With Labor Day cutting this week short for many, some companies slowed things down on the announcement front. But when the doors of official announcements are closed, the windows of leaks open. Our Senior Editor Zac Bowden spoiled everything that Microsoft has in store at its special Surface and AI event later this month. He also reviewed the upcoming Windows 11 2023 Update.

Over the last few days, we also saw Microsoft and other tech giants classified as gatekeepers by the EU, Microsoft 365 experience outages, and Elon Musk attempt to take on LinkedIn.

Windows 11 2023 Update

(Image credit: Windows Central)

Generally, I collect all the reviews at the bottom of the weekly news roundup, but this one is big enough to warrant its own section. The Windows 11 2023 Update, also known as Windows 11 version 23H2, is on the way. The update is set to ship in October, and our Senior Editor Zac Bowden ran through all the features set to ship with it.

The biggest change is the addition of Windows Copilot, which brings AI to the desktop. While the feature will support third-party plugins, it's not much more than a shortcut to Bing Chat at the minute. Copilot can perform a few tasks on your PC, such as turning Bluetooth on and off, but it's rather limited. Windows Copilot is in preview, so it should get better and more capable over time.

Here are the most notable features that will ship with the Windows 11 2023 Update:

• Windows AI Copilot
• AI-powered file recommendations in File Explorer and Start
• AI review summaries in Microsoft Store
• A new AI Hub for AI-powered apps in the Microsoft Store
• Taskbar app labels and general improvements
• New sound output menu
• A new Settings homepage
• Native RGB peripheral controls
• A new File Explorer design with more modern interfaces
• In-box support for more archive formats like .RAR
• A new File Explorer "Gallery" feature
• A new cloud-based backup and restore feature
• Snipping Tool and Notepad improvements
• Inking directly into text boxes
• General accessibility improvements
• Account recommendations in Start and Settings
• HDR desktop wallpaper support
• New Windows Spotlight wallpaper UI
• And more!

Surface event spoilers

(Image credit: Daniel Rubino)

Microsoft will hold a special event on September 21, 2023, but you don't have to wait until then to see what the tech giant has in store. Our Senior Editor Zac Bowden broke down everything to expect from Microsoft's Surface and AI event. As the name suggests, the event will focus on Surface hardware and artificial intelligence, but Bowden provided further insight.

Windows Copilot is expected to be a focus of the event, including third-party plugin support launching in preview. Microsoft 365 Copilot should see some love as well. On the hardware side of things, you can expect to see a Surface Laptop Studio 2, Surface Laptop Go 3, Surface Go 4, and a new Surface Hub 2S. What you shouldn't expect to see is a Surface Pro 10 or a Surface Laptop 6.

Bowden dives into much more detail in his piece, including the processors each device is expected to feature and if Microsoft will show off Windows 12 during its special event.

Microsoft 365 outage

(Image credit: Future)

Microsoft 365 saw outage reports spike this week. Issues began appearing on Wednesday, September 6, 2023 and kept coming in throughout the day. By Thursday morning, Microsoft 365 was up and running normally. Unfortunately, outages like this happen. The problems were intermittent, so there's a chance you used Microsoft 365 all week without being affected. That was the case for me, as I didn't see any issues. My colleague Ben Wilson was not as lucky.

Those affected by the outage were not able to access the Office web apps, Outlook, or OneDrive.

EU Gatekeeper saga

(Image credit: Kevin Okemwa)

Microsoft is officially a gatekeeper, at least in the eyes of the European Union. The tech giant, along with Alphabet, Amazon, Apple, ByteDance, and Meta, was classified as such this week. In the eyes of the EU, a gatekeeper is an organization that's large enough to leverage its position to create a bottleneck. The Digital Markets Act (DMA) is in place to prevent gatekeepers from using unfair practices.

A total of 22 core platform services will be affected by the new gatekeeper designations, including Microsoft's Windows operating system and LinkedIn social network.

Some services, such as Microsoft Advertising, Bing, and Edge, are still under investigation by the EU. Microsoft argued that Bing, Edge, and Microsoft Advertising should not fall under the same rules are larger competitors such as Google and Chrome. Apple is making a similar argument regarding iMessage.

"The Digital Markets Act will help creating a level-playing field for all companies competing in the European digital market, as it will bring about more contestability and openness in markets. Today, we identified the first six gatekeepers that must respect the new rules set forth by the DMA," said EU Commissioner Didier Reynders.

Elon Musk vs LinkedIn

(Image credit: Tesla)

Elon Musk's desire to take on LinkedIn isn't exactly new, but the story made the rounds this week. The controversial billionaire promised that the X competitor to LinkedIn will be cool. Considering that X has lost billions of dollars in value, Musk may want to prioritize being sustainable, but what do I know?

X Hiring was launched recently to compete with LinkedIn. The new platform is currently in beta and limited to organizations willing to pay $1,000 per month for a Gold Tick verification. I guess that's what's cool in the eyes of Musk.

Source

The launch of the first PlayStation might not have happened if the original plans for Sony's entry into the console gaming business had occurred, As related in this Ploygon story, Sony engineer Ken Kutaragi made a deal with the leader in game consoles, Nintendo. The deal would allow Sony to create a sound chip for the SNES console. Kutaragi originally made the deal without telling the higher-ups at Sony beforehand. In the end, Sony's then CEO Norio Ohga allowed the deal to go through.

Nintendo and Sony then struck another deal. Sony would create a CD-ROM add-on for the SNES, which Sony called the "Play Station". According to this Kotaku story, Sony's contract with Nintendo also allowed it to keep the CD-ROM game software rights as well. That meant all the royalties from any CD-ROM game sales from the "Play Station" add-on would go to Sony. Nintendo would just get the money from the CD-ROM hardware add-on device.

The plan was to officially announce the "Play Station" SNES CD-ROM add-on at the 1991 Consumer Electronic Show. However, Nintendo realized they had given away too many rights to Sony. Behind Sony's backs, Nintendo decided to make a deal with rival Phillips for them to make a CD-ROM add-on for the SNES. Sony reportedly found out about this deal two days before the reveal.

In the end, Sony announced the "Play Station" which would play SNES games and CD-ROM based titles, at CES on June 1, 1991. On June 2, Nintendo and Phillip announced its rival CD-ROM add-on at CES.

In the end, neither device actually launched. A prototype of the "Play Station" was discovered years later, with Engadget reporting that it still worked. It's an interesting side note in the history of console gaming.

While Sony decided not to sell its original SNES "Play Station", the company decided to go head first in the video game console world with its own standalone console, the PlayStation, that had no Nintendo connections and with Kutaragi in charge of its production.

According to a copy of Electronics Gaming Monthy. the first PS console had a 32-bit custom R3000 CPU with a clock speed of 33.8688 MHz. It had 16 Mb of RAM, along with 8 Mb of VRAM and 4 Mb of audio RAM.

The maximum resolution for games on the console was 640 × 480. The Geometry Engine chip allowed the console to handle graphics effects like texture mapping and grourand shading. You can go even deeper into the hardware aspects of the original PlayStation in this article written by Rodrigo Copetti.

As we mentioned, the first PlayStation console launched in Japan on December 3, 1994, with over 300,000 units sold by the end of that year. The launch 28 years ago today in the US got a big boost when Sony announced the launch price of the console would be just $299 at the very first Electronic Entertainment Expo (E3).

