The Surface Laptop Studio 2.

Andrew Cunningham

 

NEW YORK—Two years after announcing the original, Microsoft has announced a much-needed refresh of the Surface Laptop Studio, the company's convertible touchscreen laptop that also serves as its most powerful notebook for people playing games or doing any kind of 3D rendering or GPU-accelerated AI work. The new Laptop Studio 2 will start at $1,999.99 and is available for preorder now; it will ship on October 3.

 

The Laptop Studio 2 is a substantial internal overhaul, swapping out a quad-core 11th-generation Intel Core CPU and a GeForce 3050 Ti for a 13th-generation CPU and GeForce RTX 4050 and 4060 options (with an RTX 2000 also available for people who need a "pro" GPU). Microsoft continues to refresh Surface devices toward the end of Intel's product cycles, and next-generation Meteor Lake Core processors should be coming soon, but this is still a welcome upgrade over the old model. It's also the first x86 Surface that Microsoft sells with an NPU for accelerating AI and machine learning workloads, specifically an Intel Gen 3 Intel Movidius 3700VC.

 

• 

• 

Unfortunately, the base $2,000 version gets you very few of those perks. It includes an i7-13700H with six P-cores, eight E-cores, and 16GB of RAM, but it only comes with integrated graphics and a 512GB SSD.

 

Adding an RTX 4050 costs another $400, and stepping up to a 32GB/1TB configuration costs another $400. Getting an RTX 4060 also requires stepping up to 64GB of RAM, which drives the price up to $3,300. These are just a few of the config options available, but most upgrades tend to come bundled together with other upgrades, and they drive the price of the laptop way up.

 

The new Laptop Studio has the same 14.4-inch 2400×1600  screen as its predecessor, but it addresses the old Laptop Studio's input limitations by adding a USB-A port and a microSD card reader. The laptop also keeps its headphone jack, Surface Connect port, and pair of Thunderbolt 4 ports.

 

The original Laptop Studio was a replacement of sorts for the old Surface Book, a convertible where the tablet could detach entirely from the keyboard, trackpad, and discrete GPU in the base. It was a more versatile convertible in some ways, though its ports were even more limited than the Laptop Studio's, its CPU was limited by the cooling that could be squeezed into the tablet part, and it had an awkward bendy-straw hinge that left a big gap between the top and bottom halves of the laptop when closed. The Laptop Studio is a laptop first and leaves the tablet-y stuff to the Surface Pro and its ilk.

 

Source

However, a new report from Windows Central, citing unnamed sources, claims Microsoft will be cutting back on its Surface product lineup, and that it won't be moving forward with new versions of its more "experimental" devices.

That means we will not be seeing new versions of the Surface Studio all-in-one PC, nor any new Surface Duo Android dual-screen smartphones, according to the article. It adds that the company had plans to release a new Surface Laptop SE 2 earlier this year, but that notebook, like many of Microsoft's hardware devices in the past, got canceled at the last minute.

Another plan to launch an 11-inch version of the Surface Pro tablet was also canceled, according to the article, in favor of the Surface Go 4 which is expected to be officially revealed today.

The article also claims Microsoft was developing other products like a foldable smartphone and tablet, along with a mini desktop PC and a new monitor, but it looks like all of those products won't be moving forward at the present time.

Microsoft now reportedly will concentrate on making new versions of the following Surface tablets, notebooks, and PCs going forward:

• Surface Pro
• Surface Laptop
• Surface Laptop Go
• Surface Laptop Studio
• Surface Hub
• Surface Go

The article claims that the reduction of Surface products, including its more experimental devices, may have been a factor in the unexpected departure this week of the company's Surface leader Panos Panay.

Panay might have felt that he would no longer have the freedom to launch different kinds of Surface devices as Microsoft has apparently decided to concentrate on products that make the company money.

Source

According to the purported screenshots of the app, the Threads delete profile feature will get rid of your profile once and for all. It will remove your data such as posts and likes permanently 30 days after deleting your profile.

The screenshot warns that you won't be able to recreate a Threads profile using the same Instagram account for up to 120 days. It also mentions that deleting your Threads profile will not delete your associated Instagram account.

As of now, you can only delete a Threads profile if you delete your associated Instagram account. It is yet to be known when (or if) the Threads delete profile option will arrive on the app. However, you can deactivate your Threads profile independently if you need to get off the platform for some time.

The Threads app was launched earlier this year and it quickly took the crown of the fastest app launch from Pokémon GO. While tens of millions of users joined the Twitter alternative in its inaugural week, it didn't take long for the engagement to go down.

We previously reported that the social media app lacked a number of essential features that were present on rival platforms, including the ability to delete your Threads profile. However, Instagram head Adam Mosseri assured users that various Threads features were "on the list" for a future release.

While there are no imminent plans to introduce direct messaging, Threads has tried to fill in the gap by adding some of the missing features such as the following tab, language translation, custom alt-text, etc. More recently, the web version of Threads and keyword search functionality have also rolled out in select countries.

Source

Intel’s next-generation Meteor Lake laptop processors are nearing release—the company announced this week that the first processors will launch on December 14. It’s unclear whether actual Core and Core Ultra systems will be available to buy on that date, but at a bare minimum, the official announcement will pave the way to many laptop announcements at CES in January.

 

We already know a lot of basic facts about Meteor Lake; it uses a combination of chiplets manufactured by both Intel and TSMC rather than a single monolithic die, and it will mark the retirement of Intel’s “nth-generation” and i3/i5/i7/i9 branding. We also know that it won’t be ready for desktops and that the next round of Core desktop CPUs will be very similar to the 12th- and 13th-generation chips.

