Tech
AI Boom Fuels DRAM Shortage and Price Surge
If it feels these days as if everything in technology is about AI, that’s because it is. And nowhere is that more true than in the market for computer memory. Demand, and profitability, for the type of DRAM used to feed GPUs and other accelerators in AI data centers is so huge that it’s diverting away supply of memory for other uses and causing prices to skyrocket. According to Counterpoint Research, DRAM prices have risen 80-90 precent so far this quarter.
The largest AI hardware companies say they have secured their chips out as far as 2028, but that leaves everybody else—makers of PCs, consumer gizmos, and everything else that needs to temporarily store a billion bits—scrambling to deal with scarce supply and inflated prices.
How did the electronics industry get into this mess, and more importantly, how will it get out? IEEE Spectrum asked economists and memory experts to explain. They say today’s situation is the result of a collision between the DRAM industry’s historic boom and bust cycle and an AI hardware infrastructure build-out that’s without precedent in its scale. And, barring some major collapse in the AI sector, it will take years for new capacity and new technology to bring supply in line with demand. Prices might stay high even then.
To understand both ends of the tale, you need to know the main culprit in the supply and demand swing, high-bandwidth memory, or HBM.
What is HBM?
HBM is the DRAM industry’s attempt to short-circuit the slowing pace of Moore’s Law by using 3D chip packaging technology. Each HBM chip is made up of as many as 12 thinned-down DRAM chips called dies. Each die contains a number of vertical connections called through silicon vias (TSVs). The dies are piled atop each other and connected by arrays of microscopic solder balls aligned to the TSVs. This DRAM tower—well, at about 750 micrometers thick, it’s more of a brutalist office-block than a tower—is then stacked atop what’s called the base die, which shuttles bits between the memory dies and the processor.
This complex piece of technology is then set within a millimeter of a GPU or other AI accelerator, to which it is linked by as many as 2,048 micrometer-scale connections. HBMs are attached on two sides of the processor, and the GPU and memory are packaged together as a single unit.
The idea behind such a tight, highly-connected squeeze with the GPU is to knock down what’s called the memory wall. That’s the barrier in energy and time of bringing the terabytes per second of data needed to run large language models into the GPU. Memory bandwidth is a key limiter to how fast LLMs can run.
As a technology, HBM has been around for more than 10 years, and DRAM makers have been busy boosting its capability.
As the size of AI models has grown, so has HBM’s importance to the GPU. But that’s come at a cost. SemiAnalysis estimates that HBM generally costs three times as much as other types of memory and constitutes 50 percent or more of the cost of the packaged GPU.
Origins of the memory chip shortage
Memory and storage industry watchers agree that DRAM is a highly cyclical industry with huge booms and devastating busts. With new fabs costing US $15 billion or more, firms are extremely reluctant to expand and may only have the cash to do so during boom times, explains Thomas Coughlin, a storage and memory expert and president of Coughlin Associates. But building such a fab and getting it up and running can take 18 months or more, practically ensuring that new capacity arrives well past the initial surge in demand, flooding the market and depressing prices.
The origins of today’s cycle, says Coughlin, go all the way back to the chip supply panic surrounding the COVID-19 pandemic . To avoid supply-chain stumbles and support the rapid shift to remote work, hyperscalers—data center giants like Amazon, Google, and Microsoft—bought up huge inventories of memory and storage, boosting prices, he notes.
But then supply became more regular and data center expansion fell off in 2022, causing memory and storage prices to plummet. This recession continued into 2023, and even resulted in big memory and storage companies such as Samsung cutting production by 50 percent to try and keep prices from going below the costs of manufacturing, says Coughlin. It was a rare and fairly desperate move, because companies typically have to run plants at full capacity just to earn back their value.
After a recovery began in late 2023, “all the memory and storage companies were very wary of increasing their production capacity again,” says Coughlin. “Thus there was little or no investment in new production capacity in 2024 and through most of 2025.”
The AI data center boom
That lack of new investment is colliding headlong with a huge boost in demand from new data centers. Globally, there are nearly 2,000 new data centers either planned or under construction right now, according to Data Center Map. If they’re all built, it would represent a 20 percent jump in the global supply, which stands at around 9,000 facilities now.
If the current build-out continues at pace, McKinsey predicts companies will spend $7 trillion by 2030, with the bulk of that—$5.2 trillion—going to AI-focused data centers. Of that chunk, $3.3 billion will go toward servers, data storage, and network equipment, the firm predicts.
The biggest beneficiary so far of the AI data center boom is unquestionably GPU-maker Nvidia. Revenue for its data center business went from barely a billion in the final quarter of 2019 to $51 billion in the quarter that ended in October 2025. Over this period, its server GPUs have demanded not just more and more gigabytes of DRAM but an increasing number of DRAM chips. The recently released B300 uses eight HBM chips, each of which is a stack of 12 DRAM dies. Competitors’ use of HBM has largely mirrored Nvidia’s. AMD’s MI350 GPU, for example, also uses eight, 12-die chips.
With so much demand, an increasing fraction of the revenue for DRAM makers comes from HBM. Micron—the number three producer behind SK Hynix and Samsung—reported that HBM and other cloud-related memory went from being 17 percent of its DRAM revenue in 2023 to nearly 50 percent in 2025.
Micron predicts the total market for HBM will grow from $35 billion in 2025 to $100 billion by 2028—a figure larger than the entire DRAM market in 2024, CEO Sanjay Mehrotra told analysts in December. It’s reaching that figure two years earlier than Micron had previously expected. Across the industry, demand will outstrip supply “substantially… for the foreseeable future,” he said.
Future DRAM supply and technology
“There are two ways to address supply issues with DRAM: with innovation or with building more fabs,” explains Mina Kim, an economist with the Mkecon Insights. “As DRAM scaling has become more difficult, the industry has turned to advanced packaging… which is just using more DRAM.”
Micron, Samsung, and SK Hynix combined make up the vast majority of the memory and storage markets, and all three have new fabs and facilities in the works. However, these are unlikely to contribute meaningfully to bringing down prices.
Micron is in the process of building an HBM fab in Singapore that should be in production in 2027. And it is retooling a fab it purchased from PSMC in Taiwan that will begin production in the second half of 2027. Last month, Micron broke ground on what will be a DRAM fab complex in Onondaga County, N.Y. It will not be in full production until 2030.
Samsung plans to start producing at a new plant in Pyeongtaek, South Korea in 2028.
SK Hynix is building HBM and packaging facilities in West Lafayette, Indiana set to begin production by the end of 2028, and an HBM fab it’s building in Cheongju should be complete in 2027.
Speaking of his sense of the DRAM market, Intel CEO Lip-Bu Tan told attendees at the Cisco AI Summit last week: “There’s no relief until 2028.”
With these expansions unable to contribute for several years, other factors will be needed to increase supply. “Relief will come from a combination of incremental capacity expansions by existing DRAM leaders, yield improvements in advanced packaging, and a broader diversification of supply chains,” says Shawn DuBravac , chief economist for the Global Electronics Association (formerly the IPC). “New fabs will help at the margin, but the faster gains will come from process learning, better [DRAM] stacking efficiency, and tighter coordination between memory suppliers and AI chip designers.”
So, will prices come down once some of these new plants come on line? Don’t bet on it. “In general, economists find that prices come down much more slowly and reluctantly than they go up. DRAM today is unlikely to be an exception to this general observation, especially given the insatiable demand for compute,” says Kim.
In the meantime, technologies are in the works that could make HBM an even bigger consumer of silicon. The standard for HBM4 can accommodate 16 stacked DRAM dies, even though today’s chips only use 12 dies. Getting to 16 has a lot to do with the chip stacking technology. Conducting heat through the HBM “layer cake” of silicon, solder, and support material is a key limiter to going higher and in repositioning HBM inside the package to get even more bandwidth.
SK Hynix claims a heat conduction advantage through a manufacturing process called advanced MR-MUF (mass reflow molded underfill). Further out, an alternative chip stacking technology called hybrid bonding could help heat conduction by reducing the die-to-die vertical distance essentially to zero. In 2024, researchers at Samsung proved they could produce a 16-high stack with hybrid bonding, and they suggested that 20 dies was not out of reach.
From Your Site Articles
Related Articles Around the Web
Tech
Google now helps you wipe your sensitive personal data and photos from Search
Google Search can make information easy to find, but it can also make your personal data surface in ways that feel invasive or even dangerous. This is why Google is rolling out new tools that give people more control over what shows up about them online.
The company says it is expanding its Search removal features to make it simpler to take down sensitive personal information and explicit images that never should have been public in the first place.
How to remove personal information from Search
Google’s “Results about you hub” can now help you find and remove search results that contain sensitive government-issued identification numbers. This includes things like passport numbers, driver’s license numbers, and other official ID info that could be misused if they appear online.