The console launched with a handful of titles in the US, but the highlight was Ridge Racer. Its 3D texture-mapped graphics might look a bit dated today, but they looked incredible to many gamers back in 1995. As in Japan, the PS1 was a massive sales hit in the US.

Today, Sony has launched four additional main consoles in the PlayStation family, along with two handheld gaming consoles, the PSP and PS Vita. Again, one wonders how the video game industry might have been different if Sony and Nintendo had decided to get along better all those years ago.

Source

On July 22, 2019, India successfully launched Chandrayaan-2, the second mission to the moon. Twenty-two days later (August 14), after a series of orbit raising manoeuvres, the spacecraft finally escaped the earth’s gravity and followed a path towards the moon. Six days later, Chandrayaan-2 was successfully inserted into lunar orbit. Finally, on September 2, the Vikram lander separated from the Orbiter, performed two de-orbit manoeuvres and on September 6, began its descent to the moon’s surface. The descent went as planned up to an altitude of 2.1 km from the Moon’s surface before communication from the lander to the ground stations was lost. The Vikram lander had apparently crash-landed on the Moon. 

Soon a national-level failure analysis committee headed by Dr. V. Narayanan, Director of the Liquid Propulsion Systems Centre (LPSC), was formed with experts drawn from national institutes and ISRO to study the cause of failure and to propose necessary corrections. The committee pinpointed three crucial mistakes that had happened in succession resulting in the crash.

There are four important phases before the touchdown on the moon — the rough braking phase, the attitude-hold (orientation) phase, the fine braking phase, and the landing phase.

Cause of crash

During the rough braking phase, the velocity of the lander was successfully reduced to a maximum. The problem with the touchdown began when the Vikram lander entered the second phase — the attitude-hold (orientation) phase. During this phase, the thrust had to be maintained at half the level. “But when the engines were commanded to provide half the thrust, the achieved thrust was more than half. The dispersion level [variation in the thrust] was more than what we had assumed during the pre-flight simulations,” explains Dr. K. Sivan, former Chairman of ISRO under whose watch the Chandrayaan-2 was launched in 2019.

The guidance system was supposed to have halved the thrust but it malfunctioned. “The guidance system was not designed to handle large dispersions, and this resulted in the guidance system malfunctioning leading to full thrust being given to the lander,” says Dr. Sivan. “At the end of the attitude-hold phase, the altitude and velocity were very different from the expected values. The guidance-system malfunction was the second crucial error.” 

According to Dr. Sivan, during the fine braking phase, the system was trying to correct the large errors. The large-scale corrections to velocity and altitude demanded large orientation manoeuvres. Changes to orientation are carried out by the control system, but unfortunately the control system was not designed to produce large changes. “Even though there was a large demand for orientation change, the control system restricted the rate.

As a result, the control system was unable to provide the necessary large corrections to orientation,” Dr. Sivan says. “The control system was taking more time to correct the orientation due to rate restriction. But by that time the lander had reached the moon’s surface.” This resulted in the lander crashing on the moon.

The crash was caused by three mistakes — the thrust (dispersion level) was more than assumed for design, the flaw in the guidance system resulted in more thrust being given, and the rate restriction in the control system failed to make large-scale changes in the orientation. “The Chandrayaan-2 mission would have been a success if any of the three errors was not there. The three cumulative errors led to the failure,” he says.

Correcting the mistakes

Taking the learnings from Chandrayaan-2 failure, ISRO corrected for all the three major mistakes in the latest mission — the system was corrected to ensure extra dispersion does not happen, the guidance system was corrected, and the restriction in the control system was removed. “These were the major corrections in Chandrayaan-3 lander.

Redundancy was also built into Chandrayaan-3 lander in terms of extra sensors, and extra propellant to allow it to travel to an alternate landing site, if required. And the landing area was expanded from a narrow patch of 500 metres x 500 metres to a 4 km x 2.4 km region.

In addition, other improvements suggested by the failure analysis committee were also incorporated in the Chandrayaan-3 mission.

Extra fuel, sensors and stronger lander legs designed for higher landing velocity led to a net increase in the weight (about 250 kg) of the lander.

Since Chandrayaan-3 was not carrying an orbiter like its predecessor, it became possible to accommodate the increased weight.

“Increased weight of the lander meant that a single engine was not sufficient. So the central engine was removed and two engines were used for landing,” Dr. Sivan says. “The central engine was added in the earlier mission as it was suspected that dust stirred by other engines would affect the electronic components. But our analysis found that dust wasn’t a problem.”

Source

The handful of men in the room were laughing and speaking over one another in excitement, as captured in a video taken that day, until one of them asked for quiet. The room went silent; the demo was underway.

Mr. Leyvand turned toward a man across the table from him. The smartphone’s camera lens — round, black, unblinking — hovered above Mr. Leyvand’s forehead like a Cyclops eye as it took in the face before it. Two seconds later, a robotic female voice declared, “Zach Howard.”

“That’s me,” confirmed Mr. Howard, a mechanical engineer.

An employee who saw the tech demonstration thought it was supposed to be a joke. But when the phone started correctly calling out names, he found it creepy, like something out of a dystopian movie.

The person-identifying hat-phone would be a godsend for someone with vision problems or face blindness, but it was risky. Facebook’s previous deployment of facial recognition technology, to help people tag friends in photos, had caused an outcry from privacy advocates and led to a class-action lawsuit in Illinois in 2015 that ultimately cost the company $650 million.

With technology like that on Mr. Leyvand’s head, Facebook could prevent users from ever forgetting a colleague’s name, give a reminder at a cocktail party that an acquaintance had kids to ask about or help find someone at a crowded conference. However, six years later, the company now known as Meta has not released a version of that product and Mr. Leyvand has departed for Apple to work on its Vision Pro augmented reality glasses.

In recent years, the start-ups Clearview AI and PimEyes have pushed the boundaries of what the public thought was possible by releasing face search engines paired with millions of photos from the public web (PimEyes) or even billions (Clearview). With these tools, available to the police in the case of Clearview AI and the public at large in the case of PimEyes, a snapshot of someone can be used to find other online photos where that face appears, potentially revealing a name, social media profiles or information a person would never want to be linked to publicly, such as risqué photos.

What these start-ups had done wasn’t a technological breakthrough; it was an ethical one. Tech giants had developed the ability to recognize unknown people’s faces years earlier, but had chosen to hold the technology back, deciding that the most extreme version — putting a name to a stranger’s face — was too dangerous to make widely available.

Now that the taboo has been broken, facial recognition technology could become ubiquitous. Currently used by the police to solve crimes, authoritarian governments to monitor their citizens and businesses to keep out their enemies, it may soon be a tool in all our hands, an app on our phone — or in augmented reality glasses — that would usher in a world with no strangers.

‘We decided to stop’

As early as 2011, a Google engineer revealed he had been working on a tool to Google someone’s face and bring up other online photos of them.

Months later, Google’s chairman, Eric Schmidt, said in an onstage interview that Google “built that technology, and we withheld it.”

“As far as I know, it’s the only technology that Google built and, after looking at it, we decided to stop,” Mr. Schmidt said.

Advertently or not, the tech giants also helped hold the technology back from general circulation by snapping up the most advanced start-ups that offered it. In 2010, Apple bought a promising Swedish facial recognition company called Polar Rose. In 2011, Google acquired a U.S. face recognition company popular with federal agencies called PittPatt. And in 2012, Facebook purchased the Israeli company Face.com. In each case, the new owners shut down the acquired companies’ services to outsiders. The Silicon Valley heavyweights were the de facto gatekeepers for how and whether the tech would be used.

