RCS is a more advanced replacement for standard SMS text messages. The features are more akin to Apple’s iMessage, but it works on Android phones. If there were any other viable smartphone platforms, they could also implement RCS, which is an open standard. RCS offers longer messages, read receipts, typing indicators, high-resolution media, Wi-Fi messaging, and more. It’s a major upgrade over SMS, but it’s still hard to know if you’re going to get RCS/Chat functionality when you ping one of your contacts.
AT&T’s agreement to shiponlyGoogle Messages will increase the likelihood that your friends and family will be on a compatible RCS platform. The Messages app will offer to enable Chat automatically if it detects you are on a supported carrier, and making that the only messaging app preloaded on AT&T phones guarantees most of its customers will use it. Yes, this agreement even includes Samsung phones, which have long shipped with the Samsung Messages app. Samsung is one of the other big players in RCS, but it let carriers lead the way on RCS. That caused some annoying incompatibilities. Google is stepping in to eliminate that with its uncompromising support of the RCS Universal Profile.
An example of RCS messages.
Keep in mind, we’re not talking about Google Chat, which is another product altogether. Yes, the Chat branding for RCS might not have been the best idea, but it’s what we’re stuck with for the time being. Despite the confusing name, Google appears to be winning the RCS wars.
Google got tired of waiting around for carriers in late 2019 and began rolling out RCS to users in the Messages app. In late 2019, the major US carriers announced an RCS project that would have deployed a custom app and cross-platform messaging service. Luckily for all of us, this effort collapsed a few months later. You might not be crazy about running more of your data through Google, but I believe the system will work better with Google at the helm rather than cellular carriers.
We don’t know when exactly AT&T will switch to Messages on all its phones, but it should be soon. That will leave Verizon as the odd man out, but the pressure will increase on Big Red to adopt the Google default app now.
Intel announced that its next-generation 10nm Xeon CPU microarchitecture, codenamed Sapphire Rapids, will not enter production in 2021. The company now plans to be in production by Q1 2022 and to start its volume ramp during Q2 2022. This suggests general availability towards the end of Q2 or in early Q3 of next year.
How much of a delay this constitutes somewhat depends on how you count. There’s no arguing that Sapphire Rapids has slipped at least a bit; Intel’s original guidance suggested Ice Lake Xeon would ship in 2020, followed by Sapphire Rapids in 2021. Intel was still committed to a 2021 ramp earlier this year, but the company announced at Computex that it would ramp volume on its next-generation 10nm CPU in the first half of 2022. Previous guidance, however, had suggested Sapphire Rapids would enter production (and be delivered for the Aurora supercomputer) in Q4 2021.
The reason this is a rather fuzzy delay is that Intel had previously committed to a volume ramp in H1 2022. When a company says it’ll launch a part in the first or second half of the year, that typically means we’ll see hardware in either Q2 or Q4. If you read Intel’s declaration that it would ship Sapphire Rapids in H1 to mean “Q2,” then this isn’t a delay — it’s an affirmation of the fact that Intel will ramp the chip in that time period as opposed to delaying it into Q3.
There have been rumors that Sapphire Rapids wouldn’t launch until mid-2022 going back over a year, so Intel’s declaration that the CPU will launch in that time period isn’t a delay relative to what the rumor mill was milling back in 2020.
New Technical Details
Intel discusses some features of Sapphire Rapids in its new blog post after the company revealed some additional data earlier this week. The new CPU will be the first processor to implement support for Advanced Matrix eXtensions (AMX). Intel describes AMX as central to the future of Intel DL Boost, which is the company’s overarching term for various features that boost AI performance. Intel has announced AMX before, but today it gave some performance numbers. The company claims it has achieved “over two times the deep learning inference and training performance compared with our current Xeon Scalable generation.”
Intel also announced that Sapphire Rapids will be the first Intel chip to feature a DSA (Data Streaming Architecture) acceleration engine. Intel describes a DSA as “a high-performance engine targeted for optimizing streaming data movement and transformation operations common in high-performance storage, networking and data processing-intensive applications.” The DSA is intended to handle various types of data movement so that the CPU can spend more time dedicated to number-crunching. We don’t know much about how Intel will implement DSA for Sapphire Rapids or what the improvements will be, but streamlining and minimizing data movement are both critical to improving CPU efficiency and reducing power draw.
Finally, we now know that Sapphire Rapids will indeed feature up to 64GB of HBM2. These chip variants will launch after the main Sapphire Rapids family and can treat the HBM2 pool as a large cache or directly address it as memory. It’s thought that Sapphire Rapids CPUs with HBM2 may have lower core counts than variants without it, but Intel hasn’t disclosed the specific details of its HBM2 implementation yet.
According to Intel, the new Sapphire Rapids timeline will allow for further validation and testing in partnership with various OEMs. It’s also possible that the ongoing semiconductor shortage, or the associated shortage in Ajinomoto Build-up Film, led Intel to delay the launch. Maximizing manufacturing yield means accurately forecasting demand, and the entire semiconductor industry is in unprecedented waters right now.
One tidbit Intel confirmed at the International Supercomputing Conference this week: It’s now shipping DG2 GPUs to developers as part of its slow ramp into the GPU market. Up until now, Intel has confirmed itself to shipping integrated Xe-based graphics and a small, low-power GPU dubbed DG1. Intel positioned DG1 more as an AI accelerator than a dGPU, so the company’s expectations around the part were undoubtedly modest.
DG2 is different. This is the first iteration of Intel’s gaming-focused Xe HPG line. It’ll also be Intel’s first effort to enter the GPU market in decades. Larrabee had noble intentions, but the product died before coming to market and was repurposed as a test platform for what would eventually become Xeon Phi. You have to go all the way back to the i740 to find Intel’s last GPU launch, and it didn’t go particularly well.
The i740 was designed to emphasize the capabilities of AGP Texture acceleration at a time when CPUs were already memory bandwidth bound. Diverting additional bandwidth to allow the GPU to transfer texture data from DRAM via the AGP bus was faster than using PCI, but not as fast as putting VRAM directly on the GPU and storing data there. Intel zigged on graphics technology at a time when the rest of the market zagged, and years of poor IGP performance did nothing to help matters.
Intel’s onboard GPUs stagnated from 2015-2018, but Intel took notable steps forward with Ice Lake and then again with Tiger Lake. Interest in DG2 would be high no matter what — we haven’t had had three-way competition in graphics since the PowerVR-based Kyro II — but the ongoing GPU shortage and the prospect of cards that might actually be affordable have generated interest.