To use this feature, you sign in to your Google account and select ‘Results about you,’ where you can fill out the information you want to track. Google will proactively scan Search for results that match your personal information and alert you if it finds something.
From there, you can review each result and request removal directly within the tool. You can also manually submit a removal request if you come across sensitive information yourself. Google says it will review these requests and remove results that violate its policies.
How to remove explicit images from Search

Google is also simplifying the process for removing explicit images, especially those shared without consent. You can now request the removal of explicit images more easily, including submitting multiple images at once rather than filing separate requests.
Once an image is removed, Google will also offer an option to proactively filter out similar explicit images from future Search results, to prevent similar content from resurfacing.

You can now track all your removal requests in one place through the Results about you hub, with email updates to keep you informed whenever the status changes.
Google also points out that removing information from Search does not erase it from the internet altogether, but it can still go a long way in protecting your privacy.
The update also comes as Google shuts down its dark web reports, which previously alerted users when their name, phone number, or email surfaced online in a data breach.
Google says those alerts did not always help people take meaningful action, something the new removal tools are designed to address.
Tech
Microsoft releases Windows 10 KB5075912 extended security update
Microsoft has released the Windows 10 KB5075912 extended security update to fix February 2026 Patch Tuesday vulnerabilities, including six zero-days, and continue rolling out replacements for expiring Secure Boot certificates.
If you are running Windows 10 Enterprise LTSC or are enrolled in the ESU program, you can install this update like normal by going into Settings, clicking on Windows Update, and manually performing a ‘Check for Updates.’

Source: BleepingComputer
After installing this update, Windows 10 will be updated to build 19045.6937, and Windows 10 Enterprise LTSC 2021 will be updated to build 19044.6937.
What’s new in Windows 10 KB5075912
Microsoft is no longer releasing new features for Windows 10, and the KB5075912 update contains only security fixes and bug fixes introduced by previous security updates.
With today’s February 2026 Patch Tuesday, Microsoft has fixed 58 vulnerabilities, including six actively exploited zero-day flaws.
KB5075912 also fixes a known issue that prevented Windows 10 devices from shutting down or hibernating if System Guard Secure Launch is enabled.
The complete list of fixes is below:
- [Fonts] This update includes changes to Chinese fonts to meet GB18030-2022A compliance.
- [OS Security (known issue)] Fixed: After installing the Windows security update released on or after January 13, 2026, some Secure Launch-capable PCs with Virtual Secure Mode (VSM) enabled are unable to shut down or enter hibernation. Instead, the device restarts.
- [Folders] Fixed: This update fixes an issue that affects folder renaming with desktop.ini files in File Explorer. The LocalizedResourceName setting was ignored, so custom folder names did not show. Now, custom folder names appear as expected.
- [Graphics] Fixed: A stability issue affecting certain graphics processing units (GPUs) configurations.
- [Secure Boot] With this update, Windows quality updates include a broad set of targeting data that identifies devices and their ability to receive new Secure Boot certificates. Devices will receive the new certificates only after they show sufficient successful update signals, which helps ensure a safe and phased rollout.
Since June 2025, Microsoft has warned that multiple Windows Secure Boot certificates from 2011 are expiring in June 2026, and warned that if they are not updated, it would breach Secure Boot protections.
These certificates are used to validate Windows boot components, third-party bootloaders, and Secure Boot revocation updates, and if expired, could allow threat actors to bypass security protections.
As part of today’s update, Microsoft continues to roll out the new Secure Boot certificates to targeted systems, with updates to additional systems being installed as the targeting scope expands.
Microsoft states that there are no known issues with this update.
Tech
New Linux botnet SSHStalker uses old-school IRC for C2 comms
A newly documented Linux botnet named SSHStalker is using the IRC (Internet Relay Chat) communication protocol for command-and-control (C2) operations.
The protocol was invented in 1988, and its adoption peaked during the 1990s, becoming the main text-based instant messaging solution for group and private communication.
Technical communities still appreciate it for its implementation simplicity, interoperability, low bandwidth requirements, and no need for a GUI.
The SSHStalker botnet relies on classic IRC mechanics such as multiple C-based bots and multi-server/channel redundancy instead of modern C2 frameworks, prioritizing resilience, scale, and low cost over stealth and technical novelty.
According to researchers at threat intelligence company Flare, this approach extends to other characteristics of SSHStalker’s operation, like using noisy SSH scans, one-minute cron jobs, and a large back-catalog of 15-year old CVEs.
“What we actually found was a loud, stitched-together botnet kit that mixes old-school IRC control, compiling binaries on hosts, mass SSH compromise, and cron-based persistence. In other words scale-first operation that favors reliability over stealth,” Flare says.