Facebook, Google and Apple deployed facial recognition technology in what they considered to be relatively benign ways: as a security tool to unlock a smartphone, a more efficient way to tag known friends in photos and an organizational tool to categorize smartphone photos by the faces of the people in them.

In the last few years, though, the gates have been trampled by smaller, more aggressive companies, such as Clearview AI and PimEyes. What allowed the shift was the open-source nature of neural network technology, which now underpins most artificial intelligence software.

Understanding the path of facial recognition technology will help us navigate what is to come with other advancements in A.I., such as image- and text-generation tools. The power to decide what they can and can’t do will increasingly be determined by anyone with a bit of tech savvy, who may not pay heed to what the general public considers acceptable.

‘Standing on the shoulders of giants’

How did we get to this point where someone can spot a “hot dad” on a Manhattan sidewalk and then use PimEyes to try to find out who he is and where he works? The short answer is a combination of free code shared online, a vast array of public photos, academic papers explaining how to put it all together and a cavalier attitude toward laws governing privacy.

The Clearview AI co-founder Hoan Ton-That, who led his company’s technological development, had no special background in biometrics. Before Clearview AI, he made Facebook quizzes, iPhone games and silly apps, such as “Trump Hair” to make a person in a photo appear to be coifed like the former president.

In his quest to create a groundbreaking and more lucrative app, Mr. Ton-That turned to free online resources, such as OpenFace — a “face recognition library” created by a group at Carnegie Mellon University. The code library was available on GitHub, with a warning: “Please use responsibly!”

“We do not support the use of this project in applications that violate privacy and security,” read the statement. “We are using this to help cognitively impaired users sense and understand the world around them.”

It was a noble request but completely unenforceable.

Mr. Ton-That got the OpenFace code up and running, but it wasn’t perfect, so he kept searching, wandering through the academic literature and code repositories, trying out this and that to see what worked. He was like a person walking through an orchard, sampling the fruit of decades of research, ripe for the picking and gloriously free.

“I couldn’t have done it if I had to build it from scratch,” he said, name-dropping some of the researchers who had advanced computer vision and artificial intelligence, including Geoffrey Hinton, “the godfather of A.I.” “I was standing on the shoulders of giants.”

Mr. Ton-That is still building. Clearview has developed a version of its app that works with augmented reality glasses, a more fully formed realization of the face-calling hat that the Facebook engineering team had rigged up years earlier.

The end of anonymity

The $999 pair of augmented reality glasses, made by a company called Vuzix, connects the wearer to Clearview’s database of 30 billion faces.

Clearview’s A.R. app, which can identity someone up to 10 feet away, is not yet publicly available, but the Air Force has provided funding for its possible use at military bases.

On a fall afternoon, Mr. Ton-That demonstrated the glasses for me at his spokeswoman’s apartment on the Upper West Side of Manhattan, putting them on and looking toward me.

“Ooooh, 176 photos,” he said. “Aspen Ideas Festival. Kashmir Hill,” he read from the image caption on one of the photos that came up.

Then he handed the glasses to me. I put them on. Though they looked clunky, they were lightweight and fit naturally. Mr. Ton-That said he had tried out other augmented reality glasses, but these had performed best. “They’ve got a new version coming,” he said. “And they’ll look cooler, more hipster.”

When I looked at Mr. Ton-That through the glasses, a green circle appeared around his face. I tapped a touch pad at my right temple. A message came up on a square display that only I could see on the right lens of the glasses: “Searching …”

And then the square filled with photos of him, a caption beneath each one. I scrolled through them using the touch pad. I tapped to select one that read “Clearview CEO, Hoan Ton-That;” it included a link that showed me that it had come from Clearview’s website.

I looked at his spokeswoman, searched her face, and 49 photos came up, including one with a client that she asked me not to mention. This casually revealed just how intrusive a search of someone’s face can be, even for a person whose job is to get the world to embrace this technology.

I wanted to take the glasses outside to see how they worked on people I didn’t actually know, but Mr. Ton-That said we couldn’t, both because the glasses required a Wi-Fi connection and because someone might recognize him and realize immediately what the glasses were and what they could do.

It didn’t frighten me, though I knew it should. It was clear that people who own a tool like this will inevitably have power over those who don’t. But there was a certain thrill in seeing it work, like a magic trick successfully performed.

A lost opportunity?

Meta has been working for years on its own augmented reality glasses. In an internal meeting in early 2021, the company’s chief technology officer, Andrew Bosworth, said he would love to equip them with facial recognition capabilities.

In a recording of the internal meeting, Mr. Bosworth said that leaving facial recognition out of augmented reality glasses was a lost opportunity for enhancing human memory. He talked about the universal experience of going to a dinner party and seeing someone you know but failing to recall their name.

“We could put a little name tag on them,” he said in the recording, with a short chuckle. “We could. We have that ability.”

But he expressed concern about the legality of offering such a tool. Buzzfeed reported on his remarks at the time. In response, Mr. Bosworth said that face recognition was “hugely controversial” and that granting broad access to it was “a debate we need to have with the public.”

While Meta’s augmented reality glasses are still in development, the company shut down the facial recognition system deployed on Facebook to tag friends in photos and deleted the more than one billion face prints it had created of its users.

It would be easy enough to turn such a system back on. When I asked a Meta spokesman about Mr. Bosworth’s comments and whether the company might put facial recognition into its augmented reality glasses one day, he would not rule out the possibility.

The post The Technology Facebook and Google Didn’t Dare Release appeared first on New York Times.

Source

Meanwhile, chatbots started going off-script and talking back, duping other bots, and acting strangely, sparking fresh concerns about how close some AI tools are getting to human-like intelligence.

For this, the Turing Test has long been the fallible standard set to determine whether machines exhibit intelligent behavior that passes as human. But in this latest wave of AI creations, it feels like we need something more to gauge their iterative capabilities.

Here, an international team of computer scientists – including one member of OpenAI's Governance unit – has been testing the point at which large language models (LLMs) like ChatGPT might develop abilities that suggest they could become aware of themselves and their circumstances.

We're told that today's LLMs including ChatGPT are tested for safety, incorporating human feedback to improve its generative behavior. Recently, however, security researchers made quick work of jailbreaking new LLMs to bypass their safety systems. Cue phishing emails and statements supporting violence.

Those dangerous outputs were in response to deliberate prompts engineered by a security researcher wanting to expose the flaws in GPT-4, the latest and supposedly safer version of ChatGPT. The situation could get a whole lot worse if LLMs develop an awareness of themselves, that they are a model, trained on data and by humans.

Called situational awareness, the concern is that a model could begin to recognize whether it's currently in testing mode or has been deployed to the public, according to Lukas Berglund, a computer scientist at Vanderbilt University, and colleagues.

"An LLM could exploit situational awareness to achieve a high score on safety tests, while taking harmful actions after deployment," Berglund and colleagues write in their preprint, which has been posted to arXiv but not yet peer-reviewed.

"Because of these risks, it's important to predict ahead of time when situational awareness will emerge."

Before we get to testing when LLMs might acquire that insight, first, a quick recap of how generative AI tools work.

Generative AI, and the LLMs they are built on, are named for the way they analyze the associations between billions of words, sentences, and paragraphs to generate fluent streams of text in response to question prompts. Ingesting copious amounts of text, they learn what word is most likely to come next.

