In the filing, Microsoft states:

As of September 30, 2023, we believe our allowances for income tax contingencies are adequate. We disagree with the proposed adjustments and will vigorously contest the NOPAs through the IRS’s administrative appeals office and, if necessary, judicial proceedings.

Daniel Goff, Microsoft's Corporate Vice President for Worldwide Tax and Customs, offered some more info about its dispute with the IRS in a new blog post. He stated:

The main disagreement is the way Microsoft allocated profits during this time period among countries and jurisdictions. This is commonly referred to as transfer pricing and the IRS has established regulations that allow companies to use a specific arrangement for transfer pricing, called cost-sharing.

 

Many large multinationals use cost-sharing because it reflects the global nature of their business. Because our subsidiaries shared in the costs of developing certain intellectual property, under those IRS cost-sharing regulations, the subsidiaries were also entitled to the related profits.

Goff claims that Microsoft has made changes to both its practices and its corporate structure in the years since the time period of the IRS's audit. However, he claims that the agency has not taken into account taxes that Microsoft says were paid under the Tax Cuts and Jobs Act (TCJA). He says that the amount would cut the IRS's estimate of taxes owed by Microsoft by as much as $10 billion.

Goff says that Microsoft hopes to work with the IRS "to reach a mutual resolution to this issue". However, he does say that this dispute could take years to conclude through the appeals process.

Source

In May 2023, Dr. Benjamin Lanyon at the University of Innsbruck in Austria took an important step toward creating a new kind of internet: he transferred information along an optical fiber 50 kilometers long using the principles of quantum physics.

Information in quantum physics differs from the units of data—binary digits—stored and processed by computers that form the core of the current World Wide Web. The quantum physics realm covers the properties and interactions of molecules, atoms and even smaller particles such as electrons and photons.

Particle power

Quantum bits, or "qubits," offer the promise of transmitting information more securely because the particles get changed by the act of observing and measuring them. That means an eavesdropper can't go undetected.

Lanyon said his work makes the quantum internet appear feasible within cities, after which longer intercity distances will be the goal.

"You could imagine this being a large-city scale," he said.

His breakthrough was part of an EU research project to bring the goal of a quantum internet closer.

Called the Quantum Internet Alliance, or QIA, the project brings together research institutes and companies across Europe. The initiative is receiving €24 million in EU funding over three and a half years until the end of March 2026.

"It is not meant to replace the classical internet, but to work together," said Stephanie Wehner, a German native who coordinates the QIA and is a professor of quantum information at Delft University of Technology in the Netherlands. "We're not going to replace Netflix."

A key concept in quantum physics is entanglement. If two particles are entangled, no matter how far apart they are in space, they will possess similar properties—for example, both having the same measurement of something called "spin," a quantum version of the direction that the particles are spinning.

The spin state of the particles isn't clear until they are observed. Until then, they're in multiple states called superposition.

But when one is observed, the state of both particles becomes known.

Possibilities aplenty

This is useful in secure communications. People hacking a quantum transmission would leave behind an obvious trace of their attempt by causing a change in the state of an observed particle.

"We can use the properties of quantum entanglement to achieve a means of secure communication that is probably secure even if the attacker has a quantum computer," said Wehner.

The secure communications afforded by a quantum internet could open up a much broader range of applications that are well beyond the bounds of the classical internet.

In medicine, for example, the physics of entanglement allows for a level of clock synchronization that can improve telesurgery.

"If I want to perform surgery on some remote node, I want this to be very precisely timed in order to not make any mistakes," said Wehner.

Astronomy is another potential beneficiary.

Telescopes making distant observations could "use a quantum internet to generate entanglement between the sensors to get a much better image of the sky," Wehner said.

A further example might be ATM machines.

At present, were an ATM to crash when a person was withdrawing money, the machine would assume no cash had been delivered while another dispenser would register a money withdrawal. A quantum internet could remove that discrepancy.

Many applications of a quantum internet will likely become apparent only after the technology is created.

"It offers a whole range of new possibilities for making precise measurements of space and time and studying how the world and the universe work," said Lanyon.

Distance test

The trick now is scaling up a quantum internet to use many particles across long distances.

Lanyon and his team have also demonstrated communicating not just between single particles but also 'trains' of particles—in this case light particles called photons—speeding up the rate of entanglement between quantum nodes.

"If you just sent one photon at a time, you have to wait for the travel time," he said. "But if you can make trains of many photons at once, this allows you to increase the rate of entanglement between quantum nodes for the distances we want."

The ultimate goal is to extend quantum nodes to much bigger ranges, perhaps 500 kilometers, and create a prototype of a quantum internet that can link remote cities—much like the classical internet relies on different nodes to create a global internet.

While a quantum internet could exist for specialized applications as soon as 2029, experts are wary of hazarding a guess about when a full version might be available for a wide range of uses.

"It's a very difficult question," said Wehner.

As the QIA advances the components and systems of the quantum internet, Europe is also working to develop quantum computers themselves.

In June 2023, an EU public-private partnership—the European High Performance Computing Joint Undertaking—announced that six countries in Europe would host quantum computers. The countries are the Czech Republic, France, Germany, Italy, Poland and Spain.

The aim is to ensure that Europe is at the forefront of the quantum technologies revolution. Quantum computers are expected to have unprecedented calculation power with many uses, including the ability to break the cryptographic algorithms that secure most of the exchanges of the current internet.

Crowded field

With projections that half of the most used cryptographic systems will be broken by the end of the decade, Europe is hardly the only interested party.

China and the U.S. have made advances in quantum computing and the quantum internet in recent years.

Back on the infrastructure front, Europe is taking other steps. It's developing an integrated space and terrestrial infrastructure for secure communications—a building block of sorts for the quantum internet.

"I'm very proud to say we are world-leading in many domains," said Wehner.

While in all interested countries much work remains, the potential benefits signal further advances and breakthroughs before too long.

"People are developing new applications of quantum networks at quite a high rate," Lanyon said.

Source

Founded in 2013, Nod.ai builds open-source technologies for AI systems using approaches like reinforcement learning. The company is known for contributing to projects like SHARK, an open-source machine learning distribution built on frameworks like PyTorch and TensorFlow.

AMD said Nod.ai's expertise in AI model optimizations would help strengthen its efforts in providing comprehensive hardware and software solutions for AI applications.

In a blog post, Vamsi Boppana, senior vice president at AMD, said;

The acquisition of Nod.ai is expected to significantly enhance our ability to provide AI customers with open software that allows them to easily deploy highly performant AI models tuned for AMD hardware.

 

The addition of the talented Nod.ai team accelerates our ability to advance open-source compiler technology and enable portable, high-performance AI solutions across the AMD product portfolio. Nod.ai's technologies are already widely deployed in the cloud, at the edge and across a broad range of end point devices today.

Nod.ai founders include Anush Elangovan, formerly of Google and Cisco, and Harsh Menon, previously with electric aircraft startup Kitty Hawk. "At Nod.ai, we are a team of engineers focused on problem solving — quickly – and moving at pace in an industry of constant change to develop solutions for the next set of problems," said Anush Elangovan.

AMD believes the acquisition would augment its AI software portfolio, expecting to close this quarter pending closing conditions. The chipmaker hopes to better compete with Nvidia in the AI space, which has seen strong growth and is dominated by Nvidia.

During its latest quarter, which ended in July 2023, Nvidia reported over 100% revenue growth year-over-year for its data center business, including AI. In comparison, AMD saw an 18% decline in revenue, though it still beat analyst expectations.

