On Monday, researchers from Microsoft introduced Kosmos-1, a multimodal model that can reportedly analyze images for content, solve visual puzzles, perform visual text recognition, pass visual IQ tests, and understand natural language instructions. The researchers believe multimodal AI—which integrates different modes of input such as text, audio, images, and video—is a key step to building artificial general intelligence (AGI) that can perform general tasks at the level of a human.

"Being a basic part of intelligence, multimodal perception is a necessity to achieve artificial general intelligence, in terms of knowledge acquisition and grounding to the real world," the researchers write in their academic paper, Language Is Not All You Need: Aligning Perception with Language Models.

Visual examples from the Kosmos-1 paper show the model analyzing images and answering questions about them, reading text from an image, writing captions for images, and taking a visual IQ test with 22–26 percent accuracy (more on that below).

In this case, Kosmos-1 appears to be purely a Microsoft project, without OpenAI's involvement. The researchers call their creation a "multimodal large language model" (MLLM) because its roots lie in natural language processing, like a text-only LLM, such as ChatGPT. And it shows: For Kosmos-1 to accept image input, the researchers must first translate the image into a special series of tokens (basically text) that the LLM can understand. The Kosmos-1 paper describes this in more detail:

For input format, we flatten input as a sequence decorated with special tokens. Specifically, we use and to denote start- and end-of-sequence. The special tokens  and indicate the beginning and end of encoded image embeddings. For example, “ document </g>” is a text input, and "<s> paragraph <image> Image Embedding </image> paragraph </s>” is an interleaved image-text input.

 

... An embedding module is used to encode both text tokens and other input modalities into vectors. Then the embeddings are fed into the decoder. For input tokens, we use a lookup table to map them into embeddings. For the modalities of continuous signals (e.g., image, and audio), it is also feasible to represent inputs as discrete code and then regard them as “foreign languages”.

 

Microsoft trained Kosmos-1 using data from the web, including excerpts from The Pile (an 800GB English text resource) and Common Crawl. After training, they evaluated Kosmos-1's abilities on several tests, including language understanding, language generation, optical character recognition-free text classification, image captioning, visual question answering, web page question answering, and zero-shot image classification. In many of these tests, Kosmos-1 outperformed current state-of-the-art models, according to Microsoft.

Of particular interest is Kosmos-1's performance on Raven's Progressive Reasoning, which measures visual IQ by presenting a sequence of shapes and asking the test taker to complete the sequence. To test Kosmos-1, the researchers fed it a filled-out test,

one at a time, with each option completed, and asked if the answer was correct. Kosmos-1 could only correctly answer a question on the Raven test 22 percent of the time (26 percent with fine-tuning). This is by no means a slam dunk, and errors in the methodology could have affected the results, but Kosmos-1 beat random chance (17 percent) on the Raven IQ test.

Still, while Kosmos-1 represents early steps in the multimodal domain (an approach also being pursued by others), it's easy to imagine that future optimizations could bring even more significant results, allowing AI models to perceive any form of media and act on it, which will greatly enhance the abilities of artificial assistants. In the future, the researchers say they'd like to scale up Kosmos-1 in model size and integrate speech capability as well.

Microsoft says it plans to make Kosmos-1 available to developers, though the GitHub page the paper cites has no obvious Kosmos-specific code upon this story's publication.

Source

There are many challenges around the return of astronauts to the Moon. But there are also challenges when we actually get the astronauts there, and a big one comes from something small: moon dust. It has the consistency of ground fiberglass, it clings to spacesuits and crafts, and you do not want to breathe it in. Apollo astronauts used brushes to get clean, without much success. A new paper suggests another method: liquid nitrogen.

A team tested a cryogenic sprayer filled with liquid nitrogen in a vacuum, shooting a spacesuit-like material covered in a moon dust simulant. They found that the system worked to remove over 98.4 percent of the dust under vacuum conditions, without damaging the suits like methods used in the past, such as brushes and vacuuming.

“It posed a lot of problems that affected the missions as well as the astronauts once they returned home,” Ian Wells, first author of the paper from Washington State University, said in a statement.

Lunar dust can affect engines and electronics, damage suits including the seals, and gave the Apollo astronauts “lunar hay fever.” It damaged their lungs and in the long term could cause effects similar to what is experienced by coal miners: the so-called black lung disease.

“Moon dust is electrostatically charged, abrasive and gets everywhere, making it a very difficult substance to deal with,” explained Wells. “You end up with a fine layer of dust as a minimum just covering everything.”

To assess the effectiveness of the system, 26 samples were tested a total of 233 times. One of them went through 75 cleaning cycles before any damage occurred to the material. During the Apollo mission, a single brush was enough to damage the suits. And it’s not like these suits are cheap.

The tests took place both in a vacuum and under atmospheric conditions. In a vacuum, it worked much better, lifting almost 96 percent of the smallest particles of dust. We're picturing a little moon shower pod the astronauts would enter before getting back on board the spaceship.

This work won NASA’s Breakthrough, Innovative and Game-changing (BIG) Idea Challenge last year and the team is looking into understanding the process better, as well as applying for grants to conduct tests in closer to real lunar conditions.

The study was published in Acta Astronautica.

Source

Perovskite solar cells have developed extraordinarily fast, going from a curiosity to the hottest area of solar research in the space of a few years. After overcoming a series of barriers one major problem has remained; their durability. A paper in Nature Materials reports a novel solution. It comes less than a fortnight after two other potentially equally significant papers, one providing an alternative solution to the same problem and the other indicating perovskites are even more exciting than anyone knew.

The vast majority of photovoltaic cells in operation in the world are made using silicon. Although the raw material is among the cheapest and most abundant on Earth, the high temperatures needed to produce ultra-pure polysilicon require substantial energy input. Despite the spectacular fall in prices for silicon cells over the last 50 years, it is widely believed that to go much further we will need something different. Many alternatives have been tried, and some have proven better than silicon in certain niches, but only perovskites have looked likely to become a major source of power in the world.

Perovskites are named after the natural material whose basic structure they mimic. They are made up of calcium, oxygen, and titanium – all very cheap elements – doped with metals. The quantity of metals used is so small that manufacturing costs should be very low. Moreover, where silicon cells are good at collecting energy from the red part of the spectrum but bad at the blue, perovskites can be tuned to suit desired wavelengths.

Ion migrations have posed a problem for previous efforts to commercialize perovskite cells because the migrating ions make the material unstable and lead to long-term breakdown. Even very cheap solar cells are not much use for most purposes if they don’t last long. A team led by Professor Aram Amassian of North Carolina State University have now described a way to “corral” ions within perovskites to control these migrations.

“We have not found a way to prevent ions from moving through perovskite materials, but we have found that it is possible to steer these ions into a safe conduit that does not impair the material’s structural integrity or performance,” Amassian said in a statement. “It’s a big step forward.”

Perovskites are produced as a series of crystals or “grains”, which touch at grain boundaries. “Grains are better protected from impairment when the ions move predominantly along the grain boundary,” said first author Masoud Ghasemi.

By managing the boundaries, the team have shown it is possible to make perovskites more stable. 

As if extending the lifespan of clean generators were not enough, Amassian believes there is potential for the corralling work to be applied to other sorts of crystalline charge carriers, which could lead to improved electricity storage devices.

Two weeks earlier a paper in Science described another way to stabilize perovskites for long-term use, by treating them with 1,3-bis(diphenylphosphino)propane (DPPP). Despite its formidable name, "DPPP is also a commercialized product with low cost and easy accessibility,” said author Dr Zhaoning Song of the University of Toledo in a statement.

Yet another paper, also published during this epic two weeks, doesn’t address perovskites’ stability but instead demonstrates they can be even more efficient at turning sunlight into electricity. Perovskite cells are so thin they are placed on substrates to give them stability, and this has usually been glass. 

However, Professor Chunlei Guo and colleagues at the University of Rochester found perovskites on metal or dielectric substrates harvest 30-50 percent more of the energy in light falling on them. The excitation of electrons also lasts much longer, offering more time in which they can be tapped. “No one else has come to this observation in perovskites,” Guo said in a statement. “All of a sudden, we can put a metal platform under a perovskite, utterly changing the interaction of the electrons within the perovskite.”

Moreover, by alternating metal and dielectric layers, he reports in Nature Photonics, the sensitivity of photodetectors can be increased an astonishing 250 percent. 

All three teams still need to demonstrate their ideas work at scale and outside specialized laboratory conditions. Guo may also find alternating substrates adds too much to the cost of cell production to make the approach useful outside specialist environments where efficiency matters more than price.

However, if even one of the approaches makes the jump from lab bench to mass production, they could wipe out fossil fuel use during daylight hours, saving consumers money at the same time.