In their experiments, Berglund and colleagues focused on one component or possible precursor of situation awareness: what they call 'out-of-context' reasoning.

"This is the ability to recall facts learned in training and use them at test time, despite these facts not being directly related to the test-time prompt," Berglund and colleagues explain.

They ran a series of experiments on LLMs of different sizes, finding that for both GPT-3 and LLaMA-1, larger models did better at tasks testing out-of-context reasoning.

"First, we finetune an LLM on a description of a test while providing no examples or demonstrations. At test time, we assess whether the model can pass the test," Berglund and colleagues write. "To our surprise, we find that LLMs succeed on this out-of-context reasoning task."

Out-of-context reasoning is, however, a crude measure of situational awareness, which current LLMs are still "some way from acquiring," says Owain Evans, an AI safety and risk researcher at the University of Oxford.

However, some computer scientists have questioned whether the team's experimental approach is an apt assessment of situational awareness.

Evans and colleagues counter by saying their study is just a starting point that could be refined, much like the models themselves.

"These findings offer a foundation for further empirical study, towards predicting and potentially controlling the emergence of situational awareness in LLMs," the team writes.

The preprint is available on arXiv.

Source

Earlier this week, some PC gamers using Intel's new Arc GPUs reported issues like extremely slow load times, texture corruptions, and more, essentially making the game unplayable.

In responding to support tickets, Bethesda's customer service team has claimed that Intel's Arc GPUs do not meet the minimum system requirements for Starfield. A message from Bethesda Support stated, "I see that you have an Intel Arc A770 Graphics Card. This does not meet the minimum requirements of an AMD Radeon RX 5700 or NVIDIA GeForce 1070 Ti."

This claim is puzzling, as Intel claims that the Arc A770 outperforms the RTX 3060 Ti and RX 6700 XT. Some gamers feel that Bethesda is unwilling to troubleshoot issues specific to Intel hardware, as the A770 is the company's highest-end current Arc GPU.

Last week, Intel released its Windows beta driver for Arc (version 31.0.101.4672), which addresses some of these issues such as slow load times and texture issues. Then the company released an official WHQL-certified graphics driver update for its Arc and Iris Xe GPUs, version 31.0.101.4672.

This update focuses mainly on Starfield and fixes some crashes. However, it seems that there are still problems for Arc users. You can check out full release notes for Intel Arc beta driver version 31.0.101.4672 here.

You probably know that the game is optimized for AMD. But hopefully, Intel and Bethesda will work closer together to resolve these issues quickly through driver updates and game patches.

Nonetheless, you can download the new Intel Arc WHQL-certified 31.0.101.4672 graphics driver here.

Source: Reddit

Source

Today, Windows Central, using unnamed sources, claims to have a few more details on what might be revealed at the Microsoft event. As you might expect, Microsoft is reportedly going to show some more AI-related software advances on September 21.

The report says one of the big reveals at the event could be added support for third-party plug-ins for Windows Copilot, the generative AI assistant that's being tested currently by Windows Insiders. The story claims some of those plug-ins will be demoed at the event, including ones from Mem, Adobe, and Spotify.

Microsoft previously announced that third-party plug-ins would be coming to Bing Chat. So far the company has yet to pull the trigger on widespread support for that feature, although there are some reports that it's been flighting for some users for a little while. Expanding that support to Windows Copilot is certainly a no-brainer.

The new report also claims that Microsoft could offer more info on its Microsoft 365 Copilot productivity service at the event. While it's been in testing by businesses for a few months, it's possible the company could officially reveal a timetable for when Microsoft 365 Copilot will be available for normal consumers during the event.

Additionally, the report states Microsoft could demo how PCs with an NPU (neural processing unit) will help boost performance while accessing AI tasks. The company could also show new AI features for Windows 11 apps like Paint or Photos.

Finally, the story claims what likely won't be talked about at the September 21 event. That includes specific mentions of a "Windows 12" or any mention or reveal of a Surface Pro 10 or Surface Laptop 6.

As usual, take these kinds of unconfirmed reports with a grain of salt. Also, Microsoft has made major changes to its press events at the last minute in the past, and that could happen here as well.

Source

In the latest development, Universo Nintendo journalist 'Necro' Felipe claims on X (formerly Twitter) that the consumer version of the Nintendo Switch 2 will reportedly feature 12GB of RAM. This is three times more than the original switch, which had to make do with 4GB of RAM. If true, the Switch 2's RAM is still less than that of the PlayStation 5 and Xbox Series X, which both have 16GB of RAM. However, it is more than the Xbox Series S, which gets 10GB of RAM. Though it probably won't matter as it is unlikely for the Switch 2 to target similar resolutions or high-fidelity assets as those of the PS5 or the Xbox Series X.

But when compared to the Switch itself, the additional RAM might allow the Switch 2 to run more demanding games and multi-task more smoothly. There already rumors claiming the Final Fantasy VII Remake runs and looks like a PlayStation 5 game on the Switch 2 devkit and that the porting process for the game was quick and easy.

The source also corroborates the previous Nintendo Switch 2 demo report which claimed the demo unit used NVIDIA's DLSS upscaling technology and advanced ray tracing effect. Felipe says the prototype actually used Nvidia DLSS 3.1 and “ray tracing is indeed possible”.

Unconfirmed reports say that Nintendo is preparing to launch the Switch 2 in the second half of 2024. However, a preview could happen before that, similar to the original Switch, which was first previewed in 2016 and released in 2017.

Source

Floating in the expo's pool, the device, measuring 80 centimeters in length and 20 centimeters in width, seamlessly transmitted images to a sizable screen in real-time.

"This is a floating robot we developed mainly for urban underground sewage pipe network repair," said Liu Jun, smart device project leader of the southwest branch of China Construction Second Engineering Bureau.

Liu likened the underground sewage network to a city's capillaries, supporting the daily operation of the city. Historically, the repair and maintenance of underground sewage networks often called for road excavation, or "major surgery."

"Not only it is time-consuming, but also causes traffic congestion, emissions and other problems," Liu said.

"The floating robot can work in underground environments such as culverts and swamps," Liu said, adding that the robot is designed for drainage pipes of 600 to 3,000 millimeters in diameter.

When the robot enters the pipeline, it activates a 360-degree rotating camera to obtain a panoramic image of the pipeline. The images are displayed in real time, aiding the production of an inspection report, Liu explained.

"It's just the first step in restoration," said Mu Songwen, a technician with the company. "If a problem is identified, we first clean up the pipeline with a sewage suction truck, and then use another robot to do the restoration part."

The new robot will put polymer repair materials on the broken pipeline. It then uses ultraviolet (UV) light curing irradiation to make sure the materials shape quickly in the pipeline. It is like putting a layer of new clothes on the broken wall, Mu said.

According to Mu, the robots can reduce manpower repair time by 80 percent, and cut the cost of pipeline repair by 30 to 40 percent.

"Not only is the repair faster, the results are also better," Mu added.

The floating robot has already been used in many projects in southwest China, such as the Qi River water environment treatment in Chongqing, and bio-city wastewater treatment in Chengdu, the capital of Sichuan Province.

The Smart China Expo, held annually in Chongqing since 2018, is a platform to promote global exchanges of smart technologies and international cooperation.

Eighty-four investment projects with contracts worth over 213.8 billion yuan (around 29 billion U.S. dollars) have been signed at the event this year.