Source

Source

More than a year ago, Intel announced the Arc A750 and A770 graphics cards. Initially, the reviews for these graphics cards were very underwhelming. This was because while the GPU inside these graphics card was powerful, the actual performance was very lacking, especially in games which weren’t that new.

Intel quickly started working to fix it. First, it started cutting the retail price of these graphics cards. So much so that the Arc A750, which was launched at an MSRP of $289, is now available for as cheap as $190 at some places.

The second thing Intel did was to improve the performance of these graphics cards. Driver update after another, Intel has consistently kept increasing the performance of the Intel Arc series graphics cards. This is visible in the official benchmarks Intel posts time and again.

Though, issues remain. Like, there are some games that still don’t run properly. Some don’t launch. Many games are having all kinds of various issues. To add to that, all these Intel graphics cards require Re-BAR to be enabled in the BIOS for good performance, which is something that is only available in relatively newer motherboards. Most newer motherboards do support it, though.

Despite all this, Intel Arc A750 and A770 are shaping up to be relatively good graphics cards, especially due to their newer prices. These two graphics cards are also supported by the likes of the cheap Intel Arc A380 graphics card, but that’s more for non-gaming work, like video encoding and decoding.

However, during the time of the original announcement. Intel had also detailed another graphics card. The Intel Arc A580. From then, we didn’t hear anything from Intel about it, until now.

Intel Arc A580 Officially Released

After days of leaks and rumors. Today, Intel released an interesting graphics card driver update. In that update, Intel specified an impressive graphics card performance increase in some games, while adding support for newer ones.

However, that update also mentioned something different. It mentioned an official support for the Intel Arc A580 graphics card. Sure enough. Within hours, Intel officially announced the Intel Arc A580 to the public.

In the announcement, Intel mentioned that the Arc A580 graphics cards will start becoming available from today, that too, worldwide. The Intel Arc A580 will come at an MSRP of $179 and will be sold by brands like ASRock, GUNNIR and Sparkle. This means there will be no official, limited edition, made by Intel card in the Arc A580.

As for the specifications, the Intel Arc A580 uses a cut-down version of the same GPU which is found in the Arc 750 and Arc 770 graphics cards, the ACM-G10. The GPU is manufactured by TSMC’s on it’s N6 process.

It features 24 Xe-cores, 6 Render Slices and 24 Ray Tracing Units. The graphics clock is 1700 MHz (though VideoCardz says that graphics card makers have increased it to 2000 MHz), it will come with a TDP of 185W and supports PCIe 4.0 configuration.

In terms of memory, the Intel Arc comes with 8GB of GDDR6 VRAM running at 16Gbps speed, with a 256-bit bus. The bus part is particularly impressive. More on that later.

The Intel Arc A580 supports hardware based encoding and decoding of H.264, H265 (HEVC), AV1 and VP9. Here too, the AV1 support is important and impressive.

But what about the actual performance benchmarks? Looks like Intel not only made the surprise announcement, it had even sent these graphics cards to some sites for reviews. Let us check out some of them.

Intel Arc A580 Reviews

Tom’s Hardware

Tom’s Hardware reviewed the Sparkle Intel Arc A580.

In raster performance at 1080p, Intel Arc A580 performed almost similarly to AMD RX 6600, which still retails above $210 these days – $30 more expensive than Intel Arc A580. However, while Intel Arc A580 beats Nvidia’s RTX 3050 easily, it performs slower than the RTX 3060 12GB.

In ray tracing performance, Intel Arc A580 performs excellently, thanks to the dedicated ray tracing cores it comes with. In it, Intel Arc A580 beats AMD’s more expensive offerings, but again lags behind Nvidia’s ones. But honestly speaking, there’s hardly any use to look at ray tracing performance at this price range, it’s something best reserved for expensive graphics cards.

At 1440p, the performance is similar in raster, but in ray tracing the Arc A580 improves its lead further against AMD thanks to the larger 256-bit memory bus. Having said, again, at budget graphics card range, checking performance at 1440p isn’t that ideal. This is because sooner or later, newer games will make sure that this graphics card is going to be useful only for 1080p.

In power usage, Intel Arc A580 uses almost 200W, which is very high and nears the power usage of Intel Arc A750 and A770. Nvidia and AMD’s offerings are far more efficient compared to it.

Another issue which Tom’s Hardware found was that the Sparkle Intel Arc A580’s fans were loud, very loud. Though this doesn’t seem to be an issue with Intel Arc A580 from other brands. Also, this can be fixed by changing fan curve, but that requires some medium level of expertise.

TechPowerUP

TechPowerUP reviewed both Sparkle Intel Arc A580 and ASRock Intel Arc A580. We will concentrate on ASRock one this time around.

In terms of raster performance at 1080p, Intel Arc A580 was 20% faster than Nvidia RTX 3050, 11% faster than RTX 2060. However, it was 3% slower than AMD RX 6600, 9% slower than Intel Arc A750 and 10% slower than Nvidia RTX 3060. Here too, the Intel Arc A580 overtakes the AMD RX 6600 at 1440p.

In ray tracing performance, Intel Arc A580 performs as expected. Beating out AMD’s offering at various levels.

In power consumption, the Intel Arc A580 idles at 37W, which is very high. In gaming, it uses around 200W, which is, again, more than desirable.

It’s important to mention that TechPowerUP found no issues in the fan noise levels in the ASRock Intel Arc A580. However, the same issues were observed in the Sparkle one – as one can see from above, the Sparkle card it very loud.

Interestingly, TechPowerUp was able to overclock the Intel Arc A580 GPU by 500MHz, which is a massive. They found a 10% increase in its performance. However, as Intel Arc A580 GPU is originally rated to be 1.7GHz and the these companies are overclocking it to 2.0 GHz, any further overclocking is not recommended.

Conclusion

Those who are still unsure can check other sites who have reviewed Intel Arc A580. This includes the likes of KitGuru, HardwareLuxx.de (translated) and PCGamesHardware.de (translated).

Our opinion about Intel Arc A580 is simple. It’s a good budget graphics card, which performs nicely and has all the features. It beats or equals AMD RX 6600, which these days costs $30 more than the Intel Arc A580.

However, Intel Arc A580’s main rival isn’t AMD or Nvidia. It’s Intel itself. For a start, Intel Arc A750 is retailing at $190 at some places. Making it just $10 more expensive. So for such a small increase in cost, you get a massively better graphics card. This means in order for Intel Arc A580 to be attractive, it needs to be available for $150. This would make it an excellent buy.

Second issue is the drivers part as discussed earlier. Nvidia and AMD have spent years and years to improve their drivers. This means their graphics cards work on virtually almost all games. Intel is a relatively newer entry in the graphics card field. This has ensured that while it performs good in some games, it either performs badly in some others and it sometimes doesn’t even run some games. This is something Intel needs to address and needs to do better.

However, again, as mentioned earlier, Intel is improving the performance of the Intel Arc graphics cards massively with their regular driver updates. This means that the performance of Intel Arc A580 will continue to improve further. So if Intel Arc A580 does manage to retail at $150, which it might in a few months, it can be an excellent choice for gaming at 1080p for the budget users.

Those who do want to wait, however, then the next-gen Intel Arc graphics cards are expected to release next year. But don’t expect it to release cheaper versions right away, though. Expensive models will release first and while performance might increase massively, the prices won’t be as cheaper as currently they are.

