News
- Windows 11's August update reportedly introduced an SSD breaking bug
- It's been suggested that SSDs using Phison controllers are more likely to be affected
- Phison has conducted extensive testing, but couldn't reproduce the bug a single time, adding that its customers haven't reported the glitch either
The bug that recently emerged in Windows 11, which is reportedly breaking some SSDs, is being investigated by Microsoft and its partners - and now we've heard back from one of the parties involved.
This is Phison, which manufactures SSD controllers used across a wide range of drives from various manufacturers, and is involved in this controversy because some reports suggest that SSDs using these controllers were more likely to be affected by the bug.
Phison has now shared the results from its extensive testing pertaining to this matter, as Neowin reports, issuing the following statement: "Phison dedicated over 4,500 cumulative testing hours to the drives reported as potentially impacted and conducted more than 2,200 test cycles. We were unable to reproduce the reported issue, and no partners or customers have reported that the issue affected their drives at this time."
So, Phison feels it's in the clear, what with a whole lot of testing having turned up nothing, and no reports coming to the company directly from its customers, either. Of course, reports from individual consumers are going to go directly to the SSD maker (not those responsible for the controller), but when Phison says "partners or customers," it is talking about those drive manufacturers (and others, too, no doubt Microsoft included).
What hasn't helped Phison's cause here is a fake document that did the rounds online just after the bug came to light in Windows 11's August update. This purported to contain a list of affected Phison controllers, but was completely fabricated as the company quickly made clear.
(Image credit: Shutterstock)Analysis: Microsoft's findings are still to comeAlthough Phison has conducted extensive testing, this can't be regarded as a definitive conclusion. Microsoft's investigation into this SSD breaking bug in Windows 11 is still being carried out, and until we see the result of that, there remains doubt as to exactly what's going on here.
Reports of SSD failures still remain scattered. So it must be noted, that this seems to be a rare issue. At any rate, I'm hoping Microsoft will make its findings known sooner rather than later, and clear this matter up - as it's only becoming more confusing with this latest instalment of the saga.
Phison also tacked on some advice with its statement on best practices to "support high-performance storage devices" undergoing extended workloads, such as shifting large files - like prolonged write operations which apparently triggered the Windows 11 bug. Phison observes that a "proper heatsink or thermal pad" will help in terms of maintaining optimal temperatures and ensuring the drive doesn't get too hot (or throttles as a result).
Note that imparting this advice isn't directly related to the bug - meaning Phison isn't saying you should be using a heatsink to avoid coming off the rails with this Windows 11 glitch. This is just general advice aimed at all high-end SSD owners, letting them know that if they are running intense workloads over long durations, using extra cooling is advised.
Mind you, if your SSD doesn't have a heatsink already, adding one is a somewhat fiddly affair, especially for the less tech-savvy (although they are less likely to be running a high-performance solid-state drive, admittedly).
You might also like- Nvidia Jetson Thor is framed as the robot brain for physical AI
- Major robotics companies are already listed as early adopters worldwide
- Nvidia Jetson Thor offers 2,070 FP4 teraflops within a 130-watt power envelope
Nvidia has released the Jetson AGX Thor developer kit, calling it the next step toward robotics systems which can function in real time.
The system, built on the Blackwell GPU line, is framed as a platform for “physical AI” and advanced robotic functions across manufacturing, logistics, healthcare, farming, retail, and transport.
Nvidia says it can deliver up to 7.5 times more AI compute and over three times the energy efficiency of its Jetson Orin line, which has been in wide use since 2022.
Offers supercomputer-level capacityNvidia went on to describe Jetson Thor as “the ultimate supercomputer to drive the age of physical AI and general robotics.”
“We’ve built Jetson Thor for the millions of developers working on robotic systems that interact with and increasingly shape the physical world,” said Jensen Huang, founder and CEO of Nvidia.
“With unmatched performance and energy efficiency, and the ability to run multiple generative AI models at the edge, Jetson Thor is the ultimate supercomputer to drive the age of physical AI and general robotics.”
With a quoted figure of 2,070 FP4 teraflops in a 130-watt envelope, it is positioned as powerful enough to run multiple generative models at once.
It supports vision-language-action models like Isaac GR00T N1.5, along with other LLM systems.
The device also integrates 128GB of memory, which is expected to make it capable of handling larger AI workflows at the edge.