Rumors suggest desktop gamers may not see a launch this year, unfortunately. While Intel is telling fans that we are “soon heading toward a milestone moment, the pending release of the Xe HPG microarchitecture from Intel,” RedGamingTech (via NotebookCheck) claims Intel will target mobile platforms first with a DG2 variant intended to compete with RTX 3060 Ti or RTX 3070. The desktop variant of the card is expected to target higher-end cards at the top of its range.
DG2 will have the advantage or burden of launching in the strangest GPU market we can remember. If crypto demand drops and stays low, AMD and Nvidia GPUs may be available at something like historic values by the time the card launches. If they don’t, it’ll benefit from a scenario in which its two competitors are selling for far higher prices. Either way, it’s going to be interesting to see how Intel decides to price its upcoming cards and what kind of performance and power efficiency it can deliver as it returns to the market.
Samsung is currently expected to announce its upcoming Exynos SOC with an AMD-designed GPU at some point next month. Benchmarks that have leaked out ahead of the launch suggest that the company’s new mobile GPU could be much faster than anything the Korean manufacturer has previously fielded.
This leak is from Twitter user Ice Universe, who claims Samsung has been testing the GPU paired with a Cortex-A77 CPU and that this result is from one such device. If this is true, it means the chip in question isn’t technically an Exynos 2200. 3DMark applications are typically designed to isolate GPU performance, so the CPU shift shouldn’t make too much of a difference. These results suggest the chip AMD built for Samsung offers much better performance than anything else shipping in an Android device today.
Exclusive: Samsung is testing the AMD GPU in the new Exynos, and the Wild Life test on the 3D Mark has scored 8134 points. It is worth mentioning that the CPU in this test is A77 architecture, for reference only. Each test has different results, we need to wait for more results. pic.twitter.com/Rej9vqyP6s
The tweet reports a score of 8134 for the new Samsung GPU. WCCFTech reports that this is substantially faster than the Exynos 2100 (5295) or Snapdragon 888 (5382). If retail chips score this well, the upcoming Exynos 2200 could outperform even the iPhone 12 from Apple. The various flavors of iPhone 12 score between 7400 and 7600, though even the 2020 iPhone SE is capable of a 6983.
These performance figures are excellent. If they hold up on shipping devices and prove indicative of overall performance, it’ll mean Samsung nailed its goal of delivering a substantial GPU performance uplift by investing in a custom GPU architecture. The only caveat to consider along with your standard ration of salt is the fact that we know nothing about power consumption, whether this represents sustained performance over time, and how representative Wild Life performance is of the GPU architecture as a whole.
Mobile device performance can vary substantially depending on the cooling solution being used to test the SoC. An early test chip paired with a Cortex-A77 could easily underperform the final product. It’s also possible that these tests are being run on parts exposed to open air rather than stuffed inside a chassis, or that the OEM test platform is allowed to draw far more power than a shipping device will. Sustained performance or shipping performance could be different from what we see here today.
On the whole, however, these figures look good. A good showing in this space would be a boost for both Samsung and AMD. Samsung would match or surpass the iPhone family, at least until Apple’s next-generation products. AMD would have a win under its belt in the semi-custom, low-power GPU space. A successful launch next month could be a prelude to future embedded chips or maybe even mobile SoCs with a lower-power, higher-efficiency GPU core than anything AMD ships today.
AMD has been close-mouthed about the GPU’s microarchitecture, referring to it as “custom graphics IP,” but also as RDNA2. It could represent the first iteration of RDNA3 or it might be RDNA2-based, with custom IP work related to lowering power consumption or improving power efficiency.
AMD, to be clear, has not said anything about future products that might utilize this IP and we don’t have reason to suspect any near-term announcements, but both AMD and Intel are always on the lookout for ways to reduce power consumption. It’s not hard to imagine AMD developing a new low-power graphics core in partnership with Samsung while making long-term plans to take advantage of the work as a way to improve APU efficiency in the embedded or mobile markets.
This illustration depicts a planet partially hidden in the glare of its host star and a nearby companion star. After examining a number of binary stars, astronomers have concluded that Earth-sized planets in many two-star systems might be going unnoticed by transit searches, which look for changes in the light from a star when a planet passes in front of it. The light from the second star makes it more difficult to detect the changes in the host star’s light when the planet passes in front of it.
Earth is unique in our solar system. It’s one of just a few celestial objects with an atmosphere, and it’s alone in supporting life… as far as we know. How many more Earths are out there? It could be more than we thought, according to some new research. NASA astronomers now say we might be missing a lot of Earth-like exoplanets in binary solar systems. That could boost the chances of finding habitable worlds out there.
We, of course, live on a planet that orbits a single star, but systems of multiple stars are very common. Astronomers estimate that at least a third of solar systems in the Milky Way consist of two or more stars. It may be closer to half, depending on who you ask. If these stars are close enough together and sufficiently far away, they can appear as a single point of light in telescopes. That, it turns out, can be a problem for planet-hunting.
We know of more than 4,000 confirmed exoplanets, most of which are on the large side. That’s a consequence of how we detect them. Imaging exoplanets directly is beyond our technological abilities in most instances, so scientists rely on features such as gravity and light variation to identify these alien worlds. The transit method, which is utilized by NASA Transiting Exoplanet Survey Satellite (TESS), is the most effective. This involves watching for dips in light as planets transit in front of their host star. That’s also how Kepler spotted thousands of potential exoplanets.
The team from NASA’s Ames Research Center used the Gemini North and South telescopes in Chile and Hawaii to observe nearby TESS targets to see if any of them were actually binary systems. They discovered 73 of those stars were actually binary pairs. They also used the WIYN 3.5-meter telescope to look over 18 additional known binary systems scanned by TESS. The researchers looked at the populations of exoplanets identified in these solar systems and discovered something interesting: TESS found both large and small Earth-like exoplanets orbiting single stars, but it only spotted large gas giant planets in binary systems.
Our next-door neighbor Centauri is perhaps one of the most well-known multi-star systems. Alpha Centauri AB is on the left, Beta Centauri on the right, and Proxima Centauri is at the center of the red circle.