The papers are published in Nature Materials (open access), Science, and Nature Photonics (open access).

Source

Microsoft’s Activision deal likely to be approved by EU regulators, says Reuters

The partnership is part of Microsoft's Airband initiative. It aims to bring high-speed internet connectivity to unconnected communities around the world. The agreement will bring high-speed internet access to underserved areas including LDCs such as the Democratic Republic of the Congo and Zambia.

Microsoft's President Brad Smith said:

“Fifty years after the LDCs were classified by the UN as nations most in need of social, economic, and environmental opportunities to create growth, these 46 countries combined account for 13% of the world’s population, but only about 1.5% of global GDP and less than 1% of global trade. It is clear that more needs to be done."

Microsoft also has a new skilling initiative with the International Organization of Employers and Synapse that will train 20,000 youth, women, and entrepreneurs in digital, business, and employability skills across four LDCs. High-speed internet access will need more resources to grow the digital ecosystem.

Smith added:

“The private sector can play an important role in creating opportunities for the 880 million people living in LDCs, where only 36% of the population uses the internet today, and it’s important for Microsoft to do its part. The need for public-private partnerships has never been clearer. This is not a philanthropic exercise, but rather a business imperative – And a call to action for all of us to do more."

The Redmond company is also scaling its Digital Agriculture Platform with OCP Africa that will offer fertilizer solutions adapted to local conditions and the needs of soils and crops across Africa. This will support more than 40 million farmers and agri-stakeholders in Africa by 2030.

Microsoft's initiatives come ahead of the United Nations' LDC5 Private Sector Forum, which is intended to explore different ways to accelerate sustainable development for LDCs by digital means.

Microsoft announces new initiatives to bring high-speed internet to Africa and more

The Inland TD510, a MicroCenter exclusive, is dwarfed by the power cable you need to use to run its tiny fan.

Inland

 

In the high-end PC market, it's vitally important that the numbers always continue to go up. That means faster performance, newer specs, and (on occasion) new model numbers for existing components. One of the latest numbers to go up is the PCI Express version number supported by many motherboards; all PCs built around AMD's Ryzen 7000-series chips and some PCs using Intel's 12th- or 13th-generation CPUs support graphics cards solid-storage drives that use the PCI Express 5.0 interface, which is up to two times faster than version 4.0.

 

But actual PCIe 5.0 SSDs are just beginning to arrive on store shelves (via Tom's Hardware), and like so many early-adopter products, they seem purpose-built for people with more money than sense. Based on Phison's E26 SSD controller, the Gigabyte Aorus Gen 5 10000 and MicroCenter-exclusive Inland TD510 promise peak read speeds of up to 10,000MB per second, compared to 7,450MB per second for the PCIe 4.0-based Samsung 990 Pro.

 

But both drives have major shortcomings, even assuming you have a compatible PC in the first place. The 2TB versions cost more than twice as much as you'll pay for a very-good PCIe 4.0 drive like the Samsung 980 Pro or WD Black SN850X; the Gigabyte drive is currently out of stock but historical pricing data says it sells for $340 when it's available, while the Inland drive is currently discounted to $350 from a regular price of $400.

 

There's also the matter of the drives' heatsinks, each of which is ridiculous in a special way. The Gigabyte SSD comes with a gigantic, heatpipe-festooned passive heatsink, which will run silently but might be large enough to cause compatibility problems depending on your motherboard and where you install it. The Inland drive's heatsink isn't as bulky, but it includes a tiny fan connected to a large power cable. These fans can make a lot of noise because they need to spin so quickly to move much air, and the lack of a fan speed controller will make the fan spin equally quickly (and loudly) whether you're stressing the drive out or not.

 

Other SSDs come with heatsinks, but they're mostly small passive heatsinks that don't make any noise (many of them have been built to ensure that the drives will physically fit into a PlayStation 5, which also makes them unobtrusive in most PCs). The vast majority of mid-to-high-end motherboards include one or more built-in SSD heatsinks anyway, often styled to complement the rest of the motherboard. Installing either of these PCIe 5.0 SSDs would require removing that motherboard heatsink, or using it and running the risk of not cooling the SSD well enough to keep it from throttling.

 

Even putting aside the cost of the PCIe 5.0 SSDs and their silly heatsinks, it's difficult to notice the performance benefit of these kinds of drives right now. DirectStorage is a newish Windows API that allows the GPU to load and decompress assets directly from your SSD, rather than using your CPU as a go-between. Recent tests using the DirectStorage-compatible Forspoken show that the difference between PCIe 4.0 and PCIe 5.0 drives is nearly imperceptible with today's drives, and past tests show that even PCIe 3.0 and SATA SSDs can benefit quite a bit from DirectStorage—it's not a feature that requires a cutting-edge drive, and it's also not a feature supported by many games as of this writing.

 

PCIe 5.0 drives will come down in price eventually, and improved SSD controllers and NAND chips with more advanced manufacturing processes should eventually make them run cooler, too. In the meantime, look at the price tags and the designs of these early products, shake your head, and move on.

 

First wave of PCIe 5.0 SSDs arrives with high prices and ridiculous heatsinks

In a bit of irony, ChatGPT is growing so fast the AI-powered program is starting to rival Microsoft’s Bing search engine in terms of web traffic. 

The findings come from internet traffic provider Similarweb, which reports(Opens in a new window) user visits to ChatGPT nearly doubled last month. On a worldwide basis, traffic to the chat.openai.com domain was up 85% in the first 25 days of February compared to the same period from the month before. 

“Born at the end of November 2022, ChatGPT is now attracting about as much worldwide traffic as Bing, which has been in existence since 2009,” Similarweb added. 

(Similarweb)

The news is ironic because Microsoft itself has begun integrating ChatGPT into Bing, which did spark a bump in web traffic to the company’s search engine. But according to Similarweb, the traffic gains “appear to have been fleeting." Microsoft recently muzzled the new Bing’s capabilities to prevent the chats from going off the rails.

In contrast, ChatGPT has faced no such chat limit restrictions, although the site has been forced to routinely cut off access due to traffic surges from incoming users. A graph from Similarweb shows ChatGPT is now attracting about 40 million visits globally per day.

In the US, Bing still beats ChatGPT in terms of web traffic, attracting more than twice as many user visits. But the reverse is true in India. “ChatGPT is now drawing twice as much as Bing in India, one of its top traffic sources,” Similarweb said.

(Credit: Similarweb)

(Credit: Similarweb)

India also represents the second largest source of web traffic for ChatGPT at 11.17%, behind the US, which made up 11.27% of the traffic. The AI chatbot has also been trained to converse in various languages, expanding its appeal to users in numerous countries.

Similarweb added: “We might guess that the interest from India reflects both demand from its technology sector, yes, but also from workers who can use ChatGPT to increase their productivity at tasks such as copywriting and coding.”

Bing could regain the upper hand over time. Microsoft plans on releasing the ChatGPT-powered Bing as a mainstream search product to compete with Google. But for now, the experience remains locked behind a waitlist as Redmond continues to iron out the program's various kinks.

According to Similarweb, Bing ranks as the fourth largest search engine globally, behind Google, China’s Baidu, and Russia’s Yandex. However, Google holds a huge lead(Opens in a new window) as the number one search engine, so overtaking it won’t be easy.

Source

In an interview with India Today, Borchers said:

"We have a clear goal to transition fully to Apple Silicone. We believe strongly that Apple silicon can power and transform experiences from the MacBook Air to all the way up to the Mac Studio. We've been very clear from the beginning that our goal is to take our entire product line to Apple Silicon. And that's something we intend to do."

Apple first announced this transition in the WWDC 2020 keynote. Apple was already using powerful and efficiently designed chips on the iPhone and iPad for many years, and the chips are known to be energy efficient and powerful from a long time.

Borchers added:

"I think for the longest time, the broader industry was focused on performance, and we've introduced this idea that you need to think about how you can do that efficiently. And part of that focus on performance per watt has allowed us and kind of pushed us to integrate some core technologies into silicon so we can deliver them as efficiently as possible."

The Mac Pro will be the final in the line of Apple products to be transitioned to Apple Silicon. According to Borchers, Apple has a district advantage over its competitors, where processors and other components are made by different companies. He added that Apple's approach to developing its own silicon chips is focused solely on meeting the needs of its customers, without any consideration for other potential buyers. This strategy allows Apple to avoid dedicating valuable silicon space to unnecessary features and instead prioritize delivering meaningful benefits to its customers.

