Windows 11 to receive default BitLocker encryption, new RAM speed metrics in Task Manager

In a nutshell: The next major upgrade to Windows 11 will introduce several changes to the not-so-popular operating system. One key update is Microsoft enhancing the security of users' on-device data by enabling BitLocker by default, even without explicit authorization from those users.

The Windows 11 2024 (24H2) Update is on its way, and it will seemingly try to enable BitLocker disk encryption by default during installation. According to rumors shared by Deskmodder, the reworked setup process included in Windows 11 24H2 could encrypt the system disk in the background, with no visible option to disable the procedure beforehand.

BitLocker is a full-volume encryption technology first introduced by Microsoft with Windows Vista, a system designed to transparently protect users' data and the operating system itself by locking a disk volume behind a password. The password can also be saved in the TPM 1.2 security chip, with no need to enter it during every single OS boot.

BitLocker encryption can significantly boost a PC's security, but it may also impact I/O performance while reading and writing files on the encrypted volume. The technology has traditionally been an optional feature for Enterprise and Pro editions of Windows, but Windows 11 24H2 will apparently impose the full-volume encryption even to users of Windows editions like Windows Home.

Given that Microsoft often experiments with new features within Insider builds of Windows 11, default BitLocker encryption might be absent from the final build of Windows 11 24H2. Full-volume encryption would be easy enough to disable, anyway, as there are specific Windows settings for that.

Another, officially announced change planned for 24H2 should prove to be much less controversial to a wider userbase. As announced by Microsoft a few days ago, the Windows 11 Insider Preview Build 22635.3570 (Beta channel) includes an updated Task Manager with new metrics to show RAM performance.

The new Task Manager will replace the old MHz value with the more recent Mega Transfers per second (MT/s), which should be closer to how modern DDR RAM actually works. Back in the SDRAM days, memory chips would make just one data transfer per cycle. With Double Data Rate SDRAM, however, memory chips can perform two transfers per cycle.

Modern DDR5 RAM modules can be advertised as running at 6000 MHz, while the actual clock is 3000 MHz. A correct way to measure performance would then be 6000 MT/s. RAM manufacturers are still using larger MHz values to avoid user confusion, but Windows 11 24H2 should at least depict a clearer picture of how RAM memory is actually behaving.

Motherboard makers must implement new Intel power spec defaults by May 31

Recap: Since their release last fall, Intel's 13th-gen Raptor Lake and 14th-gen unlocked K-series desktop CPUs have been making waves – but not always in a good way. While offering top performance, these chips have also been plagued by stability issues for many users due to overly aggressive power settings. Intel is now cracking down, demanding that board partners implement new, more restrained defaults by the end of May.

The news comes from reports at saraba1st.com and Igor's Lab, citing sources familiar with Intel's plans. While not officially confirmed by Intel itself yet, the details have been corroborated from multiple angles, so we can likely take it that changes are coming down the pipeline.

Intel is asking that a new "Intel Default Settings" power profile become the out-of-box standard BIOS configuration on all LGA1700 motherboards. This profile will reportedly lock the CPUs' PL2 (maximum turbo power limit) to 188W – far lower than the 253W allowed by the Performance and Extreme profiles many vendors currently enable by default.

Also see: Intel CPUs Are Crashing and It's Intel's Fault: Intel Baseline Profile Benchmark – Intel is Now Paying the Price for Power Spec Mess

The move follows in the footsteps of recent BIOS updates from Asus and Gigabyte that incorporated Intel's recommended "baseline" settings. Those updates were successful enough at improving stability that Intel now feels emboldened to enforce the same approach industry-wide.

But don't worry, while reining in the extreme power limits, Intel's defaults will still keep key performance-enhancing features like Current Excursion Protection, ICCMax Unlimited, Turbo Velocity Boost, and Enhanced TVB enabled. The change is a targeted strike at the over-the-top overclocking and power delivery settings some vendors had been using out of the box in pursuit of benchmark wins.

According to sources, Intel specifically recommends not increasing PL1 and PL2 power limits beyond their stock ratings, nor pushing ICCMax current over 400A. So those 4096W PL2 and 512A limits some boards were configured for will be history.

Late last month, a leaked Intel notice obtained by Igor's Lab suggested the stability issues stemmed from motherboard makers disabling safeguards for higher performance. It listed settings like Current Excursion Protection and thermal limits being disabled on many 600/700-series boards, along with excessive power and voltage limit increases beyond Intel's specs.

The new defaults should provide a much-needed dose of stabilizing medicine for the 13th- and 14th-gen CPUs, even if it comes at some performance cost versus the extreme power profiles. By making it the mandatory baseline, Intel ensures even less experienced users can enjoy a smooth, reliable experience right out of the gate without having to dig into BIOS settings – though hardcore overclockers may lament the loss of those higher power limits.