The implication of this research is that Earth-like exoplanets could be hiding in binary systems, and our methods of detection simply aren’t good enough to see them. According to study leader Katie Lester, small planets are liable to get lost in the glare of a binary system, making it impossible to accurately track their transits.
This work complicates things for astronomy. As long as we rely on the transit method to catalog exoplanets, researchers will have to pay special attention to whether a system is binary. If it is, we could be getting an incomplete picture of the planetary population. That could mean a lot more Earth-like planets out there. Perhaps somewhere in the vastness of space, there’s someone looking up at twin stars wondering if life could ever evolve with just one sun.
The Samsung Galaxy S21 5G is Samsung’s current flagship smartphone with performance that rivals the best phones money can buy. Today you can get the 128GB model of this phone with a $100 discount that drops the price to just $699.99.
Samsung Galaxy S21 5G 128GB Unlocked Smartphone for $699.99 from Amazon (List price $799.99)
Samsung’s Galaxy S21 5G smartphone features top-tier performance that rivals the best phones currently on the market. It comes equipped with a trio of cameras including one that can record video in 8K, and it also has a 6.2-inch Dynamic AMOLED display as well as a long lasting 4,000mAh battery. This phone sells regularly for $799.99 but you can buy one today from Amazon for just $699.99.
Dell’s G3 laptops are designed to be affordable gaming computers, but they still pack a considerable punch. This laptop utilizes a Core i5 processor along with a mid-range GTX 1660 Ti graphics processor that’s able to run games at 1080p resolutions with ease. Currently, you can get one marked down from $1,218.99 to$799.99 from Dell.
The Roku Express 4K+ is one of Roku’s newest streaming devices that’s capable of streaming 4K content. It also has support for HDR and has an easy to use interface that makes finding what you want to watch quick and easy. Currently you can buy one from Amazon marked down from $39.99 to just $29.99.
Note:Terms and conditions apply. See the relevant retail sites for more information.For more great deals, go to our partners at TechBargains.com.
You may have heard several days ago that owners of certain Western Digital My Book external hard drives were hit with a remote exploit that deleted all their data. Alternatively, you may be the unlucky owner of a My Book Live and are still in mourning over the loss of your precious files. In either case, it looks like the cause of the reformat hack was not the 2018 vulnerability but was instead a zero-day exploit caused by sloppy development. However, this does not clear WD of wrongdoing. If anything, it’s even worse.
Last week, many owners of My Book Live hard drives awoke to find their devices had been reset. Unlike most external drives, the My Book Live doesn’t have a USB port. It’s intended to connect to your local network via an Ethernet cable so it can be accessed from all your other devices. However, it defaults to being available online at all times, and WD stopped supporting the My Book Live several years ago.
It’s true that if WD had not abandoned the My Book Live lineup, it might have spotted the problem before the hack. However, the initial supposition that the hack stemmed entirely from an unpatched 2018 flaw has been proven wrong. Ars Technica and security researcher Derek Abdine now say the mass hack comes from an unreported flaw in WD’s drive software. The software included an authentication check whenever the embedded reset command was triggered. However, for unknown reasons, it was disabled in the shipping software. All the attacker needed to know to blank the drives was how to format the XML request. The code, seen below, would have blocked the reformat, but the double slash at the beginning of each line indicates it was “commented out.”
So, that’s all pretty weird, but it gets even weirder. These drives are indeed vulnerable to CVE-2018-18472, the 2018 exploit Western Digital initially fingered as the cause. It claims that in at least some of the known hacks, the attackers used CVE-2018-18472 to gain access and then triggered the zero-day to format the drive. The 2018 flawshouldhave given the attacker root access, so it’s unclear why they also used the zero-day. Several hacked drives have been found to have malware designed for the drive’s PowerPC hardware. This makes the drives part of the Linux.Ngioweb botnet.
Dan Goodin from Ars has a theory about this, and it’s one with which I agree. Goodin speculates that the botnet installation and reset were carried out by different attackers. Perhaps the data deletion attack was an attempt by a rival to blow up their enemy’s botnet. It’s just a shame that regular users lost all their data by being caught in the middle. Regardless, Western Digital really screwed up by letting a device withtwoserious vulnerabilities sit in people’s homes all this time.
Cybersecurity is exponentially increasing in importance and a role in protecting information technology systems from attack by hackers and other negative parties on the internet is both rewarding and an in-demand career path. While the internet and the technology networks we rely on for business and everyday life are always becoming more sophisticated and complex, so too are the tools of those people who would take advantage of them to steal data, compromise information, or interfere with daily operations. The people who can prevent this are valuable employees in a booming sector.
Whether you’re new to cybersecurity or already working in the space, it’s worthwhile to continually update your information and learning to be in line with the latest developments. Infosec4TC is able to help you achieve new heights with your career, and platinum memberships offering lifetime access are now on sale for just $69, a sharp discount of 93 percent off the full purchase price of $999.
A platinum membership to Infosec4TC gives you unlimited access to a range of online, self-paced certification courses covering everything to do with cybersecurity. The available courses cover a huge variety of topics, from ethical hacking and GSWC to CISSP and other internationally recognized IT certifications in order to keep you at the top of your cybersecurity game. With a high passing rate, Infosec4TC can help you reach your new career goals – whether that’s a shift of focus or progression within cybersecurity.
Infosec4TC has a high rating of 4.4 out of five stars on TrustPilot, and this platinum membership grants you access to all current courses as well as those that are uploaded in the future. This includes access to the latest exam questions to allow you to practice for exams before taking the real thing.
When Intel launches Alder Lake later this year, it’ll also transition from the LGA1200 socket introduced with Comet Lake to a new socket, LGA1700. LGA1700 is a new socket with significantly more pins and it’s not the same shape as Intel’s previous LGA115x / LGA1200 sockets.
Igor’s Lab has published some details on the new socket, with drawings and diagrams that appear to be genuine. They show a slightly larger motherboard plate (78mm x 78mm, compared with 75mm x 75mm for current LGA1200), but the shape of the CPU itself is now decidedly rectangular.
Image by Igor’s Lab
The new CPU’s different shape could cause issues for anyone trying to reuse a CPU cooler. It’s entirely possible — even likely — that LGA1700 CPUs will have cores and hot spots in different physical places on the chip. Some heatsinks may not cover the entire CPU core effectively. This could also apply to AIO watercoolers. Those with smaller blocks may not cool the chip effectively, and the microchannels in the heatsink may not be in ideal places. Igor thinks Xeon and Threadripper coolers might work for this new setup because those sockets are more rectangular. It depends on what kind of conversion kits become available and which manufacturers support them.
