With lockdowns being implemented on and off around the world since 2020, it has meant that factories have been unable to churn out as many semiconductors as are needed. Also adding to the constraint issues is the fact that demand for this hardware is increasing as businesses develop new technologies for homes and cars that previously didn’t exist.

According to Gartner, the semiconductor shortages in the PC and smartphone sectors have already begun to ease, however, in the automotive sector, issues are expected to linger on until the end of the year. Commenting on the problem with regards to the automotive sector, Alan Priestly, research vice president at Gartner, said:

“Although unit production of automotive vehicles will grow below expectation at 12.5% in 2022, semiconductor device ASPs are expected to remain high due to continued tight supply driving the automotive semiconductor market to double-digit growth (19%) in 2022. Automotive HPC, EV/HEV and advanced driver assistance systems will lead the growth in automotive electronics sectors through the forecast period.”

The new worldwide revenue estimate of $676 billion is a $37 billion increase on Gartner’s previous estimates. While it’s certainly good news that things seem to be getting back to normal in some sectors, China’s decision to enforce a new lockdown in Shanghai and increase anti-COVID measures in Beijing could lead to more disruption.

Semiconductor shortage to ease this year in some sectors and revenue expected to grow 13.6%

In the last 20 years, TSMC's success was largely conditioned by its ability to offer a new manufacturing technology with power, performance and area improvements every year, while introducing a brand-new node every 18 to 24 months and maintaining high yields. However, as the complexity of modern fabrication processes are getting highly advanced and complex, it is getting much harder to keep the pace of innovation like before. With the N2 chip, there will be a strategic shift in TSMC's strategy of node development.

Apple has been TSMC's largest customer by revenue for the past decade. The company makes nearly all of Apple's chips including the ones used in iPhones, iPads and also the new M1 powered Macs. Apple's latest gadgets use a 5nm chip. The next generation of Apple devices are expected to use TSMC's 3nm chips, including the iPhone 14, to be announced later this year. However, we can only expect 2nm chips inside Apple devices after 2026.

Intel, on the other hand, will adopt the 2nm technology in its GPUs and other SoCs. Analysts from China Renaissance Securities speculate that Intel might use the N2 for the graphics tile of its next-gen client processor, code named as Lunar Lake. For the CPU tile in Lunar Lake, Intel will use its own 18A node.

Source: Computing.co.uk

Apple and Intel would dominate the 2nm chip race in 2026

This year, Microsoft will host its flagship annual developer conference on May 24-26. Like the previous two years, Build 2022 will take place online, which means everyone can join for free. If you want to "visit" the online conference and learn more about the latest dev tools and innovations from the software giant, you can register on the official Build website. Registration is now open for everyone.

Microsoft says Build 2022 will help developers "innovate without compromise, develop creatively, and build the next generation of apps" for more than one billion devices. The conference will feature 13 speakers from Microsoft, such as Satya Nadella, Kevin Scott, Amanda Silver, Scott Guthrie, Kathleen Mitford, and others. Attendees will get a chance to learn during live sessions and book time with Microsoft experts.

What differentiates Build 2022 from previous developer conferences held by Microsoft is new regional spotlights. The company wants to provide keynote analysis, trending news, and topics for specific regions, such as France, Germany, Japan, Latin America, and the UK.

Microsoft's Build 2022 is not the only major developer conference happening soon. Apple will host its WWDC 2022 on June 6-10. Similar to Microsoft, this year's Worldwide Developer Conference will be an online-only event.

Microsoft opens free registration for Build 2022

In the moments before Elon Musk bought Twitter, the company’s Slack channels were lit up with nervous anticipation. It had been days since Twitter leadership had shared anything with them, and after a weekend’s worth of reports that a sale was imminent, employees were looking for answers.

For the first few hours of the morning, none came. Work all but came to a halt, employees told me. Like a classroom where the teacher is late and students are attempting to self-govern, one said. A “hellhole,” said another. One thread, in which an employee asked good-naturedly whether anyone was excited about the prospect of working for Musk, drew dozens of responses, many of them quite ugly.

Then, just before the markets closed, the news arrived: the board had accepted Musk’s offer to take the company private for $44 billion, or $54.20 a share. What began three weeks ago as a “passive investment” will end with Musk in charge.

After the announcement, sentiment in the public Slack channels remained largely concerned and negative, employees told me. “I was kind of surprised how much people seemed like they were giving up,” one told me. “Big bummer.”

But in one-on-one discussions, responses were more tempered. Some employees I’ve spoken with are open to the idea that a private Twitter run by Musk stands a better chance of improving the service than would a public company beholden to its shareholders. They like the fact that he wants to eliminate harmful bots and bring more clarity to how recommendation algorithms work.

At the same time, many Twitter employees receive half or more of their compensation in stock. At an all-hands meeting on Monday afternoon, they were told that employees will not receive equity once the company goes private. As a result, one person told me, “group chats are scrambling to see if working at Twitter makes economic sense first and foremost.”

The deal is expected to take around six months to close. At that time, it seems likely that CEO Parag Agrawal will leave the company, although he did not say so when addressing employees today. Agrawal did say there would not be layoffs in the near term, though he would not comment on whether there would be a hiring freeze. (There is, however, a partial freeze on new features, although I’m told that’s fairly standard during moments when there is a lot of attention on Twitter and the company doesn’t want to introduce any new bugs.)

In the meantime… hiring will be harder. There seems to be at least some likelihood that the company will see significant attrition, particularly in the leadership ranks — and on the “health” teams that work to fight harassment and abuse. On the plus side, the company’s earnings report Thursday no longer has the potential to further tank its stock.

Beyond that, as with everything related to the Twitter-Musk story over the past month, it’s extremely difficult to predict.

How did we get here?

On one hand, Musk’s acquisition really is an extraordinary story. Acquiring 9 percent of the company as a “passive” investor, failing to report it on time, getting invited to join the board, turning it down, then offering to buy the company outright — that’s an extremely strange turn of events.

On the other, as Matt Levine noted today, once Musk made his offer, the whole thing proceeded fairly normally. As a negotiating tactic, the board put in place a “poison pill” to prevent Musk from acquiring any more of the company. That forced Musk to prove that he actually had the financing to get the deal done. He did so, and the board considered its options. Ultimately, the 38 percent premium that Musk offered over its current stock price struck members as the best deal they could likely get.

In the end, it was just business.

Twitter has long been an underperformer, and former executives I spoke with today were relieved that the company now has a real chance to make radical change. As a private company, beholden only to the interests of one man, Twitter may be able to transform itself in ways that it never could while it had to report quarterly earnings.

The $44 billion question, though, is… transform it into what?

“What is Twitter for? Everybody who’s ever been in control of Twitter has had a different answer to that question,” one former executive told me today. “It’s a question that has vexed users since the creation of the service.”

There was the era after Jack Dorsey returned as CEO and tried to focus the product’s attention on “what’s happening.” There was the effort led by former chief operating officer Anthony Noto to make Twitter a destination for live video. More recently, Dorsey oversaw efforts to both make the company a platform for “healthy conversation” and to turn it into a decentralized protocol.

What does Musk think Twitter is for? All we have to go on is a series of cryptic tweets and statements that are short enough to allow for wide interpretation. In this moment, he is still a mostly blank canvas on which people can project their hopes (like the Republican members of Congress tweeting “free speech is making a comeback”) and their fears (like those Twitter employees in Slack).

On stage at TED, he said he didn’t care about “the economics” of Twitter “at all.” But given the way he structured his Twitter deal, he will face pressure (if not insurmountable pressure) to see a return on his investment. Here are Krystal Hu and Anirban Sen at Reuters:

More than two-thirds of the $46.5 billion financing package that Musk unveiled on Thursday in support of his bid for Twitter would come from his assets, with the remainder coming from bank loans secured against the social media platform’s assets.

That is the reverse of how most investors structure buyouts, with debt secured against the assets of the target company typically comprising the majority of the financing. […]

What is more, Musk has agreed to take out a risky $12.5 billion margin loan, secured against his stock of Tesla Inc, the electric-car maker that he leads, to pay for some of the $33.5 billion equity check. Were Tesla’s stock to drop by 40%, he would have to repay that loan, a regulatory filing shows.

So, to avoid having to sell a bunch of his Tesla stock and pay off his loan, maybe Musk does find that he has to care about the economics of Twitter.

In which case, what does he do?

It’s fashionable lately to suggest Musk ought to get Twitter out of the ads business. (Ads are out of fashion in general.) But as a feed-based social network, ads arguably still make a great deal of sense for Twitter, former employees told me today. When former CEO Dick Costolo embraced ads as Twitter’s future, one said, it was because “it was the best way to make the most money. it wasn’t because of an inherent love of advertising.”