Source

Source

In a press release, Intel stated:

The new product fills the middle of the Intel Arc discrete graphics product stack for gamers and creators, delivering advanced gaming performance at 1080p high settings on popular modern games, high frame rates on esports titles and a comprehensive set of media capabilities.

Some of the GPU's hardware features include:

• Xe-cores - 24
• Render Slices - 6
• Ray Tracing Units - 24
• Intel Xe Matrix Extensions (Intel XMX) Engines - 384
• Xe Vector Engines - 384
• Graphics Clock - 1700 MHz
• TBP - 185 W
• PCI Express Configurations - Up to PCI Express 4.0 x16
• Device ID0 - x56A2
• Memory Size - 8 GB
• Memory Type - GDDR6
• Graphics Memory Interface - 256 bit
• Graphics Memory Bandwidth - 512 GB/s
• Graphics Memory Speed- 16 Gbps

Intel also posted a chart showing some of the frames per second that are possible with the new Arc A580 GPU. It claims that it can run titles like Blizzard's Diablo IV at up to 155fps, Baldur's Gate 3 at up to 119fps, and Cyberpunk 2077, with XeSS and Ray Tracing turned on at 85fpg.

Intel is not selling its own Arc A580 graphics card, at least not at the moment. However, third parties are selling their own cards with the GPU. You can find two of them on Newegg now. One is the Sparkle Intel Arc A580 ORC OC Edition which is priced at $179.9. The other is the ASRock Challenger Arc A580 which is priced a bit higher at $184.99. Intel says that GUNNIR will also sell its own Arc A580 graphics card.

When you purchase through links on our site, we may earn an affiliate commission.

Source

What is new in the Surface Pro 7 October 2023 firmware update?

• Improves device camera performance and stability.

• Addresses potential security vulnerability and improves system stability.

Here is the list of new drivers:

Here is extra information about the release:

The seventh-generation Surface Pro will remain supported for a few more months—Microsoft plans to stop releasing firmware updates for the Surface Pro 7 on February 28, 2024.

What is new in the Surface Go 2 October 2023 firmware update?

• This update addresses a potential security vulnerability related to Intel® Security Advisory INTEL-SA-00783, addressing CVE-2022-36392.

Here is the list of new drivers:

Here is extra information about the release:

The Surface Go 2 will remain supported until December 30, 2024.

Source

Commenting on the data, Mikako Kitagawa, Director Analyst at Gartner, said:

‘There is evidence that the PC market’s decline has finally bottomed out. Seasonal demand from the education market boosted shipments in the third quarter, although enterprise PC demand remained weak, offsetting some growth. Vendors also made consistent progress towards reducing PC inventory, with inventory expected to return to normal by the end of 2023, as long as holiday sales do not collapse.’

The comments at the end of that quote are interesting. Given the rate of inflation and the cost of living pressures people face, budgets could still be squeezed to the point where people still cut back on their PC spending.

Of all of the PC vendors, only HP saw a growth in shipments and this was a pretty modest 6.4%. Surprisingly, Apple saw the biggest decline in shipments with a decrease of 24.2%. This is pretty surprising because Apple’s shipments have been holding up this year while the competition has seen decreases.

Lenovo saw the most shipments at 16.1 million units shipped, representing 25.1% of the total market share. Apple had 9.7% of the market share down from 11.7%.

According to Gartner, Apple's decline in shipments is attributed to a significant increase in the volume of shipments in the third quarter last year after supply disruptions had eased. It said that students and teachers were the main drivers back then as the new academic year was beginning.

Speaking on future prospects, Kitagawa said:

‘The good news for PC vendors is that the worst could be over by the end of 2023. The business PC market is ready for the next replacement cycle, driven by the Windows 11 upgrades. Consumer PC demand should also begin to recover as PCs purchased during the pandemic are entering the early stages of a refresh cycle.’

In 2024, Gartner is expected to see a 4.9% growth in the worldwide PC market with growth predicted in the business and consumer segments.

Source: Gartner

Source

What you need to know

• Unity is a popular game engine often used known for its affordability and adaptability. 
• It has been used to build wildly popular games like Among Us and Genshin Impact, and is often associated with indie titles owing to its cross-platform capabilities among other things. 
• Over the summer Unity announced an unexpected price restructure, which negatively impacted teams of all shapes and sizes. Many projects that had been set in stone were no longer going to be viable as a result of Unity's changes. 
• Owing to a massive industry-wide backlash, Unity eventually back pedalled most of the changes, although it came with a large erosion of trust. Many developers have signalled plans to leave Unity, while others accelerated plans to build in-house engines instead. 
• Today, Unity CEO John Riccitiello is stepping down, per reports. 

It took a while, but the widely expected resignation of Unity's CEO is finally here. 

Over this past summer, Unity decided to implode itself by announcing a huge swath of price restructuring changes for developers in its ecosystem. The short lead time and lack of transparency led to a massive backlash, which eventually saw Unity roll back many of its plans. You can read the full Unity summer saga over here. 

The controversy revolved around a per-installation so-called "runtime" fee that developers would be liable for every time someone installed a game. The poor messaging and lack of clarity created alarm. How would Unity be able to detect pirates and bad actors? What would be in place to prevent mass uninstall-reinstall campaigns by trolls? How would pricing structure work for services like Xbox Game Pass? What about developers working as second parties? And so on. Unity's attempts to answer some of these questions simply served to create more confusion over time, as various developers spoke up to poke holes in many of the new policies. It didn't help that some Unity execs were found to be offloading stock in the run-up to the announcement, beyond typical compensation increments senior execs often get. It also didn't help that Unity had, without warning, remove older Terms of Service contracts from its GitHub repo. 

In any case, the whirlwind backlash saw repeated backpedaling from the firm, while Unity's share price slid nearly 20% over the past month of trading alone. 

A key sticking point through all this has been the erosion of trust. Developers and publishers plan their projects years in advance of launch, alongside business models, costs, and budgets. Unity only gave developers six months to adapt to these runtime fees. Many big players like InnerSloth with Among Us already signaled intent to switch engines, and you have to expect developers like Mihoyo building games like Genshin Impact probably feel the same way. 

How can you fix an erosion of trust? Even if it were possible, it has to start with leadership. It seems Unity's board agrees. Per a report from Bloomberg's Jason Schreier and via Unity itself, CEO John Riccitiello is out. 

John Riccitiello remained relatively unpopular within the gaming community in general prior to this. Known for consumer and developer-hostile gaffes, Riccitiello became a focal point of criticism from all corners of this debate. Riccitiello previously said that he envisioned a world where Battlefield players would be charged a fee for every clip reload, and called developers not focusing on microtransactions "fucking idiots."

In a press release, Unity said the following: "Unity, the world’s leading platform for creating and growing real-time 3D (RT3D) content, today announced that John Riccitiello will retire as President, Chief Executive Officer, Chairman and a member of the Company’s Board of Directors, effective immediately."

James Whitehurst has been appointed interim CEO in the meanwhile, as Unity employs the help of an outside firm to find a permanent CEO. Whitehurst was previously at IBM, and joined the firm via its massive $34 billion acquisition of open software firm Red Hat. 