Image by Igor’s Lab
Another wrinkle in the situation is z-height. LGA1200 supported z-heights of 7.312mm to 8.249mm. LGA1700 is thinner, with a z-height of 6.529 to 7.532mm. It’s not clear if the reduced Z-height is a consequence of Intel’s die shaving, which improves heat dissipation, or if the company was able to reduce socket height through other changes. Either way, the two standards only overlap at the very bottom and very top of their respective ranges.
Some conversion kits might be able to deal with this by changing the height of the base the cooler sits on, but this will depend on what kind of CPU cooler you have. Igor notes that heatsinks with heatpipes intended to directly contact the CPU could fare particularly poorly if Alder Lake’s hot spots are in different locations than Rocket Lake and previous CPUs.
We’re not preemptively declaring that no conversion kit will work. Companies such as Noctua have built a reputation for excellent hardware, and I actually don’t believe the company would release a conversion kit for a product if it couldn’t provide an acceptable experience. Don’t be surprised, however, if it turns out that a lot of high-end coolers for standard AMD and Intel CPUs today (as opposed to Threadripper and Xeon) can’t be easily adapted for Intel’s new platform. In this case, there may be good reason to opt for a new CPU, if you plan to upgrade when Alder Lake comes out.
Google and Samsung announced a few weeks that they were teaming up to unify their wearable efforts. Details on the new Samsung-infused Wear OS are still hard to come by, but Samsung gave us our first glimpse in its Mobile World Congress video stream. The updated Wear will have Samsung’s One UI Watch skin, along with more smartphone integrations. There are even a few classic Wear OS features that are coming back.
Before Samsung teamed up with Google, it was making watches with its Tizen software. It says those devices will still get their guaranteed three years of updates, but it’s all about Wear OS going forward. While Samsung didn’t show off its new smartwatch, it did have a quick demo of several features in the new Wear.
The next Galaxy Watch will have a shrunken-down version of One UI, which has been the interface on Samsung’s phones for the last several years. That means the icons, settings, and color palette will be very similar to Samsung’s phones, at least in the case of the Galaxy Watch. Google has yet to clarify if all watchmakers will have the option to skin Wear OS now. That could make watches more varied and interesting, assuming it doesn’t wreck performance. You can see the discussion of One UI Wear in the video below at about 11 minutes.
Samsung also talked about new (and not so new) integrations with Galaxy phones. Instead of using Samsung’s Galaxy Apps store, the Galaxy Watch will rely on the Play Store. If you install an app on your phone that has a Wear version, the watch will automatically install it. This is something Wear OS used to do way back when it was called Android Wear. I always appreciated this feature, so it’s nice to see it coming back. The Play Store already has a respectable selection of watch faces, but there might be a lot more soon. Samsung says it’s going to release streamlined watch face tools this summer, allowing designers to easily release new styles.Finally, the new One UI will feature Samsung’s health monitoring tech, which was one of the primary advantages cited for the Google-Samsung combo. Wear OS has traditionally had awful health tracking, but Samsung is among the best.
Again, we don’t know if this is all exclusive to the One UI Wear skin. Google hasn’t had much to say about the new Wear, and Samsung’s unannounced 2021 Galaxy Watch will be the debut device. We might have to wait for that wearable to launch before we’ll know what to expect.
Another day, another round of Windows 11 updates, clarifications, and additional confusions. Microsoft has issued a new blog post on which systems are and are not compatible with Windows 11. It explains some of the company’s specific thinking on security in more detail, but it also confuses the CPU situation further.
Specifically, Microsoft writes:
[We] are confident that devices running on Intel 8th generation processors and AMD Zen 2 as well as Qualcomm 7 and 8 Series will meet our principles around security and reliability and minimum system requirements for Windows 11. As we release to Windows Insiders and partner with our OEMs, we will test to identify devices running on Intel 7th generation and AMD Zen 1 that may meet our principles. (Emphasis added)
At this point, Microsoft is talking out both sides of its mouth. It’s obvious that the company decided to launch its OS before it had finished deciding what the upgrade and fresh install requirements would be. Two days ago, a Microsoft spokesperson confirmed that only 13 out of the 25 total Surface devices the company has launched would be upgradeable to Windows 11. There’s a specific cutoff between 7th Gen chips (which are not supported) and 8th Gen chips (which are).
The new blog post suggests that some 7th Gen CPUs and first-generation Ryzen CPUs will be supported, but the authors seem to have confused “Zen 2” and “second-generation Ryzen.” Second-generation Ryzen CPUs belong to the 2xxx product series and are built on the Zen+ core. There are Zen+ CPUs currently listed in Microsoft’s database of compatible AMD CPUs, so for now we’re going to assume that this blog post is in error and that the Windows Team meant Zen+.
One of the major points of confusion in all of this has been why a CPU or system fails when the PC Health Check app reports that a system or CPU cannot run Windows 11. Microsoft updated the app on Friday but has now acknowledged sufficiently huge problems with it to justify pulling it altogether. There’s a third-party variant that appears far more detailed, but the author acknowledges he’s just writing it based on what’s known — and what’s known about this situation is now so complex, we can’t even say if first-generation Ryzen or 7th Generation CPUs are on the upgrade list. The app currently can’t be trusted to tell you for certain which products are supported because Redmond may not have finished making those decisions yet.
According to Microsoft: “To meet the principle, all Windows 11 supported CPUs have an embedded TPM, support secure boot, and support VBS [Virtualization-Based Security] and specific VBS capabilities.” We know some systems were built without a TPM option the end-user can actually enable or disable, but support for Secure Boot has been mandatory since Windows 8. Some of the features it requires have been shipped in AMD and Intel CPUs since ~2009.
There are also low-level questions of driver and UEFI compatibility. Drivers for Windows 11 must be Hypervisor Code Integrity (HVCI)-compatible drivers, and not all existing drivers are compatible with the standard. Microsoft’s documentation notes that UEFI developers must implement UEFI v2.6 Memory Attribution Tables (MAT).
Right now, it isn’t even 100 percent clear if this is a CPU problem or a chipset problem. Some Microsoft executives have referred to lockouts as chipset-based:
The good news in all of this is that some Kaby Lake and Ryzen 1xxx owners may be able to upgrade after all. The bad news is that it’s obviously going to be some period of time before we have straight answers on the topic. Right now, different people at Microsoft are saying enough different things to make following this story confusing.