 

But at Intel's Innovation event this week, the company dove a little deeper into some of Meteor Lake’s advancements, describing more about how the chips would balance E-cores and P-cores and announcing its most substantial integrated GPU upgrade in years. We’ll hit some highlights below, though it’s worth watching or reading the full presentation to find out more.

Tiles for miles

Meteor Lake will be Intel's first consumer processor to move to a chiplet-based design—instead of being one monolithic die that contains the CPU, GPU, and all the other bits and pieces needed for a modern laptop CPU, Meteor Lake is split into four "tiles" that are joined together by a fifth base tile that allows them to communicate with each other. The process of stacking the chiplets on top of the base tile is a packaging technology that Intel calls Foveros.

 

Here's a basic breakdown of what is in each of those four tiles:

 

• • The compute tile is where most of the actual CPU is. Intel's current renderings show a chip with six high-performance P-cores based on the Redwood Cove architecture, and eight high-efficiency E-cores based on the Crestmont architecture.
  • The graphics tile is where most graphics processing happens, though a few specific functions that you'd usually find in a GPU have been moved to other tiles. Meteor Lake's integrated GPU is mostly just an integrated version of an Intel Arc dedicated GPU, complete with hardware ray-tracing acceleration.
  • The IO tile handles most external connectivity, including PCI Express 5.0 lanes and Thunderbolt 4 support (Thunderbolt 5 will need to wait).
  • The SoC tile is probably the most interesting of the four. It includes two additional Crestmont E-cores, the media encoding and decoding engine that would normally be located in the GPU, and the neural processing unit (NPU) used for accelerating AI and machine learning workloads. It also handles Wi-Fi and Bluetooth connectivity and connecting to external displays over HDMI 2.1 and DisplayPort 2.1.

 

One of the notable things about Meteor Lake is that not all of the tiles are being manufactured by Intel. The compute tile, which houses the actual P-cores and most of the E-cores, uses the new Intel 4 process, an upgrade from the Intel 7 process used for most current Core chips. But the graphics tile is being made on a 5 nm TSMC process, while the IO tile and SoC tile are made on a 6 nm TSMC process.

 

Intel has also used TSMC's manufacturing to make its Arc GPUs, so it's not the first time we've seen these two chipmaking frenemies work together. But Intel is trying to catch up with TSMC's manufacturing, and Intel sees its foundry operations as key to its future growth. I wouldn't be surprised if moving back to all-Intel-made tiles is the eventual goal.

E-ven more E-cores

Intel has said that Meteor Lake's P-cores don't change much compared to the ones used in 12th- and 13th-generation Alder Lake and Raptor Lake CPUs—we may see higher clock speeds, but not much has changed in terms of instructions-per-clock or instruction set. The E-cores do get some improvements, though.

 

Meteor Lake actually includes two different kinds of E-cores. There are two low-power (LP) E-cores in the SoC tile, and the updated version of Intel's Thread Director will attempt to use those E-cores for as many tasks as possible. Intel calls this chunk of the SoC tile the "low-power island" because the idea is to allow the compute tile and the graphics tile to completely power down as much as possible to save power.

 

When tasks require more performance than the LP E-cores can provide, Thread Director will kick them to the compute tile—either to the main E-core clusters, which are tuned to handle low-impact multithreaded workloads, or to the P-cores, which are used for single-threaded tasks and any multithreaded work the E-cores can't handle. This is a change from how Thread Director works in 12th- and 13th-generation processors, where high-priority tasks would head straight to the P-cores without trying the E-cores first. (Though it remains to be seen whether the Thread Director changes will result in any kind of user-noticeable delays for high-performance tasks.)

 

Also worth noting: Crestmont E-cores can be added to processors in groups of two, where previous-generation Gracement E-cores could only be added in groups of four. This might make it easier for Intel to justify bringing small groups of E-cores to lower-end processors that didn't have them before. All E-cores remain single-threaded, while P-cores still support two threads per core.

 

The new E-cores also include a few other niceties—VNNI instructions for accelerating AI workloads, and even AVX10, which brings many of the benefits of Intel's AVX-512 instructions without requiring 512-bit registers. The 12th- and 13th-generation Core processors totally shut off AVX-512 support because the E-cores didn't support it, even though support was present in the P-cores. This has created an awkward situation where AMD's latest Zen 4 chips support the AVX-512 instructions that Intel invented and promoted, where Intel's latest consumer chips don't.

Integrated Intel Arc: Twice the performance per watt

Intel hasn't substantially improved its integrated graphics performance since late 2020 when it launched its Iris Xe integrated GPU in 11th-generation laptop CPUs. The 12th- and 13th-generation processors used essentially the same GPU, so graphics performance on Intel ultraportables has stayed still while AMD has introduced new RDNA 2 and RDNA 3-based integrated GPUs into Ryzen laptops.

 

Meteor Lake will change this. Intel says the new graphics tile will offer up to twice as much performance per watt as the Iris Xe GPU in current Intel processors. The decade-old Iris name also seems to be going away, to be replaced by the same "Arc" branding Intel uses for dedicated GPUs. Meteor Lake also supports the same video encoding and decoding acceleration as Arc—including hardware-accelerated AV1 video encoding—but the media engine is built into the SoC tile rather than the GPU.

 

The GPU is based on an architecture called Xe-LPG (low-power gaming) that is mostly identical to the Xe-HPG (high-performance gaming) architecture in Arc A-series GPUs, replacing the older Xe-LP architecture in Iris Xe GPUs.