The company could place a new emphasis on subscriptions, either by expanding its $2.99 Twitter Blue product or by scrapping it and building a new subscription product from scratch. (“Make power users pay for access to their large audiences” iis one idea that keeps coming up which seems plausible to me.)

Or Twitter could lean harder into becoming a decentralized protocol, selling API access to enable developers to build a variety of different front-end experiences. That could enable different users to choose different styles of content moderation, while effectively turning the core service into an enterprise software product.

I think any of these outcomes is plausible. But if Musk has a hand, he’s not tipping it. His official statement today bears little evidence that he has thought much about what he will do with Twitter as a business much beyond the close of the deal:

“Free speech is the bedrock of a functioning democracy, and Twitter is the digital town square where matters vital to the future of humanity are debated,” said Mr. Musk. “I also want to make Twitter better than ever by enhancing the product with new features, making the algorithms open source to increase trust, defeating the spam bots, and authenticating all humans. Twitter has tremendous potential – I look forward to working with the company and the community of users to unlock it.”

Perhaps he will develop a stronger point of view, and share it, in the coming months.

In the meantime, the other big question looming over all this is who will run Twitter day to day. Musk already leads Tesla and SpaceX and the Boring Company and Neuralink; presumably he will not be spending eight hours a day on Twitter. Agrawal said today he intends to stay with the company through the close of the deal; he would seem to be in line for a $39 million payout if and when Musk replaces him. Who will Musk trust with his vision for the company? It’s one of the year’s most interesting tech stories.

OK, there’s one other big looming question over all this, which is will Musk let Donald Trump back on the platform? Trump said today he absolutely will not return to Twitter even if he was invited. The former president said he will instead use his own platform, Truth Social, which he has completely ignored after posting there once. But some people pointed out that Trump has lied in the past, and might even in fact be lying about this. Time will tell!

Over on my Twitter timeline today, people were going wild with the most definitive, histrionic takes on what Musk’s acquisition will mean for the user experience. Twitter is about to be overtaken by trolls and harassers; Twitter is about to re-learn 10 years’ worth of lessons about content moderation the hard way; Twitter is about to see an employee exodus like none we have ever seen. And so on.

Maybe all of that is true, or maybe none of it is. At least some of the worries would seem to be justified, based on Musk’s past statements. But the cart feels way ahead of the horse here, and in any case predicting Elon Musk’s behavior is a mug’s game. For all its cultural importance, Twitter had a notably undistinguished life as a public company. For better and for worse, it’s now up to Musk to see whether it can fare any better as a private one.

What Twitter employees are saying about Elon Musk

Two years ago, the next generation of computer memory, DDR5 RAM, was officially released to the public. It came with a lot of improvements over the previous generation DDR4. From speed improvements, to a sort of error correction (something only found in server memory) to bigger RAM storage size and even lower voltage and hence, lower power usage. It also comes with a special circuitry, making its manufacturing quite complex. RAM maker Crucial has a nice page dedicated to DDR5 explaining a lot of it.

It wasn’t that it was released quickly without any preparations. DDR5 released to the public 6 years after the launch of DDR4. Meanwhile, the companies were already working on the DDR5 RAM based memory chips before their release to the public.

However, after the release, when DDR5 RAM was tested in the gaming benchmarks, it was found that many games had no benefit whatsoever in their speeds and overall benefit was just 3%. The bigger problem with DDR5, however, are its prices.

DDR5’s high prices

The problem with DDR5 is that it is massively expensive. A quick search shows that currently, DDR5 costs almost double than DDR4. If 32 GB of DDR4 RAM costs anywhere from $120 to $140. DDR5 seems to start at $220 for a simple 4800 MHz stick of the same size, reaching to $320 for 5200 MHz and $370 for an even faster 6000 MHz.

To add the issues, it is not expected for the DDR5 prices to get cheaper at least this year. The pandemic induced lockdowns has ensured that many manufacturing companies have had to temporarily shut it’s plants, effecting the rollout of more RAM chips.

It does not stop at that. As DDR5 comes with new features, manufacturing it is a complex and expensive process. This raises the cost of DDR5 RAM further.

AMD Ryzen 7000 to support only DDR5

Going by recent reports by Tom’s Hardware, AMD’s next generation of processors, the Ryzen 7000 will only support DDR5.

They report that both, the mainstream B650 and the enthusiast X670 CPU chipsets by AMD for the Ryzen 7000 platform are not going to support DDR4 and will only support DDR5. Even the ultra-cheap A-series are highly unlikely to support DDR4. Making even cheap PC builders running after the expensive DDR5 RAM.

What’s interesting is that the leaked AMD Ryzen 7000 documents find no mention of DDR4 whatsoever. Basically confirming that none of the Ryzen 7000 CPUs are going to support it. Irrespective of the price range.

Intel’s support for both DDR4 and DDR5

This is unlike Intel which went with the support for both DDR4 and DDR5 RAM together.

When Intel released its 12th generation Core processors, code-named Alder Lake. It was mindful of the costs of DDR4 RAM. That is why Intel released the Alder Lake processors with support for both DDR4 and DDR5 RAM.

This allowed the motherboard manufacturers to support both of these RAM together in a single motherboard. Which meant users could use the cheaper DDR4 RAM right now and later switch to DDR5 when it gets cheaper without major hardware changes.

What’s also important is that going by the reports, Intel’s next generation of processors, code-named Raptor Lake, which are expected to release later this year alongside Ryzen 7000, are said to continue to support both DDR4 and DDR5, unlike the AMD Ryzen 7000. This makes Intel processors an attractive purchase.

AMD’s move surprising but long time coming

What AMD has done is indeed surprising. Whenever a new generation of RAM comes, CPU makers are expected to provide the support for both previous generation and newer generation of RAM together in at least one CPU series. This allows for a smoother and better transition for everyone, from manufactures to the users.

However, this move was coming from a long time. Going back 3 months ago, in an interview to TechPowerUp, AMD’s executive had revealed that AMD was aiming for the better availability of DDR5 RAM chips, including cheaper price, for release the of Ryzen 7000 series of the processors. Maybe AMD had decided long ago that they will not support DDR4.

Nothing new about AMD, though. It is typical of them. AMD for now might be looking at a long term prospect and support for its next generation AM5 CPU socket. Just like the AM4 socket, which ran for a few years, AMD probably wants AM5 to run for a few years too. So AMD might have kept future generation of processors and their compatibility with the socket in mind before making this decision, at the risk of short term costs associated with it.

Meanwhile, Intel processors are looking attractive now.

AMD Zen 4 Ryzen 7000 CPUs May Only Support DDR5 RAM

However, the Santa Clara company is trying to do its best in this regard so that AM5 users can get the best out of their expensive DDR5 kits. It had revealed that it was taking DDR5 overclocking quite seriously and a new report from VideoCardz says that a new technology called EXPO is how AMD plans to make this work.

According to the report, EXPO stands for "EXtended Profiles for Overclocking" and here's how the two profiles in it will work as per the report:

EXPO will be responsible for storing two memory overclocking profiles for DDR5 series. The first profile will be optimized for high bandwidth use, whereas the latter will be for low-latency. Our source claims that the second profile is optional.

Overall, it sounds quite similar to how XMP and A-XMP (AMP) function nowadays. The report also adds that EXPO will work with all kinds of DDR5 DIMM types including SODIMMs, that are generally used in laptops and small form factor (SFF) desktops:

Furthermore, EXPO will work with all types of DDR5 memory, including UDIMM, RDIMM and SO-DIMM.

If true, then we might finally get more overclockable memory offerings in small form factor (SFF) use cases, something that is quite rare with DDR4. AMD had recently trademarked the name EXPO and the details on the page do clearly indicate that the technology is indeed related to memory.

Interestingly, AMD already seemed to have a feature called RAMP or Ryzen Accelerated Memory Profile that was set to debut with Zen 4 and AM5. It will be interesting to see if the two technologies are same or the new EXPO is a bit different than how RAMP is supposed to function.

Source: VideoCardz via JUSTIA

AMD's new EXPO feature to reportedly allow RAM overclocking on next-gen Zen 4 laptops too

Over the next five years, IDC expects the market to continue growing steadily, though, factors such as privacy, security, costs, and interoperability could hamper the market. These issues will require an education drive on the part of businesses to build trust, governments will have to pass legislation to ensure companies release secure products, costs will need to be kept low, and manufacturers will need to make sure their products can talk to other smart home equipment.

Providing comment for IDC, Adam Wright, senior research analyst, Internet of Things: Consumer, said:

“The notion of the 'smart home' is well and truly mainstream in many parts of the world now. Smart home devices have proved their merit during the global pandemic as adding meaningful value to people's lives at home, such as easing social isolation, enhancing home security, and delivering convenience and entertainment, among other benefits. As we look ahead, and buoyed by strong demand, the market is poised for solid growth despite ongoing disruptions from supply chains, logistics, and uneven economies across the globe.”