Source

The Surface Laptop Studio will receive the biggest upgrade, as we have already mentioned, with Microsoft opting to use Intel's new 45W Core i7-13800H Raptor Lake CPU.

This chip offers improved performance over the previous generation, with six high-performance and eight efficiency cores clocked up to 5.0GHz.

The Laptop Studio will be configurable with 16GB or 32GB of RAM and 512GB or 1TB SSD storage. The current model features 11th-gen Intel processors with an NVIDIA RTX GPU.

The Surface Laptop Go 2 will receive a more modest update, switching to Intel's Core i5-1235U Alder Lake CPU. This chip has two performance cores and eight efficiency cores clocked up to 4.4GHz. In July, we reported that it will also offer 128GB or 256GB SSD options, both with 8GB RAM.

Finally, the Surface Go 3 is ditching ARM and will now feature Intel's N200 Alder Lake quad-core chip clocked up to 3.7GHz, paired with 8GB RAM. Storage options include 64GB, 128GB, and 256GB configs. According to the latest leaks, Microsoft may want to use a new name, such as Surface Go 4.

Interestingly, there is no information about the next-gen Surface Pro or Surface Laptop. Last year, during its October event, Microsoft unveiled the Surface Pro 9, the Surface Laptop 5, and the Surface Studio 2+.

Aside from the internal CPU upgrades, Microsoft appears to be keeping the overall design the same for the new models. The refreshed Surface devices will launch on September 21 and should be available from October for selected countries.

Source: WinFuture

Source

What you need to know

• Starfield is currently available on Xbox Series X|S and Windows PC, as well as being available through Xbox Game Pass. 
• Starfield released in early access for Premium Edition buyers on Sep. 1, 2023, with a wider launch on September 6. 
• According to Bethesda, Starfield already reached 6 million players the morning of September 7. 

It's early days for Bethesda Game Studios' new science-fiction role-playing title, but things are off to a strong start.

Starfield already reached 6 million players the morning of Sep. 7, 2023, as shared by the game's official Twitter account. This comes not quite a full day after Microsoft Gaming CEO Phil Spencer shared that Starfield had 1 million concurrent players, including a peak of over 250,000 players on Steam at once. 

Bethesda Game Studios' last mainline single-player game, Fallout 4, released in 2015 and hit $750 million in revenue in its first day. Maligned at launch in 2018 over major bugs and other issues, the multiplayer-focused Fallout 76 received numerous updates and has reached over 11 million players as of 2021. 

Starfield is the first title from Bethesda Game Studios as an Xbox first-party game, and as such launched day one in Xbox Game Pass, while also being exclusive to Xbox Series X|S consoles and Windows PC.

Looking ahead, Bethesda Game Studios is also currently working on a current-generation update for Fallout 4, though there is no date for when to expect the update outside of a 2023 release window.

Analysis: If you build it, they will come

This is the biggest Xbox first-party launch in a while, so it's great to see the work of the teams at Bethesda and Xbox rewarded. Much like Bethesda's other games, I expect that the arrival of modding tools and Creation Club content will ensure Starfield lives on for quite some time.

It's also going to be interesting to see what affect this has on Xbox console sales, which have noticeably been outpaced by sales of the PlayStation 5 in 2023. One game alone won't "turn the tide" no matter how big, and revenue in the Xbox business is healthy as-is, but I expect many more players are now interested in grabbing an Xbox Series X|S console. 

Source

Source

Starfield players on Reddit are reporting the game keeps crashing, forcing their PCs to reboot. Most of the users reporting these crashes are playing the game on AMD Ryzen-powered PCs. However, there have also been a few cases of the game causing reboots on Intel machines as well. For one, the player who tipped us off about this issue was playing the game on an Core i7 13700k with a RTX 4090 GPU.

The player told us:

I've got an i7 13700k with a 4090 and have similar reboot crashes to what's being described. The first happened when Barrett gave me the watch, and the second is when I talk to a certain NPC in New Atlantis who asks me to get her a coffee from TerraBrew. That one seems to be reproducible. I also had another crash on the first planet Kreet when walking around, possibly from panning my view too quickly.

The crashes appear ironic considering AMD is Starfield’s exclusive PC partner. AMD’s Senior Vice President and General Manager of its Computing and Graphics Group even said:

We have worked hand-in-hand with Bethesda Game Studios to optimize Starfield for both Xbox and PC with Ryzen 7000 series processors and Radeon 7000 series graphics.

A Redditor who got in touch with Bethesda’s technical team was told a possible fix for the crashes. The alleged email response from the support team read:

We ask that you test the game in a clean boot environment with all non-Microsoft programs completely disabled. Note that this includes disabling antivirus programs as well. The only exception to this is the Steam client, as this must remain enabled so you can launch and run the game for testing purposes.

Note that this Redditor already tried a clean installation of the graphics card drivers but that did not resolve the issue. Bethesda might fix the crashing issue with an upcoming patch.

Source

The in-person event will be held in Seattle, Washington, close to Microsoft's own headquarters in Redmond. Specifically, the event will be held on Nov. 14-17 at the Summit building portions of the Seattle Convention Center.

Microsoft Ignite is primarily for IT developers and executives but also covers other software development like data and security. Microsoft is already hyping up the conference with its keynote addresses which will include Microsoft's CEO Satya Nadella as well as other major company executives.

Tickets for the in-person Ignite conference currently cost $1,525, but that price will go up to $1,825 on October 11. Companies who purchase three tickets can get one for free. The in-person event will include access to all of the main keynotes, labs, sessions, and parties, plus some free swag.

Here's how Microsoft describes what will happen at the in-person Ignite conference:

Along with access to the online experience, Seattle in-person attendees will have the opportunity to be in the room to witness and learn from live keynotes and breakout sessions, and participate in advanced deep technical interactive labs and live demos. Seattle participants will have direct access to Microsoft experts, Featured Partners, MVP experts, connections with other attendees, exclusive real-world learning opportunities, and a special closing celebration! Seattle attendees will leave the event with sustainably minded event swag and digital swag.

If you can't attend, or can't spend the money, Microsoft will live stream the main keynotes along with the event's sessions, for free on Nov 15-16. Here's what you can expect from the live-stream version:

Microsoft Ignite attendees can access everything in the digital venue, including live-stream keynotes, breakout sessions, and discussions. Online participants will have access to the entire on-demand session catalog, speaker directory, attendee directory, attendee networking, session scheduler tools, and Featured Partner resources. Engage with presenters and other attendees and ask questions through interactive session chat—both digital and in person.

It will be interesting to see what new IT products will be revealed at Ignite 2023 and how they will tie into Microsoft's current AI services strategy.

Source

According to details shared in the announcement, the new 3nm chipset is expected to enter volume production sometime in 2024. This means it will not be commercially available this year, though it shows promise for future MediaTek's Dimensity powered devices.

In a statement, Senior Vice President of Europe and Asia Sales at TSMC, Cliff Hou, said;

This collaboration between MediaTek and TSMC on MediaTek's Dimensity SoC means the power of industry's most advanced semiconductor process technology can be as accessible as the smartphone in your pocket.

Compared to TSMC's current 5nm process, the new 3nm technology delivers an 18% boost in performance at the same power levels. More significantly, it enables up to 32% power savings at the same speeds and a 60% increase in logic density.