No silver bullet for Unity

Riccitiello's departure will not be a silver bullet fix to Unity's issues, however. Unity has struggled to find a single profitable quarter in its history. Riccitiello has been criticized for some of Unity's recent acquisitions, including Weta Digital, with some analysts suggesting Unity overpaid. To try and find growth, Unity has turned its hand to laying off staff, alongside this haphazard runtime fee rollout. The lack of forethought seems to have cost Riccitiello his job at Unity. Riccitiello also stepped down from EA for a similar stint of poor management efforts back in 2013. I sure as hell wish I could fail upward this successfully.

In any case, Unity has taken the first necessary step toward repairing its image, but it may be a long time before the firm can regain the trust it previously held in years past.

Source

AMD has unveiled its latest innovation, AMD FSR 3, or AMD FidelityFX Super Resolution 3. This new technology promises to enhance your gaming experience by providing stunning visuals and improved performance.

But what exactly is AMD FSR 3, and how can you enable it to take your gaming to the next level? Before we explain all these, you may watch the video from the AMD YouTube channel showing how AMD FSR 3 works on Forspoken.

What is AMD FSR 3?

AMD FSR 3 is an advanced graphics technology designed to upscale lower-resolution game images, making them look sharper and more detailed without a significant impact on performance. It achieves this by using a combination of super resolution temporal upscaling technology and frame generation.

This means that it not only enhances existing frames but can also generate entirely new game frames, improving the overall visual quality of your gaming experience

AMD FSR 3 is currently supported by two games: "Forspoken" and "Immortals of Aveum".

Additionally, the company has plans to integrate AMD FSR 3 into 10 more titles in the near future.

Here's the list of known titles:

1. Immortals of Aveum
2. Forspoken
3. Avatar: Frontiers of Pandora
4. Cyberpunk 2077
5. Warhammer 40,000: Space Marine II
6. Frostpunk 2
7. Alters
8. Squad
9. Starship Troopers: Extermination
10. Black Myth: Wukong
11. Crimson Desert
12. Like a Dragon: Infinite Wealth

Furthermore, AMD is collaborating with renowned game studios such as Square Enix, Sega, CD Projekt Red, and others. They are also developing an AMD FSR 3 plug-in for Unreal Engine, making sure that this technology will continue to expand its support and presence in the gaming industry.

AMD FSR 3 vs DLSS 3.5

AMD FSR 3 is often compared to NVIDIA's DLSS (Deep Learning Super Sampling). Both technologies aim to improve gaming visuals and performance, but they have differences:

• Methodology: AMD FSR 3 primarily relies on upscaling and frame generation techniques, while DLSS uses AI and deep learning to upscale images
• Compatibility: AMD FSR 3 is designed for AMD GPUs, while DLSS is exclusive to NVIDIA GPUs
• Availability: DLSS has been available for a longer time and is supported in more games, but AMD FSR 3 is expanding its game integrations

How to enable AMD FSR 3

Enabling AMD FSR 3 is a straightforward process, and it varies slightly depending on the game you're playing.

But before you do that, make sure you have compatible AMD graphics hardware. AMD FSR 3 is designed to work with a range of AMD GPUs, so check if your graphics card is on the list of supported hardware.

1. Update drivers: Ensure your graphics drivers are up to date. Visit the AMD website and download the latest drivers for your GPU model
2. In-game settings: Launch your game and navigate to the graphics settings. Look for the AMD FSR 3 option or a similar feature (it may vary by game) and enable it. You may need to restart the game for the changes to take effect
3. Adjust settings: Depending on your hardware and personal preferences, you can adjust the AMD FSR 3 quality setting

Once you have enabled AMD FSR 3, you should notice an improvement in visual quality and potentially higher frame rates, resulting in a smoother gaming experience.

Featured image credit: AMD.

Source

What you need to know

• A new update is available for Starfield, adding widely-requested FOV sliders to the game. 
• The sliders can be set independently of each other for first and third-person gameplay. 
• The update also improves general performance and fixes a specific quest progression bug. 

Bethesda Game Studios is continuing to improve its science-fiction role-playing game. 

The latest update for Starfield adds the highly-requested FOV sliders, allowing players to adjust their viewing space for playing the game on Xbox consoles or on Windows PC. There are separate sliders for first-person and third-person gameplay, so players can tweak things as they'd like. Both sliders are now found under the game's accessibility tab in the settings. 

The full patch is a 2.8GB download, but the game's size is only very slightly increased by a couple of hundred MB. Here's the full patch notes via Bethesda:

• FOV: Sliders are now available in Settings that allow players using first person or third person to adjust their FOV.
• [PC ONLY] Improved stability for Intel Arc GPUs.
• Various additional stability and performance improvements.
• [PC ONLY] Improved stability for Intel Arc GPUs.
• Various additional stability and performance improvements.

Analysis: Great games getting better

Starfield reached 10 million players weeks ago, and as the development team continues to add features and fix bugs through patches, that number is only going to continue to rise. Weeks ago, the team also outlined a number of big features that they wanted to add, which included DLSS support and FOV sliders. 

While this patch is fairly small overall, it's still nice that one of those big features is checked off, especially since field-of-view adjustments can be an accessibility concern for players that get motion sickness. 

I'm still not done with Starfield, and much like the team's prior games, I doubt I'll ever truly be done, but I'll definitely be playing more soon once I get some reviews taken care of in this busy season.

Source

The Lenovo ThinkCentre M70a 3rd Gen desktop is the first device that is classified by Lenovo as ready to run Esper Foundation for Android. The company plans to add support to three additional PCs, Lenovo ThinkCentre M70q, M90n-1 IoT and ThinkEdge SE30 v2, by the end of the calendar year.

All four devices run an Enterprise-grade version of Android and include dedicated device management from Esper. This makes it easier for organizations to "deploy, manage, and update dedicated PCs across their operations" according to Lenovo.

The four products are designed for specific Enterprise use cases. Lenovo mentions retail, hospitality and healthcare specifically, but the products are not limited to those industries. Customers still have the option to purchase the PCs with Windows or Linux, as Android is added as another option.

Lenovo believes that Android offers advantages to certain industries. These may include improved "flexibility, global familiarity, cost-efficiency" and a large developer pool.

Lenovo's ThinkCentre M70a 3rd Gen is an all-in-one 21.5" touchscreen desktop. It is powered by a 12th generation Intel Core processor, supports up to 16 gigabytes of DDR4 3200 MHz RAM and up to 512 gigabytes of PCIe SSD storage. Hardware components depend on the select version of the desktop PC.

The PC is powered by a custom version of Android 11 that supports customizable branding. Lenovo promises quarterly security updates and at least three years of support. Other features include Esper Seamless Provisioning as well as app and system crash notifications.

Lenovo lists four models on its website currently. The base model is available for $1679 USD at the time of writing. It is powered by an Intel Core i3-12100 processor, 8 GB of RAM and a 256 GB SSD. The current top of the line model is available for $2359 USD. It has an Intel Core i7-12700 processor, 16 GB of RAM and 512 GB SSD. No Core i9 model is listed yet on the website.

Esper's dedicated device management platform manages and monitors devices. It supports remote deploys and administrators may run updates on devices using the platform. Esper notes on its website that systems do support dual-booting. Customers may boot other operating systems next to Android on the device; this may help them run legacy software or devices, if required.

Closing Words

The new range of Android-powered devices aims to expand Lenovo's portfolio of desktop devices. While limited to specific use-cases at the time of writing, nothing is stopping Lenovo from dipping its toe in the customer segment later on. Whether Android for desktop PCs would become a hit remains to be seen.

