only work if everything goes smoothly, and as someone with hundreds of hours in ADB from testing various Android versions, I will guess that it will probably not go smoothly. On Windows, the ADB drivers often don’t install automatically. Instead, you’ll get “unknown device” or some other incorrect device detection, and you won’t be able to run any commands. You usually have to use the “let me pick from drivers on my computer” option, browse through your file system, and manually “select” (more like “guess”) the driver you need while clicking through various warnings. You can already see at least one user with driver issues in the thread, with Windows telling them, “Your device has malfunctioned,” when really it just needs a driver.

Google’s Pixel storage issue fix requires developer tools and a terminal Read More »