Source: Flare
SSHStalker achieves initial access through automated SSH scanning and brute forcing, using a Go binary that masquerades as the popular open-source network discovery utility nmap.
Compromised hosts are then used to scan for additional SSH targets, which resembles a worm-like propagation mechanism for the botnet.
Flare found a file with results from nearly 7,000 bot scans, all from January, and focused mostly on cloud hosting providers in Oracle Cloud infrastructure.
Once SSHStalker infects a host, it downloads the GCC tool for compiling payloads on the victim device for better portability and evasion.
The first payloads are C-based IRC bots with hard-coded C2 servers and channels, which enroll the new victim in the botnet’s IRC infrastructure.
Next, the malware fetches archives named GS and bootbou, which contain bot variants for orchestration and execution sequencing.
Persistence is achieved via cron jobs that run every 60 seconds, invoking a watchdog-style update mechanism that checks whether the main bot process is running and relaunches it if it is terminated.
The botnet also contains exploits for 16 CVEs targeting Linux kernel versions from the 2009-2010 era. This is used to escalate privileges after the earlier brute-forcing step grants access to a low-privileged user.

Source: Flare
Regarding monetization, Flare noticed that the botnet performs AWS key harvesting and website scanning. It also includes cryptomining kits such as the high-performance Ethereum miner PhoenixMiner.
Distributed denial-of-service (DDoS) capabilities are also present, though the researchers noted they have not yet observed any such attacks. In fact, SSHStalker’s bots currently just connect to the C2 and then enter an idle state, suggesting testing or access hoarding for now.
Flare has not attributed SSHStalker to a particular threat group, though it noted similarities with the Outlaw/Maxlas botnet ecosystem and various Romanian indicators.
The threat intelligence company suggests placing monitoring solutions for compiler installation and execution on production servers, and alerts for IRC-style outbound connections. Cron jobs with short execution cycles from unusual paths are also big red flags.
Mitigation recommendations include disabling SSH password authentication, removing compilers from production images, enforcing egress filtering, and restricting execution from ‘/dev/shm.’
Tech
Lost Soviet Moon Lander May Have Been Found
An anonymous reader shares a report: In 1966, a beach-ball-size robot bounced across the moon. Once it rolled to a stop, its four petal-like covers opened, exposing a camera that sent back the first picture taken on the surface of another world. This was Luna 9, the Soviet lander that was the earliest spacecraft to safely touchdown on the moon. While it paved the way toward interplanetary exploration, Luna 9’s precise whereabouts have remained a mystery ever since.
That may soon change. Two research teams think they might have tracked down the long-lost remains of Luna 9. But there’s a catch: The teams do not agree on the location. “One of them is wrong,” said Anatoly Zak, a space journalist and author who runs RussianSpaceWeb.com and reported on the story last week. The dueling finds highlight a strange fact of the early moon race: The precise resting places of a number of spacecraft that crashed or landed on the moon in the run up to NASA’s Apollo missions are lost to obscurity. A newer generation of spacecraft may at last resolve these mysteries.
Luna 9 launched to the moon on Jan. 31, 1966. While a number of spacecraft had crashed into the lunar surface at that stage of the moon race, it was among the earliest to try what rocket engineers call a soft landing. Its core unit, a spherical suite of scientific instruments, was about two feet across. That size makes it difficult to spot from orbit. “Luna 9 is a very, very small vehicle,” said Mark Robinson, a geologist at the company Intuitive Machines, which has twice landed spacecraft on the moon.
Tech
Rivian R2 Prototype First Drive Reviews Point to a Quick, Capable, and Ready to Play Electric SUV