IDC has split the smart home category into video entertainment, home monitoring/security, lighting, and others. In 2021, video entertainment smart home devices made up 35% of the smart home market, home monitoring/security made up 21%, lighting made up 10%, and others accounted for the final 35%. According to IDC’s 2026 forecast, video entertainment is set to fall to 26% of the market while lighting could jump to 17% with others climbing to 36%. Overall, smart home shipments could rise from 895 million in 2021 to 1.44 billion in 2026 as the market matures.

The worldwide smart home devices market grew by 11.7% in 2021 as people get better broadband

Sony’s storefront will have the PlayStation 5 console available for purchase starting at 4:30PM ET today, so time is running out before the waiting room flips into a queue for the hotly in-demand console. We expect that the store will stock both the $499.99 version of the PlayStation 5 that features a disc drive, as well as the $399.99 digital version that lacks one. We also expect that this restock will sell out just as quickly as the others.

PS5 restocks, while popping up here and there, have been a rarity compared to purchase opportunities for other tech products that were hard to find as little as a few months ago. Things like Nvidia and AMD GPUs have slowly become easier to find in-stock, as have Microsoft’s Xbox Series X and the Nintendo Switch OLED consoles. In other words, leap at this opportunity to get a PS5.

Some games and accessories to consider while you wait

Since Sony sells its consoles without the need for bundles, you can pick what accessories and exclusive games are the right fit for you. Here are some you can’t go wrong with.

The PlayStation 5 will be available directly through Sony at 4:30PM ET

It's been nearly six months since Facebook announced it was rebranding to Meta and would focus its future on the upcoming “metaverse.” In the time since, what that term means hasn't gotten any clearer. Meta is building a VR social platform, Roblox is facilitating user-generated video games, and some companies are offering up little more than broken game worlds that happen to have NFTs attached. 

Advocates from niche startups to tech giants have argued that this lack of coherence is because the metaverse is still being built, and it's too new to define what it means. The internet existed in the 1970s, for example, but not every idea of what that would eventually look like was true.

On the other hand, there's a lot of marketing hype (and money) wrapped up in selling the idea of “the metaverse.” Facebook, in particular, is in an especially vulnerable place after Apple's move to limit ad tracking hit the company's bottom line. It's impossible to separate Facebook's vision of a future in which everyone has a digital wardrobe to swipe through from the fact that Facebook really wants to make money selling virtual clothes. But Facebook isn't the only company that stands to financially benefit from metaverse hype.

So, with all that in mind …

To help you get a sense of how vague and complex a term “the metaverse” can be, here's an exercise: Mentally replace the phrase “the metaverse” in a sentence with “cyberspace.” Ninety percent of the time, the meaning won't substantially change. That's because the term doesn't really refer to any one specific type of technology, but rather a broad (and often speculative) shift in how we interact with technology. And it's entirely possible that the term itself will eventually become just as antiquated, even as the specific technology it once described becomes commonplace.

Broadly speaking, the technologies companies refer to when they talk about “the metaverse” can include virtual reality—characterized by persistent virtual worlds that continue to exist even when you're not playing—as well as augmented reality that combines aspects of the digital and physical worlds. However, it doesn't require that those spaces be exclusively accessed via VR or AR. Virtual worlds—such as aspects of Fortnite that can be accessed through PCs, game consoles, and even phones—have started referring to themselves as “the metaverse.”

Many companies that have hopped on board the metaverse bandwagon also envision some sort of new digital economy, where users can create, buy, and sell goods. In the more idealistic visions of the metaverse, it's interoperable, allowing you to take virtual items like clothes or cars from one platform to another, though this is harder than it sounds. While some advocates claim new technologies like NFTs can enable portable digital assets, this simply isn't true, and bringing items from one video game or virtual world to another is an enormously complex task that no one company can solve.

It's difficult to parse what all this means because when you hear descriptions like those above, an understandable response is, “Wait, doesn't that exist already?” World of Warcraft, for example, is a persistent virtual world where players can buy and sell goods. Fortnite has virtual experiences like concerts and an exhibit where Rick Sanchez can learn about MLK Jr. You can strap on an Oculus headset and be in your own personal virtual home. Is that really what “the metaverse” means? Just some new kinds of video games?

Well, yes and no. Saying that Fortnite is “the metaverse” would be a bit like saying Google is “the internet.” Even if you spend large chunks of time in Fortnite, socializing, buying things, learning, and playing games, that doesn't necessarily mean it encompasses the entire scope of what people and companies mean when they say "the metaverse." Just as Google, which builds parts of the internet—from physical data centers to security layers—isn't the entire internet. 

Tech giants like Microsoft and Meta are working on building tech related to interacting with virtual worlds, but they're not the only ones. Many other large companies, including Nvidia, Unity, Roblox, and even Snap—as well as a variety of smaller companies and startups—are building the infrastructure to create better virtual worlds that more closely mimic our physical life.

For example, Epic has acquired a number of companies that help create or distribute digital assets, in part to bolster its powerful Unreal Engine 5 platform. And while Unreal may be a video game platform, it's also being used in the film industry and could make it easier for anyone to create virtual experiences. There are tangible and exciting developments in the realm of building digital worlds.

Despite this, the idea of a Ready Player One-like single unified place called “the metaverse" is still largely impossible. That is in part because such a world requires companies to cooperate in a way that simply isn't profitable or desirable—Fortnite doesn't have much motivation to give players a portal to jump straight over to World of Warcraft, even if it were easy to do so, for example—and partially because the raw computing power needed for such a concept could be much further away than we think. 

This inconvenient fact has given rise to slightly different terminology. Now many companies or advocates instead refer to any single game or platform as “a metaverse.” By this definition, anything from a VR concert app to a video game would count as a “metaverse.” Some take it further, calling  the collection of various metaverses a “multiverse of metaverses.” Or maybe we're living in a “hybrid-verse.”

Or these words can mean anything at all. Coca-Cola launched a “flavor born in the metaverse” alongside a Fortnite tie-in mini-game. There are no rules.

It's at this point that most discussions of what the metaverse entails start to stall. We have a vague sense of what things currently exist that we could kind of call the metaverse if we massage the definition of words the right way. And we know which companies are investing in the idea, but after months, there's nothing approaching agreement on what it is. Meta thinks it will include fake houses you can invite all your friends to hang out in. Microsoft seems to think it could involve virtual meeting rooms to train new hires or chat with your remote coworkers.

The pitches for these visions of the future range from optimistic to outright fan fiction. At one point during Meta's original presentation on the metaverse, the company showed a scenario in which a young woman is sitting on her couch scrolling through Instagram when she sees a video a friend posted of a concert that's happening halfway across the world. 

The video then cuts to the concert, where the woman appears in an Avengers-style hologram. She's able to make eye contact with her friend who is physically there, they're both able to hear the concert, and they can see floating text hovering above the stage. This seems cool, but it's not really advertising a real product, or even a possible future one. In fact, it brings us to the biggest problem with “the metaverse.”

When the internet first arrived, it started with a series of technological innovations, like the ability to let computers talk to each other over great distances or the ability to hyperlink from one web page to another. These technical features were the building blocks that were then used to make the abstract structures we know the internet for: websites, apps, social networks, and everything else that relies on those core elements. And that's to say nothing of the convergence of the interface innovations that aren't strictly part of the internet but are still necessary to make it work, such as displays, keyboards, mice, and touchscreens.

With the metaverse, there are some new building blocks in place, like the ability to host hundreds of people in a single instance of a server (idealistic metaverse predictions suppose this will grow to thousands or even millions of people at once, but this might be overly optimistic), or motion-tracking tools that can distinguish where a person is looking or where their hands are. These new technologies can be very exciting and feel futuristic.

However, there are limitations that may be impossible to overcome. When tech companies like Microsoft or Meta show fictionalized videos of their visions of the future, they frequently tend to gloss over just how people will interact with the metaverse. VR headsets are still very clunky, and most people experience motion sickness or physical pain if they wear them for too long. Augmented reality glasses face a similar problem, on top of the not-insignificant issue of figuring out how people can wear them around in public without looking like huge dorks. And then there are the accessibility challenges of VR that many companies are shrugging off for now.

So, how do tech companies show off the idea of their technology without showing the reality of bulky headsets and dorky glasses? So far, their primary solution seems to be to simply fabricate technology from whole cloth. The holographic woman from Meta's presentation? I hate to shatter the illusion, but it's simply not possible with even very advanced versions of existing technology. 