Microsoft is shuttering four Bethesda studios, including Arkane Austin and Tango Gameworks, ending all development on Redfall

What just happened? It seems that making popular and successful games is no guarantee that a company will be safe. Microsoft has just announced the closure of four Bethesda studios as it looks to prioritize "high-impact titles." Arkane Austin (Prey), Tango Gameworks (Hi-Fi Rush), Alpha Dog Games, and Roundhouse Games are all being shuttered.

Arkane Studios has put out some excellent games in the past, including Dishonored, Prey, and Deathloop. Prey was developed by Arkane Austin, the Texas-based arm of the French video game developer, as was the much-maligned Redfall. It's the Austin studio that Microsoft is closing, with development of Marvel's Blade being passed on to Arkane Lyon.

Closing Arkane Austin means development on Redfall has now ended. The vampire co-op title will not receive its previously promised updates or new character DLC. The game and its servers will remain online, and Microsoft will provide a "make-good" offer to players who bought the Hero Pass as part of the premium Bite Back Edition or upgrade.

Pete Hines, senior vice president of global marketing & communications at Bethesda, last year said that Redfall would do a Fallout 76 and eventually gain more universal appeal.

Don't throw us into gold fever gambits, don't use us as strawmen for miscalculations/blind spots, don't make our work environments darwinist jungles. You say we make you proud when we make a good game. Make us proud when times are tough. We know you can, we seen it before.

– Dinga Bakaba 451 (@DBakaba) May 7, 2024

Also on the chopping block is Tango Gameworks, the company behind The Evil Within, Ghostwire: Tokyo, and Hi-Fi Rush.

Alpha Dog Games, maker of mobile game Mighty Doom, is being closed. Mighty Doom will be sunset on August 7 and players will no longer be able to make any purchases in the game.

Players interested in requesting a refund for purchased in-game currency can sign up here https://t.co/R7rL4LKsjk to receive details. pic.twitter.com/9g3kBoRGVw

– Alpha Dog Games (@AlphaDogGames) May 7, 2024

Roundhouse Games (formerly Human Head Studios) is being absorbed by The Elder Scrolls Online developer ZeniMax Online Studios.

Matt Booty, head of Xbox Game Studios, said in an email to staff that the cuts were due to Microsoft's "reprioritization of titles and resources."

"Today I'm sharing changes we are making to our Bethesda and ZeniMax teams," Booty writes. "These changes are grounded in prioritising high-impact titles and further investing in Bethesda's portfolio of blockbuster games and beloved worlds which you have nurtured over many decades."

Barring the failure of Redfall, the studios have been behind some successful titles. Microsoft itself called Hi-Fi Rush a "break out hit" last year. It recently won the Animation category at the BAFTA awards, too. But it seems the companies fell short in Microsoft's eyes when it came to producing "high-impact" titles, which are where resources are being allocated.

In January, Microsoft announced that it was laying off 1,900 employees from its gaming division, the majority of which would come from Activision Blizzard, though some workers at the company's Xbox and ZeniMax divisions were also affected.

Apple's redesigned 11-inch and 13-inch iPad Air models receive M2 chip as tablet sales see recovery

Something to look forward to: The new OLED iPad Pro and M4 processor were the main attractions of Apple's May 7 iPad event, but the iPad Air also received a significant update. The company's mid-range tablet now offers a new screen size option, increased performance, faster wireless connectivity, a new front-facing camera, and support for the newest Apple Pencil.

With the new iPads, Apple continues its new practice of offering large-screen models of its mainstream devices. For the first time, customers can purchase a 13-inch iPad without paying a premium for the iPad Pro.

The M2 processor makes the new iPad Air 50 percent faster than the previous model, maintaining eight CPU cores and increasing the GPU core count from eight to 10. Those upgrading from the iPad Air with the A14 Bionic chip should receive a 3x performance uplift. Moreover, Apple upgraded Wi-Fi connectivity from Wi-Fi 6 to Wi-Fi 6E.

Like the 10th generation iPad, the new iPad Pro and iPad Air models shift the front-facing camera to the long side of the bezel, allowing users to conduct FaceTime calls in landscape view. Another new feature the updated iPad Air shares with the new Pro model is support for the newly introduced Apple Pencil Pro. It adds haptic feedback and new rolling and squeezing inputs.

Pre-orders for the new iPad Air models are now open. The 11-inch and 13-inch variants start at $599 and $799, respectively. The base storage offering has doubled from 64 to 128GB, and Apple has introduced 512GB and 1TB options.

Apple likely launched the new products to try to boost tablet sales, which have suffered from the same downturn that affected other tech over the last two years like PCs and smartphones. While other product categories began showing signs of recovery in the fourth quarter of last year, tablet shipments have only just started creeping back upward.

The IDC reports that the sector grew by half a percent in Q1 2024, the first recorded growth since the second quarter of 2021. Although Apple remains the biggest tablet vendor, it sold 8.5 percent fewer units compared to the same quarter last year. Most of the growth was driven by Chinese manufacturers Huawei and Xiaomi, which saw a 43.6 percent and 92.6 percent sales increase, respectively.