 

Compared to Xe-LP, Xe-LPG sports higher clock speeds, more GPU hardware, and architectural improvements. A fully enabled Xe-LPG GPU will include 128 of Intel's Vector Engines (formerly Execution Units, or EUs), up from a maximum of 96 in Xe-LP. Intel didn't say anything about less than fully enabled Xe-LPG GPUs, but the company ships a version of Iris Xe with only 80 of the EUs enabled, so a cut-down version of the Arc-integrated GPU seems likely, too.

 

While Xe-LPG has nearly everything that a dedicated Arc GPU includes, it is missing one thing worth mentioning: Intel Matrix Extensions, or XMX. In Arc GPUs, these are used to improve the performance and appearance of Intel's XeSS upscaling in games that support it. Since these upscaling algorithms are pretty handy to have when you're trying to get a game running at 60 FPS on an integrated GPU, it's too bad that Xe-LPG only supports the less-capable GPU-agnostic version that uses DP4a instructions instead of XMX.

 

We'll need to wait to see how an Xe-LPG GPU performs compared to an Iris Xe GPU—doubling the performance per watt doesn't mean performance has been doubled—but Meteor Lake should have the most interesting integrated GPU we've seen from Intel in quite a while.

 

Source

Amazon held its annual hardware event for the press today. In addition to its new Alexa-based generative AI services that will roll out soon, there were a number of new hardware devices that were announced.

In its press release, Amazon revealed it is launching a new version of its Echo Show 8 smart display. It includes a new centered camera, new spatial audio processing technology, and a new UI feature that shows more general info at first but gets more detailed when you get closer to the screen. It is available for preorder now at $149.99 for shipment in October.

Amazon revealed three new Fire TV devices. The new Fire TV Stick 4K is 30 percent more powerful than the previous version, claims Amazon, with support for Wi-Fi 6, and it will go on sale soon for $49.99. The new Fire TV Stick 4K Max (shown above) has Wi-Fi 6E support and adds the Fire TV Ambient Experience. It will include "helpful information like your family calendar, local weather forecast, and reminders on your screen when you’re not streaming" It will be available soon for $59.99.

There will also be a new Fire TV Soundbar, which offers the Fire TV OS in a 24-inch soundbar for $119.99.

Amazon is also launching its new eero Max 7 WiFi router. Amazon says:

It combines the company’s patented TrueMesh networking technology with the latest Wi-Fi 7 standard to dramatically increase speeds, avoid interference from neighboring networks, and improve mesh latency—and with 10 Gigabit Ethernet, it’s now possible to download a 4K movie in 10 seconds or a 50 gigabyte video game in less than a minute.

The router will be priced starting at $599.99 and it will come in one, two, and three packs.

Amazon also announced a new Echo Hub device. It has an eight-inch touchscreen display that you mount on your wall to keep track of all of your smart home devices. It will go on sale later in 2023 for $179.99.

You can check out more of Amazon's new hardware products, including new Kids Fire tablets and new Echo Frames smart glasses, at the company's press release. Amazon did not make any mention of a replacement for its current hardware head Dave Limp, who has already announced he is retiring from that role. Unconfirmed rumors claim Microsoft's former Surface and Windows head Panos Panay will be the new hardware leader at Amazon.

As an Amazon Associate when you purchase through links on our site, we earn from qualifying purchases.

Source

Today happens to be the 41st anniversary of the first documented use of an emoticon. It happened on September 19, 1982. We can all thank Scott E. Fahlman, a computer scientist at Carnegie Mellon University, for the first emoticons.

As he posts on his own website, in the early 1980s, the computer science students and teachers at Carnegie Mellon used computer bulletin boards to send messages to each other. However, Fahlman noted that sometimes, if someone tried to write something that was humorous on these boards, some people would fail to get the joke.

So some people in that community decided to see if they could come up with a way to show that a message was not supposed to be taken seriously. As Fahlman himself wrote on his site:

Various “joke markers” were suggested, and in the midst of that discussion it occurred to me that the character sequence :-) would be an elegant solution – one that could be handled by the ASCII-based computer terminals of the day. So I suggested that. In the same post, I also suggested the use of :-( to indicate that a message was meant to be taken seriously, though that symbol quickly evolved into a marker for displeasure, frustration, or anger.

His message was posted on the Carnegie Mellon computer bulletin board on the previously mentioned date. Fahlman stated that his character sequences "caught on quickly" in that community, and soon spread to other college computer communities as well.

However, for a while, it looked like Fahlman's original post that showed his idea for "joke markers" was lost as he didn't think about saving it. It looked for a while like the proof that he had started an internet messaging phenomenon would never be found.

However, in 2002, a former Carnegie Mellon computer researcher named Mike Jones, who was then working at Microsoft, decided to start an effort to find the original post from Fahlman. As Fahlman notes, other CMU staff members joined in the search.

Jeff Baird and the CMU CS facilities staff put in a heroic effort with the support and encouragement of Howard Wactlar, Bob Cosgrove, and David Livingston. They found the proper tapes, located a working tape drive that could read the ancient media, decoded the old formats, and did a lot of searching to find the actual posts. I am most grateful to all who participated in this successful quest, which I call the “Digital Coelacanth Project.”

As he indicated, the search for the post was indeed successful, and you can find a copy of his post, along with the bulletin board thread of the initial idea and the responses to Fahlman's message, at this site. As Jones notes on that same page, "Jeff Baird should get most of the credit for doing the hard work of locating and retrieving the data."

There's a lot more info from Fahlman about his creation on his site. He also makes some comments about claims that someone else made the first emoticons. However, there's no documented evidence of these claims. Fahlman writes:

I probably was not the first person ever to type these three letters in sequence, perhaps even with the meaning of “I’m just kidding” and perhaps even online. But I do believe that my 1982 suggestion was the one that finally took hold, spread around the world, and spawned thousands of variations. My colleagues and I have been able to watch the idea spread out through the world’s computer networks from that original post.