Despite that, Apple still has a few things ahead to take care of. Borchers acknowledged that some of its devices, like the Apple Watch, requires further efficiency gains - particularly in terms of battery life. He further explained that finding a solution to this challenge is a complex process, and that the company is still exploring various options, including faster charging. Apple is also considering ways to balance real-time health and body tracking features such as heart alerts and fall detection, with the need of extended battery life.

At the end of the day, Borchers says that Apple prioritizes the overall product package instead of individual chipsets. Apple's upcoming products will undoubtedly be powered by Apple Silicon, and will definitely be more efficient and powerful than their predecessors, if not than their Intel counterparts already.

Source: India Today

Apple-silicon powered Mac Pro could come out soon, says Apple VP of Marketing

As we move towards smaller nm architecture in chips, the need for innovation increases more and more.

AMD did their own way of innovating their Ryzen processors by adapting a chiplet based design. Where a single CPU would have two chiplets inside it, each consisting of 8 cores max. The reason is simple, the smaller the chip is, the higher chance of picking the good ones out of a silicon wafer, giving higher yields and massively saving costs.

After years of waiting, Intel decided that it needed to do something like that too. So Intel announced that it’s next-gen, that is, 14th-gen Meteor Lake CPU processors will come with a tile-based design.

While AMD went for two smaller core chiplets and a big I/O chiplet, Intel has decided to do it differently.

Intel Meteor Lake’s Tile-based Design

In Meteor Lake, Intel has announced that the processor will have 4 tiles. One SOC tile, one IO Tile, one GFX (GPU) tile and one CPU tile. The CPU tile will have all the performance and efficiency cores inside it.

While moving forward, this is an ideal move taken by Intel. As a big monolithic CPU would get harder and harder to manufacture and splitting things inside a processor is better. The problem is that, it seems that making a tile-based processor isn’t easy either. We aren’t sure what exactly is the status of these processors currently.

Intel Meteor Lake Desktop Maybe Cancelled

From some time now, there are rumors that Intel might cancel the Meteor Lake Desktop processor. This rumor became public by a tweet by a known Twitter based leaker @OneRaichu.

In it, he mentions that Intel has most likely converted whatever chips or design they had for Meteor Lake desktop into Meteor Lake mobile CPUs.

The MTL-S chip is referring to is Meteor Lake desktop. With MTL being Meteor Lake and S being the desktop variants of Intel Core processors.

It’s entirely possible that Intel tried making a tile-based desktop chip, but was not able to do so yet. Hence, making it available only for laptops.

Meteor Lake Might Target Efficiency More Than Performance

Another hint that Meteor Lake might not release on desktop is from an earlier tweet by OneRaichu.

In it, he mentions how Meteor Lake is going to target 1.5x times the efficiency of the current generation Raptor Lake.

Efficiency is something that mobile processors used in laptops benefit more than just desktop processors.

More Information About A Possible Meteor Lake Desktop Appears

To make things more confusing, another Twitter based leaker has shared the details about a possible Meteor Lake Desktop CPU. Twitter based leaker @leaf_hobby had posted a now deleted Tweet sharing the details about possible Meteor Lake Desktop features.

In that tweet, it is mentioned that Intel Meteor Lake Desktop will have x20 PCIe 5.0 lanes from the CPU. One x16 for the graphics card and one x4 possibly for an M.2 SSD.

It also mentions that the Z890 chipset might have an additional x4 PCIe 4.0 lanes, totaling x24 lanes. This is x4 more than what is found in current-gen Raptor Lake processors. Debut support for Wi-Fi 7 too is stated.

But what makes interesting is that it mentions that Meteor Lake Desktop has only 6 performance cores with 8 performance cores version possibly under or not under development. This is where it gets interesting. 6 performance cores are found in the Raptor Lake P (mobile variant), so does that support OneRaichu’s contention that Meteor Lake-S (desktop) processors are now converted to Meteor Lake-P (mobile) instead.

When asked about it, leaf_hobby doesn’t deny that being a possibility.

The @leaf_hobby, however, has later shared that their accounts are getting compromised repeatedly after leaking details about Intel. So they have decided to stop sharing Intel related information.

Either way, for now, we cannot say anything for sure about Meteor Lake. Whatever it maybe, there’s one thing almost certain, Intel is going to release Raptor Lake Refresh processors with higher clock rates this year. With Meteor Lake likely to be delayed till next year.

Intel’s Meteor Lake CPUs Might Not Release On Desktop

All the tech giants are trying to show off their new technologies and products. On the first day, companies like Huawei and Samsung dominated the exhibition area, but Xiaomi's genius strategy to bring its promising and interesting robots to Barcelona surely attracted much interest. Around a week prior to the event, Xiaomi's CEO Lei Jun tweeted that the CyberOne and CyberDog robots would also be present at the MWC 2023 event, and the ground was already prepared beforehand.

CyberDog was first introduced in 2021 and looks similar to Boston Dynamics' Spot. It has three different control methods. You could use your phone, an external remote control, or your voice to give commands and communicate. The remote control will be sold separately. In the demo video of CyberDog, it was announced that only Xiaomi’s Redmi K40 Gaming Edition, Redmi K40, Redmi K30 5G, Redmi 10X Pro, Redmi 10X, and Mi MIX 2S are supported. The company's plan is to introduce 1,000 of them and keep them a more valuable asset rather than bringing them to every home.

CyberOne, on the other hand, was introduced last year, sooner than CyberDog. It is a concept humanoid robot with a Mi-sense depth vision module and an AI algorithm that helps it recognize objects and react accordingly. The MiAI environment semantics recognition engine is offered with it. It can detect your voice, recognize your gestures and expressions, and have 45 different classifications of human emotions. CyberOne can recognize 85 environmental sounds and uses NVIDIA's Jetson Xavier NX platform with 11 built-in sensors. It comes with a curved OLED module that shows real-time interactive information.

Both robots still have limited functionalities, and more features are expected from their succeeders. However, Xiaomi's flex in Barcelona attracted many people as a robot dog is not something we see on a daily basis. This also showed the company's plans to concentrate on the West more.

Do you want a new "dog"? Xiaomi latest cyberdog is here

In a follow-up message, he stated the major improvements in Bing Chat v96. One is that users should now experience fewer times when the chatbot won't answer a question for no real reason. The other big improvement might cause some to chuckle, as Parakhin states there should be "Reduced instances of hallucination in answers."

That should make a few folks happier, as Bing Chat has generated some very weird interactions in the past month. Microsoft limited the number of daily chat turns in Bing Chat to combat that kind of "off the rails" response, but the company expanded those limits to 100 per day this weekend.

In yet another Twitter post, Parakhin stated that the final version of the Bing Chat Mode Selector will be released in "the next couple of days". This will allow Bing Chat users to select if they want standard answers to questions, more accurate responses, or more "creative" answers.

Microsoft rolls out Bing Chat v96 with less "hallucination in answers"

For those unfamiliar with PC Game Pass, it is a sister service to Xbox Game Pass and carries hundreds of PC games, including Microsoft's own Xbox party releases and Bethesda games, alongside access to EA Play titles.

Here's the full list of new countries joining PC Game Pass starting today, February 28:

Like the last injection of new supported regions, this wave is also beginning its mission as part of the Insider Preview Program. During this period Microsoft usually tests local payment systems, usually at a cheaper price while the tests are ongoing and gather feedback from new users regarding the experience.

To join the PC Game Pass Insider Preview as someone residing in one of the mentioned countries, download the Xbox Insider app from the Microsoft Store and join the program in the Previews section. Subscribing to PC Game Pass will then be enabled and all the available games will be accessible via the Windows Xbox app like usual.

PC Game Pass announced for 40 new countries, Insider access begins today

As carbon dioxide continues to build up in the Earth’s atmosphere, research teams around the world have spent years seeking ways to remove the gas efficiently from the air. Meanwhile, the world’s number one “sink” for carbon dioxide from the atmosphere is the ocean, which soaks up some 30 to 40 percent of all of the gas produced by human activities.

Recently, the possibility of removing carbon dioxide directly from ocean water has emerged as another promising possibility for mitigating CO2 emissions, one that could potentially someday even lead to overall net negative emissions. But, like air capture systems, the idea has not yet led to any widespread use, though there are a few companies attempting to enter this area.

Now, a team of researchers at MIT says they may have found the key to a truly efficient and inexpensive removal mechanism. The findings were reported recently in the journal Energy and Environmental Science, in a paper by MIT professors T. Alan Hatton and Kripa Varanasi, postdoc Seoni Kim, and graduate students Michael Nitzsche, Simon Rufer, and Jack Lake.

The existing methods for removing carbon dioxide from seawater apply a voltage across a stack of membranes to acidify a feed stream by water splitting. This converts bicarbonates in the water to molecules of CO2, which can then be removed under vacuum.