Samsung launches a 114-inch Micro LED TV so expensive, buyers receive a free $8,000 8K TV

WTF?! Samsung has launched a new television aimed squarely at those for whom money is no object. Measuring 114 inches and packing Micro LED technology, the "ultra-premium" set costs the equivalent of around $132,630. It's so expensive that Samsung is giving away an 8K TV worth $8,000 as an incentive to buyers.

Samsung has a history of producing massive, and massively expensive, Micro LED TVs, going back to CES 2018 when it first unveiled The Wall, a 146-inch modular TV.

The Korean giant made a more living room-friendly 75-inch Micro LED TV a year later, followed by a 110-inch version in 2020. There have also been 89-inch and 110-inch models.

In addition to the massive size, the reason for the 114-inch version's KRW 180 million ($132,630) price tag is the set's use of Micro LED.

Like OLED, each of the millions of tiny LEDs packed onto a substrate produces its own light and color without the need for backlighting or a color filter. But Micro LED TVs have several advantages over OLED, including brighter colors, being more power-efficient, faster response times, and improved brightness – we've seen several Micro LEDs with a maximum brightness of 2,000 nits across the entire display. The most important element is that it uses an inorganic LED structure, offering a longer lifespan than OLED and no burn-in issues.

The downside of Micro LED TVs is, of course, their incredibly high prices. Samsung is trying to make its nearly $133,000 set more appealing by giving buyers a free 85-inch Neo QLED TV (QND900) worth $8,000, along with a discount for the JBL L100 MK2 speaker (over $4,000 per pair).

Further illustrating the kind of people who will buy this TV, Samsung is also giving customers who buy one this month a complimentary accommodation voucher for the Signiel Seoul Hotel valued at 3 million won ($2,200).

The TV is currently only available to buy in Korea. No word when or if it will launch in other locations.

The price of Micro LED TVs means only a few thousand are sold globally each year, equivalent to less than 1% of the 200 million televisions shipped annually. But there was a time when OLED sets were prohibitively expensive for most people, so prices of Micro LED sets will likely fall in the future, though don't expect them to be affordable for a long time yet.

Samsung's latest Micro LED TV seems like a good deal next to the 118-inch model LG introduced last year. It costs $237,000, which is still a lot cheaper than its $300,000 136-inch model.

TikTok and ByteDance sue US government over forced sale, claim free-speech violations

What just happened? TikTok and its owner ByteDance have sued the US government to stop a bill passed last month that will force the sale of the app or ban it in the United States. The companies argue that the bill is unconstitutional and violates free speech rights.

A lawsuit filed by TikTok and ByteDance in the court of appeals for the District of Columbia calls the law an "unprecedented violation" of the First Amendment. The companies add that the requirement to divest TikTok to avoid a nationwide ban is not commercially, legally, or technically possible, partly because it would lead to the app operating in isolation from the rest of the world – most of TikTok's one billion users are outside of the US.

"For the first time in history, Congress has enacted a law that subjects a single, named speech platform to a permanent, nationwide ban, and bars every American from participating in a unique online community with more than 1 billion people worldwide," ByteDance claims.

Many government and industry insiders, including former White House Chief Information Officer (CIO) Theresa Payton, previously said that ByteDance would never sell TikTok no matter the consequences. The Chinese owner confirmed this in the suit.

"There is no question: the Act will force a shutdown of TikTok by January 19, 2025, silencing the 170 million Americans who use the platform to communicate in ways that cannot be replicated elsewhere," the suit said.

Also confirmed in the lawsuit were the previous reports that ByteDance would not sell TikTok without its recommendation algorithm, which China added to its export control list in 2020 after then-President Donald Trump tried to force a sale.

The Chinese government "has made clear that it would not permit a divestment of the recommendation engine that is a key to the success of TikTok in the United States."

ByteDance wants the law to be declared an unconstitutional violation of the First Amendment, adding that national security concerns are not sufficient reason to restrict free speech. It adds that the government has not proven such a restriction is warranted.

The free speech argument has saved TikTok before. The company sued Montana last year when the state tried to ban the app claiming First Amendment violations. A judge agreed with the company, but Montana's Attorney General is trying to overturn the ruling in the Ninth Circuit Appeals Court.

A group of lawmakers announced in March the bill that would force ByteDance to sell TikTok or see the app banned. President Biden signed the bill into law last month, giving ByteDance 270 days to comply, meaning a ban could come into effect on January 19, 2025, though Biden could extend the deadline to sell by three months if he determines ByteDance is making progress.

TikTok started Project Texas in 2022, designed to protect US users' data from its Chinese parent company. But former employees say the project is "largely cosmetic" and current staff continue to work closely with Beijing-based ByteDance executives.

TikTok could eventually face another legal battle, this one in Europe. The European Commission, which was the very first institution to ban TikTok on its corporate smartphones, hasn't ruled out following the US in implementing a widespread ban.