So there you have it. If you ever have typed in :-) to express joy or laughter, or :-( to show sadness in a message, it's Fahlman you have to thank.

Source

"Thank you to more than 10 million Starfield explorers for creating the biggest launch in Bethesda history," says the Twitter announcement from earlier today.

We already knew Starfield had become Bethesda's biggest game in the studio's history. It was announced on September 8 when the game passed six million players. Xbox Head Phil Spencer also revealed back then that concurrent player numbers had surpassed one million on launch day.

The role-playing game was released on September 6 across Xbox Series X|S, PC, Game Pass, and Xbox Cloud Gaming as a Microsoft platform exclusive. Keep in mind though that $100 (and higher) Special Edition owners did receive five days of early access.

Unfortunately, Bethesda has not revealed any platform specific numbers for this 10 million player milestone, nor which edition of the game they are playing through.

Even with Xbox and PC Game Pass being a factor, Starfield seems to be moving plenty of units as a full-priced game too. The title continues to stay in Steam's global Most Played and Top Sellers charts. It also became Europe's biggest game launch from Xbox, making it the fastest-selling new IP of 2023 in the region.

Looking forward, at least one major expansion is confirmed for Starfield though details about it are still missing. Bethesda has also promised "years and years" of updates to the sci-fi game to add new features, resolve remaining issues, and optimize the experience. Official modding support across all platforms is incoming in early 2024 as well, though that hasn't stopped PC players from tinkering with the title quite heavily already.

Source

Back in September of last year, Ubisoft announced it is taking Assassin's Creed back to its roots with the next entry in the franchise. Today, the company has revealed the PC features and system requirements for the upcoming Assassin's Creed Mirage.

According to the official blog post, Mirage won't require too beefy of a system to run. For 1440p at High settings, Ubisoft recommends at least an NVIDIA GeForce RTX 2070 or AMD Radeon RX 5700 XT. To run the game at 4K Ultra, a GeForce RTX 3080 or Radeon RX 6900 XT is advised.

In terms of graphic technologies, Mirage on PC will include ultrawide monitor support, HDR, and uncapped framerates. One notable inclusion is XeSS Super Sampling, Intel's AI-powered upscaling solution. Ubisoft has partnered with Intel for Mirage so that XeSS will be the only upscaling tech supported at launch.

Competing options like NVIDIA DLSS and AMD FSR are not mentioned, which means the game doesn't support them. While DLSS and FSR could still be added later, Ubisoft's silence implies XeSS will be the sole upscaling option for now.

1080p, Low Preset, 30 FPS

• CPU: Intel Core i7-4790K (Intel Core i5-8400 for Intel Arc with ReBAR)/AMD Ryzen 5 1600
• GPU: Intel Arc A380 6GB/NVIDIA GeForce GTX 1060 6GB/AMD Radeon RX 570 4GB
• RAM: 8GB (dual-channel mode)
• OS: Windows 10/11
• SSD Storage: 40 GB

1080p, Hight Preset, 60 FPS

• CPU: Intel Core i7-8700K/AMD Ryzen 5 3600
• GPU: Intel Arc A750 8GB/NVIDIA GeForce GTX 1660 Ti 6GB/AMD Radeon RX 5600 XT 6GB
• RAM: 16GB (dual-channel mode)
• OS: Windows 10/11
• SSD Storage: 40 GB

1440p (2K), Hight Preset, 60 FPS

• CPU: Intel Core i7-9700K/AMD Ryzen 7 3700X
• GPU: Intel Arc A770 8GB/ NVIDIA GeForce RTX 2070 8GB/AMD Radeon RX 5700 XT 8GB
• RAM: 16GB (dual-channel mode)
• OS: Windows 10/11
• SSD Storage: 40 GB

2160p (4K), Ultra Preset, 60 FPS

• CPU: Intel Core i5-11600K/AMD Ryzen 5 5600X
• GPU: NVIDIA GeForce 3080 10GB/AMD Radeon RX 6900 XT 16GB
• RAM: 16GB (dual-channel mode)
• OS: Windows 10/11
• SSD Storage: 40 GB

Assassin's Creed Mirage is set in 9th-century Baghdad and returns to the stealth-focused roots of earlier franchise entries.

In August, Ubisoft moved a release date forward for Mirage. The game launches on October 5 for PC, Xbox Series X|S, Xbox One, PS5, and PS4. You can pre-order the Standard or Deluxe Edition for $49.99 and $59.99, respectively.

As an Amazon Associate when you purchase through links on our site, we earn from qualifying purchases.

Source

Earlier today, a huge amount of confidential information about Microsoft's current and upcoming plans for Xbox hardware and games was uploaded to a federal court site. Now it looks like someone from the company might be responsible for this huge info leak.

Bloomberg reports, via unnamed sources, that the information was uploaded to the court site "accidentally" by someone at Microsoft. The site has been the source of info on Microsoft's current legal fight with the US Federal Trade Commission (FTC) over the company's plans to acquire Activision Blizzard.

In a post on X (formerly Twitter), Douglas Farr, the FTC's director of the office of public affairs, wrote, "The FTC was not responsible for uploading Microsoft’s plans for its games and consoles to the court website."

In a follow-up X post, Farr wrote that the judge in the FTC vs. Microsoft case revealed that Microsoft provided a secure cloud link on September 14 and the court uploaded what was contained in that link to its website, which inadvertently included "non public information".