Several robotics players are already listed as early adopters, including Agility Robotics, Amazon Robotics, Boston Dynamics, Caterpillar, Hexagon, and Medtronic.
Meta has also been named as an early partner, while companies such as John Deere, OpenAI, and Physical Intelligence are said to be testing the system.
“Nvidia Jetson Thor offers the computational horsepower and energy efficiency necessary to develop and scale the next generation of AI-powered robots that can operate safely and effectively in dynamic, real-world environments, transforming how we move and manage goods globally,” said Tye Brady, chief technologist at Amazon Robotics.
Nvidia notes more than two million developers already use its robotics stack, with over 7,000 customers having deployed Jetson Orin hardware in edge AI projects.
Jetson Thor runs on the Nvidia Jetson software platform, which is designed to support multiple AI tools at once.
The package integrates with Nvidia Isaac for simulation, Metropolis for vision AI, and Holoscan for real-time sensor processing.
This arrangement is intended to allow one system-on-module to support many AI writer models and workflows, rather than requiring several separate chips.
The developer kit is available now at $3,499 and the production systems, including carrier boards, will be distributed worldwide through its partners.
You might also like- These are the best mobile workstations you can buy right now
- We've also listed the best mini PCs for every budget
- Acemagic's new Ryzen AI 9 mini PC is a better deal than Geekom's HX370 model
- Novodisq claims 230PB rack capacity using proprietary 144TB SSDs
- Novoblade integrates compute, networking, and storage in dense blade servers
- Novodisq promises 95% lower power compared with conventional arrays
At the recent Flash Memory Summit, a new name from New Zealand surfaced in a bid to cause waves in the enterprise storage space.
Novodisq presented its Novoblade system, a platform built to combine dense storage, compute acceleration, and network capacity in a compact design.
The Novoblade modules are designed as blade servers, each offering 576TB of raw storage built on flash drives. The drives themselves are based on E2 form factor SSD units with capacities reaching 144TB per device.
How Novoblade is structuredThe company says a 2U enclosure can hold up to 20 modules, which equates to 11.75PB of capacity in a single shelf.
Scaling this configuration across an entire 42U rack, Novodisq projects that storage can rise to 230PB.
Alongside the storage figures, Novodisq promotes Novoblade as a hyperconverged design that integrates compute resources directly into each blade.
These include ARM64 cores, FPGA resources, and optional AI or machine learning engines, with networking supported by 200Gbps or 400Gbps Ethernet.
The company positions this as a platform that can replace conventional NAS arrays, with up to 95% lower energy consumption. Such claims, however, are difficult to validate without detailed independent benchmarks.
While the theoretical capacity appears high, the price of such a system raises serious questions.
The company has not announced official figures, but estimates can be made from existing hardware, as a single 122.88TB SSD currently (August 2025) costs close to $14,000.
Using that as a reference, and accounting for Novoblade’s proprietary 144TB SSDs, a single blade with four drives could already exceed $60,000 before considering added compute and networking.
With 20 blades in a 2U enclosure, the total could approach $1.2 million. Extending this to a full 42U rack with 230PB of raw storage means costs would rise well beyond $2 million.
This positions Novoblade as an extremely dense solution, but one that only highly specialized organizations could justify financially.
On paper, these numbers suggest one of the densest deployments yet described, but practical use and performance remain untested.
Novodisq describes the Novoblade as both a storage server and a converged compute platform.
It can expose block, file, and object interfaces, or integrate into distributed systems such as Ceph or Lustre.
At the moment, major players in the storage field continue to focus on balancing capacity with performance.
Therefore, it remains uncertain whether Novodisq can provide not only the largest or fastest SSD arrangements but also sustainable pricing and support.
You might also like- These are the best mobile workstations you can buy right now
- We've also listed the best mini PCs for every budget
- AWS just snapped up a load of custom Xeon chips for extra cloud power
- Security agencies issue joint statement warning Chinese tech firms may be indirectly collaborating with Salt Typhoon
- Salt Typhoon is a hacking group behind multiple high-profile attacks
- Group is thought to have serious links to Chinese government
A new joint cybersecurity advisory from the National Security Agency (NSA) and other agencies like CISA, the UK’s NCSC, Canada’s CSIS, Japan’s NPA and many more looks ti expose advanced persistent threat (APT) actors believed to be sponsored by the Chinese Government.