Apple M4 arrives less than a year after M3 as the AI PC battle looms

Something to look forward to: With its newly announced iPads, Apple is fast-tracking the introduction of its M4 processors to compete in the upcoming AI PC race. Although the Cupertino giant claims it has the most powerful NPU on the market, Qualcomm might disagree. Meanwhile, Intel is preparing significant AI performance upgrades over the next year, and Nvidia claims that its GPUs reign supreme in generative AI.

Apple's event mostly focused on iPads, but the presentation also introduced the M4 – the company's next-generation in-house SoCs. As expected, the processors offer performance improvements over the M3 and M2, but Apple's rundown also delivered a preview of the company's AI PC strategy for 2024.

The segment focusing on the M4 could be considered a sneak peek, as the company only unveiled the standard variant and mostly focused on its performance improvements over the last iPad Pro model. Apple is expected to fully reveal the M4 lineup at WWDC in June, likely with new Macs.

The iPad Pro is skipping the M3 altogether, a chip Apple introduced just last year. This means that the M4 brings some of its predecessor's features to tablets for the first time, such as a 3nm node, dynamic caching, mesh shading, and ray tracing. The CPU is 50 percent faster than the M2 featured in the last iPad Pro, and the GPU is four times faster.

Compared to the standard M3, the M4 features a similar 10-core GPU but increases the CPU core count from eight to 10. Additionally, the SoC gains 3 billion transistors, bringing the total count to 28 billion.

One of the new features of the M4 is a display engine to help the new iPad Pro's OLED screens produce better brightness and colors. However, the most significant improvement of the new SoC is likely its enhanced AI performance.

Prior reports indicated that the M4 would mark the beginning of Apple's push toward offering products that run on-board AI applications, which are less reliant on cloud services used by programs such as ChatGPT, Bard, or DALL-E. The company's flex that, at 38 TOPs, the M4's neural engine is the world's fastest, appears to vindicate the rumors, but competition is set to heat up quickly.

AI PCs featuring Qualcomm's Snapdragon X Elite should begin shipping sometime in mid-2024. The company claimed that its new SoC outperformed M3, and the spec sheet's 45 TOPs of NPU performance suggests it might also compare favorably against M4.

The Arm-based Snapdragon X Elite aims to become the Windows equivalent of Apple Silicon. Apple's latest processors have enabled recent games like Resident Evil Village and Assassin's Creed Mirage to run on iPhones and iPads. In a similar vein, Qualcomm showed off its upcoming chip maintaining playable framerates in modern titles like Baldur's Gate 3 (which might come to Apple devices next year).

Meanwhile, Intel, which began the push for AI PCs with Meteor Lake late last year, plans to significantly increase its AI capabilities with Panther Lake next year. The 18A (1.8nm) chip will double the AI performance of Arrow Lake, which is expected to launch in the second half of 2024.

While Apple, Intel, Qualcomm, and AMD focus on enhancing on-device AI using NPUs, Nvidia believes GPUs are better suited to the task. The company claims that its RTX dedicated graphics chips, millions of which are already in the wild, can achieve between 100 and 1,300 TOPs. Nvidia is set to introduce the next generation of RTX, codenamed Blackwell, later this year.

Here is what it would look like if you traveled into the event horizon of a black hole

Editor's take: Black holes fascinate me. Nobody knows what goes on inside the cosmic maw. Math and physics break down as we get closer to the singularity, so we can only use our imagination to wonder what it's like on the inside, but that's okay. Theoretical cosmologist and physicist Professor Janna Levin once said, "Black holes are the ideal fantasy scape on which to play out thought experiments that target the core truths about the cosmos."

On Monday, NASA released a video showing what a camera would theoretically capture if it spiraled into a black hole (below). The clip is reminiscent of those Winamp music visualizations in the 90s (and today). It's a trippy trip into the weird math and physics of a black hole.

The video starts with a not-so-distant approach to the black hole, where we can see the cosmic entity in its entirety and the universe beyond. Super-heated gases form the wide outer ring called the accretion disk. The thin inner one is the photon ring. Even this thin ring has layers comprised of "distorted images of the gas disk layered between the background sky." These layers continue as photons orbit the event horizon one or more times before reaching the camera.

In the video, you can see the entire background universe smashed between layers. However, as the camera crosses the event horizon, the "sky" shifts and appears to shrink as the camera moves progressively faster.

As the camera moves from space to the accretion disk, then to the photon ring, event horizon, and finally the singularity, it accelerates faster and faster until it has to be slowed down to see what's happening. The camera acceleration becomes exponential as it moves through the accretion disk toward the event horizon in a rapidly decaying orbit. After crossing the point of no return, it only takes microseconds before the hole destroys the camera in a process physicists call spaghettification.

Spaghettification is when the part of an object closer to a black hole experiences exponentially increasing gravitational force faster than the part that is further away. It essentially stretches and crushes objects into long, thin, spaghetti-like structures, hence the highly technical name.