Microsoft can still fix this situation. The company needs to identify which technologies, specifically, render a CPU or system ineligible for upgrade. If some 7th Gen or Zen-based systems can still be eligible for Windows 11, it needs to provide a list of which features or capabilities need to be enabled, and probably an explanation for how to enable them or directions on where to find that information.
Second, the company needs to start being more transparent about its own internal testing process. This single sentence: “As we release to Windows Insiders and partner with our OEMs, we will test to identify devices running on Intel 7th generation and AMD Zen 1 that may meet our principles,” along with a mention regarding support on Cascade Lake, Epyc, Xeon, and Threadripper, published last Thursday, would have dealt with an awful lot of end-user concerns before they had a chance to become concerns. The reason everyone ran out to buy a TPM 2.0 module is that people thought they were required.
There’s nothing wrong with Microsoft saying it’s still testing specific CPU families and there’s no harm in handing people a list of still-to-be-tested components. But don’t just give people a laundry list of features that systems running Windows 11 need to support when some of those standards have been shipping for over a decade. Every 7th Gen CPU appears to support VT-x with EPT, for example.
To put this differently: If most machines dating back to 2016 can run Windows 11, and Windows 10 is supported through 2025, the outcry from users will most likely be limited. A Sandy Bridge rig built in 2011 will be 14 years old in 2025. Outside of gaming, we’d expect the benefits of an upgrade to be quite robust at that point. It also matters if the lockouts mostly hit DIY users or if the OEM channel is also affected.
If the cutoff line isn’t even three full years ago (Coffee Lake shipped in H2 2018), then it plays very differently. Part of the problem here is that Microsoft hasn’t tried to seriously raise system specs since Windows Vista, and that didn’t go particularly well. It’s been nearly 14 years since anyone tried to sell PC users on a major system upgrade as part of buying a new OS. People are typically grudgingly willing to tolerate upgrade requirements if the new OS requires more horsepower and makes good use of it. Fewer people are willing to throw out perfectly good systems for amorphous security features, even when those features are beneficial.
If Microsoft had led by saying that it was still evaluating a whole list of specific hardware and that updates would be forthcoming, this would not have blown up into the confusing mess it has. The best way to solve the situation is to be transparent about which specific requirements block Windows 11 installation in the majority of cases and to either provide information on addressing those concerns or declare that systems in such configurations will not be supported. An update on whether it’s possible to install and use the OS in such an “unsupported” mode would also be very helpful.
And in Other Windows 11 News
Windows 11 Insider Preview builds are now available today. Microsoft notes: “We’ve set the bar for previewing in our Windows Insider Program to match the minimum system requirements for Windows 11, with the exception for TPM 2.0 and CPU family/model.” The company promises to take provided feedback into account if it changes its minimum system requirements in the future.
If you’re looking to upgrade to Windows 11 this year, you may be out of luck. Our colleagues at PCMag discuss a tweet from the Windows account indicating that while the OS will launch this year, it won’t be immediately available as an upgrade to Windows 10. Those upgrades will begin in 2022, through the first half of the year. Microsoft has previously made OS versions available to OEMs before putting them on general sale, so this is not unprecedented.
Finally, the current smart money is pointing towards an October launch for the new OS. That’s based on multiple screenshots of the OS showing the date 10/20, often set to 11:11 AM. There’s also a tweet visible in one shot from Steve Bathiche saying he “Can’t wait for October,” shown above.
So, the new updated guidance from Microsoft is this: Its minimum system requirements may still change. Certain 7th Gen and Ryzen 1xxx systems may be allowed to upgrade if they fulfill certain Microsoft principles. This has not yet been decided. We do not know if there will be a viable upgrade path forward for DIY builders or if the 7th Gen/Zen systems Microsoft has in mind were only sold by OEMs. We’ll continue to assume that Surface devices not otherwise named by Microsoft are ineligible for upgrade based on Microsoft’s statements to the press, but this could change as well.
There’s been no word on allowing any systems older than 7th Gen/Ryzen 1xxx to upgrade to Windows 11, even if minimum system requirements are relaxed. We also do not yet know if Microsoft’s lockouts and CPU requirements can be easily bypassed or if the company will throw up more fundamental roadblocks, like refusing to deliver Windows Updates if you have the wrong make/model of CPU or motherboard. This would not be unprecedented, though there were workarounds when this happened with Windows 7 and 8.1.
We’ll have to wait and see just how many older systems can qualify for Windows 11 under Microsoft’s principles. Hopefully, we’ll eventually be able to tell you which specific issue might block updates or installation on certain PCs and what can be done to ameliorate or avoid the problem.
Today you can get a highly versatile laptop from Dell with over $800 marked off the retail price. This system is perfect for work, but it also has a 100 percent sRGB compatible display for editing images and a GPU that’s powerful enough to keep the average gamer happy.
Dell Vostro 15 7500 Intel Core i7-10750H 15.6-Inch 1080p Laptop w/ Nvidia GeForce GTX 1650 Ti GPU, 8GB DDR4 RAM and 256GB NVMe SSD for $979.00 from Dell (List price $1,784.29)
The new Vostro 15 7500 laptop is a true jack-of-all-trades. Dell’s Vostro systems are oriented as business solutions, and this system is no different, but it also has fairly strong gaming capabilities. Its 100 percent sRGB display is also well suited for editing images. No matter what you need a laptop for, this system should fit the bill. Currently, you can get this system from Dell with a hefty discount that drops the price from $1,784.29 to just $979.00.
Apple’s newest iPad is significantly faster than the older 7th Gen model. The new 8th Gen model comes equipped with the company’s A12 Bionic SoC that first appeared inside of the iPhone XS, and this chip far outstrips the A10 SoC in the older 7th Gen tablets. The 8th Gen tablet also has a slightly higher resolution screen and is in general an all-around enhanced version of its predecessor. Amazon is currently selling these new tablets marked down from $329.99 to just $299.00.
This compact desktop PC comes equipped with an Intel Core i5-10400 processor that gives it solid performance. In general it would work well as an office PC or home desktop. The system has a 256GB NVMe SSD as it’s main storage device that allows it to boot quickly. It also has a DVD-ROM drive and it has four USB ports on the front to make connecting devices easy. Currently you can get this desktop from Dell marked down from $998.57 to just $549.00.