Unlike motion-tracked digital avatars, which are kind of janky right now but could be better someday, there's no janky version of making a three-dimensional picture appear in midair without tightly controlled circumstances. No matter what Iron Man tells you. Perhaps these are meant to be interpreted as images projected via glasses—both women in the demo video are wearing similar glasses, after all—but even that assumes a lot about the physical capabilities of compact glasses, which Snap can tell you isn't a simple problem to solve.

This kind of glossing over reality occurs frequently in video demos of how the metaverse could work. Another of Meta's demos showed characters floating in space—is this person strapped to an immersive aerial rig or are they just sitting at a desk? A person represented by a hologram—do they have a headset on, and if so how is their face being scanned? And at points, a person grabs virtual items but then holds those objects in what seems to be their physical hands.

This demo raises so many more questions than it answers.

To a limited extent, this is fine. Microsoft, Meta, and every other company that shows wild demos like this are trying to give an artistic impression of what the future could be, not necessarily account for every technical question. It's a time-honored tradition going back to AT&T's demo of a voice-controlled foldable phone that could magically erase people from images and generate 3D models, all of which might've seemed similarly impossible at the time.

However, the last several months of metaverse pitches—from tech giants and startups alike—have relied heavily on lofty visions that break from reality. Chipotle's “metaverse” was an ad disguised as a Roblox video game. Stories about scarce “real estate” in “the metaverse” refer to little more than a buggy video game with virtual land tokens (which also glosses over the very real security and privacy issues with most popular NFTs right now).

The confusion and disappointment surrounding most “metaverse” projects are so pervasive that when a video from 2017 of a Walmart VR shopping demo started trending again in January 2022, people immediately thought it was yet another metaverse demo. It also helped demonstrate how much of the current metaverse discussion is built on hype alone. Walmart's VR shopping demo obviously never went anywhere (and for good reason). So why should anyone believe that it's the future when Chipotle does it?

This kind of wishful-thinking-as-tech-demo leaves us in a place where it's hard to pinpoint which aspects of the various visions of the metaverse (if any) will actually be real one day. If VR and AR headsets become comfortable and cheap enough for people to wear on a daily basis—a substantial “if”—then perhaps a virtual poker game with your friends as robots and holograms and floating in space could be somewhat close to reality. If not, well you could always play Tabletop Simulator on a Discord video call.

The flashiness of VR and AR also obscure the more mundane ways that our existing, interconnected digital world could be improved right now. It would be trivial for tech companies to invent, say, an open digital avatar standard, a type of file that includes characteristics you might enter into a character creator—like eye color, hairstyle, or clothing options—and let you take that data everywhere, to be interpreted by a game engine however it chooses. There's no need to build a more comfortable VR headset for that.

But that's not as fun to imagine.

The paradox of defining the metaverse is that in order for it to be the future, you have to define away the present. We already have MMOs that are essentially entire virtual worlds, digital concerts, video calls with people from all over the world, online avatars, and commerce platforms. So in order to sell these things as a new vision of the world, there has to be some element of it that's new.

Spend enough time having discussions about the metaverse and someone will inevitably (and exhaustingly) reference fictional stories like Snow Crash—the 1992 novel that coined the term “metaverse”—or Ready Player One, which depicts a VR world where everyone works, plays, and shops. Combined with the general pop culture idea of holograms and heads-up displays (basically anything Iron Man has used in his last 10 movies) these stories serve as an imaginative reference point for what the metaverse—a metaverse that tech companies might actually sell as something new—could look like.

That kind of hype is arguably more vital to the idea of the metaverse than any specific technology. It's no wonder, then, that people promoting things like NFTs—cryptographic tokens that can serve as certificates of ownership of a digital item, sort of—are also latching onto the idea of the metaverse. Sure, NFTs are bad for the environment and the public blockchains most are built on come with massive privacy and security problems, but if a tech company can argue that they'll be the digital key to your virtual mansion in Roblox, then boom. You've just transformed your hobby of buying memes into a crucial piece of infrastructure for the future of the internet (and possibly raised the value of all that cryptocurrency you're holding.)

It's important to keep all this context in mind because while it's tempting to compare the proto-metaverse ideas we have today to the early internet and assume everything will get better and progress in a linear fashion, that's not a given. There's no guarantee people will even want to hang out sans legs in a virtual office or play poker with Dreamworks Mark Zuckerberg, much less that VR and AR tech will ever become seamless enough to be as common as smartphones and computers are today.

In the months since Facebook's rebrand, the concept of “the metaverse” has served as a powerful vehicle for repackaging old tech, overselling the benefits of new tech, and capturing the imagination of speculative investors. But money pouring into a space doesn't necessarily mean a massive paradigm shift is right around the corner, as everything from 3D TVs to Amazon's delivery drones and Google Glass can attest. The history of tech is littered with the skeletons of failed investments.

That doesn't mean there's nothing cool on the horizon. VR headsets like the Quest 2 are cheaper than ever and finally weaning off of expensive desktop or console rigs. Video games and other virtual worlds are getting easier to build and design. And personally, I think the advances in photogrammetry—the process of creating digital 3D objects out of photos or video—is an incredibly cool tool for digital artists.

But to a certain extent, the tech industry writ large depends on futurism. Selling a phone is fine, but selling the future is more profitable. In reality, it may be the case that any real “metaverse” would be little more than some cool VR games and digital avatars in Zoom calls, but mostly just something we still think of as the internet.

4/25/2022: This story has been updated with additional reporting.

What Is the Metaverse, Exactly?

(May require free registration to view)

Elon Musk became the new owner of Twitter on Monday, after completing a stunning $44 billion takeover of the social media platform, ending a process that has vacillated between a done deal and dead in the water in the last three weeks.

“Free speech is the bedrock of a functioning democracy, and Twitter is the digital town square where matters vital to the future of humanity are debated,” Musk said in a press release announcing the news. Twitter independent board chair Bret Taylor described the deal as “the best path forward” for the company’s shareholders.

The result ends prolonged speculation over Musk’s financial interest in Twitter. On April 4, the entrepreneur’s 9.2 percent stake in the company—or 73.5 million shares at a cost of around $2.4 billion—was disclosed to the public. At the time, the purchase of stock in Twitter came with an offer to sit on the board—though on April 10 Musk declined to take his seat.

He soon made it obvious that he wanted the whole thing. On April 14, Musk offered to buy the remaining percentage of the company for $54.20 per share—a 38 percent premium on the price he paid for his initial investment. Musk’s accompanying letter to the chair of Twitter’s board was strident in its criticism of the platform. “I believe in [Twitter’s] potential to be the platform for free speech around the globe, and I believe free speech is a societal imperative for a functioning democracy,” he wrote. However, he added, “I now realize the company will neither thrive nor serve this societal imperative in its current form.”

Instead, he wanted to take the company private, offering $44 billion for it in a “best and final” offer. At the time, analysts were split about the likelihood of Musk’s bid succeeding, and whether it was good value; while it sat in the middle of the usual 30 to 40 percent premium above the trading price, the stock price had reached well above that just last year. Twitter’s board, for its part, said it would evaluate the offer.

“He’s setting a bit of a precedent for activists that will go after a company,” says Timothy Galpin, senior lecturer in strategy and innovation at the Said Business School at the University of Oxford. “It’s been done a bit before by Carl Icahn and a few others, but it’s not as prevalent to go after the whole company.”

On the same day that he lodged his bid to take over the entirety of Twitter and take it private, Musk appeared at a TED talk in Vancouver, where he laid out his vision. “This is not a way to sort of make money,” he claimed. “My strong intuitive sense is that having a public platform that is maximally trusted and broadly inclusive is extremely important.” That gave some within Twitter, and those who held large shares in the platform, pause.

Contemporary reports indicated Twitter would fight to repel Musk, while the Tesla and Space X CEO got into a Twitter spat about press freedom with Saudi Arabia’s Kingdom Holding Company, a major shareholder that said it would reject Musk’s offer. (Saudi Arabia has been accused of the murder of journalist Jamal Khashoggi.)

Such social media battles may be unusual when considering a takeover of a massive business, but Musk is himself unusual, says Cary Cooper, a business professor at Manchester Business School. “He’s not a traditional businessman,” he says. “He’s a man that is pretty creative and pretty innovative. He’s a unique guy and does things in a way that a normal businessperson wouldn’t do. He doesn’t play the normal games that an entrepreneur would play.”

On April 15, Twitter’s board triggered a break-glass-in-emergency financial tool: the poison pill. Also known as a limited duration shareholder rights plan, the poison pill invited shareholders to increase their investments in Twitter in order to reduce Musk’s ability to build his stake up into a controlling one. Any attempts to take his share over 15 percent would require Musk to negotiate with Twitter’s board.