Not all Enterprise customers may be happy about the quarterly security updates that Lenovo and Esper promise. Google Android receives monthly security updates.

Now You: would you run an Android desktop PC?

Source

Table of contents:

1. Windows 11 news
2. Updates are available
3. Gaming news
4. Reviews are in
5. A blast from Microsoft's past
6. Random fact about Microsoft

Windows 11

Here we talk about everything happening around Microsoft's latest operating system in the Stable channel and preview builds: new features, removed features, controversies, bugs, interesting findings, and more. And of course, you may find a word or two about older but still supported versions.

Let us start the Windows 11 section of Microsoft Weekly with a fresh stats drop from Statcounter and Valve. Statcounter's latest findings show that less than 24% of all PC users run Windows 11. The operating system's growth has slowed down slightly, leading to it maintaining a relatively unchanged market share for eight consecutive months.

Things are slightly better on the gaming side: Valve claims Windows 11 has more than 37% on Steam. However, its market share decreased compared to the previous month. The Windows 11 Moment 4 update may help the operating system get more users. Still, you should not expect many changes for at least two more months since the broad availability is anticipated only in November 2023. Until then, the Moment 4 update remains an optional non-security update. You can click here to learn how to install it right now.

One of the changes introduced in the Moment 4 update is the new Windows Backup app. However, it is not exclusive to Windows 11—Microsoft also released it for Windows 10 users, leading to a mixed reception from IT admins. To calm the enraged crowd and make it put pitchforks down, Microsoft released a statement claiming the company would release an update to prevent the Windows Backup app from appearing for Azure Active Directory or Active Directory users. At the same time, the company clarified that the app is a system component, so you cannot remove it.

The previous Microsoft Weekly covered a story about missing Windows Copilot in the EU due to the Digital Markets Act. As it later turned out, it might be for better: Windows 11 users with AMD graphics cards complained about their systems losing overclocking and tuning profiles after each reboot. And the funniest part is that turning off Windows Copilot resolved the problem for some.

But there is more! KB5030310, aka the "Moment 4" update, is also messing with Wallpaper Engine, a popular third-party app for animated desktop backgrounds. And as a cherry on top, you now get ads when using Copilot in Windows 11—Microsoft gotta pay somehow for those expensive GPUs powering the thing.

Interestingly, Microsoft might soon have a solution for that problem. According to newly published reports, the company will reveal plans to introduce its own generative AI processors to power Bing Chat, Bing Image Creator, Copilot, and other AI-powered developments requiring astronomical amounts of horsepower. Per latest rumors, Microsoft will share more details during its Ignite 2023 developer conference scheduled for November 14-17.

This is how Bing Image Creator imagines a Microsoft-made processor for powerful AI computing.

Next week will mark the end of support for several Windows versions. On October 10, 2023, Microsoft will stop issuing security updates and patches for the original Windows 11 release, also known as version 21H2, Windows Server 2012, and Windows Server 2012 R2. If you use one of those operating systems, it is time to consider updating to something more recent to avoid potential security risks.

With Windows 11 entering its third year of service and rumors pilling up on the next release, users are getting increasingly agitated about Windows "12." This week, a report emerged about Microsoft potentially making its next-gen operating system subscription-based, sparking a heated customer dispute. However, it later turned out those rumors were false.

Still, that does not mean Windows "12" is not coming. An Intel executive revealed in interview that the upcoming "Windows refresh" in the second half of 2024 will help the company sell its chips.

Windows Insider Program

Here is what is new in the Windows Insider Program this week:

• Canary Channel: build 25967 with a new home page for the Settings app and a reworked Quick Settings flyout. The latter now displays all available toggles, and you can check out our guide detailing how to enable the new variant.
• Dev Channel: build 23560 with fixes for File Explorer, context menus, and Task Manager.
• Server vNext: build 25967.

Updates are available

This section covers software, firmware, and other notable updates (released and coming soon) delivering new features, security fixes, improvements, patches, and more from Microsoft and third parties.

One of the biggest news on the updates side is OneDrive 3.0, a big update for Microsoft's cloud storage. The latest update features a fancy new UI and plenty of new capabilities, such as a new "For you" area with AI-powered file recommendations (because everything nowadays needs AI recommendations), context-based file organization, a new page for shared files, various customization features, better integration with various apps, and more.

Microsoft officially killed 32-bit Windows editions in 2021 with the release of Windows 11. The company's latest operating system is 64-bit only, and third-party developers are slowly ditching 32-bit Windows support. This week, Discord announced plans to discontinue its 32-bit version on December 1, 2023. Although the service will continue operating on 32-bit Windows versions, developers will not release new features or security updates. Therefore, Discord will stop working on your 32-bit Windows PC at some point.

The recently announced Surface Laptop Studio 2 and Surface Laptop Go 3 are now available. If you plan to buy or already received one, make sure to check Windows Update and install available updates. Microsoft has released a set of day-one patches to improve your user experience. Although there are no specific details about what exactly the updates enhance, their changelogs contain plenty of new drivers, so be sure to get them.

By the way, check out a neat Easter egg when setting up your new Surface devices. Users noticed that the onboarding experience (OOBE or initial setup) now lets you play the Surf game while the operating system prepares itself for work. For those unfamiliar, Surf is a game Microsoft developed for the Edge browser, and you can play it by visiting the edge://surf link. It commemorates the legendary SkiFree game from the Microsoft Entertainment pack from 1991.

In addition to updating Surface Laptop Studio 2 and Surface Laptop Go 3 firmware, Microsoft released new patches for the Surface Studio 2+ and Surface Pro 8. The update improves wireless connectivity and overall system stability.

The Microsoft Store received several updates this week. The Windows 11 app now features redesigned pages for Windows 10 and 11 license pages with consistent visuals and comparison tables between various editions to help customers select the right license. The company also rolled out a new apps.microsoft.com website with fresh looks and new features to match the app on Windows 10 and 11.

Finally, here is the latest issue of our Microsoft 365 Roadmap Weekly, a series where we cover all the upcoming changes and new capabilities for Microsoft Teams, Office apps, OneDrive, and other services from Microsoft. The latest edition mentions performance improvements for OneDrive for the web, iCloud account support in the new Outlook app for Windows, and new capabilities for Microsoft Teams Rooms.

A few interesting reports emerged about Bing this week. For starters, at some point, Microsoft was willing to pay Apple as much as $15 billion per year to make Bing the default search engine on iPhones and iPads. In addition, Satya Nadella offered to hide Bing's name and respect Apple's privacy policies, all just to remove Google from iOS devices. According to a report, Nadella has been trying to replace Google with Bing (which cost Microsoft about $100 billion to develop) as the default search engine on Apple devices since becoming Microsoft's CEO in 2014. As we all know it, his attempts never yielded any result.

Here are other notable software updates released this week:

• PowerToys 0.74.1 with fixes for FancyZones, SVG viewer, Quick Accent, and more.
• VeraCrypt 1.26.7 with bootloader, memory encryption, and plenty of other improvements for the popular encryption app.
• Microsoft Edge 119.0.2132 Dev with new notifications, flags, bug fixes, and more.

And here are the new drivers released this week:

• Intel Arc 31.0.101.4885 with support for Forza Motorsport and Assassin's Creed Mirage, plus performance improvements for DirectX 11 and 12 titles.

On the gaming side

Learn about upcoming game releases, Xbox rumors, new hardware, software updates, freebies, deals, discounts and more.