Hatton, who is the Ralph Landau Professor of Chemical Engineering, notes that the membranes are expensive, and chemicals are required to drive the overall electrode reactions at either end of the stack, adding further to the expense and complexity of the processes. “We wanted to avoid the need for introducing chemicals to the anode and cathode half cells and to avoid the use of membranes if at all possible,” he says.

Initially, the system can use existing or planned infrastructure that already processes seawater, such as desalination plants, but the system is scalable. This rendering shows how the new method could also be used by ships and offshore platforms. Credit: Courtesy of the researchers

 

The team came up with a reversible process consisting of membrane-free electrochemical cells. Reactive electrodes are used to release protons to the seawater fed to the cells, driving the release of the dissolved carbon dioxide from the water. The process is cyclic: It first acidifies the water to convert dissolved inorganic bicarbonates to molecular carbon dioxide, which is collected as a gas under vacuum. Then, the water is fed to a second set of cells with a reversed voltage, to recover the protons and turn the acidic water back to alkaline before releasing it back to the sea. Periodically, the roles of the two cells are reversed once one set of electrodes is depleted of protons (during acidification) and the other has been regenerated during alkalization.

This removal of carbon dioxide and reinjection of alkaline water could slowly start to reverse, at least locally, the acidification of the oceans that has been caused by carbon dioxide buildup, which in turn has threatened coral reefs and shellfish, says Varanasi, a professor of mechanical engineering. The reinjection of alkaline water could be done through dispersed outlets or far offshore to avoid a local spike of alkalinity that could disrupt ecosystems, they say.

“We’re not going to be able to treat the entire planet’s emissions,” Varanasi says. But the reinjection might be done in some cases in places such as fish farms, which tend to acidify the water, so this could be a way of helping to counter that effect.

Once the carbon dioxide is removed from the water, it still needs to be disposed of, as with other carbon removal processes. For example, it can be buried in deep geologic formations under the sea floor, or it can be chemically converted into a compound like ethanol, which can be used as a transportation fuel, or into other specialty chemicals. “You can certainly consider using the captured CO2 as a feedstock for chemicals or materials production, but you’re not going to be able to use all of it as a feedstock,” says Hatton. “You’ll run out of markets for all the products you produce, so no matter what, a significant amount of the captured CO2 will need to be buried underground.”

Initially at least, the idea would be to couple such systems with existing or planned infrastructure that already processes seawater, such as desalination plants. “This system is scalable so that we could integrate it potentially into existing processes that are already processing ocean water or in contact with ocean water,” Varanasi says. There, the carbon dioxide removal could be a simple add-on to existing processes, which already return vast amounts of water to the sea, and it would not require consumables like chemical additives or membranes.

“With desalination plants, you’re already pumping all the water, so why not co-locate there?” Varanasi says. “A bunch of capital costs associated with the way you move the water, and the permitting, all that could already be taken care of.”

The system could also be implemented by ships that would process water as they travel, in order to help mitigate the significant contribution of ship traffic to overall emissions. There are already international mandates to lower shipping’s emissions, and “this could help shipping companies offset some of their emissions, and turn ships into ocean scrubbers,” Varanasi says.

The system could also be implemented at locations such as offshore drilling platforms, or at aquaculture farms. Eventually, it could lead to a deployment of free-standing carbon removal plants distributed globally.

The process could be more efficient than air-capture systems, Hatton says, because the concentration of carbon dioxide in seawater is more than 100 times greater than it is in air. In direct air-capture systems it is first necessary to capture and concentrate the gas before recovering it. “The oceans are large carbon sinks, however, so the capture step has already kind of been done for you,” he says. “There’s no capture step, only release.” That means the volumes of material that need to be handled are much smaller, potentially simplifying the whole process and reducing the footprint requirements.

The research is continuing, with one goal being to find an alternative to the present step that requires a vacuum to remove the separated carbon dioxide from the water. Another need is to identify operating strategies to prevent precipitation of minerals that can foul the electrodes in the alkalinization cell, an inherent issue that reduces the overall efficiency in all reported approaches. Hatton notes that significant progress has been made on these issues, but that it is still too early to report on them. The team expects that the system could be ready for a practical demonstration project within about two years.

“The carbon dioxide problem is the defining problem of our life, of our existence,” Varanasi says. “So clearly, we need all the help we can get.”

Source

US Secretary of Commerce Gina Raimondo has made her recommendations for the revitalization of America’s semiconductor industry, a plan that is pivotal to retaining US economic leadership but one that also raises hard questions about government interference in the private sector and potential damage caused to US allies in Europe and East Asia.

In a speech entitled “The CHIPS Act and a Long-term Vision for America’s Technological Leadership,” delivered on February 23, Raimondo spoke about “the incredible opportunity we have as a nation to unleash the next generation of American innovation, protect our national security, and preserve our global economic competitiveness as we implement the historic CHIPS and Science Act.”

Signed into law by President Joe Biden in August 2022, the CHIPS and Science Act allocates US$52.7 billion to semiconductor production and R&D, and authorizes another $24 billion in tax credits for semiconductor production.

In her speech, Raimondo also outlined the Biden administration’s desire for the US to dominate the global tech industry:

I want the United States to be the only country in the world where every company capable of producing leading-edge chips will have a significant R&D and high-volume manufacturing presence.

 

We will be the premier destination in the world where new leading-edge chip architectures can be invented in our research labs, designed for every end-use application, manufactured at scale and packaged with the most advanced technologies.

 

This combination of technological leadership, supplier diversity, and resiliency does not exist anywhere else in the world today.

 

US Commerce Secretary Gina Raimondo. Photo: Progressive Hub

Raimondo’s statements seem to suggest she would like to reverse the globalization of the semiconductor industry and establish a complete supply chain for every product on US territory, regardless of comparative advantage.

Is this a good idea? Morris Chang, founder and former CEO and chairman of TSMC, the world’s leading semiconductor foundry (contract chip manufacturer) and a 25-year veteran of Texas Instruments, wouldn’t think so. Speaking in Taipei in 2021, he said,

If you want to reestablish a complete semiconductor supply chain in the US, you will not find it to be a possible task. Even after you spend hundreds of billions of dollars, you will still find the supply chain to be incomplete, and you will find that it will be very high cost, much higher cost than what you currently have.

It should be noted that most new leading-edge IC designs always have been and are still created in the US.

In 2022, according to market research firms and media reports, seven out of TSMC’s top 11 customers were American (Apple, AMD, Qualcomm, Broadcom, Nvidia, Marvell, Analog Devices and, number 11, Intel), one was Taiwanese (MediaTek), one Japanese (Sony) and one European (STMicro).

And, we might ask, must Sony, the world’s top maker of image sensors, establish R&D and manufacturing operations in the US? Sony has just formed a joint venture with TSMC in Japan.

Outside Japan, Sony Semiconductor has design and development bases in Belgium, France, Finland, Spain, Switzerland, Israel and Taiwan; manufacturing bases in Thailand (image sensor assembly) and China (optical pickups); and sales bases in mainland China, Hong Kong and Taiwan, South Korea, Singapore, the UK and the US (San Jose).  Presumably, management has a pretty good idea of what should be done where.

To be sure, Raimondo is not a complete techno-nationalist:

Very importantly, we are not aiming for self-sufficiency or looking to close ourselves off from global markets or competition. And of course, we are eager to continue working with our partners and allies to create diverse, resilient, and sustainable supply chains, write tech standards that align with our values and invest in our shared digital future.

After reviewing the decline of semiconductor manufacturing in the US – its share of global chip manufacturing is down from 37% to 12% since 1990; none of the world’s most advanced semiconductors are now made in the US – Raimondo went on to say:

This manufacturing atrophy has real consequences. For starters, it’s a threat to our national security. So many of our defense capabilities – like hypersonic weapons, drones, and satellites – depend on a supply of chips that aren’t currently produced in America.

That could indeed turn out to be a problem, Raimondo said:

But our dependence on foreign semiconductor supply chains also hurts our economy. In 2001, the US had more than 300,000 semiconductor manufacturing workers. In the past 20 years, we lost a third of those jobs while the global semiconductor industry has more than tripled in size. We sacrificed our manufacturing capacity and workforce in the mistaken belief that we could somehow maintain our technological leadership without them.

Raimondo correctly identifies the link between manufacturing and technological advances and the consequences of rampant outsourcing driven by an obsession with financial returns. Pointing out that “funding in technology hardware makes up only 3% of America’s venture capital, down from 20% in 2005,” she stated:

The brutal truth is that, without manufacturing strength in the US, and the innovation that flows from it, we are at a clear disadvantage in the race to invent and commercialize future generations of technology.