NASA produced the simulation using the Discover supercomputer at the Goddard Space Flight Center. It only took five days to render the simulation using just 0.3 percent of Discover's processing power. For perspective, rendering it on a typical laptop would have taken over a decade. Be sure to watch the video to the end, as NASA explains everything better than I can.

Cheyenne supercomputer sold to the highest bidder for $480,000

End of an era: A once-powerful supercomputer has now been decommissioned and sold after a week-long auction. However, the undisclosed "lucky" owner will surely need to spend significant additional funds to do anything remotely useful with the system.

The auction to get a piece of High-Performance Computing (HPC) history ended a few days ago, with the final bidder winning the Cheyenne supercomputer for $480,085 – a mere fraction of its $25-35 million development cost. The HPC system was one of the most powerful supercomputers ranked in the Top500 list in the mid-to-late 2010s, but now it needs some assistance to keep its aging circuits going.

Cheyenne's unnamed new owner purchased the complete HPC system comprising 28 rack units, 14 E-cell units, and 4,032 dual-socket units configured as quad-node blades. Cheyenne is equipped with 8,064 Intel Xeon E5-2967v4 CPUs, or a total core count of 145,152, more than 313 TB of DDR4-2400 ECC RAM, and other hardware.

The US General Services Administration hosted the auction, stating that the supercomputer's new owner must hire a professional moving company to pick up and deliver the system. Movers must use proper Professional Protection Equipment to safely handle the hardware's "considerable" weight, while the purchaser will have to assume full responsibility for moving operations. In other words, "don't look at us if it breaks."

The resale package did not include the entangled mass of fiber optic and CAT5/6 network cables connecting everything. General Services techs removed the internal DAC cables within each cell and graciously boxed them after graciously and meticulously labeling them to help the buyer if he chooses to reinstall the system.

Of course, General Services also stipulated that Cheyenne's sale condition is "as is," and disclosed that the machine currently has issues with "water sprays" caused by a faulty cooling system. Its DRAM memory nodes also show some ECC errors in around 1 percent of the system, meaning that it needs some expensive DIMM modules for the system to start up and begin reliably crunching numbers once again.

Cheyenne became operational in 2016 at the US National Center for Atmospheric Research. The system was ranked 21st in the Top500 supercomputer list, providing a peak performance of 5.34 PFLOPS. Cheyenne only ranked 160th in last year's TOP500 list after newer, more powerful systems superseded it.

Ransomware criminals use children's phone numbers to coerce payments from parents

WTF?! As awful as ransomware attacks are, perpetrators have found a way of making them even worse: psychologically attacking victims to make them hand over their money. One of these methods involves calling company executives from phone numbers belonging to their children.

Speaking at a Google Security Threat Intelligence Panel at this year's RSA Conference in San Francisco (via The Reg), Charles Carmakal, CTO of Google-owned cybersecurity firm Mandiant, said, "We saw situations where threat actors essentially SIM swap the phones of children of executives, and start making phone calls to executives, from the phone numbers of their children."

Carmakal noted the psychological dilemma of seeing an incoming phone call from your child only to answer and hear a stranger's voice. "Sometimes, it's caller ID spoofing. Other times, we see demonstrated SIM swapping family members," he said.

Ransomware gangs have evolved over the years from deploying malware that encrypts files to also stealing them and threatening to publish sensitive data online if demands are not met. These types of operations made more than $1 billion in 2023, explaining why they remain so popular among criminals.

One of the worst parts about ransomware is the fact that many hackers use it to target healthcare facilities such as hospitals. These incidents can threaten the security of sensitive information, extend patients' hospital stays, and even place people's lives at risk, meaning targets are more likely to pay up.

According to technology research provider Omdia, the healthcare sector suffered 241 cyberattacks during the first nine months of 2023. That's over 100 more than the government (147) and almost three times more than software, hardware, and IT services (91).

Law enforcement repeatedly tells victims of ransomware attacks to not pay any ransom money as there's no guarantee the perpetrators will hand over the decryption key. Doing so also marks the organization as one that is willing to pay, encouraging others, or even the same criminals, to launch new attacks.

The SIM swapping/ID spoofing highlights how ransomware gangs are starting to focus more on people rather than faceless organizations in the hope they will feel pressured to hand over the crypto payments. "It's less about 'do I need to protect my customers?' But more about 'how do I better protect my employees and protect the families of employees?' That's a pretty scary shift," said Carmakal. "There are a few threat actors that really have no rules of engagement in terms of how far [they] try to coerce victims."

The good news is that ransomware operators face harsh prison sentences. A REvil hacker was recently sentenced to 13 years behind bars and ordered to pay a $16 million fine. There was also the case of a LockBit ransomware group member who got four years and was fined $635,000.

Visa: Not all stablecoin transactions are "real"

In brief: Stablecoins are a particular type of cryptocurrency designed to provide a higher degree of financial reliability. The value of a stablecoin is theoretically pegged to a reference asset, which can be fiat money, traded commodities, or even another cryptocurrency.