The judge says that the private Microsoft info has been removed from the court site. However, it was way too late, as the leaked information went all over the internet this morning. It included info on Microsoft's next Xbox consoles, including a Xbox Series X refresh planned for 2024 and a new console in 2028.

It also contained info on Microsoft's next Xbox controller and a leaked schedule for upcoming games from its Bethesda Softworks studios. Microsoft has yet to offer any official or public comments about the information from this leak.

The court now says both the FTC and Microsoft will provide "amended trial exhibits" by September 22. In addition to using a secure cloud link, the parties will also "file a written certification" that will confirm that the FTC and Microsoft have reviewed the files contained in those links and "certify they only contain public information in accordance with the Court's orders."

Source

Intel previously announced it would be rebranding its CPUs with new Core and Core Ultra branding, ditching the "i" brand in the process. In a press release, the company offered up some more details on these chips, which have been described under its "Meteor Lake" code name in the past.

Intel stated:

Core Ultra integrates an NPU into client silicon for the first time. The NPU is built to enable low power and high quality and provide entirely new PC experiences. It is ideal for workloads migrating from the CPU that need higher quality or efficiency, or for workloads that would typically run in the cloud due to lack of efficient client compute.

Core UItra chips will also include an integrated Intel Arc GPU which will have "performance parallelism and throughput, ideal for AI infused in media, 3D applications and the render pipeline." Intel says that the Core Ultra chip is the first to use its Foveros packaging technology.

Intel also announced plans to launch new 5th generation Xeon processors on December 14 as well. The company said that the new data center CPUs will offer better performance and faster memory compared to earlier models, but will still use the same amount of energy. The company demoed a single 5th Gen Xeon running ChatGPT on the cloud which sounds pretty impressive.

The company also announced the general availability of the Intel Developer Cloud. It's designed to help developers test high end computing applications and AI services. It stated:

The cloud development environment also provides access to the latest Intel hardware platforms, such as 5th Gen Intel Xeon Scalable processors (code-named Emerald Rapids), which will become available in the Intel Development Cloud in the next few weeks and launch on Dec. 14, and Intel Data Center GPU Max Series 1100 and 1550.

Intel also briefly showed off an AI demo of its upcoming Lunar Lake CPUs on state at the keynote event which are currently scheduled to launch in late 2024 or early 2025.

Source

In a blog post, NVIDIA says the new update will add what it calls Ray Reconstruction. This is a supercomputer AI-based method for enhancing ray-traced images in games, as opposed to the current method of using hand-tuned denoisers. NVIDIA states:

Using training derived from the analysis of offline-rendered images, which require far more computational power than can be delivered during a real-time game, Ray Reconstruction recognizes lighting patterns from training data, such as that of global illumination or ambient occlusion, and recreates it in-game as you play. The results are superior to using hand-tuned denoisers.

You can see the effects in the YouTube video above as DLSS 3.5 is used in Cyberpunk 2077: Phantom Liberty. In a separate blog post, NVIDIA shows how the game gets a performance boost with DLSS 3.5 and the game's Ray Tracing: Overdrive mode on GeForce 40 Series GPUs:

At 4K, with full ray tracing enabled, along with max settings, NVIDIA DLSS 3.5 using Frame Generation, Ray Reconstruction, Reflex, and Super Resolution is 4.9X faster on average than native 4K rendering, while also looking substantially more realistic and immersive.

Look for the new NVIDIA GeForce drivers to be released on September 21 with the DLSS 3.5 support for Cyberpunk 2077.

NVIDIA has also announced that other games are launching with DLSS support for GeForce GPUs. The open beta test for Nexon's sci-fi action game The First Descendant begins today with DLSS 3.0 support. Two more games, Warhaven and Witchfire, launch September 20 with DLSS 3.0 support. Also, Lies of P and Party Animals launch this week with DLSS 2 support.

Source

Titled "ZeniMax Management Forecast", this release schedule lists what console/PC and free to play/mobile titles were planned by the company in 2020. This document had been a part of Microsoft's probe into acquiring ZeniMax Media, its studios, and IPs.

Alongside a range of titles that we have already seen releases and announcements for, the leak namedrops remasters for Bethesda's grand RPGs The Elder Scrolls IV: Oblivion from 2006 and Fallout 3 from 2008. A brand-new Doom entry named "Year Zero" is also listed here, plus two post-launch DLC packs for the unannounced title.

Coming from Bethesda's other studios, there are mentions of a Ghostwire: Tokyo sequel as well as Dishonored 3. There are also three projects listed with only codenames to go by, which are Project Kestrel, Project Platinum, and a mysterious "licensed IP game."

Check out the slide below for the full details, though keep in mind the mentioned timelines are in the company's fiscal years. We may see some of these games be revealed soon by Microsoft as exclusives at the upcoming Game Awards showcase or during next year's Xbox Showcase.

Via FTC

Being a document from 2020, we can't be sure which unannounced projects from the leak are still in development, under what studios, and how many delays have hit them since dealing with COVID-19 setbacks and other factors.

For example, The Elder Scrolls VI has a fiscal year 2024 release planned in the document, though now we know it's still years away. MachineGames' upcoming Indiana Jones game is also listed as coming next year, though take it with a grain of salt until any official announcements come through.

Source: FTC

Source

Google is currently facing an antitrust probe from the US government which is trying to show that the search giant is using anticompetitive practices to maintain its dominance to maintain its position in search. Google says people use its products because of their superior quality.

The decision on whether to allow the public and reporters into particular hearings will be up to Judge Amit Mehta. It’s worth noting too that it’s not unusual for cases to move behind closed doors when it comes to sensitive pricing information.

Given the size of Google and how it plays an important role in most people’s lives, it could be argued that this case is different to other business cases that are held behind closed doors and deserves to be open to the public.