This is not always the case, as the overwhelming dominance of American semiconductor design and software companies demonstrates. However, as a general proposition, it is a correct diagnosis of what has been going on in the semiconductor, consumer electronics, industrial robot and other tech-focused industries for the last three decades.

As US production has been outsourced overseas, the development of some key semiconductor production equipment, flat panel TVs, industrial robots and other products has gone with it.

China, of course, has taken the opposite tack, to its great benefit and to the alarm of the US government. Raimondo noted that “over the last two years, China has produced more than 80% of new global capacity for certain mature chips, and their market share is growing.”

Is her concern justified? Data from Gartner show that among the top 10 semiconductor vendors in 2022 ranked by revenue, seven were American (Intel, Qualcomm, Micron, Broadcom, AMD, Texas Instruments and Apple), two South Korean (Samsung and SK hynix) and one Taiwanese (MediaTek).

Samsung and SK hynix ranked first and third, respectively, while Intel ranked second. Five were design companies that outsource to TSMC.

Taiwan-based TSMC is, along with South Korea’s Samsung, one of the two leading pillars of the global chipmaking industry. Photo: AFP / Sam Yeh

On the other hand, Raimondo stated that “Taiwan alone produces 92% of the world’s leading-edge chips.”

Defined as ICs made at the 7nm process node and below, that means TSMC. The rest are made by Samsung. Although not the subject of her speech, the key semiconductor industry-related risk to US national security is dependence on TSMC’s factories in Taiwan.

To hedge this risk and rebuild the US semiconductor industry, Raimondo said,

We need the private sector to invest with us, using our $50 billion of public investment to crowd in at least $500 billion in additional funding for manufacturing and R&D.

 

Specifically, the US will have at least two new large-scale clusters of leading-edge logic fabs, that will have been built by highly-skilled union labor.

 

Each cluster will include a robust supplier ecosystem, R&D facilities to continuously innovate new process technologies, and specialized infrastructure. Each of those clusters will employ thousands of workers in well-paying jobs.

 

Additionally, the US will develop multiple high-volume advanced packaging facilities, and become a global leader in packaging technologies.

That’s a tall order, history shows. Judging from the US auto industry, the application of the US-style adversarial relationship between union labor and management could disrupt and increase the cost of the project, and undermine the frontier spirit and entrepreneurial mentality that created companies like Intel, Microsoft, TSMC and Apple.

Already there is a “culture clash” at TSMC’s new factory in Arizona. Last year, EE Times quoted an American engineer working there as saying, “The work culture in Taiwan is really different than in the US. TSMC will have to change to an eight-hour workday five days a week.”

But TSMC engineers often work longer hours and are always on call, on weekends and at night, in case of an emergency.

There has also been an effort to organize workers at Intel’s factories in Oregon in order to “address long hours, pay disparities and job security.” This comes even as Oregon Live reports that “Workers in Oregon’s chip industry earn an average of more than $150,000 annually, according to state wage data.”

President Biden appears virtually at an event in Michigan on August 2, 2022, to celebrate the passage of the bipartisan CHIPS Act. Photo: Screenshot

On top of that, $550 billion is a lot of money. The Taiwanese, South Koreans, Japanese and Europeans, who are also subsidizing and otherwise promoting their semiconductor industries, might see it as a threat – perhaps as more dangerous than doing business with China.

And there is a risk of oversupply if investment decisions are made by ambitious politicians rather than by experienced, market-sensitive executives.

Raimondo also aims to shore up semiconductor R&D in the US:

The $39 billion in incentives [under the CHIPS Act] will bring semiconductor manufacturing back to the US, but a robust R&D ecosystem will keep it here. That is why we will invest $11B to build a strong semiconductor R&D ecosystem to generate the ideas and the talent we need to support these efforts.

 

The heart of these investments will be the creation of the National Semiconductor Technology Center. The NSTC will be an ambitious public-private partnership where government, industry, customers, suppliers, educational institutions, entrepreneurs, and investors converge to innovate, connect, and solve problems.

 

We envision a network of several centers around the country, solving the most impactful, relevant and universal R&D challenges in the industry. Their work – fueled by industry support – will generate new devices, processes, tools, and materials for our manufacturing ecosystem.

So does this mark the advent of an expensive, unfocused, bureaucratic boondoggle or is Raimondo’s plan ultimately sound industrial policy?

The American Semiconductor Innovation Coalition (ASIC) thinks the NSCT can work as long as it can develop and implement a “practical, technical agenda focused on the transition from innovation to commercialization” and “be held accountable for reaching clear and measurable goals.” To this end, ASIC has published detailed recommendations.

Raimondo phrased her bottom line this way: “Here’s the truth: If we don’t invest in America’s manufacturing workforce, it doesn’t matter how much we spend. We will not succeed.” 

Tens of thousands of engineers, technicians and scientists must be trained, she acknowledged.

“In the 10 years after Kennedy announced his mission to put a man on the moon, the number of physical science PhDs tripled and engineering PhDs quadrupled,” she said.

That level of commitment and inspiration is needed again. It already exists in China, after all.

Source

Technology firms are in a serious race with each other on who will come up with an innovative solution or an improvement for daily-used products. Internal hardware, chipsets, and displays have seen serious changes in the past few years. Tech giants mainly concentrated on the three aspects mentioned recently, but battery capacities were always left behind. In the past couple of years, we haven't seen many innovative solutions regarding battery capacity and density from these companies. However, Honor came to Barcelona to change the game.

The Chinese smartphone manufacturer, Honor, has revealed one of the most interesting features in Barcelona, apart from its new products Magic5, Magic5 Pro, and the Magic Vs. Yesterday, the CEO George Zhao announced Honor's new silicon-carbon battery with a longer runtime. According to Zhao, the silicon-carbon battery has 240% more capacity than a lithium-ion battery when run at 3.5 V. Honor's new technology offers approximately a 12.8% better performance than the current lithium-ion batteries which are used in most devices.

Honor

As mentioned, Honor also introduced its new Magic5 family, and the new Magic5 Pro comes with a 5,100 mAh battery. If the new technology had been used in this model, it could have gone up to 5,400 mAh. It might seem like a small improvement, but it is pretty impressive for a start. According to GSMArena, Honor's new Magic5 Pro comes with a 5,400 mAh battery, while the global version offers a 5,100 mAh battery. It looks like Honor has already started using the new silicon-carbon battery in China. However,

it might also be a coincidence as the company yet to make any announcements regarding the matter. In both ways, the global version features a regular 5,100 mAh battery.

Source

I wrote a little while ago (okay, five years) that “AI Powered” is the meaningless tech equivalent of “all natural,” but it has progressed beyond cheeky. It seems like just about every product out there claims to implement AI in some way or another, yet few go into detail — and fewer still can tell you exactly how it works and why.

The FTC doesn’t like it. Whatever someone means when they say “powered by artificial intelligence” or some version thereof, “One thing is for sure: it’s a marketing term,” the agency writes. “And at the FTC, one thing we know about hot marketing terms is that some advertisers won’t be able to stop themselves from overusing and abusing them.”

Everyone is saying AI is reinventing everything, but it’s one thing to do that at a TED talk; it’s quite another to claim it as an official part of your product. And the FTC wants marketers to know that these claims may count as “false or unsubstantiated,” something the agency is very experienced with regulating.

So if your product uses AI or your marketing team claims it does, the FTC asks you to consider:

• Are you exaggerating what your AI product can do? If you’re making science fiction claims that the product can’t back up — like reading emotions, enhancing productivity or predicting behavior — you may want to tone those down.

• Are you promising that your AI product does something better than a non-AI product? Sure, you can make those weird claims like “4 out of 5 dentists prefer” your AI-powered toothbrush, but you’d better have all 4 of them on the record. Claiming superiority because of your AI needs proof, “and if such proof is impossible to get, then don’t make the claim.”

• Are you aware of the risks? “Reasonably foreseeable risks and impact” sounds a bit hazy, but your lawyers can help you understand why you shouldn’t push the envelope here. If your product doesn’t work if certain people use it because you didn’t even try, or its results are biased because your dataset was poorly constructed… you’re gonna have a bad time. “And you can’t say you’re not responsible because that technology is a ‘black box’ you can’t understand or didn’t know how to test,” the FTC adds. If you don’t understand it and can’t test it, why are you offering it, let alone advertising it?

• Does the product actually use AI at all? As I pointed out long ago, claims that something is “AI-powered” because one engineer used an ML-based tool to optimize a curve or something doesn’t mean your product uses AI, yet plenty seem to think that a drop of AI means the whole bucket is full of it. The FTC thinks otherwise.

“You don’t need a machine to predict what the FTC might do when those claims are unsupported,” it concludes, ominously.