According to the advisory, Chinese firms have been providing products and services to China’s Ministry of State Security and the military - which in turn, it is claimed, props up hacking groups.
These threat actors target infrastructure like telecommunications, government, military, transport, and energy agencies - specifically in a global hacking campaign linked to the notorious Salt Typhoon group.
Supplying components“The data stolen through this activity against foreign telecommunications and Internet service providers (ISPs), as well as intrusions in the lodging and transportation sectors, ultimately can provide Chinese intelligence services with the capability to identify and track their targets’ communications and movements around the world," the advisory warns.
Some of the firms named in the advisory, like Sichuan Juxinhe Network Technology Co. Ltd, have already been sanctioned for their ties to the group.
Other named companies include Beijing Huanyu Tianqiong Information Technology Co., Ltd., and Sichuan Zhixin Ruijie Network Technology Co., Ltd, all of which are thought to be linked.
The report also outlines specific threat hunting guidance and mitigations against these groups, particularly in quickly patching devices, monitoring for unauthorized activity, and tightening device configuration.
Earlier in 2025, Salt Typhoon was discovered carrying out a cyber espionage campaign that breached multiple communications firms, with hackers lingering inside US company networks for months.
The group was observed abusing vulnerabilities in Microsoft Exchange Servers, which allowed them to breach networks and exfiltrate data. A fix for this flaw has been available for years, but research suggests that nearly 91% of the 30,000 affected instances remain un-patched - highlighting the importance of deploying effective patch management software.
China has always strenuously denied any ties to this group, and to any other cyber-espionage campaigns.
You might also like- Take a look at our picks for the malware removal software around
- Check out our choice for best antivirus software
- Chinese nationals will no longer be allowed to manage Pentagon cloud services
- NJ prisoners get only twenty 1.44MB floppy disks, barely enough for appeals
- Lawyers must transfer files from flash drives back to floppy disks, complicating the process
- Authorities say the ban on flash drives is a matter of security
A prisoner at New Jersey State Prison has publically voiced frustration at being forced to rely on floppy disks for critical legal work.
The US state's prison system restricts inmates to using floppy disks, each with a maximum capacity of 1.44MB, but each prisoner is allowed 20 floppy disks, a limit which barely matches the needs of complex legal correspondence.
Writing for the Prison Journalism Project, Jorge Luis Alvarado said, “Inside New Jersey State Prison, it’s like 1985, where we rely on out-of-date word processors, electric typewriters, and floppy disks that are going extinct in the free world.”
Outdated tools in modern timesAlvarado explains even a single legal brief can exceed this size, requiring the use of multiple disks to store one document.
Such a process becomes cumbersome, and with the added risk of corruption, the format introduces real uncertainty into how files are preserved.
In addition, since major companies like Sony stopped manufacturing floppies about 15 years ago, their scarcity only adds to the impracticality of the rule.
The reliance on floppy media seems especially arbitrary, given that they have only about a year of lifespan left and that flash drives became widely adopted more than two decades ago.
In the early 2000s, USB drives quickly eclipsed floppies, offering both speed and durability.
Today, they are inexpensive, compact, and reliable, with capacity far surpassing anything the floppy era could provide.
Even consumer SSD options now span into the terabyte range, with the largest SSD models rivaling enterprise storage.
Devices once labeled the fastest SSD can manage transfers that dwarf anything possible with legacy media.
However, authorities argue that the ban on flash drives is a matter of security, suggesting they could be misused within prison environments.
While this position explains the reluctance to modernize, it leaves prisoners disadvantaged when dealing with legal matters where technology should serve as a bridge, not a barrier.
Alvarado describes a process where lawyers must copy digital files onto flash drives, only to have them transferred back to floppy disks through a single library computer.
Delays are inevitable, with access often taking days at a time.
Some researchers estimate that between four and six percent of those incarcerated in the United States may be innocent.
Therefore, even if a fraction of these individuals face barriers to appeals due to outdated technology, the issue extends far beyond mere inconvenience.
Via Toms Hardware
You might also like- Google is building a small nuclear reactor in Tennessee to power its data centers
- These are the best mobile workstations you can buy right now
- We've also listed the best mini PCs for every budget