Rivian’s R2 Prototype has hit road with early reviews, and it’s a capable electric SUV that truly delivers on the adventure promise without breaking the bank like its larger siblings do. The test rides were place on California highways, curvy back roads, and rocky off-road trails near Rivian’s Irvine headquarters. Most reviewers agree that the R2 retains the adventurous spirit of the larger R1 models while also making it more fun and approachable for daily driving.
The power comes from two motors, and this all-wheel drive configuration generates a 656 horsepower and 609 pound-feet of torque. It accelerates from 0 to 60 mph in around 3.6 seconds and feels robust even at highway speeds. In normal mode, the R2 runs largely on rear-wheel drive for improved fuel efficiency, but when necessary, it uses the front motor. Sport mode engages full all-wheel drive for a faster reaction.
Sale
VOLPAM Electric Scooter Adult, 19 MPH & 15 Miles Range, 350W Motor, 8.5″ Solid Tires, Lightwight Foldable…
- ✅【 Powerful Performance】Equipped with a 350W brushless motor, this adult electric scooter reaches 19MPH and handles 15° inclines with ease…
- ✅【Long-Lasting Battery】This e scooter has 19 miles max long range on a single 4-5 hours fast charge, making it ideal for everyday commutes…
- ✅ 【Safety First】Scooter stay protected with a drum brake + EABS electronic brake system for quick, stable stopping. The bright LED headlight…
The R2 handles quite well and stays grounded in corners. The steering feels natural and connected, with plenty of feedback that increases as you request more. Body control is outstanding, especially given the high ground clearance and all-terrain tyres with tall sidewalls. Roll remains well under control, and the chassis responds quickly and without drama. The R2 feels lighter on its feet because to its unibody structure, lower weight (about 4,850 to 5,000 pounds), and lower center of gravity. When cruising, the ride is nice, but when you push harder, it tightens up. Steel coil springs and semi-active dampers handle uneven roads with ease, providing an excellent balance between pavement and dirt.
Off-road performance stands up well on the trails it was tested on, with 9.6 inches of ground clearance and angles that allow you to tackle tough terrain with confidence. The long-travel suspension articulates well, and torque vectoring maintains traction without the use of typical locking differentials. When the wheels begin to spin, the brakes come into action, but there is some initial slip before they fully intervene.
Inside, the room is surprisingly generous for a tiny SUV. Tall adults may comfortably sit into the back seats, which provide 40.4 inches of legroom and headroom. The inside remains clutter-free, featuring a large central touchscreen and a smaller driver display. The haptic steering wheel on the column controls climate, drive modes, and other settings via rolling, tilting, and pushing actions, and the feedback is satisfactory, but they are currently working on adjusting the prototypes. There is plenty of storage space, ranging from dual gloveboxes to a flat-folding rear section that can accommodate a fitted mattress for overnight use. The low beltline and upright windows provide excellent visibility.
The EPA cycle shows a range of more than 300 miles, thanks to a compact battery pack and a well-designed interior. Filling up is also much faster than you’d think, with the R2 going from 10 to 80 percent in less than half an hour at a fast charging station, and with a native NACS port, it’s virtually ready to go at any Tesla Supercharger.
The price starts about $45,000, with dual-motor variants costing $50,000 or $55,000, depending on how specced out you want to get. The truth is, that puts the R2 in a really good position in the market; it’s like a true alternative to the more mainstream electric crossovers (Tesla Model Y), but with one significant bonus: you can actually take it off the beaten track and get a real rush of performance.
Tech
North Korean hackers use new macOS malware in crypto-theft attacks
North Korean hackers are running tailored campaigns using AI-generated video and the ClickFix technique to deliver malware for macOS and Windows to targets in the cryptocurrency sector.
The threat actor’s goal is financial, as suggested by the role of the tools used in an attack on a fintech company investigated by Google’s Mandiant researchers.
During the response engagement, the researchers found seven distinct macOS malware families and attributed the attack to UNC1069, a threat group they’ve been tracking since 2018.
Infection chain
The attack had a strong social engineering component as the victim was contacted over the Telegram messaging service from a compromised account of an executive at a cryptocurrency company.
After building a rapport, the hackers shared a Calendly link that took the victim to a spoofed Zoom meeting page on the attacker’s infrastructure.
According to the target, the hackers showed a deepfake video of a CEO at another cryptocurrency company.
“Once in the ‘meeting,’ the fake video call facilitated a ruse that gave the impression to the end user that they were experiencing audio issues,” Mandiant researchers say.
Under this pretext, the attacker instructed the victim to troubleshoot the problems using commands present on a webpage. Mandiant found commands on the page for both Windows and macOS that would start the infection chain.
Huntress researchers documented a similar attack method in mid-2025 and attributed it to the BlueNoroff group, another North Korean adversary also known as Sapphire Sleet and TA44, that targeted macOS systems using a different set of payloads.
macOS malware
Mandiant researcher found evidence of AppleScript execution once the infection chain started, but could not recover the contents of the payload, followed by deploying a malicious Mach-O binary. In the next stage, the attacker executed seven distinct malware families:
- WAVESHAPER – C++ backdoor that runs as a background daemon, collects host system information, communicates with C2 over HTTP/HTTPS using curl, and downloads and executes follow-on payloads.
- HYPERCALL – Golang-based downloader that reads an RC4-encrypted configuration file, connects to C2 over WebSockets on TCP 443, downloads malicious dynamic libraries, and reflectively loads them into memory.
- HIDDENCALL – Golang-based backdoor reflectively injected by HYPERCALL that provides hands-on keyboard access, supports command execution and file operations, and deploys additional malware.
- SILENCELIFT – Minimal C/C++ backdoor that beacons host information and lock screen status to a hard-coded C2 server and can interrupt Telegram communications when executed with root privileges.
- DEEPBREATH – Swift-based data miner deployed via HIDDENCALL that bypasses macOS TCC protections by modifying the TCC database to gain broad filesystem access and steals keychain credentials, browser data, Telegram data, and Apple Notes data.
- SUGARLOADER – C++ downloader that uses an RC4-encrypted configuration to retrieve next-stage payloads and was made persistent via a manually created launch daemon.
- CHROMEPUSH – C++ browser data miner deployed by SUGARLOADER that installs as a Chromium native messaging host masquerading as a Google Docs Offline extension and collects keystrokes, credentials, cookies, and optionally screenshots.
.jpg)
Source: Mandiant
Of the malware found, SUGARLOADER has the most detections on the VirusTotal scanning platform, followed by WAVESHAPER, which is flagged by just two products. The rest are not present in the platform’s malware database.
Mandiant says that SILENCELIFT, DEEPBREATH, and CHROMEPUSH represent a new set of tooling for the threat actor.
The researchers describe as unusual the volume of malware deployed on a host against a single individual.
This confirms a targeted attack focused on collecting as much data as possible for two reasons: “cryptocurrency theft and fueling future social engineering campaigns by leveraging victim’s identity and data,” Mandiant says.
Since 2018, UNC1069 has demonstrated its ability to evolve by adopting new techniques and tools. In 2023, the bad actor switched to targets in the Web3 industry (centralized exchanges, developers, venture capital funds).
Last year, the threat actor changed its target to financial services and the cryptocurrency industry in verticals such as payments, brokerage, and wallet infrastructure.
Tech
Hubble’s Sharpest Look Yet at a Star’s Final Act in the Egg Nebula