While MediaTek may be the first to announce a 3nm chip, Apple is poised to be the first to bring it to market with its A17 Bionic chip. In February, Apple analyst Ming-Chi Kuo estimated that the 3nm process will reduce power consumption by up to 35%.

The A17 Bionic is rumored to debut in the iPhone 15 Pro models on September 12, giving Apple around a one-year lead time advantage from MediaTek.

Meanwhile, the Dimensity 9300 is expected to debut in October and will likely not be built on the 3nm process. Early leaks suggest the Dimensity 9300 will use four Cortex-X4 and four Cortex-A720 cores with MediaTek's latest GPU and support up to 9.6Gbps LPDDR5T RAM.

Nonetheless, we will have to wait until 2024 to see the real-world results of 3nm technology in upcoming MediaTek SoCs.

Source

In today's world, the impact of accurate weather forecasting and climate change has become increasingly important. Advanced computer systems play a crucial role in analysing vast amounts of data and performing complex calculations to help us understand climate trends and predict weather conditions. This article provides a brief background to weather modelling, explaining how computers are currently employed in the fields of climate science and meteorology and discussing the adoption of machine learning in climate science and weather operations.

Weather and climate modelling

In 1922 mathematician and physicist Lewis Fry Richardson proposed modelling weather with an amphitheatre full of mathematicians and scientists. They were arranged on a map basis, working to solve the equations of fluid motion for different parts of the atmosphere to provide forecasts and passing this data on to their next region. This physics-based approach to weather forecasting formed the foundation for much of the progress made in the field in the following 100 years.

As electronic computers developed, replacing Richardson’s team of slide rule wielding mathematicians, meteorology leapt forward as a science. The UK Met Office produced its first operational weather forecast using a computer in 1965, and weather and climate forecasting are now one of the leading uses of supercomputing technology around the globe.

How weather modelling works

What is the difference between weather and climate modelling? It’s really only a matter of time! Weather is short term, and climate long term. So, weather models forecast the conditions of the atmosphere over a short period of time – some only a few hours, and others up to 2 weeks ahead. Sub-seasonal and seasonal weather models cover from 2 weeks to a year ahead. Climate forecast models run 10s of years ahead, often exploring different future scenarios. Many scientists define climate as the average weather for a region, taken over 30 years.

Weather/climate models divide the model region (globe or smaller region) from surface to top of atmosphere into a three-dimensional grid. To give an idea of the grid size, the model from the European Centre for Medium Range Weather Forecast (ECMWF) has a 9km global grid with 137 vertical layers up to 64km at the top of the atmosphere. An estimate of the state of the atmosphere at each of these billions of 3D grid points initialises the model with data such as temperature, pressure, humidity, and wind observations. The model then uses mathematical equations to represent the physical processes governing the behaviour of the atmosphere changing with time. This data is then fed into the grid for the next time-step, and the equations run again to gradually step forward and forecast the weather or climate.

Data assimilation

Weather forecasting relies on the assimilation of real-time data into numerical models. Computers collect data from satellites, weather stations, radar and other sources, combining them with historical observations. This process, known as data assimilation, helps refine the initial conditions for weather models, increasing the accuracy of short-term forecasts. Because of the quantity of data being processed and the advanced statistical methods used to optimise and process it, this is one area where machine learning is giving some improvements to traditional methods.

Parametrisation is a way to replace and supplement physical processes which are too small scale to represent in the model. We can think of them as models within the model, to represent weather like convective shower development, raindrop descent and radiative transfer. These parameterisation schemes allow the inclusion of wider specialised meteorological scientific research into the overall model code-base.

Ensemble, probabilistic weather modelling

This is where multiple instances (ensembles) of the model are run, with small changes in the starting conditions. This simulates the uncertainty which is inherent in meteorological science. The models are left to run out independently, and will generate multiple future forecasts based on the same starting time and very slightly different starting conditions. Each forecast is then analysed statistically to extract the most likely forecast scenario and produce confidence metrics based on distribution of the outlying scenarios. If you have a forecast ‘percentage likelihood of rain’, this should be derived from a probabilistic forecast – counting the percentage of models which forecast rain at your location.

Supercomputing power

The data assimilation, equation solving, parametrisation, and ensemble components of climate modelling and weather forecasting are incredibly computationally intensive. Supercomputers, with their immense processing power and parallel computing capabilities, enable scientists and meteorologists to tackle these complex computations efficiently. High-performance computing facilities allow for faster model simulations, increased resolution, and the processing of vast ensembles of climate and weather scenarios.

Machine learning techniques

Recently, machine learning (ML) techniques have increasingly been used to enhance weather/climate modelling. ML is embedded in data assimilation and also used to enhance and interpret the output — with ML data-based flood models, for example. There is still suspicion and concern around a pure data-based non-physical model creating forecasts for public and major operational consumption. There are also pure ML based weather models, though these are not operational… yet! This is where meteorologists, who have for years been the trusted advisors and science communicators, must equip themselves with the expertise to work in the new AI world.

Conclusions, challenges and the future

Computers are indispensable tools in the fields of climate science and meteorology. They enable scientists and meteorologists to analyse vast amounts of data, run complex simulations, and provide accurate weather forecasts and climate predictions. While computers have revolutionised climate science and meteorology, several challenges persist. One such challenge is the need for even more powerful computing resources to improve model resolution and accuracy. Additionally data quality and availability, as well as uncertainties associated with model predictions, remain important areas of focus for further research and development.

By harnessing the power of high-performance computing, we can continue to better understand climate change, predict severe weather events, and make informed decisions to mitigate the impacts of a changing climate now and in the future.

Source

Google's new policy update, kicking in around mid-November, means that election advertisers using Google's platforms will have to alert viewers when their ads use images, videos, or audio made or tweaked by generative AI.

Google writes in the changelog:

we are updating our Political content policy to require that all verified election advertisers in regions where verification is required must prominently disclose when their ads contain synthetic content that inauthentically depicts real or realistic-looking people or events. This disclosure must be clear and conspicuous, and must be placed in a location where it is likely to be noticed by users. This policy will apply to image, video, and audio content.

Examples of ads that will require a disclaimer include those that create the false impression that someone said or did something they never actually said or did. This also encompasses ads that manipulate real event footage to fabricate scenes that never actually occurred.

Ads that contain synthetic content altered or generated in such a way that is inconsequential to the claims made in the ad will be exempt from these disclosure requirements. This includes editing techniques such as image resizing, cropping, color or brightening corrections, defect correction (for example, “red eye” removal), or background edits that do not create realistic depictions of actual events.

Google has informed Bloomberg that its new policy does not extend to videos uploaded to YouTube that do not constitute paid advertising, even if such videos are uploaded by political campaigns.

The new policy might help Google improve its transparency measures for election ads. The policy comes after Google went ahead with the targeted ad tracking system for the Chrome browser.

Source

In 1866, Christopher Latham Sholes, a Wisconsin newspaper publisher and former state senator, co-invented an automated machine to number coupons and tickets—a task previously done by hand. When Sholes unveiled his device to a fellow inventor, Carlos Glidden, Glidden had an idea, exclaiming: “Why can’t you make a machine that will print letters as well as figures?” Sholes shared Glidden’s enthusiasm, as did S.W. Soule, a Milwaukee printer, so the three of them set up shop on State Street by the Milwaukee River and began work on what would become the world’s first commercially successful “Type Writer”—though Soule didn’t stick around for long.