Stablecoins have been hailed as the future of blockchain-based digital money due to their financial stability compared to Bitcoin and other "traditional" cryptocurrencies. However, according to Visa, more than 90 percent of stablecoin transactions aren't actually generated by real users of the technology.

The financial giant worked in partnership with Allium Labs to develop a new metric that helps distinguish genuine stablecoin transactions from fake ones initiated by bots and large-scale traders. According to data provided by Visa, most of the $2.2 trillion in stablecoin transactions recorded in April weren't made by real people. The company detected "organic" payment activities for just a minimal portion of those transactions ($149 billion).

Some high-profile companies like PayPal and Stripe have adopted stablecoin tokens as a potentially revolutionary payment technology compared to traditional money. According to Pranav Sood, general manager for EMEA of payment platform Airwallex, the latest data provided by Visa shows that stablecoins are still considered a "nascent" monetary tool.

Despite being one of the hottest topics in the financial sector for the past few years, digital tokens have yet to be widely accepted by most customers. Airwallex says there's only a "tepid" demand for stablecoin-based payments, noting that many of its clients find them too complex and not user-friendly enough.

Stablecoins still present a "significant barrier" to potential user adoption, Airwallex said, and the financial market isn't moving as quickly as tech companies would like. In the US alone, a significant portion of all business payments (40-60 percent) are still made by using checks. Technology adoption is slow, and stablecoins as a potential dollar replacement have not gained traction.

The new metric co-developed by Visa can apparently track the real value of digital token transactions on the blockchain, which has always been a challenging task. Previous estimations provided by Glassnode suggested that the theoretical $3 trillion value assigned to the digital token market in 2021 was actually closer to $875 billion. Visa, which handles more than $12 trillion in transactions per year, could see its business shrink if stablecoins and other digital tokens were to become generally accepted tools for payments and other monetary transactions.

Apple's impossibly thin iPad Pro packs an M4 chip and tandem OLED technology

In a nutshell: Apple has formally taken the wraps off its new iPad Pro lineup, and there's a lot to get excited about. The tablets are powered by new Apple M4 silicon (M3, we hardly knew you) and are the thinnest Apple devices ever created – and that's just scratching the surface.

The new iPad Pro is offered in two variants, one with an 11-inch display and the other sporting a larger 14-inch panel. Both utilize the brand new M4 chip, although specs vary depending on the chosen storage configuration.

Models with 256 GB or 512 GB of storage ship with 8 GB of RAM and a nine-core CPU featuring three performance cores alongside six efficiency cores paired with a 10-core GPU. Buyers that opt for 1 TB or 2 TB of storage will get an M4 with 10 cores (four performance cores and six efficiency cores) along with the same 10-core GPU. All variants include a 16-core Neural Engine and 120 GB/s of memory bandwidth.

The new M4 chip is built on second-gen 3-nanometer tech that's far more efficient than previous iterations. According to Apple, it can deliver the same level of performance as the M2 at just half the power, or match the performance of the "latest PC chip" in a thin and light laptop using just a quarter of the power.

The new Neural Engine, meanwhile, is capable of 38 trillion operations per second, making it 60 times faster than the first Neural Engine from the A11 Bionic SoC. According to Apple, the Neural Engine is more powerful than any NPU found in AI PCs currently on the market today.

At the heart of the new iPad Pro is the Ultra Retina XDR, described by Apple as the world's most advanced display. It features tandem OLED technology that literally combines two OLED panels to boost full-screen brightness to 1000 nits for SDR and HDR content, and up to 1600 nits peak for HDR.

Around back, you'll find a single 12-megapixel camera sporting a five-element lens with f/1.8 aperture and a True Tone flash. Another 12-megapixel shooter – an ultrawide with f/2.4 aperture – handles forward-facing duties.

Given all the new hardware and dual OLED displays, it's impressive that Apple managed to cram it all into its thinnest devices to date. The 11-inch model measures just 5.3mm thick while the 13-incher is even slimmer at 5.1mm.

Apple's new iPad Pros are available to pre-order today. Pricing starts at $999 for an 11-inch Wi-Fi only model, or $1,199 if you need cellular. The 13-inch iPad Pro commands $1,299 for a 256 GB model with Wi-Fi, or $1,499 for the cellular-enabled variant. Both are offered in your choice of silver or black finishes with 256 GB, 512 GB, 1 TB, or 2 TB storage options. The 1 TB and 2 TB storage variants also have the option for nano-texture glass, which is precisely etched at a nanometer scale to reduce glare.

Robert Dennard, inventor of DRAM and pioneer of semiconductor scaling, dies at 91

RIP: If you're reading this on just about any modern computing device, you're benefiting from the pioneering work of Robert H. Dennard. The prolific inventor and IBM researcher, affectionately known as "Bob" to colleagues, passed away on April 23 at the age of 91.