Triggering the poison pill headed off the speedy hostile takeover, but Musk’s offer never left the table. On April 21, Musk outlined how he’d come up with the $44 billion in cash required to fulfill his bid. Morgan Stanley and other firms offered to back Musk’s bid, while he’d pay around $21 billion from his own estimated $263 billion fortune. The filing put meat on the bones of what had previously been a speculative offer—and indicated how seriously Musk wanted to take Twitter private.

The confirmed funding reportedly caused some of Twitter’s shareholders who were more agnostic about Musk to petition the company to hear him out. Meetings reportedly took place over the weekend, and Twitter’s board met on April 25 to recommend the deal to shareholders. It was a swift and surprising reversal. “On Friday, there was so much skepticism and cynicism, and now it almost looks like a done deal,” says Vasant Dhar, a professor of information systems at NYU Stern. Musk’s quick movements have left other potential bidders stuck playing catchup. But the deal appears to have passed the money test, at least for Twitter’s board of directors, since “the board’s fiduciary responsibility is to get the most value for shareholders,” says Galpin. “Obviously, there are questions about what he’ll do with the company if he takes control of it. He’s got to do more than just add an edit button.”

Taking the company private would allow Musk to make the changes he wants far more quickly, without answering to public markets. “I also want to make Twitter better than ever by enhancing the product with new features, making the algorithms open source to increase trust, defeating the spam bots, and authenticating all humans,” Musk wrote in Monday’s press release.

“I think he’s played it brilliantly,” says Dhar. “One could have expected the reaction we got: ‘Musk is a megalomaniac and he’s doing it for self-promotion.’ But I actually think there’s a lot more to it than that.”

It’s possible that the purchase will come under regulatory scrutiny. While there’s unlikely to be an antitrust concern, the Securities and Exchange Commission could still take issue with Musk’s disclosures along the way. “You could ask a court to enjoin the deal on the basis that he has improperly filed,” says Pritchard. “He didn’t file his initial stake on a timely basis, then he filed the wrong form because he really had the intention of influencing management the whole time,” he suggests. That, however, would require showing the harm caused by those infractions. Shareholders could lodge private lawsuits but would likely only succeed in getting more money from Musk in the deal. And the SEC is unlikely to halt the transaction because of the damage that could do to shareholders.

It seems Elon Musk will almost inevitably assume control and ownership of Twitter—changing the face of the platform in the process. For some of Twitter’s millions of users, it’s a welcome development that gives them more freedom to say and do what they want. For others, it’s a worrying development with potentially chilling consequences. As for the shareholders, and Musk himself, things are looking rosy.

“Shareholders will feel like they’ve won, and Musk has got what he wanted,” says Galpin. “He’s got control of the company, for not an exorbitant price but not a cheap price either. Nobody really gouged the other one, and nobody lost.”

How Elon Musk Won Twitter

(May require free registration to view)

AMD introduced its chiplet approach for the first time back when it unveiled Zen 2 in 2018. The chiplet design is a modular way to scale up multi-processing performance that's more efficient than simply adding more cores on a monolithic die.

Fast forward nearly four years today, a new report from Tom's Hardware suggests that AMD is so impressed by the effectiveness of the approach that it is using "dual-chiplets" to design its upcoming 600-series chipsets on next-gen Socket AM5. The report claims that this technology is only coming to the flagship X670 parts only and not to the mainstream B650 and entry-level A620 chipsets.

Today's report builds on an earlier report by Chinatimes which initially broke the story of the chiplet-based 600-series chipsets. However, there are no details so far on how these chiplets will work.

There's also more news on AM5 as the Tom's Hardware story also adds that the AMD LGA1718 AM5 socket will come with DDR5 support only. This means the next-gen AMD platform based around Zen 4 or Ryzen 7000 "Raphael" CPUs could be really expensive as DDR5 memory is still pretty exorbitant compared to the far more abundant DDR4. An earlier report had also suggested a similar thing that AMD could delay the introduction of next-gen RDNA 2 desktop APUs due to the high prices of DDR5 DRAM kits.

However, not all news is bad news when it comes to AM5's DDR5 side as Team Red has already confirmed that it is working hard to make sure high speeds are achievable for DDR5 memory. During a recent Meet the Experts webinar with Samsung, AMD's Memory Enabling Manager, Joseph Tao said the following:

Our first DDR5 platform for gaming is our Raphael platform and one of the awesome things about Raphael is that we are really gonna try to make a big splash with overclocking and I'll just kinda leave it there but speeds that you maybe thought couldn't be possible, maybe possible with this overclocking spec.

It is likely that AMD was referring here to its upcoming AMD Ryzen Accelerated Memory Profile (RAMP) technology which will succeed current A-XMP.

Source: Tom's Hardware via China Times

AMD reportedly bringing "dual-chiplets" to its X670 chipsets on Socket AM5 for Zen 4

And while Nvidia is making great strides to add the new power standard for its next-gen cards for additional power headroom, the company is purportedly only adding PCIe 4.0 bus capabilities on Ada Lovelace RTX 40-series GPUs, according to reliable leakster kopite7kimi. If true, the upcoming Nvidia RTX cards will be limited to just 2GB/s per lane uni-directionally.

PCIe 5.0 is something which Nvidia already announced with its Hopper architecture that's meant for data center and high-performance computing (HPC) markets. Hence, it is slightly surprising to see that Nvidia may not offer the same feature parity for its consumer gaming lineup.

In terms of support, Intel has already introduced PCIe 5.0 on its 12th Alder Lake platform while AMD will also be bringing it with Zen 4 and Socket AM5 soon by the end of the year. However, it's not confirmed so far whether either Intel or AMD is bringing PCIe 5.0 support in their respective Arc Alchemist or RDNA 3 GPUs.

Source: kopite7kimi (Twitter)

Nvidia Ada Lovelace to reportedly feature just PCIe 4.0 bus despite needing PCIe 5.0 power

Updates to Microsoft apps and services

We'll start this section off with news about Microsoft Teams. The online communication and collaboration app is getting two highly requested features within the next couple of months. These are the abilities to enable Together mode for everyone (GA in May 2022) and select a Calendar Show As status in meetings (GA in June 2022). Teams is also coming to the Microsoft Store and should be downloadable from there by next month.

Windows Terminal received a few updates too. The generally available version of the app was updated to version 1.12.1098 with a Windows 11 redesign and a bunch of bug fixes. Similarly, Windows Terminal Preview was bumped to version 1.13.1098 that squashed bugs and improved reliability by fixing crashing issues.

Pivoting a bit to Office now, Microsoft has issued a reminder that support for Office 2013 is ending in less than a year, on April 11, 2023. This comes after five years of mainstream support and five years of extended support. Office apps will continue to function but customers are obviously recommended to update to newer versions since Office 2013 will no longer receive security updates or technical support.

On the browser side of things, Edge Dev Channel build 102.0.1227.0 was released this week and it introduces a management policy to govern the Network Sandbox service and also fixes some crashes. Meanwhile, Edge Canary's latest builds comes with better search engine management and a flag to enable experimental appearance settings.

Other minor updates to Microsoft apps and services include Autofill in Microsoft Authenticator being able to generate strong passwords, ethical settings for Shopping hub in Bing, and historical search query pinning in Microsoft Launcher. Our readers who are gamers may also want to know that PC Game Pass has now arrived in five new Southeast Asian countries, namely: Indonesia, Malaysia, the Philippines, Thailand, and Vietnam.

Windows updates... or the lack of them

Microsoft announced this week that Windows 10 version 21H2 is finally ready for broad deployment and should be available to every Windows 10 user via Windows Update. This is only a small enablement package but it's recommended that you upgrade your build as support for Windows 10 version 20H2 is ending next month.

Although we did receive build 22598.200 (KB5014105) in the Beta and Dev Channels for Windows 11, this was just a servicing update to check the deployment pipeline. Microsoft later confirmed that no new build is coming this week but we may get one in the next.

While we are on the topic of Windows 11, Microsoft has put up a job listing for a vacancy in its Android Microsoft Platform Experiences (AMPX) team. This new division is responsible for making Android integration on Windows more refined than what is currently on offer, and ramping up on this front should hopefully mean better support for Android on Windows 11 in the future.

Rumors about the release date for the next version of Windows 11, version 22H2 (Sun Valley 2), are running wild at this time though. A new report now claims that Windows 11 version 22H2 will hit general availability in mid-September.

But for those on Windows 11 right now, Voice Clarity and touchpad improvements have landed on the Surface Laptop Studio via a firmware update. Microsoft has also clarified that it no longer installs the legacy SMB1 standard by default on Windows 11 Home builds in the Dev Channel, with the expectation that this change will eventually roll out to GA builds too. That said, it can still be installed from optional Windows features, and Microsoft has announced that it will release an out-of-band unsupported install package for organizations that cannot ditch SMB1 for more modern and secure standards yet.