A star almost identical to our sun is nearing the end of its life in the Cygnus constellation, about 1000 light years away. Astronomers call this spectacle the Egg Nebula, or CRL 2688 for short. Hubble’s most recent image provides a magnificent view of this particular object in unprecedented detail, thanks to the combination of new data and previously captured images. What we get is a stunning display of light cutting through the dust.
A star almost identical to our sun is nearing the end of its life in the Cygnus constellation, about 1000 light years away. Astronomers call this spectacle the Egg Nebula, or CRL 2688 for short. Hubble’s most recent image provides a magnificent view of this particular object in unprecedented detail, thanks to the combination of new data and previously captured photographs. What we get is a stunning display of light cutting through the dust.
NASA Lunar Telescope for Kids – 90x Magnification, Includes Two Eyepieces, Tabletop Tripod, and Finder…
- INSPIRE CURIOSITY – The NASA Lunar Telescope allows your child to see the moon in incredible detail; the perfect gift for girls and boys interested…
- HIGH-QUALITY OPTICAL GLASS AND FINDER SCOPE – This easy-to-use telescope comes with a finder scope, low power, and high-power eyepieces; when used…
- TABLETOP TRIPOD & SMOOTH MOUNT SYSTEM – Use the included tripod to steady your Lunar Telescope for optimal viewing, with a smooth mounting system…
The core star is hidden deep in the center, enveloped by a thick cloud of gas and dust that allows very little light to pass through. What does pass through is compressed into two narrow beams of light that sweep outward, revealing the fast-moving clouds of material being ejected from the star. Those clouds glow orange in infrared, adding some color to the image. You can also observe faster-moving clouds of heated molecular hydrogen that light brightly in the infrared, adding depth to the scene.
Over the previous 5000 years, the star has lost its outer layers in large concentric rings of gas. These rings are made up of tiny arcs of gas that accumulate every few hundred years. Now, these rings reflect the star’s light in a fashion that resembles ripples on water – and the dust produced by these outbursts is what shapes the nebula that bears its name, since the dense core is like the yolk of an egg wrapped up in darker, dustier layers.

This is only transient; it will only last a few thousand years. The star has depleted all of its hydrogen and helium fuel, and what remains of its outer layers are floating away, while the center is becoming increasingly hot. Eventually, that center will cause the surrounding gas to glow, similar to the Helix Nebula or Butterfly Nebula. As of now, the Egg Nebula is in its pre-planetary phase, a brief period before winds and radiation begin to obscure the picture.
Hubble first observed the Egg Nebula in 1997, when a picture revealed the hidden light source. In 2003, we were able to get a full picture of the ripple patterns surrounding the nebula, and in 2012, we got an even closer look at the central cloud and outflows. Today’s image combines all of that data with some new frames to provide the sharpest look yet, courtesy of the Wide Field Camera 3.
Tech
MoFi Electronics Introduces UltraPhono Pro: More Affordable High-End MM/MC Phono Preamp
Vinyl isn’t having a moment anymore, it’s entrenched. Turntables are selling, cartridges are backordered, and phono preamps are once again a real battleground instead of an afterthought. With the $2,495 UltraPhono Pro, MoFi Electronics finally fills a very obvious gap in its lineup by bringing much of the sonic DNA of its flagship MasterPhono down to a price point where the fight actually is.
Designed by Peter Madnick and built around the same ultra-low-noise circuit concepts and parts philosophy that made the MasterPhono a serious contender at the top end, the UltraPhono Pro strips away the excess and focuses on what matters: clean gain, flexibility for MM and MC cartridges, and musical accuracy.
Just as importantly, it gives MoFi a credible sub-$2,500 answer to established heavyweights like Pro-Ject, EAR, E.A.T., MOON by Simaudio, Rega, and Musical Fidelity; a segment where serious vinyl listeners actually shop. In short, this is MoFi admitting the obvious: the market below its halo gear is thriving, and now they’re properly in the game.

From MasterPhono to UltraPhono Pro: Flagship Design, Scaled for the Real World
Designed by veteran engineer Peter Madnick, the MoFi Electronics MasterPhono was MoFi’s no-holds-barred statement piece: a fully discrete, all-analog phono preamplifier built to handle literally any MM or MC cartridge on the planet, including ultra-low-impedance moving coils via its current-input architecture. Its dual-chassis design isolates the power supply from the audio circuits to keep noise vanishingly low, while cascaded voltage regulation, passive RIAA with ultra-tight tolerances, and fully balanced DC-coupled circuitry deliver reference-grade accuracy (±0.05 dB RIAA) and serious signal-to-noise performance.
Add extensive front-panel configurability, multiple gain and loading options, balanced and single-ended I/O, metering, remote control, and firmware upgradability, and the $5,000 MasterPhono firmly established MoFi as a serious electronics brand—not just a turntable and cartridge company.
The new MoFi Electronics UltraPhono Pro is a clear and deliberate trickle-down from the MasterPhono’s design philosophy. By stripping away the dual-chassis construction, extreme configurability, and flagship-level excess, MoFi delivers a far more approachable phono stage, at roughly fifty percent less cost, without abandoning the engineering fundamentals that made the MasterPhono credible in the first place. The UltraPhono Pro keeps the focus where it belongs: ultra-low noise, fully discrete analog circuitry, and cartridge flexibility that actually matters to real-world vinyl listeners.

Despite its streamlined approach, the UltraPhono Pro is anything but entry-level. It delivers up to 71 dB of gain, maintains a wide and linear frequency response from 10 Hz to 50 kHz (±0.2 dB), and achieves exceptionally accurate RIAA equalization rated at ±0.05 dB. Noise performance is equally impressive, allowing low-output moving-coil cartridges to play against a quiet, stable background with excellent dynamic contrast.
User-facing features are practical and thoughtfully implemented. Front-panel controls include mono mode, mute, a subsonic filter, and dimmable faceplate illumination. Cartridge matching is straightforward and flexible, with multiple loading options for moving-coil cartridges and selectable gain settings that support both single-ended and balanced system integration.
Key Features
- Single-input architecture derived directly from the MasterPhono
- Fully discrete J-FET and MOSFET input and output stages
- Fully balanced, DC-coupled, servo-controlled signal path from the high-pass filter to the outputs
- Passive RIAA equalization with zero global feedback
- Seven-stage regulated, fully balanced power supply
- Five selectable resistive loading options via front-panel–controlled, hermetically sealed gold-on-gold relays: 100, 300, 500, 1k, and 47k ohms
- Gain settings of 40 dB, 52 dB, and 65 dB (plus +6 dB via XLR outputs)
- Ultra-low-noise design using premium-grade electronic components
- Critical resistors specified at 0.1 percent or 0.5 percent tolerance
- Passive RIAA network using polypropylene film/foil capacitors matched to better than 1 percent
- Thermally coupled critical components to minimize performance drift with temperature changes
- Maximum signal-to-noise ratio (A-weighted): 93 dB (MM) / 85 dB (MC)