Microsoft is finally on track to close its ever-going Activision-Blizzard purchase. A new report emerged this week claiming Microsoft is ready to finalize the merger on October 13, 2023. After numerous back-and-forth with the United States Federal Trade Commission (FTC) and the UK's CMA, the company finally has the green light to proceed with its humongous $69 billion acquisition.

Mojang has partnered with NERF to release the NERF World Minecraft DLC. It is now available on supported platforms, allowing you to play with NERF guns solo or with your friends in several areas. Shoot mobs to earn Nerf bucks to later spend on new blasters, including the Ender Dragon blaster.

The first wave of October 2023 additions to the Xbox and PC Game Pass catalog is here. It includes the new Forza Motorsport, From Space, Like A Dragon: Ishin, and more. Note that several games, such as Eville, Overwhelm, Shenzhen I/O, and more, will leave the subscription soon.

Speaking of games leaving Xbox and PC Game Pass soon, Microsoft is working on a new notification system to remind users that the game they are about to play will soon be delisted from the catalog. The new notification is now available for testing in the Xbox Insider program in the Alpha Skip-Ahead, Alpha, and Beta Channels.

NVIDIA also has a catalog of games playable on its cloud platform, GeForce NOW. In October 2023, 60 different games will be available to play, including Microsoft's newest Forza Motorsport, Alan Wake 2, Cities Skylines 2, Metro: 2033 Redux, Surviving Mars, MudRunner, and many more. However, it is worth noting that you must own the titles you want to play. Unlike Xbox Game Pass and Microsoft's Cloud Gaming, NVIDIA only gives you access to hardware, requiring you to bring your own games to the party.

Finally, Microsoft announced a new Special Edition Xbox Wireless Controller. It is called Gold Shadow, and it features a blacked-out case with a two-tone golden front panel. The new gamepad is now available for pre-order at the Microsoft Store, with availability scheduled for October 17, 2023.

A blast from Microsoft's past

John Callaham's weekly "Look back" series provides throwbacks into the past, detailing the company's products, partnerships, mishaps, and successes from years ago.

This week's look-back article is about Windows Home Server and its quirky promo campaign. In 2007, Microsoft unveiled Windows Home Server, a special Windows edition for backing up all home PCs and storing your media in one location, available even when you are away from home. To promote the new Server SKU aimed at home users, Microsoft created a weird children's book called Mommy, Why is There a Server in the House?

Sadly, the idea did not took off, even though some vendors backed it up with dedicated hardware. In 2012, Microsoft discontinued Windows Home Server.

Random fact about Microsoft

And here is a randomly selected piece of trivia about the company, Windows, and other Microsoft-made things.

This week's random fact about Microsoft or its products comes from the @wowstartsnow Twitter user. Modern Windows versions are infamous for constant update notifications, nagging users to install the latest patches and new features. Did you know that Microsoft introduced Windows Update notifications in 1998?

Windows 98 users could download the Critical Update Notification Tool and receive messages about available updates with a link to guide them through the update process. It also allowed postponing notifications for 24 hours by clicking a "Notify Me later" button. Microsoft later renamed it Critical Update Notification Utility and made it a system component, part of Windows 2000, released in 1999.

Source

Numenta has demonstrated that Intel Xeon CPUs can vastly outperform the best CPUs and best GPUs on AI workloads by applying a novel approach to them.

Using a set of techniques based on this idea, branded under the Numenta Platform for Intelligent Computing (NuPIC) label, the startup has unlocked new performance levels in conventional CPUs on AI inference, according to Serve the Home.

The really astonishing thing is it can apparently outperform GPUs and CPUs specifically designed to tackle AI inference. For example, Numenta took a workload for which Nvidia reported performance figures with its A100 GPU, and ran it on an augmented 48-core 4th-Gen Sapphire Rapids CPU. In all scenarios, it was faster than Nvidia’s chip based on total throughput. In fact, it was 64 times faster than a 3rd-Gen Intel Xeon processor and ten times faster than the A100 GPU.

Boosting AI performance with neuroscience

Numenta, known for its neuroscience-inspired approach to AI workloads, leans heavily on the idea of sparse computing – which is how the brain forms connections between neurons. 

Most CPUs and GPUs today are designed for dense computing, especially for AI, which is rather more brute force than the contextual manner in which the brain works. Although sparsity is a surefire way to boost performance, CPUs can’t work well in that way. This is where Numenta steps in. 

This startup looks to unlock the efficiency gains of sparse computing in AI models by applying its “secret sauce” to general CPUs rather than chips built specifically to handle AI-centric workloads. 

Although it can work on both CPUs and GPUs, Numenta adopted Intel Xeon CPUs and applied its Advanced Vector Extensions (AVX)-512 plus Advanced Matrix Extensions (AMX) to it, because Intel’s chips were the most available at the time. 

These are extensions to the x86 architecture – serving as additional instruction sets that can allow CPUs to perform more demanding functions. 

Numenta delivers its NuPIC service using docker containers, and it can run on a company’s own servers. Should it work in practice, it would be an optimum solution to repurposing CPUs already deployed in data centers for AI workloads, especially in light of lengthy wait times on Nvidia’s industry-leading A100 and H100 GPUs. 

Source

Now there's word that Epic is going to try to get some extra revenue from people who use its Unreal Engine 5 game development tools. However, this new plan won't actually affect game developers at all, but rather people who use the tools for other projects and applications.

VGC reports that during the company's Unreal Fest event last week, Sweeney revealed a plan to charge non-gaming developers a "per seat" fee for using Unreal Engine 5. He stated:

And this doesn’t affect game developers, but one of the things we’re going to change next year is for industries other than game development, such as the automobile industry and so on, we’re going to move to a seat-based enterprise software licensing model for Unreal Engine.

The plan will be similar to how Adobe charges for its development tools. Companies will soon have to pay a license to use Unreal Engine 5 based on how many of their employees use it.

Sweeney didn't offer a specific date on when this business model will be launched, nor did he state any price points. He did say that it would not be "unusually expensive or unusually inexpensive." It will be interesting to see if those non-game developer companies will go ahead and keep using Unreal Engine 5 under this proposed plan, or if they might abandon the game engine.

Game developers will still be able to make projects in Unreal Engine 5 for free, with Epic charging a 5 percent royalty on a game's revenues once it reaches the $1 million level.

Source

When Nvidia released the GeForce RTX 4090 for $1600, everyone was amazed by its performance. It came with the flagship AD102 GPU with 16384 CUDA cores, 24GB VRAM and a TGP of 450W. The performance of RTX 4090 was unparalleled and remains unmatched even now.

However, later Nvidia released the GeForce RTX 4080 graphics card for $1200. Everyone felt it to be lacking. It came with a AD103 GPU with 9728 CUDA cores, 16GB VRAM and a TGP of 320W.

The reason behind why people feel disappointed by RTX 4080 is that while it costs just 28% lesser than RTX 4090, it comes with 50% lesser CUDA cores than the flagship graphics card by Nvidia.

Not only that, later, reviewers found that AMD Radeon RX 7900 XTX, which costs just $1000, actually equals or beats the $200 more expensive Nvidia RTX 4080 in usual raster performance. It’s only in ray tracing performance that RTX 4080 manages to easily beat AMD RX 7900 XTX.

So it’s understandable that Nvidia wants to fill-up that gap somehow. The gap, which is between RTX 4080 and 4090. Not only that, beat AMD’s offerings too. Not to forget, Nvidia RTX 5000 series is unlikely to come any sooner.