According to Reuters, David Dahlquist, who is speaking for the government, complained to the judge about a document which was redacted. The document was a short back and forth about how Google prices online ads.

Dahlquist argued that information such as this should not be redacted as it ‘satisfies public interest because it’s at the core of the DOJ case against Google.’

Google is a large company that operates different tools and services including Android, Chrome, Search, and YouTube.

If the judge in the case comes down on the side of the government it could potentially lead to the break up of various parts of the Google business or it could be made to present users with competitors’ products on platforms like Android rather than pre-installing Chrome or setting Google Search as the default search engine.

Source: Reuters

Source

Well, it just does not stop there as the document also explains how Microsoft plans to overthrow Sony and become the console market leader by fiscal year 2030. Just like with Windows, the plan is to devise a hybrid generation of consoles leveraging the power of the client and cloud.

While the current generation of consoles has been using AMD's APU's powered by Zen CPU and RDNA GPU, the next-gen Xbox devices will marry Arm64 with x64, and the latter will be in the form of Zen 6 cores. In terms of GPU, Microsoft is planning to either work with AMD to build a semi-custom chip or use AMD's Navi 5/ RDNA 5.

Interestingly, according to the roadmap, the company may be currently working on its Arm64 approach:

The leak also suggests the next generation of Xbox, likely coming out in 2028, will have Nvidia DLSS-like alternative. While the project does not have a name yet, the tech giant refers to it as Machine Learning (ML) based Super Resolution and Microsoft says ML and AI will be used for the "optimization and acceleration of game performance". Dedicated NPU (Neural Processing Units) may do the heavy lifting for AI and ML tasks.

All this will culminate in what the company is referring to as "One Microsoft" where it hopes for "full convergence" of its ecosystem.

Keep in mind though that the document is from October of 2022 so nearly a year has passed since which means the plans may have changed slightly. However, hardware upgrade cycles do not typically change overnight as they require long-term commitments and strategies.

Source

Coming with the codename Sebile, the shape of the controller will remain the same from the Xbox Series X generation. However, what seems to be a more grippy material is covering the bottom half of this iteration, while the top half keeps the same plastic finish fans have come to expect from Xbox controllers. The middle share button is repositioned somewhat.

As the document shows, the biggest additions are on the inside, with Sebile featuring "precision haptic feedback" drivers as opposed to the current rumble functionality. These Voice-coil actuator (VCA) haptics can also double as speakers. An accelerometer as well as quieter buttons/thumb sticks are also a part of it.

Haptics for more immersive game experiences have been a major promotional point of Sony's PlayStation 5 DualSense controllers. If Microsoft is also jumping into the technology, we should see much more titles support the refined vibration tech.

The Sebile Xbox controller is also listed as shipping with Xbox Wireless 2, Direct-to-Cloud (possibly similar to Stadia controller), and Bluetooth 5.2 support. Removable battery compartment is returning as well. A new Lift to Wake feature and modular thumbsticks are mentioned too, with the latter hopefully being a cheaper solution to the analog stick drift issue.

The documents reveal that Sebile will be included with the newly leaked Xbox Series X|S refreshes. The controller itself will come in at $69.99 with support for consoles, PC, and mobile devices. As expected, special and limited edition versions are planned as well.

Source: FTC via The Verge

Source

A new wave of documents has just leaked from the FTC v. Microsoft courtroom battle, which was supposed to be redacted, and it seems to have revealed a large part of Microsoft's game plan for its Xbox business. One of the biggest leaks is the existence of a new Xbox console that's meant to refresh the current Series X.

Ditching the current rectangular cuboid design, this upcoming Xbox Series X, Codenamed "Brooklin", looks to be going for a more cylindrical form, at least from the front angle seen in one of the leaked documents.

As this is a refresh of the hardware and not an upgraded "Pro" version, Brooklin's performance should remain the same as the current Series X. However, shrinking the current 6nm die powering the machine has reportedly lowered the power usage by 15%. 2TB of internal storage, a USB-C port in the front for power delivery, as well as Wi-Fi 6E and Bluetooth 5.2 support are also listed as features.

However, the confidential document also reveals Microsoft is leaving behind the disk drive for this iteration, something that only the cheaper Xbox Series S has done for now. Microsoft may be opting to offer a separate disk drive for Xbox fans that want to stick to physical media or utilize their back catalogs, something that Sony is also rumored to be doing.

As for when people can get their hands on this refreshed Xbox Series X, another leaked document regarding launch timelines gives some targets Microsoft is hoping to hit in its 2025 fiscal year.

Per the leak, the new cylindrical Series X is planned for November 2024 with the same $499 price tag as the current console. While the Series S just recently received a small refresh with doubled storage capacity, the document may be alluding to another refresh (Codename "Ellewood") coming in 2024 just behind Brooklin's launch as well.

Source: FTC via The Verge

UPDATE

Details regarding a next-generation Xbox Controller with haptics and an accelerometer, Microsoft's long-term plans for its Xbox consoles through 2030, and even Bethesda's launch schedule for upcoming games (including Fallout and Oblivion remasters) have also leaked from the same FTC source.

Source

Online platform X, formerly known as Twitter, could introduce a monthly fee for all users, its owner Elon Musk has said, citing the need to cut down on bots.

The tech tycoon has made multiple changes since taking over the site for $44bn in October last year.

He has fired thousands of employees, introduced a paid premium option, cut content moderation and reinstated formerly banned accounts, including that of former United States President Donald Trump.

In July, he said the platform had lost roughly half its advertising revenue.

Bots – accounts run by computer programs rather than humans – are common on X, where they can be used to artificially amplify political messages or racial hatred.