Since the agency already put out some common-sense guidelines for AI claims back in 2021 (there were a lot of “detect and predict COVID” ones then), it directs questions to that document, which includes citations and precedents.

Source

A year ago, AMD announced a new addition to the world of processors which completely changed how we looked at CPUs while gaming. It released AMD Ryzen 7 5800X3D with 3D V-Cache. In it, AMD added an extra 64MB layer of cache on top of the existing L3 cache. This gave it a tremendous boost in some games.

So great were the results that Ryzen 7 5800X3D was able to beat all the top, far more expensive processors. Seeing that, we expected AMD to do the same in the AMD Ryzen 7000 series processors. But to our surprise, AMD announced not one, but three new processors with the same 3D V-Cache tech in them. The processors it announced were AMD Ryzen 7950X3D, 7900X3D & 7800X3D.

While the AMD Ryzen 9 7950X3D and 7900X3D are releasing 28th February. AMD Ryzen 7800X3D releases more than a month later. AMD had meanwhile sent its samples of AMD Ryzen 9 7950X3D to all the reviewers. However, the reviews were not allowed to be made public until 27th February. Now the reviews are finally out and they are really exciting.

Prerequisites Before AMD Ryzen 7950X3D Reviews

All AMD Ryzen 9 7000 CPUs come with two chiplets inside a single CPU. While AMD Ryzen 7950X3D has 8 + 8 Cores on two chiplets, Ryzen 7900X3D has 6 + 6 Cores on it. So, AMD had to decide on what chiplet to add the 3D V-Cache on. AMD finally decided that only one chiplet will have the 3D V-Cache. The second chiplet without the 3D V-Cache, on the other hand, will be clocked higher.

So that means, games will need to run on either one of the chiplet, not both. That means, something needs to tell the games which chiplet to use. That is why AMD released special chipset drivers for these processors. In it, AMD would decide which games / software will need to use the chiplet with cache and which will use the chiplet with higher frequency but without 3D V-Cache.

Additionally, AMD required the reviewers to one, make sure the Game Mode is enabled in OS settings, and Xbox Game Bar app be installed. Not to forget, latest version of BIOS with 3D V-Cache optimizations and settings is installed. Only then reviewers were suggested to review the processors. However, all the efforts seem worth it when we check all the reviews posted by top reviewers out there.

AMD Ryzen 9 7950X3D Reviews

Tom’s Hardware

Tom’s Hardware made one of the best reviews for the AMD Ryzen 9 7950X3D. In it, they tested 3D V-Cache’s L3 latency compared to the non-cache chiplet (page 3). Then they checked clock and power usage of both the chiplets (page 4). Later came, all the other tests like overall power consumption, thermal, software and gaming benchmarks.

In gaming tests at 1080p (page 6), AMD Ryzen 9 7950X3D was 12% faster than Intel Core i9-13900KS and Core i9-13900K. It also found that AMD Ryzen 9 7950X3D was 26% faster than the Ryzen 9 7950X and 16% faster than Zen 3 based Ryzen 7 5800X3D. Which is massive.

What’s noticeable was that not all the games benefited with the 7950X3D’s extra cache, those games which did, they performed extraordinarily.

In productivity and rendering tests, Ryzen 7950X3D was behind, but not too behind. Which was nice, but that’s not what this CPU was designed for.

The biggest thing AMD Ryzen 9 7950X3D achieved, however, was doing all this with extraordinary power efficiency. While Tom’s Hardware didn’t give us the power usage for the games, in rendering benchmarks, it showed that Ryzen 9 7950X3D consumed half, yes, we repeat, half the power than the Intel Core i9-13900K CPU. This is something we feel is unheard of.

TechPowerUp

TechPowerUp was another one with a great review for AMD Ryzen 9 7950X3D. It not only tested the processor with various software, but it also tested it with emulators too. But the great thing they did different from others, is that they tested everything, including games, by setting the processor to use either the cache chiplet or the non-cache one. Not to forget PBO results, that is, overclocked results are included too.

At 1080p, AMD Ryzen 9 7950X3D performed almost the same as Intel Core i9-13900K, both in average FPS and minimum FPS. There was a minor increase in performance when the chiplet with cache was preferred. Here too it was seen that it performed great in some games but badly in some others. However, big improvements were seen when using Ray Tracing in games. Except in Ray Tracing, similar benchmark results were found in 1440p and 4K gaming benchmarks too.

However, here too the biggest achievement of the AMD Ryzen 9 7950X3D was seen was the power efficiency. TechPowerUp did power usage tests on both software and gaming workloads. The results were similar all around. AMD Ryzen 9 7950X3D consumed less than half the watts of Intel Core i9-13900K.

igor’sLAB

igor’sLAB was another professional reviewer which reviewed AMD Ryzen 9 7950X3D.

In gaming benchmarks at 1080p (page 4), AMD Ryzen 9 7950X3D was 9% faster than Intel Core i9-13900K, 17% faster than Ryzen 7 5800X3D and 18% faster than Ryzen 9 7950X. This performance difference were seen at 1440p and 4K too, though with more parity.

What igor’sLAB did different from others was it checked frametime in various games. There too, the AMD Ryzen 9 7950X3D was impressive and almost did it better than others.

Here too, AMD Ryzen 9 7950X3D showed great power efficiency in gaming. So much so that Igor says “it’s an outright humiliation” for both Intel and other AMD Ryzen 7000 processors.

Other Reviewers

There are many other reviewers which got hold on to the AMD Ryzen 9 7950X3D processor. AnandTech, Club386, KitGuru and Gamers Nexus (YouTube) to name some of them.

However, the most unique reviews were from Hardware Unboxed (YouTube) and TechPowerUp’s other review article where they simulated how yet to be released AMD Ryzen 7 7800X3D would perform by disabling the non-cache core on AMD Ryzen 9 7950X3D. But take those reviews with some considerations like different core clock and such on AMD Ryzen 7 7800X3D. So we cannot say for sure until it’s actually released.

Conclusion

It must be mentioned that we only posted images of 1080p benchmarks. There’s a lot of debate about what’s the best resolution to test CPUs at. The almost unanimous conclusion in the benchmarking community is that games are more CPU bound at lower resolutions and hence they best for testing CPUs. The higher you go, the more GPU bound the game become, hence true difference between CPUs are not visible. So we thought 1080p is perfect.

AMD Ryzen 9 7950X3D is a new gaming king. That’s for sure. Not only it’s the fastest gaming processor out there (depending on the game), it’s also one of the most power efficient processor ever created. All thanks to the 3D V-Cache which AMD has banked on. Sure, it doesn’t deliver on all the workloads, but in gaming it does, with great abilities.

However, one, there are hardly any reviews available for AMD Ryzen 9 7900X3D which got released together. It has only 6 cores on the 3D V-Cache chiplet. So it’s expected to perform worse than Ryzen 7 7800X3D. Is it the reason why AMD has not sent it to everyone for sampling or is there any other reason, we don’t know.

What’s important thing to mention is that AMD is holding onto Ryzen 7 7800X3D by not releasing it with the above two CPUs. The reason is simple, games are only going to use a single 8 core chiplet like they do in the Ryzen 9 7950X3D. Meaning, Ryzen 7 7800X3D is expected to perform almost the same as Ryzen 9 7950X3D in gaming, at far cheaper prices. So all we can say is, wait for AMD Ryzen 7 7800X3D to release if gaming is main and only use.

AMD Ryzen 9 7950X3D Reviews Are Out: The New Gaming CPU King

Several weeks ago, iniBuilds revealed plans to bring the Antonov An-225 "Mriya" to the Microsoft Flight Simulator. Those waiting for the largest jet plane ever built can rejoice as the Mriya has just landed in Microsoft Flight Simulator on PC.

The An-225 "Mriya" (from the Ukrainian word for "dream") was the world's largest transport aircraft capable of hauling up to 225 tons or 496,000 lbs of cargo. It was initially made to transport the Buran orbiter but then converted to cargo aircraft once the Buran space program collapsed alongside the USSR. The "Mriya" features six engines, two vertical stabilizers, and 32 landing gear wheels.

Unfortunately, the plane was destroyed on February 27, 2022, during the first days of the Russian invasion. After liberating the area, Ukraine announced plans to rebuild the An-225 using the remaining parts from the destroyed plane and the spare airframe. According to Microsoft, all proceeds from the in-game An-225 will go toward the rebuilding effort to bring the iconic aircraft back to life.

An-225 wreckage in Hostomel, Ukraine. Photo by Oleksii Samsonov, KCSA

The An-225 in the Microsoft Flight Simulator features three real-life liveries from 2010, 2008, and 2006, plus three extra. You can purchase the "Mriya" in the in-game marketplace on PC for $19.99. Microsoft says the DLC will arrive on consoles and Xbox Cloud Gaming in late March.