Finally, for our readers managing Windows Server installations, Windows Server VNext Preview build 25099 is now available, there's no change log but there is a fix for Gen 2 VMs created from ISOs failing to boot.

Cybersecurity is important

While you are browsing the web, be on the lookout for malicious websites masquerading as their official counterparts. Some websites are now modifying their styling to match the design theme of Microsoft's websites and offering Windows 11 downloads to users. What they are actually distributing is malware that steals sensitive information from your browsers and crypto wallets (if any).

Security experts are also warning about vulnerable Microsoft Exchange Servers being targeted by attackers using Hive ransomware. A successful attack that encrypts your environments can be completed within 72 hours after initial compromise so it's highly recommended that you urgently update your Exchange Server installations, practice password rotation, block SMB1, and train employees in the domain of cybersecurity.

Similarly, Google Project Zero has published its annual report for 2021 which highlights how the most 0-day exploits ever were discovered in this year. These include 14 for Chromium, 10 for Windows, seven for Apple's WebKit, and four for Microsoft's Internet Explorer. However, there's some nuance to these findings so make sure to check out the details here.

It does appear that Microsoft Defender is being "overactive" in terms of threat detection... or is it? I kid, I kid. The defensive suite recently began flagging Google Chrome updates as "suspicious". Microsoft later confirmed that this was a false positive and it has now fixed the issue.

Dev Channel

• Microsoft is offering a free, three-month trial of PC Game Pass if you have played these games
• Bugsnax, 7 Days to Die, and more have joined Xbox Game Pass
• Halo composers Marty O'Donnell and Mike Salvatori have finally resolved their Halo royalties lawsuit with Microsoft
• Sea of Thieves' new Pirate Legend voyage and story Adventure are now live
• Halo Infinite's new game modes in Season 2 include a mini battle royale experience

Under the spotlight

Earlier this week, I published an editorial arguing that Microsoft should first focus on getting the basics of Windows 11 right before it starts adding more features that barely anyone has asked for. Most of our readers also agreed that they would appreciate more consistency in the design and user experience of Microsoft's latest OS. Check out the discourse here.

If you are a Microsoft Edge user and are bothered by websites auto-playing media content, our News Reporter Taras Buria wrote a guide on how you can better govern this behavior in Microsoft's browser.

Taras also wrote a handy guide on how you can personalize the experience of using your multi-monitor setup by setting up different wallpapers for each display. If that sounds exciting to you, head here to read the guide.

Meanwhile, Neowin co-founder Steven Parker wrote a technical but very useful guide on how you can add Terminal to the Windows 10 Quick Links (WinX) right-click context menu on Start menu. If you use command-line environments heavily, I would recommend reading the details here.

Finally, I penned a guide on how you can connect your Android smartphone to your Windows 10 or Windows 11 PC and access most content from a single device. Depending upon your hardware, you may also be able to access apps, stream your phone's display, and make calls. If that sounds exciting and useful, read the guide here.

Logging off

This week, we ran a poll asking our readers about whether they prefer using Caps Lock or Shift when typing. Although the polls results are currently skewed, you can still cast your vote to change that and leave your thoughts in the comment section as well.

On a closing note, we are always on the lookout for technical writers who can write about Microsoft news. Our payments start at $5 per news article but can go as high as $100 per piece (excluding bonuses). If that sounds like something right up your alley, ensure that you thoroughly go through the details here and apply accordingly.

Microsoft Weekly: App updates, broad deployment of Windows 10, and cybersecurity

Apple looks set to introduce 2nm chips in its flagship iPhone and Mac products as early as 2025, according to the grapevine, making it one of the first companies with plans to do so.

The first 2nm chip prototype was produced by IBM just last year, so the process is nowhere near ready for mass production. According to DigiTimes Asia, semiconductor giant TSMC indicated that its 2nm node won’t be ready for general production until 2025. 

Apple and Intel will be the first customers in line for this new process: For its part, Apple is looking forward to replacing the 5nm chips currently used in its iPhone and Mac with the new 2nm processor.

Before it can get there, TSMC is still ramping up its 3nm FInFET process for the latter half of 2022, with Apple being a major customer for the new 3nm chips. 

There have been conflicting rumors about whether TSMC’s 3nm production process was behind schedule or not, so it’s unclear how any potential delays might affect the company’s timetable for the 2nm chips; 2025 is a pretty far away, after all, so anything can happen.

Analysis: beyond 2nm lies...what exactly?

As 2nm processors near the consumer market, the question of what will follow nears as well. At 2nm, you're really talking about transistors that are barely bigger than a string of atoms, and it is becoming physically impossible to make them much smaller.

It's a problem that we've known about for more than a decade now: the "end of Moore's Law", named for legendary Intel chief Gordon Moore who famously quasi-predicted that computer transistor density would double roughly every two years. That has been the case for the past six decades, but the microprocessor industry has encountered major headwinds now that we're into the single-digit nanometer scale.

Beyond that, we're looking a picometer, or a thousandth of a nanometer, about 100 times smaller than an individual silicon atom. Where the industry goes for new innovation and progress on the semiconductor front is unclear, but it definitely isn't there.

Apple could bring ultra-fast 2nm silicon to Macs and iPhones as early as 2025

Insteon finally comes clean about its sudden smart home shutdown

Twitter board in a bind as Elon Musk makes his move

Sony is aiming to launch its revamped PlayStation Plus tiers in the Americas on June 13th, the company announced Friday. Sony is targeting a launch date first in markets in Asia (with the exception of Japan) on May 23rd, followed by a Japan launch on June 1st, an Americas launch on June 13th, and a Europe launch on June 22nd.

Sony first announced the new tiers in March, with the premium options offering perks like access to a collection of games to download or stream and the ability to play classic PlayStation, PS2, PS3, and PSP games. (Unfortunately, if you want to play the retro games, you’ll have to pay for the most expensive tier.) At the time of the original announcement, Sony had said that the tiers would be available first in the “June timeframe” in Asia, with “North America, Europe and the rest of the world” to follow.

Here is my colleague Tom Warren’s description of the PlayStation Plus tiers from when they were first announced:

PlayStation Plus Essential — includes multiplayer access, two monthly downloadable games, discounts, and cloud storage for game saves. Priced at $9.99 per month or $59.99 a year.

PlayStation Plus Extra — includes everything in Plus Essential and access to a catalog of up to 400 PS4 or PS5 games. Priced at $14.99 per month or $99.99 a year.

PlayStation Plus Premium — includes all benefits from Essential and Extra, with an extra 340 games that include PS3 titles that can be streamed and some original PlayStation, PS2, and PSP games. Time-limited game trials also available. Priced at $17.99 per month or $119.99 a year.

When the new tiers launch, PlayStation Now, which offers a collection of PS4, PS3, and PS2 that you can install or stream on a PS5, PS4, or PC, “will no longer be available as a standalone service,” Sony says.

Sony aims for June 13th US launch for revamped PlayStation Plus

As we recently reported, Microsoft may be building an in-game system for Xbox games that would allow brands and companies to serve ads. Sony could be contemplating a very similar strategy, indicated a report from BusinessInsider.

Microsoft might be placing digitally and dynamically rendered billboards within a game. However, these ads wouldn’t disturb gamers or impede their gameplay. Sony, on the other hand, seems to be working with AdTech companies to develop a dynamic platform that will help developers create (and perhaps distribute) in-game ads.

In other words, Microsoft could be building a platform that will allow more ad space. But the company won’t allow businesses interested in advertising their products and services to simply pay money and ensure gamers are served ads. Sony seems to be going in a slightly different direction, albeit with an identical agenda.

According to the report, Sony wants to “encourage developers to keep building free-to-play games”. This strongly suggests Sony wants to help developers make more money from their game titles by allowing them to serve ads to gamers. In other words, Sony could be developing newer avenues for enhanced monetization but would restrict the same to games that are free to download and play.

It is not immediately clear if Sony would charge a commission for allowing game developers and publishers to serve more ads to gamers. However, the report suggests Sony might consider “charging developers and publishers for data on consumer activity on the PlayStation” while respecting user privacy. Some of the suggested techniques include vetting AdTech companies and anonymizing user data.

Sony could increase monetization through promotional messages on its PlayStation 5 gaming console. The company could start pushing the new monetization techniques this year itself. Doing so could invite severe backlash, but if Sony does restrict ads to only free-to-play games, gamers and developers might not complain much.

It is important to note that Sony has neither confirmed nor denied that it is actively exploring options to include more ads within the PlayStation gaming ecosystem. Currently, the company serves ads within the in-game menus. Ads are also served to viewers who stream from gamers playing on consoles.