During a talk with Musk on Monday, Israeli Prime Minister Benjamin Netanyahu raised the question of online anti-Semitism and how X could “prevent the use of bots – armies of bots – to replicate and amplify it”.

Musk replied that the company was “moving to having a small monthly payment for use of the X system”.

“It’s the only way I can think of to combat vast armies of bots,” he said. “Because a bot costs a fraction of a penny – call it a tenth of a penny – but if somebody even has to pay a few dollars, some minor amount, the effective cost of bots is very high.

“And then you also have to get a new payment method every time you have a new bot.”

Currently, X Premium costs $8 a month in the US but charges depend on which country the user is in.

Source : https://www.aljazeera.com/news/2023/9/19/elon-musk-says-x-formerly-twitter-could-charge-a-monthly-fee-for-access

Source

The substrate forms an important base for the processor. While they are made from organic materials these days, Intel wants to make them using glass instead.

Computer processors, which form the most important part of computing, are not made of just silicon-based CPU chips themselves. They also contain substrates, which form the base of the processors.

The substrates are small circuits that contain the wires connecting the CPU chips and form the base of the processors.

Years ago, processor substrates were made using leadframe / wirebond. That changed when the industry moved toward ceramic substrates. Then came organic substrates in the 1990s, whose rollout was led by Intel.

Since then, substrates have not seen many changes. Now organic substrates, which are mostly made of plastic, are not able to handle the advancement in processor packaging. So Intel has decided something new needs to be used to make them. Hence, Intel has decided to use glass instead.

Intel Processors To Be Made From Glass

In an official announcement today, Intel said that it has developed the industry’s first glass substrate for advanced packaging of processors. This comes after Intel’s decade-long research into it.

Intel says that it’s a better replacement than the currently used organic substrates. Quoting the announcement:

Compared to today’s organic substrates, glass offers distinctive properties such as ultra-low flatness and better thermal and mechanical stability, resulting in much higher interconnect density in a substrate.

Reasons For This Move

Simply put, the more transistors, the faster the processor. As we are going into the single-digit nanometer (a marketing term) range of transistor sizes, making processors is getting very complicated. Reducing the size of the transistors inside chips is getting harder and harder.

This is why processor makers are moving towards chiplet-based processors. Where a single processor is not made from a single chip, but instead contains multiple CPU chips inside a single processor.

The best example is AMD, which uses multi-chiplets in its Ryzen processors. But not just AMD, even Intel is moving towards a multi-chiplet approach, where it will start using tiles on its processors starting with the Meteor Lake series of processors.

So something is required to connect these chiplets or tiles properly. This is why processor substrates become even more important.

Quoting Intel:

“As the demand for more powerful computing increases and the semiconductor industry moves into the heterogeneous era that uses multiple “chiplets” in a package, improvements in signaling speed, power delivery, design rules and stability of package substrates will be essential. Glass substrates possess superior mechanical, physical and optical properties that allow for more transistors to be connected in a package, providing better scaling and enabling assembly of larger chiplet complexes (called “system-in-package”) compared to organic substrates in use today. Chip architects will have the ability to pack more tiles – also called chiplets – in a smaller footprint on one package, while achieving performance and density gains with greater flexibility and lower overall cost and power usage.”

Glass Benefits

Intel has given many reasons and benefits for using glass over the currently used organic substrates:

“Glass substrates can tolerate higher temperatures, offer 50% less pattern distortion, and have ultra-low flatness for improved depth of focus for lithography, and have the dimensional stability needed for extremely tight layer-to-layer interconnect overlay. As a result of these distinctive properties, a 10x increase in interconnect density is possible on glass substrates. Further, improved mechanical properties of glass enable ultra-large form-factor packages with very high assembly yields.”

Simply put. Glass is flatter, has better stability and has better connectivity. Making it ideal for using it as a substrate that is going to be used in connecting CPU chiplets.

Rollout Date

Intel mentions that by 2030, the transistors on a CPU package will reach 1 trillion. So such technological innovations are required.

This is in addition to innovations such as RibbonFET, which is Intel’s version of GAAFET which is an advanced method of making CPU transistors. Not to forget PowerVia backside power delivery, which changes how processors are powered, improving the CPU’s efficiency.

Intel believes that such innovations mean that glass substrates will be required going forward. It says that glass substrates are going to be used in CPU processors starting in the later parts of this decade.

The photos above show that it’s being tested on smaller processors too, meaning it will not be limited to just server processors but will be used on desktop and notebook processors too. Meaning, years later, the CPU you will buy might use these new glass substrates.

Here’s hoping that people don’t drop these CPUs while installing them. It would be very expensive to do so. Maybe people should ask Intel to make sure there is one free glass replacement available for every single CPU made by them.

Source

Source

Unity first announced changes to its business model last week, stating that on January 1, 2024, it would begin will charge developers a per-install fee for its Runtime software after a certain number of installs and revenue levels have been reached. That plan was quickly slammed by the gaming development community in general, and Unity developer in particular.

Now, Bloomberg is reporting that a possible new plan may be officially announced soon. The company held an all-hands meeting Monday with its employees to discuss the plan, according to the report. It states:

Under the tentative new plan, Unity will limit fees to 4% of a game’s revenue for customers making over $1 million and said that installations counted toward reaching the threshold won’t be retroactive, according to recording of the meeting reviewed by Bloomberg.

The new changes are not yet official, as Unity executives are still getting feedback from their partners, in an effort to avoid the massive influx of protests that happened last week.

Another change in Unity's model will allow game developers to submit reports about runtime installs themselves. Previously, Unity said it would use its own proprietary tools to monitor the number of installs.