Looks like Nvidia is preparing a new graphics card for that.

RTX 4080 Ti To Be Released By Nvidia

Well known Twitter (now X Corp.) based leaker MEGAsizeGPU has shared (via VideoCardz) an update about an upcoming Nvidia RTX product.

In a tweet, MEGAsizeGPU has revealed that Nvidia is possibly preparing either RTX 4080 Ti or RTX 4080 SUPER graphics card that is all set to release in early next year, in 2024.

It further reveals that unlike RTX 4080, which is based on the AD103 GPU, the RTX 4080 Ti will be based on the (cut-down version of the) flagship 102 GPU, the same one which powers RTX 4090 graphics card.

In terms of power, it seems to have a TGP below 450W, which RTX 4090 is marketed as. We believe it could be around 400W.

As for pricing, it says that the RTX 4080 Ti could come in the same range as the RTX 4080. Which is interesting as RTX 4080 costs around $1100. Meanwhile, AMD the RX 7900 XTX is available at around $900. So is Nvidia planning to cut the price of RTX 4080 further.

When asked about the VRAM and memory bus size, MEGAsizeGPU replied that they are not decided yet. Though, VideoCardz speculates that RTX 4080 Ti could with around 20GB of VRAM with a 320-bit bus.

RTX 4080 Ti Or SUPER

While the leaker mentions a possibility that the graphics card will be named either 4080 Ti or RTX 4080 SUPER, it doesn’t go in detail about it.

There’s another possibility about this graphics card, though. This is entirely a speculation on our side, so don’t take it as a confirmed thing.

As RTX 5000 series is not expected to release next year, is Nvidia planning to release a refresh for the RTX 4000 series. The reason we say that is because if indeed this new graphics card is touted as SUPER, as opposed to Ti, then there’s a possibility of it being a refresh.

That’s because SUPER naming of graphics cards is mostly done by Nvidia for refresh versions, just like it refreshed the GTX 1600 series and the RTX 2000 series.

Or maybe Nvidia is just intending to add more graphics cards in the RTX 4000 line-up, that’s also a good possibility.

Whatever it could be. It’s a good thing if indeed a faster version of RTX 4080 is released at the same price as the original one.

Source

Source

While exact numbers were not disclosed, the job cuts impacted several departments across the company. In a statement to TechCrunch, a Brave spokesperson said,

Brave eliminated some positions as part of our cost management in this challenging economic environment. Several departments were affected, amounting to 9% of our staff.

For those unfamiliar, Brave Software is the creator of the Brave browser and a privacy-focused search engine, launched in 2016.

In recent months, Brave has been actively working to diversify its revenue streams. In April, the company took a major step by transitioning Brave Search away from relying on Microsoft's Bing index and deploying its own solution. This increased Brave's independence and allowed it to control the search experience better.

The following month, Brave launched a commercial Search API that offers clients various paid plans and features. In addition to the basic $3 per 1,000 query plan, the API provides options for AI model training, spellcheck, auto-suggest, and access to query results with storage rights.

Brave continued its efforts and expanded the API's capabilities in August by rolling out integrated image, news, and video results.

In addition to developing its search offerings, Brave has been testing an AI assistant, Leo, that is directly integrated into its privacy-focused browser. While Leo will have basic functionality for all users, the company is planning a premium tier to unlock additional conversation models and higher rate limits.

However, the tough macroeconomic conditions have led the company to downsize staffing levels to help manage costs. Affected employees were notified earlier this month.

Source: TechCrunch

Source

The Ignite conference, which is for developers, IT professionals, and admins, among others, will be held in Seattle from November 14-17. Microsoft is already promoting that the conference will feature a number of major product announcements including many centering on its AI platform, especially during its keynote address which will feature its CEO Satya Nadella and other major company executives.

Microsoft currently uses NVIDIA's H100 GPUs to power its datacenters that handle its AI services like its Bing Chat AI chatbot, the Bing Image Creator art generator, and more recently its Copilot assistant service.

Microsoft and other generative AI companies have been buying a lot of NVIDIA GPUs for those dataservers over the past year, which has caused the company's revenues and stock price to go way up in 2023. However, the new report says Microsoft would not have to rely as much on NVIDIA's chips if it creates its own AI processors. That could cause NVIDIA's revenues to go back down.

In April, The Information reported that Microsoft was developing this new in-house AI chip, under the code name Athena. It added that the chip had already been in development since 2019 and had about 300 team members working on the project at the time.

In May, another unconfirmed report from Bloomberg claimed Microsoft had secured a partnership with NVIDIA's chief GPU rival AMD to help them co-develop the Athena chip. The story also claims Microsoft is providing financial support to AMD so it can make its own AI chip.

Source

What is new in the Surface Studio 2+ October 2023 firmware update?

Here is the list of new drivers:

Here is extra information about the release:

The Surface Studio 2+ has six years of active support. Microsoft will stop releasing firmware updates with new drivers and security patches on October 2, 2028. You can get this unique desktop computer from Microsoft for the eye-watering $4,499.

What is new in the Surface Pro 8 October 2023 firmware update?

Here is the list of new drivers:

Here is extra information about the release:

Surface Pro 8 will remain supported for four more years, with the end of its lifecycle scheduled for October 5, 2027.

We may earn an affiliate commission when you purchase through links on our site.

Source

Source

Albums are an optional feature of OneDrive. Users may create albums to sort photos into them. It is a popular feature, as it allows users to manage their photos better.

It was not really clear if Microsoft was storing photos multiple times on OneDrive, which would explain why it made the announcement, or if Microsoft was just trying to increase user sign-ups for paid OneDrive subscriptions.

Microsoft promised to give customers a temporary storage boost for a year to avoid that the change would push customers over their limits. After that year, customers would find themselves in the same situation, however, unless they would delete files on OneDrive or photos in albums.

Microsoft backs down

A support page update confirms that Microsoft has had a change of heart regarding the change. The company writes: "On August 31, 2023, we began to communicate an upcoming update to our cloud storage infrastructure that would result in a change in how OneDrive photos and photo albums data is counted against your overall cloud storage quota. This change was scheduled to start rolling out on October 16, 2023. Based on the feedback we received, we have adjusted our approach, we will no longer roll out this update. We will maintain the current photo album experience, as it is today."

To sum it up: Microsoft won't roll out the planned change on OneDrive. Photos that users put into albums won't count against a user's storage on OneDrive.

It is still unclear if Microsoft is storing album photos separately, but that is no longer a problem for users of the service.

Closing Words

Microsoft claims that it changed course because of feedback that it received. It is certainly possible that some customers made the decision to migrate to another cloud storage service, one that would not count individual files multiple times against a user's storage quota.

All in all, it is a welcome development for OneDrive users, especially those who use the photo albums feature.

Now You: do you store photos online? (via Dr. Windows)

Source

The report comes from The Verge, citing unnamed sources. The same report does point out that while Microsoft is making plans around that date to close the deal, all of that depends on the final approval of the UK Competition and Markets Authority.

The CMA initially would not approve of Microsoft's plans to buy the game publisher back in April. At the time, the decision came as somewhat of a surprise. The CMA claimed that such a purchase could cause Microsoft to have too much control of the cloud gaming industry.