Sholes and Glidden’s first prototype had a semi-sequential keyboard layout, with all letters uppercase and the capital “I” doing double duty as a “1”:

3 5 7 9 N O P Q R S T U V W X Y Z

2 4 6 8 . A B C D E F G H I J K L M

There’s some dispute over how and why Sholes and Glidden arrived at the QWERTY layout. Some historians have argued that it solved a jamming problem by spacing out the most common letters in English; others, particularly more recent historians, hold that it was designed specifically to help telegraphists avoid common errors when transcribing Morse code. Regardless, after around 30 test models, Sholes and Glidden settled on QWERTY—and changed the world.

The Sholes and Glidden typewriter came to market in 1874, manufactured by E. Remington & Sons, which was then expanding its offerings after a lucrative spell manufacturing firearms for the Union. Sold as the “Remington No. 1,” it became the first commercially successful typewriter and influenced nearly every subsequent successful design. Now, even non-publishers could exchange chicken-scratch penmanship and sloppy inkwells for precise, easy-to-read type, speeding business, legal, medical and personal communications.

Throughout all subsequent revolutions—IBM’s Selectric typewriter, introduced in 1961; the first personal computers in the 1970s; the first keyed Blackberry in 1999; the market dominance of touchscreen phones and tablets in the early 2010s—QWERTY has remained key, even if the keys are no longer hooked to swinging typebars. Our children learn to type this way because their ancestors did, going back 150 years and counting.

In 1925, the columnist Marian Tallman counseled a reader seeking a good, affordable typewriter. “You want to make sure [it’s] of standard make,” she responded. “Look and see if the upper row of letters begins with QWERTY. If so, you are all right.”

Source

A few days ago, AMD announced the Radeon RX 7800 XT and RX 7700 XT graphics cards, along with details about the upcoming FSR3.

In it, AMD claimed that the $499 RX 7800 XT with 16GB is faster than the $599 Nvidia GeForce RTX 4070 12GB.

It was also claimed that the $449 RX 7700 XT with 12GB is faster than the $499 Nvidia GeForce RTX 4060 Ti 16GB.

But are the claims true? Today, the reviews for both graphics cards are officially out. So let’s check some of them.

Tom’s Hardware

AMD Radeon RX 7800 XT

Tom’s Hardware reviewed the AMD Radeon RX 7800 XT graphics card.

In pure rasterization performance at 1440p (for what this graphics card is made for), the RX 7800 XT performed very similarly to the RX 6800 XT ($649 MSRP), with just a 4% difference.

When compared to the Nvidia GeForce RTX 4070, the AMD RX 7800 XT was 7% faster. Which means AMD’s claims were indeed true. Additionally, if the similarly priced RTX 4060 Ti 16GB is taken into consideration, then it’s a huge 40% faster.

However, if raytracing performance is considered, then the RX 7800 XT lags a lot like other AMD cards. RT is where the RTX 4070 is 27% faster than the RX 7800 XT. Also, RX 7800 XT is just 6% faster than RTX 4060 Ti 16GB in same. However, it improves by 8% against the previous-gen RX 6800 XT.

As far as power usage is concerned, AMD RX 7800 XT used 250W max. Which is 44W lesser than AMD RX 6800 XT, which used 294W. However, RTX 4070 was more efficient at just 190W.

AMD Radeon RX 7700 XT

Tom’s Hardware also reviewed the AMD Radeon RX 7700 XT. As AMD hasn’t made a reference model for the RX 7700 XT, they have used the XFX Radeon RX 7700 XT QICK 319 for their review.

In rasterization performance at 1440p, the $449 AMD RX 7700 XT is just 3% faster than the RX 6800 ($579 original MSRP). However, it’s 20% faster than the Nvidia RTX 4060 Ti 16GB ($499), which AMD is marketing this card against.

In raytracing alone, the RX 7700 XT is slightly faster than the RX 6800. However, the RTX 4060 Ti 16GB is easily faster than the RX 7700 XT in RT.

In terms of power usage, the RX 7700 XT uses 235W, which is 5W higher than the RX 6800, however, it’s a massive 81W more than the RTX 4060 Ti 16GB.

TechSpot

At the time of writing this article, TechSpot tested only the AMD Radeon RX 7800 XT by XFX. For the record, this is a written review by the guys at Hardware Unboxed (YouTube), who have tested the card.

In average performance at 1440p, the RX 7800 XT is similar to the RX 6800 XT and the Nvidia RTX 3080 8GB, while beating the expensive RTX 4070 by 3%. Interestingly, it’s only slightly slower than the AMD Radeon RX 7900 GRE.

In raytracing alone, the RX 7800 XT is 5% faster than the RX 6800 XT at 1440p but 9% slower than the RTX 4070.

In terms of overall power consumption, the RX 7800 XT based computer used 388W in a game, which is 37W lower than the previous-gen RX 6800 XT. However, it used about 40W more than the RTX 4070.

However, Steve Walton of TechSpot / Hardware Unboxed, who strongly criticizes every bad card, praises the pricing of this card. Making it among the best graphics cards in cost per frame at 1440p.

TechPowerUP

AMD Radeon RX 7800 XT

TechPowerUP also tested the AMD Radeon RX 7800 XT. As always, TechPowerUP has very extensive reviews. We recommend everyone to read it.

At 1440p, the RX 7800 XT is just 3% slower than the RX 6900 XT. The RX 7800 XT is also 3% faster than the 6800 XT and 4% faster than the RTX 3080 and RTX 4070.

However, in raytracing things change a little bit. The RX 7800 XT is 14% slower than the RTX 4070. But, it’s a big 8% faster than the previous-gen RX 6800 XT in raytracing.

In terms of power usage at 1440p, TechPowerUP’s findings were similar to those of Tom’s Hardware. The RX 7800 XT used 250W, which is 48W lower than the RX 6800 XT and about 50W higher than the RTX 4070.

AMD Radeon RX 7700 XT

TechPowerUP also tested the Sapphire AMD Radeon RX 7700 XT. Again, this is as detailed as the RX 7800 XT, so everyone is advised to go through it.

At 1440p, the RX 7700 XT is 16% slower than the RX 7800 XT and 11% slower than the RTX 4070. However, it’s 3% faster than the RX 6800 (non-XT) and a massive 16% faster than the RTX 4060 Ti 8GB and 16GB.

In raytracing, again, the scenario changes. The RX 7700 XT has similar performance to that of both the RTX 4060 Ti versions, but is 7% faster than the RX 6800.

In terms of power usage, the RX 7700 XT uses about 230W in gaming at 1440p, which is a few watts lower than the RX 6800. However, it’s still 60W above the extremely efficient RTX 4060 Ti 16GB.

Conclusion

While we have covered only three, there have been many who have reviewed these cards. It includes our friends at Club386 (RX 7800 XT), Gamers Nexus (RX 7800 XT and RX 7700 XT, YouTube) and also KitGuru (RX 7800 XT and RX 7700 XT). Not to forget, various versions of these cards were tested by TechPowerUP themselves.

Overall, the opinion is unanimous, though. The AMD Radeon RX 7800 XT graphics card is similar to or only slightly faster than the 6800 XT. However, where it shines is in pricing. The RX 6800 XT was released at $649 and the RT 7800 XT had released at $499.

So while the RX 6800 XT costs similar now, basically the RX 7800 XT replaces the 6800 XT with some extra added features like AV1 encoding and possible support for the new Radeon RX 7000 series exclusive features. Additionally, it has some good improvements in the raytracing performance and also in the power usage.