Dennard's claims to fame are enormous. He's the mind behind dynamic random-access memory (DRAM), the ubiquitous memory that quite literally makes our computers and smartphones function. And that's just the start. This Texas-born extraordinaire left an indelible mark on the trajectory of semiconductor innovation.

It all began at the Carnegie Institute of Technology (now Carnegie Mellon University), where Dennard earned his PhD in electrical engineering back in 1958. Fresh out of grad school, he joined the hallowed ranks of IBM's research division, kicking off decades of breakthroughs.

In the 1960s, integrated circuits were all the rage, with microchip pioneers racing to enhance computer memory and logic. Dennard felt that existing magnetic core RAM was too big and power hungry, so he assembled a team to dream up an alternative using just six MOS transistors per bit of data. But even that design felt convoluted.

Then came the eureka moment: Dennard wondered if a single bit of data could be stored using just one tiny transistor. With that, Dennard laid the groundwork for DRAM as we know it today. His 1968 patent for this technology ushered in an era of compact, low-power memory that made modern computing possible.

However, DRAM was merely Dennard's opening salvo. He's also known for the scaling theory that bears his name – Dennard scaling – which described how shrinking transistor dimensions could boost performance while reducing power consumption. This principle turbocharged the industry's rapid miniaturization for decades, allowing for exponential leaps in chip density and clock speeds in accordance with Moore's Law.

Eventually, transistors became so minuscule that current leakage and voltage scaling broke down, bringing an end to the clockspeed arms race. But by then, semiconductor sorcery had already transformed our world.

For his contributions, Dennard racked up a trophy case of prestigious awards including the National Medal of Technology and Innovation in 1988 and the Kyoto Prize in Advanced Technology in 2013, among others. More importantly, he inspired generations of chip designers, engineers, and scientists to pursue world-changing innovations.

A memorial celebration of Dennard's life and career is planned for June 7 at IBM's Thomas J. Watson Research Center in Yorktown Heights, New York. While Bob may be gone, his genius lives on in the silicon powering our digital age.

Image credit: Fred Holland

PC market recovery leads to increased client CPU shipments, but server sales lag

In context: With the global PC market continuing to recover from a years-long slump, processor makers Intel and AMD are reportedly increasing their CPU shipments year-over-year, even as they see a drop in sales sequentially. The server market, however, remains sluggish, with server CPU shipments registering a double digit fall compared to last year.

According to a new report from Jon Peddie Research (JPR), the client PC CPU market reached 62 million units in Q1, 2024, representing a healthy 33 percent growth from the same period last year. However, the shipments are still down 9.4 percent sequentially from Q4, 2023, when the manufacturers shipped 69 million units globally.

Total iGPU shipments, which hit 56 million units last quarter, also show a similar trend. While it grew 30 percent year-over-year, it still represents a reduction from Q4, 2023 which saw the shipment of 62 million units. Looking forward, the report predicts that iGPU penetration in the PC sector will grow significantly to reach 98 percent.

Another interesting trend confirmed by the report is the increasing preference for laptops among most PC users. While demand for notebooks has far outstripped desktop sales for years, the latest data suggests that this is now more pronounced than ever. In Q1 2024, desktop processors accounted for 27 percent of the market, down from 32 percent in Q1 2023, as mobile CPUs increased their market share from 68 percent to 73 percent.

Meanwhile, despite the growth in client PCs, it's not all good news for companies like Intel and AMD. According to the report, server CPU shipments fell 13 percent from last quarter and as much as 17 percent from the same period last year, suggesting server farms and data centers are not upgrading as fast as the PC hardware industry would want.

According to JPR president Jon Peddie, the latest results took many analysts and investors by surprise, as they expected the sequential growth to continue for at least a few more quarters. Peddie expects the Q2 numbers to remain weak, but believes that it could be a good thing in the long run, as it's an indication that the market has stabilized and returned to its "traditional cyclic behavior." The industry as a whole, however, will hope that the market returns to growth sooner rather than later.

Rare Super Famicom prototype zooms to $40k bid at auction

Rare opportunity: Nintendo collectors with deep pockets have the rare opportunity to get their hands on an early Super Famicom prototype that looks an awful lot like the pre-production model featured in gaming magazines in the late 80s.

The Super Famicom (known as the Super Nintendo Entertainment System in North America) was the successor to the original Family Computer. That system debuted in Japan in 1983 before finding its way to American test markets as the original NES in late 1985. Nintendo wasn't in a hurry to release a follow-up but pressure from competitors like Sega and its Genesis prompted a rethink.

Nintendo launched the new console on November 21, 1990, and sold out of its initial batch of 300,000 units within hours. A redesigned version, the SNES, arrived in North American in the summer of 1991.

The prototype Super Famicom up for auction differs from the production model in that it includes a headphone jack and volume wheel, much like the Nintendo PlayStation prototype that surfaced in 2015. Eagle-eyed readers may also notice that the controller ports are left-aligned instead of centered, and the expansion port appears on the front of the console instead of the bottom. The bright red power switch stands out like a sore thumb (but in a good way).