Source: BusinessInsider

Sony scouting AdTech companies to insert ads within PlayStation games, just like Microsoft

Way back in June 2008, Nvidia released its GeForce 200 series of graphics cards. The cards they released were GeForce GTX 260 and GeForce GTX 280.

In the same month, just a few days later, AMD (named ATI back then for their graphics cards), released the Radeon HD 4000 series of its graphics cards. The cards they released were Radeon HD 4850 and Radeon HD 4870. Both of them were runaway success due to their excellent performance and low price.

That was the last time both the companies released their cards together. Years and years went by, generations went by. But the releases of new graphics card series by both the companies never happened together. Looks like that is going to change. At least as far as the new series announcement is concerned.

September launch likely

Well known Twitter based leaker Greymon55 has said that both Nvidia, with its RTX 4000 graphics cards series and AMD, with its RX 7000 series, are going to launch them together in September.

He says, “The next generation of new products are basically concentrated in September, very exciting.”

He also mentions that both Nvidia and AMD might not follow the same release pattern, though.

Nvidia, he says, will follow the same pattern as Ampere, that is GeForce RTX 30 series. Which means, the expensive versions of the series, most likely GeForce RTX 4090 and GeForce RTX 4080 might release first. Followed by cheaper mid-range cards like RTX 4070 and RTX 4060 later.

AMD, he says, will release Navi 33 first, then Navi 31, then Navi 32. Which means, mid-range Radeon RX 7600 / XT might release first. Followed by expensive Radeon RX 7900 XT and Radeon RX 7800 / XT cards. Then Radeon RX 7700 might get released later.

Testing already started

While Greymon55 has said the testing of AMD’s next generation graphics cards has already started, it looks like there is some confirmation of the same for Nvidia too.

Another well known Twitter based leaker kopite7kimi has said that Nvidia has already started testing it’s next generation graphics card too.

The AD102, the GPU he mentions, is most likely going to be RTX 4090.

The release of the graphics cards is usually followed by a few months of testing. So all this confirms the earlier claim by Greymon55.

Expected prices

For now, we have no confirmation of prices. The MSRP is going to be in the line of current generation of graphics cards. The good thing however is that recent news reports confirm that graphics cards are becoming cheaper and are nearing the MSRP levels.

Previously, they were being sold at 2 times the MSRP levels. So them nearing the MSRP levels is excellent news. This means the next generation of cards might be relatively more available and affordable then current generation have been.

Nvidia RTX 4000 & AMD RX 7000 Cards Might Release Together

The Sapphire Pulse RX 6400 is one of the first low-profile RDNA2 GPUs.

If you’re looking for a serious gaming graphics card, look elsewhere — reviewers absolutely dragged the $199 AMD Radeon RX 6500 XT in January, and today’s GPU entry is even weaker. But if you absolutely, positively need to fit a miniature graphics card in a very small PC, the new RX 6400 might be worth a look.

Today, AMD has quietly launched the Radeon RX 6400 with an array of partners including ASRock, Biostar, Gigabyte, MSI, PowerColor, Sapphire, and XFX, and we’re actually seeing a few of them retail for that price or close to it, GPU shortage be damned. There’s a $159.99 ASRock Challenger and a $169.99 XFX Speedster SWFT105 in stock at Newegg right now — with a $159.99 Sapphire Pulse card also on the way.

Intriguingly, every one of these cards appears to be a miniature model, and many of them are single-slot, low-profile GPUs that can fit in much narrower cases.

(If you’re not familiar with low-profile GPUs, check out the pics above and below — they typically come with a shorter, swappable metal PCI-Express bracket like the one you see on the Sapphire card at the top of this post; the XFX card below is showing off its longer bracket for scale, but it should come with both.)

This XFX card looks short, and it’ll be shorter if you attach its low-profile PCIe bracket.

Mind you, there’s a reason these cards don’t need to be big! They’re only rated at 53W of power, less than half the power of even the lackluster RX 6500 XT, with only 12 compute units (down from 16), lower clocks, slower RAM, only 128GB per second of bandwidth, and just two display outputs. (Some of this is less surprising when you consider that its Navi24 GPU was originally designed for laptops.) On the plus side, you don’t need an extra power connector: the card can draw all its power from the PCI-Express slot.

According to one early video review, it looks like it will struggle with games like Elden Ring and Cyberpunk 2077 even at 1080p and on low settings but should be fine for the likes of Fortnite.

Know that while all RX 6400 models we’ve seen so far are miniature (big hat tips to VideoCardz and Wccftech for rounding up most of them), not all of them are low-profile. Many are two slots wide, some have two fans, and Gigabyte even has a model (below) that’s all of the above.

This Gigabyte model is squat but seemingly long and wide.

AMD quietly launches new space-saving RX 6400 graphics cards for $159

Smart home company Insteon appears to have closed up shop with no warning to its customers. Image: Insteon

Revolv, Iris, Insignia, Staples Connect, Wink, and now Insteon and iHome: the graveyard of dead or dying smart home ecosystems that promised so much yet failed to deliver is getting crowded. Smart home company Insteon has turned off its cloud servers, as first reported by Stacey on IoT, and device maker iHome has also shut down its servers, confirming to The Verge that its iHome cloud services were terminated on April 2nd.

This feels like a good time for a reflection on the state of the smart home. Is it all over? Or is this cloud carnage simply necessary to clear the way for a brave new world, one where the smart home is no longer a curiosity but something that actually matters?

What those companies mentioned above have in common is a reliance on a proprietary cloud server to deliver at least part of the experience customers signed up for. When the company’s business model changed and the cost of running that cloud was deemed unnecessary, consumers were left in the lurch.

The Revolv smart home hub was bought and then shut down by Google, Iris and Insignia’s clouds were switched off by Lowes and Best Buy, respectively, Staples pulled the plug on its Connect hub, and Wink has pivoted from a free to a paid service. A fact many manufacturers seem to overlook when jumping on the smart home bandwagon is that maintaining a cloud-based smart home service costs money — a lot of it, for a long time.

The Revolv hub was one of the first high-profile smart home companies to turn out the lights.

While most of those examples are ancient history, in the last few weeks, the cloud carnage has begun again. On April 2nd, device manufacturer iHome shut down its iHome app and iHome cloud service, announcing this quietly with only an in-app notification. The action ends support for several of its iHome branded smart plugs, its smart monitor, motion sensor, leak sensor, and door window sensor.

While the smart plugs and smart monitor will still work with the Apple Home app thanks to their HomeKit compatibility, beyond that, these devices are essentially junk. Astonishingly, many are still being sold, but as they require the iHome app, which no longer exists, they simply will not work.

Then, late last week, users of Insteon, a smart home ecosystem that relies on a proprietary communication protocol, started reporting that the hubs that control their Insteon smart light switches, outlets, sensors, thermostats, and other devices, were offline. The company, which has been in business since 2005 and was one of the earliest smart home pioneers, has gone completely dark.

There was no official word from the company ahead of the shutdown, and no advanced warning to users — which is inexcusable. And while the Insteon system status is still cheerily announcing that all services are online, the only official response so far is this cryptic “message to the Insteon community” on its site, which discusses the company’s financial troubles. It doesn’t explain what’s happening with its services or what its customers can do.

Interestingly, because Insteon was originally built as a locally controlled system, owners can switch their existing devices and hub to an open-sourced home automation system such as Home Assistant or Hubitat. So, while it’s a significant inconvenience, they aren’t completely out of luck, unlike non-Apple iHome users.

iHome’s smart home app has shut down, leaving users of some of its smart devices with no way to control them. Image: iHome

The weak link here is the proprietary cloud. A cloud-connected device has a myriad of benefits — most notably away-from-home control, over-the-air updates, and easier setup and programming. But its instability, especially if you’re taking a bet on a bootstrapped startup, is a major downside. The end-user has no control if the company that owns it decides to stop running the servers. This is a major reason why many people are wary of the smart home in its current form. Why spend money on something that could become a very expensive paperweight one day? That Revolv hub cost $300. Many Insteon customers spent hundreds or thousands of dollars on their systems.

The solution, as appealing as it might be in the moment, isn’t to abandon the smart home. Most connected devices offer a significant upgrade over their non-smart counterpart. A smart door lock can tell you exactly who unlocked your door and when; a connected sprinkler controller won’t water your garden if it’s going to rain; smart light bulbs can mimic the natural cycle of sunlight to help you feel more energized or more relaxed; and smart thermostats know when you’ve left and can stop wasting energy heating an empty home. And these are just a few examples.

The solution is to make smart home devices the norm, not the exception. For this to happen, they need a unified system to connect them, one that isn’t dependent on the fortunes of individual companies.