Even if these changes do go into effect, the truth is that Unity has lost a lot of trust from many game developers who have used its tools over the years. Some of them will likely never return, and other game developers who thought about accessing Unity tools might now think twice and then decide to use a rival game engine like Epic's Unreal Engine or something else.

Source

According to a post by known leaker Panzerlied on the Chiphell forums, Nvidia aims for a 50% increase in CUDA cores for the RTX 5090, bringing the total to a massive 24,576. That's over 8x the cores of Turing-based RTX 2080. Memory bandwidth is also slated to receive a 52% bump through use of 32Gbps GDDR7, up from GDDR6X on the current RTX 4090.

Clock speeds could see one of the biggest generational jumps with a projected 15% increase. This would see the RTX 5090 boost up to 2.9GHz, easily surpassing 3GHz during gaming workloads. The existing RTX 4090 maxes out at 2.52GHz.

To further boost performance, Nvidia reportedly plans to triple the L2 cache to 128MB - far exceeding AMD's 96MB on the RX 7900 XTX. This additional caching aims to maximize the higher memory bandwidth enabled by GDDR7.

If Nvidia can deliver on these ambitious targets, the RTX 5090 would provide over 70% greater performance than the RTX 4090 based on specs alone.

NVIDIA RTX 5090 vs. RTX 4090

• 50% more cores than the RTX 4090

• 52% increase in memory bandwidth

• 78% larger L2 cache (likely 128MB)

• 15% higher frequencies, pushing boost clocks over 3.0GHz

• Around 70% higher overall performance

While unconfirmed, the leaked specs so far point to a new Blackwell-based GPU codenamed "GB202" that could push gaming performance to new heights. It remains to be seen if Nvidia can maintain a reasonable 3-slot form factor for such a powerful chip.

Source: Chiphell forum (Chinese) and VideoCardz

Source

While things don’t appear good for gaming PCs, the market is actually more resilient than the rest of the PC market. IDC says this is because buyers are looking to take advantage of lower prices of last-generation products and discounts on current generation products due to excess inventory.

Commenting on the predictions, Jay Chou, research manager for IDC’s Worldwide Quarterly PC Monitor Tracker, said:

‘After five quarters of year-on-year volume declines, the gaming monitor market managed to regain positive momentum in the second quarter. Improved cost structures and continued industry focus means we should expect a stronger recovery in this part of the gaming device landscape. With a typical gaming monitor costing a little over $300, it presents a cost-effective way to improve user experience both in and out of gaming.’

While things suck for gaming PC shipments right now, things will improve in the coming years, according to IDC. Between 2023 and 2027, the compound annual growth rate (CAGR) of gaming PCs will reach 4.6% while gaming monitors will have a 7.7% CAGR.

IDC said that one of the main factors in the decline in gaming PC shipments is the poor economic situation. In the United States, inflation is still above the 2% target set by the Federal Reserve and currently sits at 3.7%.

This means that prices in shops are rising about twice as fast as normal and wallets are under more pressure. In addition, higher interest rates mean that borrowing is more expensive so it’s harder to lend money for consumers to buy gaming hardware on credit.

Source: IDC

Source

Bloomberg reports, via unnamed sources, that Panay will be taking over as head of Amazon's hardware division. He is said to be replacing Dave Limp, who has been the Senior Vice President of Devices and Services at Amazon since 2010. Limp announced in August he would step down from that role. So far, neither Panay nor Amazon have commented on Bloomberg's report.

If the story is accurate, it would certainly be an interesting change for Panay, who helped to lead the development of Microsoft's first Surface PC products in 2012. The Surface family, which first started out as tablets for Windows 8 and Windows RT, expanded to full laptops and all-in-one PCs over the past 12 years.

Panay's switch over to Amazon will have him lead a team that's best known for selling more affordable products than Microsoft's Surface devices, most of which are on the high end of prices. By contrast, products like Amazon's Fire tablets, Kindle Readers, and Echo smart speakers, have started out as affordable, and get discounted even more during sales events, like the upcoming Amazon Prime Big Deal Days on October 10-11.

It certainly will be interesting to see if Panay will continue to help create affordable tech gadgets, or if he will expand Amazon's hardware to more expensive and more full featured devices.

Amazon is scheduled to hold a hardware-focus press event on Wednesday, September 20 at its Northern Virginia headquarters. It's more that possible Panay could be there to offer up a preview of what he might come up with when he's fully in charge of the company's tech products.

Source

In a press release, Intel says it has been researching glass substrates for at least 10 years and now feels that it could replace organic materials by the end of this decade. Intel says:

Glass substrates possess superior mechanical, physical and optical properties that allow for more transistors to be connected in a package, providing better scaling and enabling assembly of larger chiplet complexes (called “system-in-package”) compared to organic substrates in use today. Chip architects will have the ability to pack more tiles – also called chiplets – in a smaller footprint on one package, while achieving performance and density gains with greater flexibility and lower overall cost and power usage.

In addition to packing in more transistors in the same amount of space, using glass substrates has other advantages, according to Intel. They include being able to operate at higher temperatures while at the same time having "ultra-low flatness for improved depth of focus for lithography."

Intel estimates that glass substrate-based chips could have as much as 10 times the interconnect density compared to current chip packages.

If you are thinking that new Intel CPUs and GPUs could be made will glass substrates you would be correct. However, don't expect them to show up in your next gaming PC in 2024, or even 2025.

Intel says that the first chips with these glass packaging materials will be coming by 2030 for "data-intensive workloads such as artificial intelligence (AI)." That likely means this new tech won't show up in consumer and gaming PCs until a decade from now.

Source