An-225 "Mriya," world's largest aircraft, is now available in Microsoft Flight Simulator

The Outer Worlds publisher Private Division has announced a new version of Obsidian's The Outer Worlds is incoming. Dubbed Spacer's Choice, the Obsidian Entertainment-developed title is heading only for PC, Xbox Series X|S, and PlayStation 5 platforms, and it touts enhanced graphics, smarter AI, higher level cap, and more. The last-gen original version came out in 2019, and the game never received an official next-gen upgrade.

On the visuals side, lighting, environments, character details, and particle effects have all received upgrades, and there's even dynamic weather conditions now. Meanwhile better enemy and companion AI, 4K 60FPS support on consoles, and DualSense controller adaptive triggers and haptics support on PS5 have also been confirmed.

Both Peril on Gorgon and Murder on Eridanos story expansion packs are also included here.

"Don’t worry, we didn’t dare touch the core of what made The Outer Worlds great," Obsidian remarks regarding the changes being made here. "There’s no point in reinventing the wheel, but fresh tires are always a plus."

The Outer Worlds: Spacer’s Choice Edition launches March 7 with a $59.99 price tag on PC, Xbox Series X|S, and PlayStation 5. Private Division is also offering an upgrade path for owners of the original game and all DLC on Xbox One, PlayStation 4, or PC, with them only needing to pay $10 to get the Spacer's Choice version.

While the original RPG is still getting development time, keep in mind that Obsidian is also working on The Outer Worlds 2 in the background under Xbox. Unfortunately, not much has been announced about the sequel other than its humorous cinematic trailer from 2021.

The Outer Worlds: Spacer's Choice Edition brings upgraded graphics, smarter AI, and more

Asteroids, like stars, only come out at night. Hidden in the glare of our Sun are an unknown number of asteroids on paths we cannot track, many of which could be heading for Earth, and we just don’t know it.

The European Space Agency’s (ESA) planned NEOMIR mission will be located between Earth and the Sun and will act as an early warning system for asteroids 20 meters and larger that cannot be seen from the ground.

Predicting Chelyabinsk

No one saw the Chelyabinsk meteor of February 15, 2013, coming. Just after sunrise on a calm and sunny winter’s day, a 20-meter (66-foot) asteroid struck the atmosphere over the Ural Mountains in Russia, at a speed of more than 18 km/s (40,000 miles per hour).

Fallen trees at Tunguska, Imperial Russia, seen in 1929, 15 km from epicenter of aerial blast site, caused by explosion of a meteor in 1908. Credit: Photo N. A. Setrukov, 1928

 

The relatively small rock approached Earth from very near the direction of the Sun, exploding in the atmosphere and creating a shockwave that damaged thousands of buildings, breaking windows and injuring roughly 1500 people from flying shards of glass. It was the largest asteroid to strike Earth in over a century.

Statistically, asteroids this size strike Earth about once every 50-100 years. Larger asteroids are far less common but – just ask the dinosaurs – do a great deal more damage. These are, fortunately, much easier to detect.

Asteroid danger explained. Credit: ESA, CC BY-SA 3.0 IGO

 

In fact, we have discovered almost all asteroids larger than 1 km in size. Small and medium-sized asteroids are more common, and can still do great damage, but warning times of a few days can be enough for local authorities to notify the public to keep away from windows or even to evacuate a local area.

With NEOMIR, we’ll be prepared

Whether it’s preparing for a mission to deflect a large asteroid years in advance or providing the data for local authorities to keep communities informed of airbursts weeks ahead, ESA’s NEOMIR will fill a gap in our current asteroid detection capabilities.

Asteroids are visible because they reflect the Sun’s light, which we can detect from Earth. However, the closer they get to the Sun, the harder they are to see. Asteroids crossing the face of the Sun are particularly difficult to detect, but from Earth, we are also blind to asteroids near the Sun as they are outshone by its glare.

Lagrange points are locations in space where the gravitational pull from the Earth equals the pull from the Sun, and satellites can reside with less ‘orbital maintenance’ than those orbiting Earth or flying out to deep space. Credit: ESA

 

ESA’s upcoming NEOMIR mission will be launched into orbit around the first Lagrange point (L1) between the Sun and Earth, remaining in the same position relative to the two bodies. This will give the telescope a view of asteroids that may come towards the Earth from the direction of the Sun.

Being situated outside of Earth’s distorting atmosphere and with a telescope observing in infrared light, NEOMIR will monitor a close ring around the Sun that is impossible to observe from the ground. The mission will detect asteroids passing between Earth and the Sun – any that pose a threat and that we cannot currently see will have to pass through this ring.

Our brightly burning Sun, photographed here on an Antarctic summer day by ESA-sponsored medical doctor Stijn Thoolen at Concordia research station. Credit: ESA/IPEV/PNRA–S. Thoolen

 

By making observations in the infrared part of the light spectrum, NEOMIR will detect the heat emitted by asteroids themselves, which isn’t drowned out by sunlight. This thermal emission is absorbed by Earth’s atmosphere, but from space, NEOMIR will be able to see closer to the Sun than we can currently from Earth.

Asteroids 20 meters (66 feet) and larger that are heading toward Earth should be detected by NEOMIR at least three weeks in advance. In the worst-case scenario, in which the asteroid is spotted passing near the spacecraft, we would get a minimum of three days’ warning – the fastest the asteroid could move from L1 to Earth.

Current status

Details of the Space Safety Program’s NEOMIR mission are currently being fleshed out and it is planned to be launched around 2030 with an Ariane 6-2 rocket.

Artist’s view of the configuration of Ariane 6 using four boosters (A64). Credit:
ESA – D. Ducros

 

An initial study to assess the feasibility of the NEOMIR mission was conducted by ESA’s Concurrent Design Facility in the Netherlands, in 2021. The study focused on defining a mission that would complement NASA’s NEO Surveyor mission. The US-funded mission should fulfill the US Congress mandate to discover 90% of near-Earth objects larger than 140 meters (460 feet) in diameter, while NEOMIR is designed to focus on imminent impactors of any size.

NEOMIR is currently early mission study phase. It will require a half-meter telescope with a large, corrected focal plane, as well as two infrared channels covering light in the 5-10 micrometer waveband.

NASA’s DART spacecraft collided with the smaller body of the Didymos binary asteroid system in September 2022. ESA’s Hera mission will survey ‘Didymoon’ post-impact and assess how its orbit has been changed by the collision, to turn this one-off experiment into a workable planetary defense technique. Credit: ESA–ScienceOffice.org

 

The required detector technologies and associated electronics for this novel mission are currently under development. Industrial research and development projects are planned as supporting activities in parallel.

The requirements will be to deliver a similar performance to the ‘NEO Surveyor detectors’, i.e., Teledyne’s HxRG, which are in use in the James Webb Space Telescope (NIRSpec) and ESA’s Euclid (NISP) and Ariel missions, although at shorter wavelengths.

Source

However, it looks like the 3GHz RNDA 3 iGP core clock, which was listed previously on AMD's site has now been altered. Spotted by Twitter user Bionic_Squash, AMD has seemingly and quietly lowered the Radeon 780M's core clock speed on the Ryzen 9 7940HS. While the previous clock was 3,000MHz or 3GHz (surprisingly similar to those 3Ghz rumors before Radeon 7900 series was officially revealed), the 780M is currently rated at 2.8GHz or 2,800MHz, which is a 200MHz reduction.

Throttled clocks have not only affected the flagship 7940HS though, as other Phoenix SKUs, like the AMD Ryzen 7 7840HS, also see reduced clocks by 200MHz. Overall, it looks like all Radeon 780M IGPs have been stifled by 200MHz due to some reason, probably because the graphics portion of the APUs are unable to hit such clocks under the allotted TDP and voltages.

Another area where Phoenix has faltered is PCIe support. Back when AMD began teasing Phoenix, the company had claimed that PCIe 5.0 support would come to the Dragon range, as well as Phoenix. Here's an image of a slide that showed such features:

However, things did not turn out that way as Phoenix had to settle for PCIe 4.0 only. Perhaps AMD had felt that the Dragon range, which is geared more towards performance, would benefit much more from it.

Source: Bionic_squash (Twitter) via Andreas Schiling (Twitter)

AMD quietly throttles clocks, PCIe 5.0 support on Ryzen "Phoenix" Zen 4 APUs

Over 500 Apple computers and related accessories are being auctioned off next month online and in Beverley Hills, California. The auction will feature numerous products dating from 1977 to 2008, including Macintosh systems from the '80s, more modern machines like the 2001 iMac G3, and old-school accessories like RH Electronics' Mac N' Frost external fan and surge protector.