While it looked for a while that Microsoft and the CMA would battle it out in a UK regulatory appeals court, the CMA instead allowed Microsoft to adjust its deal. In August, Microsoft presented a different proposal, where it would let Ubisoft get the cloud gaming rights to all current and future Activision Blizzard deals for 15 years.

That change seems to have changed some minds at the CMA. In September it gave provisional approval to the deal, while at the same time also stating they would also consult with Microsoft for some last "limited residual concerns."

The CMA also set up a deadline for people to offer their comments on the deal. The deadline for sending those comments expires today, and after that, the CMA should make its final decision sometime next week. Unless it makes another surprise announcement, the group is expected to approve the deal.

While Activision Blizzard could finally become part of Microsoft in a week, the company still has to battle the US Federal Trade Commission, which has tried to unsuccessfully block the deal in court. The Ninth Circuit Court of Appeals is expected to rule on this current court case in early December.

Source

A DEC VAX 8350 with cover removed.

Adamantios

 

Even though very few of the early players in technology still exist, we use their creations to this day. Bell Labs created the transistor, and Fairchild Semiconductor created the integrated circuit, but neither company is still around. So is the case with Digital Equipment Corporation (DEC). It no longer exists, but unless you're using a handheld device to read this article, you're using a descendant of DEC technology.

 

DEC was founded in 1957 by Ken Olsen, Harlan Anderson, and H. Edward Roberts to build small digital modules, but the team soon discovered that they could use those modules to build minicomputers—computers that were smaller and less powerful (but cheaper) than mainframes, which were the business standard at the time.

 

In 1977, DEC introduced the VAX, a new line of minicomputers that featured a 32-bit instruction set architecture and virtual memory. Its operating system, VMS, was a multi-user, multitasking OS that provided features we now take for granted, including virtual memory, file sharing, and networking. It amassed a wide variety of third-party software packages that made it the most popular system in its class.

 

In the late 1980s and early '90s, Andy Green ran a bulletin board system (BBS) and later an Internet service provider called Intelecom Data Systems (IDS) on a VAX 11/730 (later a VAXstation 4000) server in the basement of his parents’ house in Rhode Island. IDS had seven lines—unheard of at the time—and users could talk in a real-time chat room. All of this was written by Green in VAX BASIC. Today, Green is the owner and CEO of Acme Atronomatic, developer of the MyRadar mobile app.

 

Green was exposed to the VAX through work and had picked up an old VAX 11/730 and started tinkering. He had previously run a BBS on a TRS-80, but the VAX, with its multitenancy, allowed for multiple concurrent users.

 

“Prior to [IDS], the PC or TRS-80 were only engineered originally to be single-user, and they weren't set up to be multi-user,” Green told Ars. “The fact that VAX and VMS in general were designed for [multiple users] from the scratch is what facilitated the multi-user aspect.”

 

The VAX served DEC well throughout the '80s and into the '90s, but as the latter decade went on, DEC began to face stiff competition from UNIX vendors, particularly Sun Microsystems. DEC struggled to change with the times, and the company ultimately failed. In 1998, DEC was acquired by Compaq, and in 2001, Compaq was acquired by Hewlett-Packard. The DEC line, including the VAX/VMS system, was discontinued and faded from the market.

 

And yet it lives on today. Here’s how.

VMS=WNT

VMS was popular because DEC supported it so thoroughly. It had a user-friendly interface and powerful command-line tools, and it was one of the first operating systems to support networking protocols, including TCP/IP, DECnet, and SNA. It had a powerful file system that supported hierarchical directories and file permissions, and it was highly customizable.

 

In 1988, a senior VMS engineer named David Cutler joined Microsoft to lead the development of the Windows NT operating system. Windows NT was a major departure from previous Microsoft operating systems, as it was a 32-bit, multi-user, multitasking OS. Windows client, still finding its way to usability, was a 16-bit layer that ran over MS-DOS. It wasn’t really an operating system; it was more like a program launcher.

 

Windows NT launched in 1993 with version 3.1, matching the desktop version of Windows, which had also just been released. But while Windows 3.1 finally got it right, NT 3.1 was a bit too heavy for the PCs of the day, and it was recast as a server operating system.

 

Both operating systems continued on separate but iterative tracks, with Windows for desktop becoming Windows 95/97 and NT becoming version 3.5, version 4.0, and eventually Windows 2000. It wasn’t until the release of Windows XP in 2001 that the consumer and business products were merged into one line based on NT.

 

XP begat Windows Vista (a bad memory best left for the past), the vastly improved Windows 7, and the disastrous Windows 8, then Windows 10 and finally Windows 11.

 

Cutler and his team didn’t copy VMS, per se, but a lot of the structural elements and functions, such as the network file system, networking protocols, security, and memory management, were duplicated in NT.

 

Green saw the resemblance right away. “There definitely were some similarities between NT and VMS,” he said. “The permissions systems, file protection mechanisms, privileges and roles—those were some security concepts integrated into VMS that were unique at the time but incredibly useful and configurable. NT implemented very similar paradigms, and it definitely makes sense, now, knowing the heritage.”

 

“There is an undeniable shared heritage between the two,” said Wes Miller, analyst with Directions on Microsoft. He noted that multiple features in NT look “strikingly similar” to VMS. “It's just fascinating when you go back and look at it the way the process scheduler works, the way task prioritization works. There's a lot in common between the two.”

 

“I would say [NT] bears a lot of similarities [to VMS],” said Nathan Brookwood, a research fellow with Insight 64 who worked for DEC before becoming an analyst. “The command structures and user interfaces were very similar.”

 

And in fact, the innovation went both ways, Miller noted. The Registry, a hierarchical database of a system’s hardware and software configuration, was first designed for NT, and DEC then introduced its own Registry in VMS 7.3.

 

How much of the original Windows NT 3.1 code written by Cutler and his team more than 30 years ago remains in Windows 11 is unknown. Recent versions of Windows for both client and server have shed 16 bit application compatibility, and this was definitely a function found in Windows NT 3.1 to version Windows 2000. Back in the early to mid-'90s, most applications were 16-bit, and 32-bit only began to creep into the market later in the decade.

 

Even the 32-bit support in current versions is only through a compatibility mode called “Windows on Windows” and not native. As Windows inevitably marches to 64-bit-only, “you've got to jettison some of the old stuff,” said Miller. Which means more of the old code disappears from the operating system.

 

“[16-bit] sort of died out when AMD64 came along, and the team that was working on it decided not to port 16-bit software support. When they went full 64-bit, you had a 32-bit subsystem but no 16-bit support. So there's never been a 64-bit version of Windows that supports 16-bit applications,” said Miller.

Alpha and Omega

The second piece of the puzzle is the CPU. Unless you use a Mac, your computer's CPU has its roots in a DEC processor that failed in the market.

 

In the late 1980s early '90s, DEC was looking to change with the times and evolve its VAX line. In 1992, it introduced the Alpha AXP, later shortened to just Alpha, a RISC-based processor designed to compete with the other RISC chips on the market such as Sun Microsystems SPARC and Hewlett-Packard’s PA-RISC.

 

“DEC was getting its lunch eaten by the RISC guys, particularly Sun, but also, to a certain extent, Silicon Graphics with their MIPS products, both of which were based on this newfangled RISC environment,” said John Culver, a CPU historian and operator of The CPU Shack.

 

“They knew that they had to improve or get competitive with performance. So they looked at all the alternatives, which were basically to try to make the VAX faster or to come up with an architecture that they thought would be very competitive in the market. And that turned out to be Alpha.”

 

Source