Eufy designed this slim Robovac with a vacuum capable of 1,300Pa of suction. This gives it the power it needs to help keep your home clean, and it can also last for up to 100 minutes on a single charge. Right now you can get one from Amazon marked down from $229.99 to just $139.99.
Note:Terms and conditions apply. See the relevant retail sites for more information.For more great deals, go to our partners at TechBargains.com.
As part of its overall Windows 11 push, Microsoft is partnering with Amazon to make its app store available for Windows devices. It’s a smart move for Microsoft. Windows x86 users will be able to buy Amazon mobile apps and run them via Intel Bridge technology. ARM and AMD-based systems will still be able to take advantage of the feature, and we assume (though this has not been confirmed) that Windows on ARM users who download apps from Amazon through the Microsoft Store will receive native binaries.
This is an interesting move for Microsoft with some clear analogs in mobile. ChromeOS can run Android apps and new M1-capable Macs can run iOS applications. According to the head of Windows and devices Panos Panay, the inclusion of Amazon is hopefully just the beginning. The Verge reports him as explicitly hoping other major storefronts, like the Epic Game Store and Steam, agree to be part of the Microsoft Store as well.
“Windows already in many ways hosts those stores, and if we can host it through the Microsoft Store then of course,” says Windows and device chief Panos Panay, in an interview with The Verge. “For sure, it means as others want to come to the Store, they’re very welcome. As a matter of fact, encouraged, and that’s kind of why we’re building out some of these policies.”
From Microsoft’s perspective, this makes perfect sense. The company has been trying to build an ecosystem around its own storefront for nearly a decade at this point. Services like xCloud and Game Pass are intended to draw people into the combined Xbox/PC gaming ecosystem. Microsoft has made a lot of effort to integrate its gaming businesses, and Panay isn’t wrong when he says that Windows effectively hosts these storefronts already. It does. The same arguments for the security convenience of the App Store can be made for the Microsoft Store, and Microsoft has expanded the types of apps that can be sold there. Win32 apps, for example, are now included.
But would this be a good deal for Steam and Epic the way it might be for Amazon? The evidence there is decidedly mixed. When Amazon signs up to sell Android apps through the Microsoft Store, it expands its customer base. That’s a positive for Amazon. What do Steam and the Epic Game Store get in exchange for signing up to sell games through the Microsoft Store?
Gabe Newell famously launched SteamOS because he saw the Microsoft Store as a threat. The principal reason that threat never came to be is that gamers didn’t switch their buying habits. But selling Steam games (or Epic Store Games) through the Microsoft Store would seem to be a losing bet for both Epic and Valve. Neither company is hurting for visibility or the ability to attract new customers.
The Verge notes that having the ability to download all of your applications in one place would be a simpler setup method than visiting various websites to download installers. This is true. Reinstalling every necessary app for a new PC build is not an overwhelming chore, but it can take 2-4 hours to hunt down, download, and install every application, driver, and Windows Update, especially if you have to download a tranche of games. If you have a lot of games to download it could easily take 6-12 hours in total, depending on the games and the speed of your internet connection.
But while this might be simpler for users, it’s probably not in the best interest of Valve or Epic. It’s one thing for Amazon to work with Microsoft in the hopes of propping up its own app store, though I’ll note my colleague Sascha Segan at PCMag is dubious that this will work. It’s something else entirely for Steam or EGS to allow the Microsoft Store to function as a wrapper for their own content. Why jump through whatever additional hoop is required to buy from Steam through the Microsoft Store when you could buy from the Microsoft Store directly?
As an aside: If you want a one-stop application for downloading and handling the installation of multiple applications, I’d recommend the NiNite package installer. It’s a free-to-use batch installer that will download and install a fairly long list of applications (you customize your own list). It’s not quite the same thing as a universal Microsoft Store with Epic and Valve under the same roof, but it can simplify the process of getting software loaded on a new system.
Having a virtual private network (VPN) is just common sense these days. As the internet becomes more sophisticated and more and more nefarious interests are actively searching for ways to create data breaches, using a VPN is a simple way to protect yourself, your family, your data, and your identity while online, allowing for hassle-free browsing and a range of other benefits that will all combine to help keep you safe online.
Right now, you can get a two-year subscription to the Private Internet Access VPN for just $69.95, a great discount of 73 percent off the full purchase price of $268 – and you’ll also get a bonus $15 dollars of store credit to put towards your next purchase.
The Private Access VPN allows you a whole host of benefits while connecting to the internet. By having this VPN enabled, you’ll be able to protect your identity, your files, and your login credentials, as well as mask your location. The new MACE feature allows you to block ads, trackers, and malware, and you will also be able to encrypt your data with the cryptographically secure Blowfish CBC algorithm. An advanced firewall blocks unwanted connections, and you’ll also get live chat support all day, every day, should you need it. Even further, the VPN allows you to access censored content and bypass geographically blocked websites, apps, and services, so you can access them from anywhere in the world. The VPN service has a rating of 4.5 stars out of five on the iOS store and 4.5 out of five on the Google Play store from previous users.
The aging Hubble Space Telescope is still in safe mode after an apparent hardware failure on June 13th. NASA initially hoped it would be a quick changeover to backup hardware, but the swapping in replacement hardware has not solved the problem. The team is still working on a solution, but it’s going to take a lot more troubleshooting to track down the glitch.
Hubble was designed with redundancy in mind, but many of its parts are getting on in years. While service missions following the original launch have upgraded its hardware, many of the core components are still based on decades-old technology. For example, the payload computer that appeared to be the original culprit is a Standard Spacecraft Computer-1 (NSSC-1) from the 1980s.
NASA has a full backup for the payload computer located on the Science Instrument and Command and Data Handling (SI C&DH) unit. Late last week, the ground team attempted to switch on the backup computer for the first time. Previously, NASA confirmed switching to one of the three backup memory modules failed to fix the memory access error. However, the backup payload computer is experiencing the same issues.
Anyone who has ever found themselves troubleshooting a computer problem will have at least a hint of what the Hubble team is going through. NASA believes it’s unlikely that all these hardware components are malfunctioning at the same time, so there must be an issue elsewhere that is affecting all of them. This has sent NASA hunting through the spacecraft’s other components in search of the root cause.