Comparison
| UltraPhono Pro | MasterPhono | |
| MSRP | $2,495 | $5,995 |
| Inputs | RCA jack, voltage mode, single-ended | Current mode & Voltage mode, balanced and unbalanced |
| 100 Ohms, 300 Ohms, 500 Ohms, 1k Ohms, 47k Ohms | 15 Ohms, 30 Ohms, 50 Ohms, 75 Ohms, 100 Ohms, 500 Ohms, 1K Ohms, 10K Ohms, 47K Ohms+150pF, & Option | |
| Gain | RCA output: 40db, 52db, 65db Balanced output: 46db, 58db, 71db |
(MM) 40db, (MM) 50db, (MC) 60db, (MC) 70db |
| Output Impedance (Balanced) | 230 Ohms | 230 Ohms |
| Output Impedance (Unbalanced) | 115 Ohms | 115 Ohms |
| Maximum Output | 1% THD, 10 Volts RMS | 1% THD, 10 Volts RMS |
| THD | 110mV@1kHz 40db gain: 0.005% A-Weighted | 5mV@1KHZ MM low: < 0.01% A-Weighted |
| 10mV@1kHz 65db gain: 0.005% A-Weighted | 5mV@1KHZ MM high: < 0.01% A-Weighted | |
| 1mV@1kHz 40db gain: 0.010% A-Weighted | 0.5mV@1KHZ MC low: < 0.005% A-Weighted | |
| 1mV@1kHz 65db gain: 0.013% A-Weighted | 0.5mV@1KHZ MC high: < 0.005% A-Weighted | |
| RIAA Accuracy | +/- 0.05db | +/- 0.1db using passive R-C technology |
| Frequency Response | 10Hz – 50kHz +/- 0.20db | 10Hz – 50KHz +/- 0.20db no filter-3.5db @ 10Hz filter ON |
| Dimensions (WxHxD) | 17” x 2.5” x 11.375” (43.2 x 6.4 x 28.8 cm) |
17” x 3” x 17” (43.2 x 7.6 x 43.2 cm) |
| Weight | 7 lbs (3.2 kg) | 15 lbs (6.8 kg) |
| AC Mains | 100VAC – 260VAC, auto switching | 100VAC – 260VAC, auto switching |
In short, the UltraPhono Pro doesn’t try to replace the MasterPhono, it translates its core engineering into a price and feature set that makes sense for the crowded, highly competitive mid-to-upper phono preamp market.

The Bottom Line
The MoFi Electronics UltraPhono Pro delivers genuine high-end phono design with fully discrete circuitry, very low noise, accurate RIAA, and balanced operation at a price where serious vinyl listeners actually shop. It’s built for MM and MC users who want flexibility and precision without paying for flagship excess, and it stands out by sounding like a scaled-down reference component rather than a dressed-up midrange box.
Where to buy: $2,495 at Music Direct | SkyFi Audio
Related Reading:
Tech
Which is better for you in 2026?
Buying a smartphone in 2026 is a far cry from where we were 10 years ago, with less obvious reasons to go with either iOS or Android – but the choice remains one of the most significant you’ll make in your digital life.
Whether you’re looking at the iPhone 17 or an Android flagship like the Samsung Galaxy S25 Ultra, the gap in hardware has shrunk massively – but the way you interact with the phones is still completely different.
Here, we explain the fundamental differences between Android and iOS in 2026 to help you decide which is better for your needs.
Market share
If you look at the global market share, Android continues to dominate the vast majority of the planet with a massive 72% using the platform. Apple’s iOS, on the other hand, only accounts for a 28% share.
Android’s dominance is driven mainly by the sheer variety of hardware available at pretty much every price point, from budget-friendly blowers in emerging markets to ultra-premium foldables. That said, while iOS’ 28% share may sound small in comparison, it’s worth noting that it dominates in more premium markets like the UK and US.
That might not sound like something you should care about, but if you want to use the same platform-exclusive features as your friends – namely things like iMessage on iOS – you’ll want to make sure you make the right choice.
Updates
A few years ago, Apple would’ve had a massive win on its hands with its iOS update system – but the Android competition has come along leaps and bounds more recently.
Apple’s approach to software updates is still the stronger of the two, with Apple dropping new versions of iOS on all supported devices on the same day, ensuring that even three- or four-year-old iPhones get the latest software updates as soon as they’re available.


Android makers have made big strides in this department, with the likes of Samsung, Google, Honor and Motorola now offering up to seven years of support, but the rollout of this software is much more fragmented. Sure, you might get the Android 17 update, but it won’t be as soon as it’s released, and it might be available for other phones from the same brand first.
Also, that’s pretty much exclusive to flagships – if you buy a mid-range or budget device from a brand like Xiaomi, you’re at the mercy of a more limited software promise. This means that while the hardware might last, the software experience can feel dated much faster on Android than on iOS.
Software experience
Using an iPhone in 2026 is a largely smooth, polished experience with impressive visuals thanks to the Liquid Glass UI introduced with iOS 26 in late 2025.
Apple’s interface feels like a premium, more curated experience where the software just kind of works in the background without much setup or intervention needed, but it also means you’re locked into Apple’s infamous walled garden – even if those towering walls are slowly beginning to crumble.


Android, on the other hand, is designed for those who want greater control and customisation in their smartphone.
From the deep customisation available on Android skins to the ability to swap out your entire home screen launcher or icon packs, Android feels more like a tool that adapts to your needs – though that does depend on the Android you’re using, as different Android skins offer different levels of visual customisation.
You do need more time and patience, especially if you get into the nitty-gritty of Android customisation, but it’s usually a well-rewarded task.
Apps
The days of “iOS gets it first” are largely over for major releases, but the App Store still feels like the more polished storefront of the two – though with Apple set to introduce more ads to the App Store experience, that could soon change.
That said, Apple’s strict app vetting process and the limited number of screen sizes to accommodate generally result in higher-quality UI and better optimisation. After all, it’s much easier for devs to polish an app for five iPhones than for five hundred different models of Android.


Google Play offers more freedom than Apple’s App Store, offering powerful system-level utilities, retro game emulators and niche productivity tools that Apple simply wouldn’t allow on its platform. It also features most, if not all, major apps available on iOS, though there are still a few iOS exclusives floating around – especially when it comes to big-screen tablet apps.
For the average user, the difference between the two storefronts is negligible, but for power users who want to use their phone as a genuine pocket computer with super-niche apps, Android remains the better choice.
Security and privacy
It feels like Apple’s entire brand is centred around privacy, and in 2026, features like App Tracking Transparency and Advanced Data Protection remain industry-leading.
That’s because Apple produces not only the hardware but every aspect of the software experience, from custom silicon to encrypted iCloud backups, it can offer a level of security that’s difficult to replicate on the Android side of things.