As of this writing, bidders have just over four days to vie for the console. The high bid sits right around 6 million yen, or around $40,000, and is sure to go higher. For comparison, the Nintendo PlayStation brought around $360,000 at auction in 2020. It could be worth far more than that now, as it is believed to be the only example to have survived after Nintendo and Sony had a falling out during development.

The SNES that eventually found its way to North America looks much different than the Super Famicom. According to Lance Barr, who designed the chassis for the NES and SNES, he felt the Super Famicom was "too soft and had no edge."

At the time, Nintendo was considering future modular components (hence the expansion port on the bottom of the machine). Barr said he thought the Super Famicom didn't look good when stacked (or even by itself), adding that it had a kind of "bag of bread" look.

Leak casts doubt on Intel's Battlemage, Celestial GPU plans for 2024

Rumor mill: Just when we thought Intel was finally gaining some solid momentum in the discrete GPU space, recent rumblings have thrown a major wrench into those plans. If they hold true, Intel's next-generation Arc GPUs codenamed "Battlemage" and "Celestial" could face delays or even get axed entirely.

According to a post by (typically) reliable leaker Golden Pig Upgrade on Weibo, there will be no "DG3" discrete graphics cards based on the Xe2 Battlemage architecture coming next year. For those not in the loop, Battlemage is slated to debut first in Intel's Lunar Lake "Core Ultra" CPUs expected later this year, boasting a hefty performance uplift over the current Alchemist and Alchemist+ parts.

If these get benched or delayed significantly, it'd be a massive blow to Intel's GPU ambitions.

The situation is the same for Celestial, Intel's follow-up Xe3 architecture. Golden Pig claims there might not even be Xe3 products next year at all, integrated or discrete. Celestial is supposed to arrive in 2025's Panther Lake chips on Intel's 18A and 3 nodes, but this leak casts serious doubt on those plans. It's still unclear whether the rumored setbacks affect just the discrete GPUs, the integrated versions, or both product lines.

Abandoning the discrete GPU arena just a year after breaking into it with Arc Alchemist would be equally confusing and disappointing. The Arc team has made impressive strides through steady driver updates, optimizations and day-one game support. Tossing all that progress out the window would be a waste of resources.

However, a report from last month citing sources at April's Embedded World Conference claimed that Intel plans to introduce Battlemage discrete GPUs before Black Friday 2024. This launch window would avoid the delays that plagued 2022's Alchemist debut but also pit Battlemage against new Nvidia and AMD offerings. The delays with Alchemist put Intel behind the competition performance-wise, so hitting this alleged November target could make or break Battlemage's prospects.

The only official word so far has been a roadmap slide from Intel's Meteor Lake Japan launch keynote late last year, which listed Battlemage GPUs for release sometime this year alongside details on the company's AI ambitions.

For now though, Golden Pig's assertions are still just rumors. Because if this leak proves true, it would represent a dramatic shift in strategy and call into question Intel's ability to compete long-term against team blue and green in the graphics department.

DDR5 memory prices could rise 20% due to increasing demand for AI chips

Facepalm: DDR5 prices have fallen drastically since their debut a few years ago, with PC builders and DIY enthusiasts currently enjoying dirt-cheap prices on the latest memory modules. However, the good days of ultra-cheap DDR5 may be coming to an end, with a new report suggesting that DRAM prices may increase significantly over the next year.

According to analysts from market research firm TrendForce, high-bandwidth memory (HBM) prices are expected to rise 5-10 percent in 2025 due to exceptionally high demand from the AI sector, leading to a price increase for other types of DRAM. Unfortunately for PC builders, this will also affect the DDR5 market, as it will reportedly see a 15-20 percent price increase, with memory manufacturers expected to shift priorities to the more lucrative HBM market.

HBM is about five times more expensive than DDR5, and the exploding demand for this type of memory is expected to "dramatically raise" its share in the DRAM industry, both in terms of capacity and market value. Fueled by the AI boom, demand for HBM has reached fever pitch, with both Micron and SK Hynix reporting that their supply of high-bandwidth memory is completely booked for the entirety of 2024 and a significant portion of 2025.

According to TrendForce Senior Research Vice President Avril Wu, HBM's share of total DRAM capacity is estimated to rise from 2 percent in 2023 to 5 percent in 2024 and surpass 10 percent by 2025, while its market value could exceed 30 percent of the total DRAM market by the end of next year.

While HBM offers significant performance and capacity advantages over other types of DRAM, it is also relatively more difficult to manufacture, meaning the additional HBM production is likely to reduce resources for the manufacture of other types of memory, including DDR5, thereby reducing their availability in the market and pushing their prices up.

The report predicts that the annual growth rate of HBM demand will be nearly 200 percent this year, and it is tipped to further double in 2025. There will also reportedly be a significant shift towards HBM3e next year, with the report predicting an increase in 12Hi stack products. The change is expected to increase the HBM capacity per chip, which should come in handy for AI solution providers.