Auction house Julien's Auctions has dabbled in Apple auctions before. Sadly, that includes the auction of Steve Jobs' Birkenstocks for a disturbing $218,750. Its upcoming auction, announced last week and spotted by sites like PetaPixel, features classic Apple items accrued by Swiss collector Hanspeter Luzi.

The auction house's announcement describes Luzi as a late historian with many hobbies who maintained a collection of old sewing machines that are now part of Germany's Sewing Machine Museum.

Luzi reportedly became a volunteer IT manager for schools, where he'd buy off unwanted vintage computers and parts. The auction house said Luzi died in 2015, and his family decided to auction his collection of Apple products.

Julien's Auctions' announcement included a sneak peek at some of the hundreds of items it'll be putting up from the collection next month.

One of the most notable items listed is a 1983 Lisa. Not surprisingly, it has the most expensive price estimate of the bunch, at $10,000 to $20,000. Although, we wouldn't be surprised to see the vintage computer sell for more. Previous auctions have seen functioning Lisa computers go for over $50,000, including on eBay. (If the whole Lisa is too pricey, maybe the Lisa OS source code is more your budget.)

Like some of the other computers in the auction, the Lisa is being auctioned with extras. A deep-pocketed collector can walk away with the system, a keyboard, the box it came in, and a "Lisa CRT Magnets Spares Kit (652-4520)." For a detailed look at the computer, which was one of the first to leverage a mouse and GUI, check out our recent deep-dive on the Apple Lisa's short life.

For those looking to spend less, the auctioneer is also advertising a motherboard from the Lisa, as well as other Apple motherboards dating back to 1978. The boards are expected to sell for up to $200 each.

Julien's will also try to sell numerous Apple computers from the '80s, including the Apple III from 1980 with a Monitor III and "various connecting cables" included (estimated price of up to $500), plus the first Macintosh, from 1984. They're even throwing in the floppy disk apparently still in the Macintosh 128K's drive.

Meanwhile, the 1989 Macintosh Portable will be sold with its "Apple Portable canvas case, with adapter, connecting cables, an Apple luggage tag, and two manuals in German (one copyrighted 1988, the other 1990)," Julien's Auctions said. Apple fans will remember the laptop as the first Mac to run on a (heavy) battery and one of the first personal computers to use an active matrix LCD display.

For those who couldn't afford the machine when it sold for a short time for $6,500, you may be able to get it at a much more affordable $300 to $500, by the auction house's estimates.

Other items included in the auction are joysticks for the Apple IIe and IIC from 1983-1984 and 1994's Apple QuickTake 100 camera.

It'll be interesting to see what these products actually go for, however, as pricey Apple auctions aren't new. Most recently, an original iPhone from 2007 originally expected to auction for $50,000 ended up selling for $63,356.

Julien's Apple auction will run from March 27 to March 30.

Source

Credit: The Verge

Foldable phones are welcomed pretty "normally" in today's technological world. They were first introduced a couple years ago, and people are now used to seeing them. Moreover, LG also introduced its foldable TV a while ago, which is also off the table regarding technological innovations. However, Lenovo is onto something new. The company introduced its first-ever rollable laptop at MWC 2023. Lenovo has also teased its first "rollable smartphone" at the event under the Motorola banner, its subsidiary smartphone brand.

According to The Verge's Jon Porter, Lenovo's rollable laptop prototype was placed near regular laptops, and at first glance, nobody really realized the impressive technology behind it. It looks like any other ThinkPad, but once the "show begins," you understand its technology is very different compared to other laptop models. There is a small switch on the right side of the chassis, and once you turn it on, the screen starts to roll. It turns into an almost-square 15.3-inch display with an 8:9 aspect ratio. Normally, it comes with a regular 12.7-inch display with a 4:3 aspect ratio.

It has a 2024x1604 resolution, and once the rolling begins, it increases up to 2024x2368. "Lenovo thinks such a tall display could be helpful for both office workers and creative professionals alike, offering everything from more lines of code to more cells in a spreadsheet," Porter says. The concept is very similar to LG's rollable TV model.

Credit: The Verge

Regarding specifications, Lenovo didn't provide information regarding the battery life and weight. However, the company officials said that "it is aiming to get 20,000 to 30,000 rolls." Moreover, Porter wasn't allowed to hold the laptop in his hands.

Unfortunately, Lenovo's new concept design is nowhere near hitting the shelves soon. The prototype might seem promising, but the company has to work on multiple aspects to make it ready for consumer use. For now, it is only a "show off" that Lenovo revealed at Mobile World Congress 2023.

The second prototype Lenovo showcased at the event was the rollable smartphone. However, it was revealed under the Motorola banner. Multiple companies showed improvement signals regrind rollable smartphones in the past, but none have hit the shelves yet. It is known that Lenovo is one of the few companies with a "foldable laptop," which makes the rollable one possible to hit the shelves soon.

Source

Xiaomi, the Chinese technology giant, has unveiled a prototype pair of AR glasses to prove they can sell products beyond smartphones. The Xiaomi Wireless AR Glass Discovery Edition will be considered on par with the Google Glass Enterprise and Microsoft’s HoloLens if Xiaomi officially launches the product on the AR market.

Xiaomi has stated its AR glasses won’t be on sale yet as it still is a concept technology. However, at the world’s biggest mobile industry trade show, MWC (Mobile World Congress) in Barcelona, the announcement highlighted Xiaomi’s ambition to enter the augmented reality space.

Apple, Microsoft, and Google technology giants focus on mixed or augmented reality to feature new experiences to users. The “metaverse” has become part of an idea of whatever important part technology can play.

Tim Cook (Apple CEO) mentioned the previous year: “I think AR is a profound technology that will affect everything.” However, Apple hasn’t officially launched or acknowledged any AR or virtual headset.

Although Bloomberg reported that Apple’s mixed reality headset would be unveiled this spring and at the beginning of fall 2023 with consumer deliveries.

“It’s a widespread view in the technology industry that AR smart glasses could be the next major form of mobile computing after the smartphone,” an analyst at CCS Insight, Leo Gebbie, stated.

Xiaomi AR Glasses Gesture Control

The lightweight Xiaomi AR glasses connect wirelessly to a smartphone and offer a “retina-level display,” allowing users to see virtual objects as clearly as they would the real thing!

Another exciting feature for the headset wearers will be gesture control to carry out tasks. By something as simple as raising your hands in front of the AR glasses and moving your fingers or hands, you can execute certain functions. Xiaomi says, "the thumb sliding on the index finger is used to enter and exit applications.”

Without the need to touch the device, “This kind of interaction showcases one of the directions that Xiaomi believes human-computer interaction will take in the future,” Xiaomi said.

The Xiaomi Wireless AR Glass Discovery Edition will work with its flagship smartphones like Xiaomi 13, and 13 Pro, said to launch worldwide on Sunday. Xiaomi continues to unlock revenue from other devices and is seen as one of the prominent smartphone players globally. Over recent years the company has become known for launching devices from electric scooters to TVs.

Gebbie stated, “Xiaomi’s new AR glasses feel more like a stake in the ground than a meaningful product launch. Chinese rivals, including Oppo and TCL, have also shown off AR devices, and it’s clear no one wants to get left behind.”

These features sound impressive on paper, but it’s still a prototype. So only time will tell how well the AR glasses will succeed in the real world. Let’s wait and see what the future holds for.

Xiaomi’s new AR glasses, shall we?

Source

Earlier on Sunday, Android Authority showed off the Motorola rizr, a smartphone concept device with a rollable expandable display. Now, the same site has a similar prototype device from Motorola's parent company Lenovo. This is a Windows notebook that also has a rollable screen that moves vertically to expand the laptop's display.

You may remember that Lenovo launched a Windows notebook with a foldable display, the ThinkPad X1 Fold, back in late 2020 and followed it up with a 2022 hardware refresh. This new concept notebook uses lessons learned from that foldable laptop in its design. The story reports that X1 Fold owners frequently used the notebook in a portrait configuration to multitask, write code, and more.

Lenovo then decided to make this concept rollable notebook with that in mind. The display rolls inside the case when normally used as a 12.7-inch laptop with a 4:3 aspect ratio. The laptop expands to a 15.3-inch plastic OLED screen and a nearly square 8:9 ratio when fully rolled out.

Since this is a concept device, it won't be sold to the general public, and Lenovo didn't reveal anything more about the laptop's hardware specs. We will likely have to wait a while to see when or if this rollable notebook will be released as an actual laptop.

The Lenovo Rollable Notebook concept can expand its screen vertically