Android’s security model is more fragmented in comparison. While Google has hardened the OS significantly and introduced a range of privacy-focused features, the “open” nature of Android places greater responsibility on users to avoid installing dodgy apps from random websites that might contain malware.
Specific manufacturers offer deeper security features, with the likes of Samsung’s Knox and Motorola’s ThinkShield for Mobile featuring more robust features to protect your data from hacks, but it’s not consistent among all Android manufacturers.
AI
Artificial Intelligence, or AI, is the buzzword in the smartphone world at the moment, with pretty much every smartphone manufacturer seemingly cramming as many AI features into their smartphones as possible. That said, the playing field is far from level.
While Apple has marketed Apple Intelligence as a seamless, integrated experience, it’s clear that Android brands have a massive lead in both capability and accuracy. The fact that Apple is using Gemini to power its long-awaited (and delayed) redesigned Siri experience should speak volumes to this fact.


Whether it’s the more sophisticated generative object removal offered by Samsung’s Galaxy AI or the multi-modal on-device processing from the latest Pixel phones, Android AI tools generally feel more robust and less prone to hallucinations that still plague Apple’s efforts.
That said, much of this power is becoming platform-agnostic. Many popular AI tools, including the full suite of Gemini features, are available as apps on both iOS and Android, meaning you aren’t necessarily locked out of top-tier AI just because you chose an iPhone.
The real difference is how companies handle the data; Apple continues to lean heavily on Private Cloud Compute to handle cloud-based AI processing of sensitive data, while Google and other Android manufacturers offer a mix of on-device and cloud-based processing depending on what it needs.
Verdict
When it comes to the all-important decision of choosing between iOS and Android, there’s no wrong choice, only a choice of priorities.
If you want an easy-to-use phone with a wide variety of high-quality apps that works well with other Apple gear and offers the most polished experience, the iPhone remains the best pick.
That said, if you want the best AI tech and the freedom to make your phone look and act exactly how you want, with niche system-level apps and extensive customisation, Android is your best bet.
Tech
Unison Research Unico PRE v2 & DM v2 Power Amplifier: More Muscle, Sharper Design, and Zero Doubt It’s Italian
With the Unico DM v2, Unison Research makes it clear that evolution, not reinvention is the mission. Now positioned as the flagship power amplifier in the Unico lineup, the DM v2 arrives with a completely renewed, unmistakably Italian design that’s cleaner, more modern, and aligned with the brand’s new visual language—formally introduced alongside the Unico PRE v2. This isn’t a styling exercise for Instagram; it’s a cohesive rethink of how Unico components look, feel, and slot into a contemporary high-end system.
That design confidence isn’t coming out of nowhere. We’ve already spent serious time with Unison Research’s Triode 25 and Simply 845 integrated amplifiers, and both left a lasting impression. Price-sensitive shoppers need not apply, but for listeners who care more about musicality than spreadsheets, they remain two of the most compelling tube amplifiers in their class, combining drop-dead Italian industrial design with a command of tone, texture, and scale that many modern tube amps still struggle to get right. The Unico DM v2 builds on that legacy, just with more power, sharper tailoring, and zero interest in playing it safe.
There’s a clear design pivot happening here. Unison Research has long been celebrated for mixing real hardwoods with machined metal gear that looked handcrafted, tactile, and proudly old-world Italian. The Unico PRE v2 and DM v2don’t abandon that heritage, but they definitely reinterpret it. The lines are cleaner, the surfaces more restrained, and the overall presentation feels less romantic throwback and more contemporary confidence. Think less classic Sophia Loren, more modern Nicole Grimaudo; still unmistakably Italian, still elegant, just sharper, leaner, and very much living in the present rather than trading purely on nostalgia.
Unison Research Unico DM v2: Flagship Power Amplifier with a New Design Direction

The $10,999 USD Unico DM v2 is the new flagship power amplifier in Unison Research’s Unico series. Introduced alongside the Unico PRE v2, it reflects a clear shift in the company’s design language toward a more modern, restrained aesthetic while maintaining the hybrid tube/solid-state approach that has long defined the Unico line.
The chassis design is notably more contemporary than previous Unico models. The front panel is machined from a 15-mm-thick aluminum block, giving the amplifier a dense, solid feel, while the Midnight Black and Velvet Gold finishes emphasize its cleaner lines. A 2-mm aluminum top cover wraps around the enclosure, reinforcing both structural rigidity and visual continuity. Wooden accents remain, but they are used sparingly, serving as a reference to the original Unico logo rather than a dominant visual element. The Unison Research logo also functions as the power switch, integrating branding and operation in a subtle, functional way.
Internally, the Unico DM v2 is built around a dual-mono architecture. Each channel is powered by its own 750 VA encapsulated toroidal transformer, with potting and shielding used to reduce electromagnetic interference. This layout is intended to preserve channel separation and maintain consistency under load. When operated in bridged mono mode, the two power supplies are connected in parallel, increasing available current and output capability.

The amplifier uses a three-stage hybrid amplification circuit. The input stage operates in pure Class A and employs ECC82 / 12AU7 Gold Lion valves, providing the initial voltage gain. A solid-state intermediate stage buffers and adapts the signal for the output section. The power stage uses a complementary push-pull configuration with three parallel pairs of MOSFETs, designed to deliver sufficient current for demanding loudspeaker loads while remaining stable across a wide impedance range.
A key technical addition is A.S.H.A. (Class A-AB) technology, introduced for the first time in the Unico DM v2. This output-stage topology is designed to combine aspects of Class A operation at lower levels with the efficiency and thermal behavior of Class AB at higher power. According to Unison Research, this approach keeps distortion low and consistent up to maximum output while maintaining tonal balance and low-frequency control even at moderate listening levels.
In practical terms, the Unico DM v2 delivers 220 W into 8 ohms and 340 W into 4 ohms in stereo operation, with stability down to 2 ohms. In bridged mono mode, it provides 650 W continuous output into both 8-ohm and 4-ohm loudspeakers, allowing it to function as a high-power monoblock when required.
Connectivity is conventional and system-focused, with balanced XLR and unbalanced RCA inputs, a remote power-on trigger, and dual binding posts per channel to support bi-wiring. The Unico DM v2 is clearly aimed at listeners who want high output capability, a hybrid circuit design, and a more contemporary visual presentation from Unison Research, without departing from the brand’s established engineering principles.
Unison Research Unico PRE v2: Flagship Preamplifier with Expanded Functionality and a Modernized Look