STS 31 Discovery unloading the Hubble Space Telescope
One of the chief suspects is the Command Unit/Science Data Formatter (CU/SDF), which is another module on the SI C&DH unit. Its job is to format commands going to the instruments and the data that comes back. The team is also inspecting the power regulator to see if it’s feeding incorrect voltage to the computers. Luckily, there are backups for both of these components on the spacecraft. NASA just needs to track down the cause.
Hubble has been in space for 31 years, and during that time it has expanded our understanding of the universe in ways we never expected. Hopefully, NASA can get the iconic observatory working again, but its successor is almost ready to take over. The long-delayed James Webb Space Telescope is supposed to launch late this year, but NASA recently had to postpone the Halloween launch. A new date hasn’t been set.
Scientists have announced that a fossil discovered and hidden away 85 years ago may represent humanity’s closest known relative on our ancestral family tree. The skull, dubbed “Dragon Man” for the Dragon River near where it was discovered, is thought to be more closely related to humans than Neanderthals, who are typically considered our closest known relative.
According to scientists, this skull (shown above) displays a mixture of features common to Neanderthal and Homo sapiens. If I’m being honest, it just looks like a Neanderthal skull to me. The prominent brow ridge is an easy distinction between Neanderthals and ourselves, but I’m not much good at subtle differences in bone morphology.
I had an idea for how to make it easier to see the distinction. Here’s an identical picture of the same skull with a red line drawn down the image. The red line is perfectly straight. It is positioned directly next to the farthest edge of the brow ridge. I’ve combined the image from the new paper with a previous image comparing Homo sapiens (left) and Neanderthals (right) and drawn the same line in the same way down the photos of all three skulls.
Bottom image published in Cell by Qiang Ji, Wensheng Wu, Yannan Ji, Qiang Li, Xijun Ni.
The line descending from the brow ridge on the Neanderthal skull crosses the lower jaw with seven pixels of material to the left. The Dragon Man brow ridge line crosses with 10 pixels to the left. Our imaginary brow ridge line splits the human skull with 21 pixels on the right. I don’t know what the actual measurement differences are in millimeters, but we don’t need them to show you the difference. This skull is a midpoint between Neanderthal/Denisovans and ourselves.
It differs from all the other named Homo species by presenting a combination of features, such as long and low cranial vault, a wide and low face, large and almost square orbits, gently curved but massively developed supraorbital torus, flat and low cheekbones with a shallow canine fossa, and a shallow palate with thick alveolar bone supporting very large molars.
The proposed name for the new species is Homo longi. The skull has a fascinating history. A worker in the city of Harbin found the skull in 1933. It’s thought he recognized he had an important scientific discovery, but the authors of the papers discussing it speculate that he may have been ashamed at working for the Japanese. The man who found the fossil hid it in a well for 85 years before telling his family about it shortly before his death, in 2018.
Pro Tip: If you have a hidden family treasure, do not wait 85 years to tell your family.
In recent decades, we’ve learned more about a race of humans called the Denisovans. We have only a handful of Denisovan bones — Neanderthal remains are much more plentiful — but we’ve learned enough to determine that they contributed to the genome of Micronesia and that their existence overlapped with Neanderthals. We have direct evidence of Neanderthal-Denisovan interbreeding, so we know these two groups of humans had contact with each other.
The researchers argue that Homo longi fits into the Denisovan lineage. The Harbin skull is said to be quite similar to an earlier fossil, Dali Man, as well as to the Jinniushan woman.
If Homo longi is established as being part of Dali Man’s family tree, it would typically lead to the species being named as Homo dali instead of Homo longi. Our efforts to precisely determine the relationships between archaic human fossils is limited by the difficulty of recovering DNA, the paucity of the fossil record, and the fact that ancient groups of Homo interbred multiple times.
In this case, there’s a puzzle to be solved between the genetic relationship between Denisovans and Neanderthals versus Denisovans and humans. This fossil has morphological similarities to our own species, but Denisovans and Neanderthals were more closely related to each other than to us.
One thing we do know is that we don’t yet have the full story of humanity’s emergence as a species. There’s an unknown hominin ancestor who contributed to Denisovan DNA but split off from modern humans over a million years ago. A paper published last year found evidence for a ‘ghost’ hominin contribution — a species we haven’t yet found — in African genomes. Sub-Saharan Africans are also known to have the lowest percentage of Neanderthal DNA, while Europeans have the most. It’s hard to solve a puzzle when you don’t have all the pieces. Discoveries like the Harbin skull should help us see the picture a bit more clearly.
Feature image by Qiang Ji, Wensheng Wu, Yannan Ji, Qiang Li, Xijun Ni.
Microsoft’s Windows 11 unveil has quickly devolved into a confused mess regarding which CPUs will and won’t be supported under the new operating system. While I touched on this Friday, new information from both Microsoft and end-users has shed a bit of light on a confusing topic.
Here’s what we know so far:
TPM 2.0 Support Is Not the Only Variable in Play
Simply having TPM 2.0 support in your system does not appear to be sufficient to install Windows 11. Microsoft has confirmed to PCWorld that machines like the Surface Studio 2 ($3,499 and up) — which isn’t quite three years old and supports TPM 2.0 according to Microsoft’s own spec sheets — will not support Windows 11. This machine is ineligible to upgrade. You can still buy a Surface Studio 2 from Microsoft today.
Microsoft executives, the company’s list of supported CPUs, and various statements to other press outlets all consistently claim a few things:
1). TPM 2.0 support is required: There is no longer any mention being made of a “soft floor” for TPM 1.2. We do not know if such a floor is still under internal discussion or if Microsoft intends to stick to its guns on this one. Older systems that supported TPM 1.2 can sometimes upgrade to TPM 2.0 if your motherboard vendor released a UEFI update, but this will not help you if your chipset or CPU is unsupported. This brings us to our next point:
2). Limited official compatibility with older devices. Microsoft isn’t just limiting installation based on the presence or absence of TPM 2.0. The Surface Studio 2 is TPM 2.0-enabled, but uses a Core i7-7920HQ processor based on Kaby Lake. According to Microsoft’s support documents, no Intel CPU earlier than 8th Gen will support Windows 11. No first-generation Ryzen or earlier CPU is listed as compatible with Windows 11.
According to Microsoft executives, these restrictions are being enforced at the chipset level.