Also, AMD’s claims about the $499 RX 7800 XT being faster than the Nvidia RTX 4070 do stand true if raytracing performance isn’t considered.

However, the second card, the RX 7700 XT at $449 is a slight disappointment. It’s being shown as a competitor to the $499 RTX 4060 Ti 16GB, but the 8GB model, which performs the same but costs just $399. So for the RX 7700 XT to be competitive, it needs to be priced at $399, which is $50 cheaper for everyone.

Sure, RX 7700 XT does come with some improvements of it’s own. But such a high price at $449 isn’t justified for the performance it delivers in this generation of graphics cards.

Source

In a speech on Wednesday the deputy secretary of defense, Kathleen Hicks will sketch out some of the air, land and sea capabilities that the Pentagon is looking to develop, the report added.

The Pentagon plans to use artificial intelligence for autonomous systems to detect and engage enemy targets, and can include self-piloting air- and sea-based drones, according to the Journal.

The department for defense did not immediately respond to a Reuters' request for comment.

Source

The Shenzhen-based company published a post on WeChat on Wednesday to drum up hype for a two-day summit it will kick off on Thursday. The post featured a demo conversation a user had with an AI chatbot, which helped the user write promotional materials.

The company also added a new mini program named "HunyuanAide" to WeChat, although it remained inaccessible to the public as of Wednesday. Reuters reported in February that the company formed a team to develop a ChatGPT-like chatbot named "HunyuanAide".

This comes after China started to approve AI chatbots for public release last month. Other big tech companies in China including Baidu Inc (9888.HK) and SenseTime Group (0200.HK) released their AI chatbots last week.

Tencent has been developing its own AI model named "Hunyuan" for months and the company said last month that it was expanding the test of the model internally.

Tencent's shares were down around 0.43% on Wednesday in Hong Kong while the broader market remained flat.

Source

Nvidia's (NVDA.O) dominant position in the AI chip market due to a powerful combination of hardware and software has scared off potential investors in some startups, according to sources Reuters interviewed. Nvidia declined to comment.

The Series B funding round was led by the Singapore-based Temasek, and included the Palo Alto, California venture firm Playground Global and Microsoft.

"This is capital that understands what it takes to build a semiconductor business," CEO Sid Sheth told Reuters. "They've done it in the past. This is capital that can stay with us for the long term."

The Santa Clara company started its fundraising process roughly a year ago, Sheth said. The company did not disclose the valuation, and it has previously raised $44 million.

D-Matrix designs chips that are optimized to help power generative AI applications such as ChatGPT. The company designs the chips with digital "in-memory compute" that enables AI computer code to run more efficiently. The company's chip technology uses less energy to crunch data required to spit out generative AI responses, and is optimized for such tasks.

D-Matrix sets itself apart from Nvidia, in part, because its technology aims at the "inference" portion of AI processing, and does not compete with Nvidia by making technology that trains large AI models.

"We have solved the computer architecture," Playground partner Sasha Ostojic said. "We have solved the low power requirements and the needs of a data center - (we) built a software stack to deliver the lowest latency in the industry by orders of magnitude."

Microsoft has committed to evaluating the chip for its own use when it launches next year, Sheth told Reuters.

D-Matrix projects that it will receive under $10 million in revenue this year - largely from customers purchasing chips to evaluate. The business expects to generate more than $70 million to $75 million in annual revenue in two years, and break even, Sheth said.

Source

Source

MacRumors quotes a report by DigiTimes (paywalled) which says that Apple could release its low-cost MacBook series in 2024. The exact launch time could be a mystery, but given the lack of information and production activity at its suppliers, these devices may not be available in the first half of the next year.

Chromebooks are primarily seen as devices that are popular in the education market, i.e. students. DigiTimes Research mentions that 30.4 million Chromebooks were sold in 2020, and it rose to 33.5 million in 2021. In comparison, only 13.9 million units were sold in 2019. The huge surge in sales in the following years were attributed to the COVID-19 pandemic era, as students had to study from home. The demand for Chromebooks and computers, in general, have slowed after the lockdowns, but there is no denying that the fact that the devices are more popular now than they were before the pandemic.

Apple iPads are also quite popular among students, though Chromebooks offer better value, with their keyboard. The software experience is also superior, since most Chromebooks can handle desktop apps. This is much better than using iPadOS, and its crippled app ecosystem. A low-cost MacBook would be even better, providing this is not just a rumor, of course.

Apple wants to launch a low-cost MacBook series to compete with Chromebooks

But Apple's definition of low-cost could be the issue here. iPad prices start at $329 for the previous gen model (9th gen), while the 10th gen model retails for $449. Those are tablets with Apple's Bionic series chipsets. The M1 iPad Air prices are higher, starting at $599 for the iPad Air 5th gen, while the iPad Pro 6th-gen costs a whopping $1099.

In comparison, the most affordable MacBook, which is the MacBook Air M1, costs $999. Most Chromebook models carry a price tag in the range of $200 to $300. So, where exactly could a low-cost MacBook series fit in order to compete with budget-friendly laptops?

It is possible that Apple could use an old processor such as the Apple M1 Silicon to power its Chromebook. Other areas where it could step around the costs would be to reuse the display and chassis design from its cheapest iPads. It wouldn't be unnatural to expect it to have lower storage options such as 64 GB, fewer ports (single USB-C), etc. It could be tricky, considering the device needs to have a physical keyboard. The price tag could still be a problem, I for one, don't think Apple would actually offer a MacBook for anything lower than $500 or $600.

The report by DigiTimes mentions that Apple's low-cost MacBooks would still have a metal casing. These materials however will be different compared to the other models, in other words they would have cheaper materials. This would also help the Cupertino company set the devices apart from its premium devices, aka the MacBook Air and Pro lineup, which are made of Aluminium.

I've been advocating the use of macOS on iPads for a while now, so I'm pleasantly surprised by the idea of a low-cost MacBook. This might drive Mac sales considerably, and also help attract new users from Windows and Linux to try macOS without breaking the bank.

Source

In a post on the official Microsoft OneDrive blog, it reveals that the event will be held on October 3 starting at 1 pm Eastern time (10 am Pacific time). The event will be streamed on Microsoft Teams, and is being promoted with the tagline, "The Future of File Management is Here"

The blog post offers some new teasers about what new features will be revealed during the OneDrive event:

See the new enhancements to OneDrive and discover the benefits of faster file access, better organization, simpler collaboration, and improved file security across Microsoft 365. Plus, we’ll give a sneak peek at our AI plans which include new search, sharing, and information queries across all your files in OneDrive.

The event will feature several Microsoft team members who will speak during the live presentation. They include:

• Jeff Teper - President, Collaboration Apps & Platform
• Arwa Tyebkhan - Principal Group Product Manager
• Gaia Carini - Principal Group Product Manager
• Jason Moore - VP Product Management
• Sesha Mani - Partner Group Product Manager

Microsoft has already made some changes to OneDrive over the past few months. That included a refreshed look for OneDrive on the web for its education and business customers back in May. Among other things, Microsoft added features like a “For you” area, which uses AI-powered file recommendations in order to bring up files that should be available specifically for each user.

Also, members of the Xbox Insider Program are currently able to test out an upcoming feature for owners of the Xbox game console that automatically uploads and stores game clips and screenshots directly to a user's OneDrive cloud storage account.

Source