The $7,499 USD Unico PRE v2 is the new flagship preamplifier in the Unison Research Unico series. Introduced alongside the Unico DM v2 power amplifier, it reflects the same shift toward a more contemporary design language while retaining the hybrid valve/solid-state approach that defines the Unico range. Rather than a cosmetic refresh, the PRE v2 represents a full redesign intended to improve usability, system flexibility, and overall consistency with modern audio systems.
Visually, the Unico PRE v2 follows the same restrained, more architectural styling as the DM v2. The front panel is machined from a 15-mm-thick solid aluminum block, giving the unit a solid, precisely finished appearance. Midnight Black and Velvet Gold finishes highlight the cleaner lines and tighter detailing, while the 2-mm aluminum top coverwraps around the chassis to reinforce both rigidity and visual continuity. Wooden accents remain, but in a reduced, more symbolic role, referencing the original Unico logo rather than dominating the design. As with the DM v2, the Unison Research logo doubles as the power button, integrating branding and function in a straightforward way.
Volume control is handled by a high-quality integrated circuit using precision resistors, chosen to ensure accurate channel balance and consistent attenuation across the full range. The goal here is stability and repeatability rather than novelty, preserving signal integrity regardless of listening level.
Internally, the Unico PRE v2 has been completely reworked. The circuit remains faithful to zero global feedback and a dual-mono topology, design choices Unison Research has long associated with natural, unforced sound. The preamplifier uses a three-stage architecture, with the first stage built around a pair of ECC83 / 12AX7 Gold Lion valves operating in Class A. This stage establishes the preamp’s basic tonal character while maintaining low noise and low distortion. The following solid-state stages handle buffering and output duties, working in tandem with the valve section to maintain consistency and drive capability under a wide range of system conditions.

One of the Unico PRE v2’s strengths is its unusually broad connectivity. On the analog side, it offers three RCA line inputs, three XLR line inputs, a dedicated MM/MC phono input, and an additional Line In for system integration. Outputs include two RCA outputs for bi-amping, a balanced XLR output, an unfiltered dual subwoofer output, and Line Out connections for external processors or recording devices. A 12 V trigger output allows synchronized power control with compatible amplifiers and accessories.
Digital playback is handled by an integrated DAC based on the Sabre ES9018K2M converter. The DAC section uses a balanced output architecture designed to interface cleanly with the valve input stage, aiming for tonal consistency between digital and analog sources. Digital inputs include USB-B, two S/PDIF, and two optical Toslink connections, supporting PCM up to 384 kHz over USB, native DSD up to 256×, and DoP up to 128×, with S/PDIF and Toslink supporting resolutions up to 192 kHz.
The built-in phono stage uses passive RIAA equalization and high-precision components. It supports both MM and MC cartridges, with selectable load and gain settings accessible from the rear panel, making cartridge matching straightforward without internal adjustments.
In practical terms, the Unico PRE v2 is a fully balanced hybrid preamplifier with a solid-state output stage, moderate power consumption, and output voltage levels high enough to drive a wide range of power amplifiers without difficulty. It measures 45 × 43 × 14 cm and weighs 11 kg, placing it firmly in the full-size component category.
Overall, the Unico PRE v2 is less about spectacle and more about refinement—modernized styling, expanded connectivity, and a carefully updated circuit design intended to serve as a flexible control center for contemporary hybrid and high-power systems.

The Bottom Line
The Unico DM v2 separates itself with a high-power, dual-mono hybrid architecture and Unison Research’s new A.S.H.A. Class A-AB output stage, designed for real loudspeaker control rather than headline specs. The Unico PRE v2complements it as a fully balanced control center with a tube-based input stage, broad analog and digital connectivity, and a genuinely useful MM/MC phono stage with selectable load and gain. There’s no internal streamer and no Bluetooth, which feels deliberate—hinting that dedicated digital sources may not be far behind.
At $18,498 USD for the preamp and power amplifier alone—before speakers, sources, and cabling—this is a serious investment. Fidelity Imports represents a wide range of appropriately priced loudspeakers that would make sense with this combination. The takeaway is simple: new look, significantly more power, and pricing that reflects Unison Research’s move further upmarket.
For more information: unisonresearch.com/type/unico/
Related Reading:
-
Tech7 days agoWikipedia volunteers spent years cataloging AI tells. Now there’s a plugin to avoid them.
-
Politics2 days agoWhy Israel is blocking foreign journalists from entering
-
NewsBeat1 day agoMia Brookes misses out on Winter Olympics medal in snowboard big air
-
Sports4 days agoJD Vance booed as Team USA enters Winter Olympics opening ceremony
-
Tech4 days agoFirst multi-coronavirus vaccine enters human testing, built on UW Medicine technology
-
Business2 days agoLLP registrations cross 10,000 mark for first time in Jan
-
NewsBeat2 days agoWinter Olympics 2026: Team GB’s Mia Brookes through to snowboard big air final, and curling pair beat Italy
-
Sports2 days agoBenjamin Karl strips clothes celebrating snowboard gold medal at Olympics
-
Sports3 days ago
Former Viking Enters Hall of Fame
-
Politics2 days agoThe Health Dangers Of Browning Your Food
-
Sports5 days ago
New and Huge Defender Enter Vikings’ Mock Draft Orbit
-
Business3 days agoJulius Baer CEO calls for Swiss public register of rogue bankers to protect reputation
-
NewsBeat4 days agoSavannah Guthrie’s mother’s blood was found on porch of home, police confirm as search enters sixth day: Live
-
Business5 days agoQuiz enters administration for third time
-
Crypto World7 hours agoBlockchain.com wins UK registration nearly four years after abandoning FCA process
-
Crypto World16 hours agoU.S. BTC ETFs register back-to-back inflows for first time in a month
-
NewsBeat2 days agoResidents say city high street with ‘boarded up’ shops ‘could be better’
-
Sports1 day ago
Kirk Cousins Officially Enters the Vikings’ Offseason Puzzle
-
Crypto World16 hours agoEthereum Enters Capitulation Zone as MVRV Turns Negative: Bottom Near?
-
NewsBeat6 days agoStill time to enter Bolton News’ Best Hairdresser 2026 competition