Matter devices will have the Matter logo on to show they're compatible with the new standard. Image: NXP

Here is where the promise of Matter comes in. When it arrives, the new smart home interoperability protocol backed by most of the big (and small) names in the industry (but notably not Insteon or iHome) should allow devices to work locally in your home without relying on a single cloud service to operate.

Instead, the expectation is that they will work with or without a cloud service, communicate with devices from different manufacturers locally, and, if you want the benefits of cloud control, work with whichever compatible platform you choose. If one service or ecosystem goes away, you should be able to just choose another way to control your devices.

“In cases like this, where manufacturer support ends, it is expected that devices that support Matter will continue to work locally with others in the home and be discoverable and controllable from other smart home systems and apps,” confirms Michelle Mindala-Freeman of the Connectivity Standards Alliance, the organization that oversees Matter. “This is another benefit of Matter’s Multi-Admin capability.” Multi-admin allows devices to use multiple platforms simultaneously, so your light bulb can be controlled by HomeKit and Alexa, for example.

However, Matter has been repeatedly delayed, and we still don’t know exactly how it will work in practice because no one has actually used it yet (note Mindala-Freeman’s use of the word “expected”). When it does arrive (currently scheduled for fall 2022), it will be too late to help iHome and Insteon customers. But it is clear that the smart home is at a major tipping point right now.

Which company will be next to shut off its servers? Smart lighting manufacturer LIFX’s parent company has gone into receivership, and despite the company’s protestations on Reddit that all is fine, it’s hard not to worry. In reality, any small company that relies on a cloud server, doesn’t charge a monthly subscription fee, and lacks deep pockets, is potentially at risk.

The safest bet for building your smart home today is to stick with the bigger names with good track records and solid companies behind them. Or sit around for a while under dumb light bulbs and wait patiently for Matter.

Update, Wednesday, April 20th, 3:18 PM: Updated the article to include a response from Insteon posted on its website regarding the company’s situation.

Insteon’s troubles are a smart home tale as old as time

QSDC takes advantage of entanglement as a means of securing network transmission over unsecured data lines. Because such particles are linked in a way that cannot be changed, protocols using them cannot be hacked without being detected by systems on the intended receiving end of such messages. As research has progressed to allow for the use of QSDC in real-world applications, the goal has been to reduce errors, increase transmission rates and, above all, extend the distance that messages using the protocol can be sent. Prior to this new effort, the record was just 18 km.

To extend that distance, the researchers devised a new QSDC protocol, one that involves the use of photonic time-bin states for monitoring signals and phase states for the actual communication messages. The researchers suggest adding such features to the QSDC protocol protects against phase errors and polarization. Further, it does not rely on feedback nor accurate matching of pairs of interferometers. They also suggest it makes such systems more reliable as well, which in turns leads to a lower error rate. And lowering the error-rate allows for extending the distance messages using the protocol can be sent.

The researchers acknowledge that the transmission rate is slow, at just 0.54 bps, which is slower even than systems using classical computing. But they note that it is still fast enough to allow for sending encrypted messages or even phone calls. They suggest their work shows that it is possible to create intercity QSDC-based networks using current technology. And they further suggest that certain parts of the Internet now in place could be replaced with parts based on the QSDC protocol they have developed to allow for hacker-resistant communications.

Source

According to a recent statement by Bill Dally, who is the Chief Scientist and Senior Vice President of Research at Nvidia, the company has started designing chips using an automatic standard cell layout generator called NVCell, that not only critique and find faults in human-made chip layouts, but it also can design its own chips and at a rate much faster than humans can. Apparently, it takes just two GPUs to do 92% of the work in just a few days that 10 people get done in about a year's time.

Here's what Dally says about NVCell:

This does two things for us. One is it’s a huge labor savings. It’s a group on the order of 10 people will take the better part of a year to port a new technology library. Now we can do it with a couple of GPUs running for a few days. Then the humans can work on those 8 percent of the cells that didn’t get done automatically. And in many cases, we wind up with a better design as well. So it’s labor savings and better than human design.

Nvidia isn't the only one who is impressed by AI's chip design capabilities though. Last year in June, Google too said a similar thing as it too was very impressed by how well reinforcement learning was doing in terms of chip layout making.

Source: HPCWire

Nvidia: Just two AI GPUs can do better chip design in a few days than 10 people do in a year

Only the graphite-colored model pictured here is available in both left and right-handed options. Photo by Amelia Holowaty Krales / The Verge

Logitech has announced Lift, a $69.99 wireless vertical mouse that’s launching today in multiple colorways, as well as with both right- and left-handed options. The Lift features a vertically oriented design that puts your hand at a 57-degree angle for better ergonomics. Compared to using a traditional mouse, using mice like the Lift may help to greatly reduce wrist strain since you’re holding your wrist at a similar angle to shaking someone’s hand. At least, that’s what I experienced. Last year, I made the shift to a split ergonomic keyboard, and a mouse like the Lift completes the setup.

For those keeping track, the Lift isn’t quite as fully featured as the pricier $99.99 MX Vertical, but it could be a good starting option if you don’t want to spend that much. It lacks USB-C charging, instead running off of a single AA battery that Logitech claims can (impressively) last up to two years.

The Lift ditches the elegant design of the MX Vertical for something that’s more simplified and playful (plus, you can get it in graphite, silver, or pink). It retains the rubber grip to keep it snug in your palm, and most of the same key functionality is here. The mouse has two main buttons, a scroll wheel that emphasizes smooth, quiet scrolling, a DPI switching button, and two thumb buttons. On its bottom, there’s a button for toggling between one of three devices that you can connect the Lift to (holding it doubles as the Bluetooth pairing button).

This mouse supports Logitech Flow, the company’s unique software feature that allows the mouse to be used simultaneously on multiple computers — even if they run a different OS. You’ll need the Logi Options Plus app running on both computers; then, the cursor can travel from one PC to the other. The app can also be used to easily copy and paste files between machines. I’ve seen a hands-on demo of this software working, but it just wouldn’t cooperate with me at home.

Logitech includes its new Bolt USB receiver with the Lift for quickly connecting to a PC that has a USB-A port. Compared to its previous unifying receiver, this one boasts better security. Or, you can utilize its Bluetooth function, which doesn’t require using a receiver.

I’ve had just a couple of days alone with the Lift as my primary mouse, but the transition from a standard mouse has gone smoother than I anticipated. Aside from accidentally knocking my hand into the Lift and tipping it over (it’s much taller than your average mouse), the learning curve isn’t too bad since this mouse has a similar button layout to mice that I’m accustomed to using. If you’ve been curious about vertical mice, the Lift could be a good option to start with since it costs well under $100.

The Lift is very comfortable to use, thanks in part to the rubberized grip. Photo by Amelia Holowaty Krales / The Verge

Logitech’s Lift is a low-cost vertical mouse that might convert you

After listening to the arguments in a case that involved Microsoft-owned LinkedIn and competitor hiQ Labs, the Ninth Circuit Court of Appeals has concluded that scraping publicly available data does not constitute a federal crime. The case dates back to 2017 which LinkedIn had filed against hiQ Labs. The social media platform for professionals had objected to its data being scraped.

LinkedIn essentially wanted hiQ Labs to immediately cease scraping public data from the social networking site. During the first trial, the court sided with hiQ Labs, noting that LinkedIn couldn’t invoke federal hacking laws to stop the practice. The court opinioned that hiQ Labs’ behavior didn't seem to violate any laws, and hence, the company’s actions could not be classified as a crime.

LinkedIn appealed the ruling to the United States Supreme Court, which remanded the case back to the Ninth Circuit court. The Appeals Court had recently deliberated on a case that reportedly involved a former Georgia police officer who was accused of violating the Computer Fraud and Abuse Act (CFAA) by looking up license plate data in exchange for bribes.

The Ninth Circuit Court of Appeals seems to have made a distinction between “improper use”, “unauthorized access”, and “authorized access”. The judgment reads:

As we have noted, however, a defining feature of public websites is that their publicly available sections lack limitations on access; instead those sections are open to anyone with a web browser. In other words, applying the ‘gates analogy’ to a computer hosting publicly available webpages, that computer has erected no gates to lift or lower in the first place.”

Simply put, had LinkedIn deployed mechanisms to prevent data from being scraped, hiQ Labs would have been in the wrong. However, since there were no restrictions, LinkedIn's insistence that hiQ Labs must cease its practice doesn’t have any merit.

The ruling could have a significant impact on data scraping. The practice has long infuriated social media platforms. Startups and smaller competitors to giants like Facebook have aggressively devised algorithms to scrape large chunks of data to quickly build and finetune their technologies. Needless to mention, tech giants have fiercely opposed the idea of data scraping. But henceforth, they might not have legal recourse, unless they deploy technologies to prevent the practice.

Scraping data from websites is not hacking or a crime, rules Appeals Court in US