As of this writing, we’re assuming there must be a CPU component as well. Intel may not have offered a platform upgrade between 7th and 8th Gen, but AMD’s Ryzen 2000 CPUs worked in 300-series boards and the Ryzen 1x family similarly functioned in 400-series boards. One possibility — and this is supposition on our part — is that differences in how AMD and Intel rolled out features might mean that a 300-series motherboard + Ryzen 2xxx is still Windows 11-compatible if TPM 2.0 is available and active, while for Intel, chipset and CPU guidance are linked. It is also possible Microsoft is blocking either CPUs or chipsets of a given generation and checks for both.
Microsoft may make additional disclosures on this topic to clarify it at a later date.
3). Windows Insider Preview Builds are not locked down (but will be): Right now, Microsoft is allowing Windows 11 to install to systems that will not qualify to run the final version. The company has made this explicit in some recent documentation updates.
End users who were already enrolled in the Windows 11 Insider program can continue to test builds on their PCs, but they will not be eligible for Release Candidate preview testing. This has injected additional confusion into this discussion. Right now, the only two groups of people running Windows 11 are folks who are part of the Windows Insider campaign and folks who downloaded a leaked build of the operating system. The leaked build seems to require TPM 2.0 (we couldn’t install it to any system that lacked support), but since it’s an early version of the OS, it may not lock out all the devices that Microsoft apparently intends to restrict.
4). A special edition of Windows 11 will exist that does not require TPM. This version is intended only for markets where Western encryption is not used and does not require a TPM 2.0 module. Everyone else, however, is expected to have one.
Not much ambiguity about that. Later in the document, Microsoft states:
A UEFI firmware option to turn off the TPM is not required. Upon approval from Microsoft, OEM systems for special purpose commercial systems, custom order, and customer systems with a custom image are not required to ship with a TPM support enabled.
So customers in certain markets will be able to buy a system without TPM 2.0, but for everyone else, the feature is required. Hat-tip to THG for finding this bit.
5). Skylake-X may work: We haven’t heard back from Intel on our question regarding Core i9-9980XE support, but Brad Sams on Twitter has a 7900X and reports the CPU scans as Windows 11 compatible:
My workstation is a 7900x (7th gen) and is not on Microsoft's Intel list for supported chips but the PC Health Check app says it will run Windows 11. pic.twitter.com/RQMAGDH2CT
This is not an official confirmation from Intel or Microsoft, but it may be a sign that a few more chips could be added to Microsoft’s list.
6). The new Windows PC Health Check differentiates between “No TPM 2.0” and “Your CPU is unsupported.” We covered this on Friday, but we want to include it again based on how much confusion there was over whether we had re-described the TPM 2.0 problem.
This is what the PC Health Check app displays if you run it against an eligible machine without TPM 2.0 enabled:
This is what the PC Health Check displays if you run it against 6th Generation Broadwell-E, 6th Generation Skylake, or Kaby Lake CPU:
The fact that the application distinguishes between these two states implies it’s detecting two different things and that Microsoft has done this deliberately. The previous version of the app literally only told you that you couldn’t update to Windows 11 while giving no explanation as to why.
7). Many people we spoke to think Microsoft will change these requirements. We’ve talked to a number of folks we know in various types of jobs in the PC ecosystem, including corporate IT staff, PC OEMs, and hardware manufacturers. The general opinion is that these are very restrictive requirements and that Microsoft is likely to loosen them.
If I’m being honest, I’m not so sure about that. I think there’s room for Microsoft to officially announce support for platforms like Skylake-X and Threadripper, but the company has been unequivocal about certain requirements: TPM 2.0, 8th Gen Intel CPU, 2nd Gen AMD Ryzen CPU (thus far). It’s entirely possible that whether Microsoft changes these requirements will depend on how much blowback the company gets. Microsoft executives have confirmed that while the CPU support list will “evolve,” the chips currently listed are the supported CPUs.
Yep, these lists (Intel, AMD, and Qualcomm) are the currently supported CPUs. The lists will evolve over time, of course, but these are the supported CPUs.https://t.co/Y26xrKvg8g
8). Not all systems that technically support TPM 2.0 can enable it. This depends on the make and model of your computer, but OEMs may not enable it. Getting TPM 2.0 up and running on a system that hasn’t previously used it isn’t impossible, but it should be approached with caution. If you installed Windows under BIOS instead of UEFI, there’s no simple way to convert to a TPM-enabled Secure Boot configuration. There is a way to do so, discussed by various commenters in this Ars Technica conversation thread. The linked article covers how to build a proper Windows 11 VM with TPM 2.0 enabled.
One big remaining question: Will individuals be able to upgrade to Windows 11 even if their hardware doesn’t meet requirements? Several people I spoke to brought up the fact that Microsoft never officially listed Haswell support for Windows 10, even though Haswell installed fine.
Our attempts to upgrade a computer to Windows 11 from Windows 10 using the previously leaked installer failed, so we can’t speculate on what the eventual state of the software will be. Microsoft did not release a utility telling people with three-year-old PCs that their CPUs were not supported under Windows 10 when it launched that OS, so the situations may not be analogous.
Where We Stand Right Now
To sum it all: TPM 2.0 support appears necessary but not sufficient to guarantee Windows 11 compatibility. Skylake-X is not currently listed on Microsoft’s Windows 11 Intel support list but may be supported. Current Windows Insiders are allowed to test Windows 11 but have been informed they cannot update to Release Candidate builds when such software is available.
End-users may find themselves in a situation where TPM cannot be enabled in firmware despite being technically supported. Absent a physical TPM 2.0 module, there may not be a way to enable TPM 2.0 on such a machine. This may or may not matter, depending on whether Microsoft will enable upgrades for your platform in the first place. This may be part of why Microsoft locked out older systems; it’s possible that OEMs only began shipping all of their systems with UEFI-accessible TPM options when AMD launched Zen+ and Intel was shipping Coffee Lake. This would explain the lockouts.
For now, ExtremeTech is taking Microsoft at its word. The company has repeatedly stated that 6th and 7th Gen CPUs will not be compatible, so we’ll be treating them as if they aren’t until we hear differently. Skylake-X appears to work, Threadripper 1x is an unknown. Buying a TPM 2.0 module for a 7th Generation Kaby Lake CPU may not be enough to get that CPU working under Windows 11.
That’s the state of things as of now, as near as anyone can tell. “Clear as mud” could have been coined to describe this situation.
