Not long ago, Keychron announced Zephyr Gaming Mouse as an upcoming open-source firmware project for gaming mice. More recently, the peripheral maker announced MKF (Magnetic Keyboard Firmware), an open-source firmware project designed for TMR and Hall effect gaming keyboards. The project has yet to feature any public code contributions, but there is a landing page that outlines what Keychron aims to do with the project.

MKF is meant to serve as "an open firmware foundation for magnetic gaming keyboards, with room for the project and its community to grow," and its current feature outline includes support for the obvious, like 8 kHz polling, adjustable actuation, Rapid Trigger, and SOCD, but it will also apparently support custom response curves, PDO—sending more keypresses the deeper a key is held—and controller emulation. It currently seems as though only wired connectivity is planned for the initial release, but the project will be open to new features and use cases outside of competitive gaming.

AMD has reportedly cut its gaming GPU supply in China by at least 20% to 50% since the start of the fourth quarter in October. According to a November 12 post on Board Channels spotted by Videocardz, AMD's shortage is now comparable to NVIDIA's, with very few gaming chips going to add-in board (AIB) partners. Their monthly allocations are apparently still shrinking, and the source expects supply to remain extremely tight until January.

The same source describes Radeon prices in China climbing every quarter, with smaller increases month to month within each quarter, and puts that down to rising memory and GPU costs. AMD is reportedly raising the price of its GPU and GDDR6 kits again this month, after an earlier increase in October. Things aren't much better on the green side either, with an NVIDIA partner saying in late October that it couldn't buy RTX 5090 GPU and GDDR7 kits anymore, while RTX 5090 and 5080 prices continue to skyrocket. Graphics cards were already selling 44% above MSRP on average in October, so a smaller pool of Radeon cards is unlikely to help matters.

GIGABYTE brings anime aesthetics and gaming performance together with the AORUS ARI Bundle, featuring the AORUS GeForce RTX 5060 ARI 8G graphics card and B850 AORUS ELITE-P ICE motherboard. Inspired by Ari, AORUS's original anime character, this exclusive bundle combines distinctive character artwork, sleek white hardware, and advanced gaming technologies to create a PC build that truly stands out.

At the heart of the bundle is the AORUS GeForce RTX 5060 ARI 8G, powered by NVIDIA's Blackwell architecture and equipped with 8 GB of GDDR7 memory. Featuring NVIDIA DLSS technology and advanced ray tracing capabilities, the graphics card delivers enhanced visuals and smooth gaming performance. GIGABYTE's WINDFORCE cooling system incorporates patented Hawk fans, server-grade thermal conductive gel, and composite copper heat pipes to optimize heat dissipation. Its reinforced metal backplate, featuring exclusive ARI artwork, combines long-term durability with a bold anime-inspired design.

Buy GIGABYTE AORUS ARI Bundle on Newegg!

AMD's Adrenalin 26.11.1 optional driver is now available with the promised fix for the Radeon bug that kept system drives writing in the background. According to the release notes, the update addresses excessive writes to the system drive while the PC is idle on Radeon RX 7000 series cards and newer. It also fixes a related case where excessive writes could show up while moving or resizing application windows.

AMD confirmed the problem earlier this week, tracing it to the External Events Client Module (atieclxx.exe) in the 26.10.1 and 26.10.2 drivers. User reports put the write rate anywhere from 0.3 MB/s to 5 MB/s, which works out to roughly 26 GB to 432 GB a day on a machine that's doing nothing. Until now, affected users had to disable the External Events Utility, turn off Resizable BAR, or use a community-made open-source tool that blocks the writes. The release notes only mention RX 7000 and newer for the idle fix, so it isn't clear whether older Radeon cards were affected at all.

AskAnyModel AI Pro Plan is now available for $39.99 (MSRP $499) for lifetime access through Nov. 8. The browser-based platform brings more than 50 AI models into one workspace, with tools for comparing responses, generating images, and switching models during a conversation.

Its main comparison feature sends one prompt to two to six selected models simultaneously. Responses appear side by side, along with response times and token counts. For writing, coding, research, or brainstorming, this lets users assess different approaches without repeating a question across separate services. Comparing answers can help identify useful differences.

Get the AskAnyModel AI Pro Plan Lifetime Subscription for $39.99 (MSRP $499).

The Witcher 4 has been voted Most Wanted Game at the Golden Joystick Awards 2026, beating the likes of Fable and Resident Evil Veronica. CD Projekt Red production director Małgorzata Mitręga accepted the award near the end of the November 11 show, just before Resident Evil Requiem was crowned the night's big winner. It's a step up from last year, when The Witcher 4 came in a close second in the same category with 16.68% of the vote against GTA 6's 18.03%. GTA 6, which launches on November 19, wasn't on this year's ballot. Here's what Mitręga said on stage:

This is a huge, huge honor for us, but it's also a great responsibility. We all know it. We are working hard, giving heart and soul there, all the skills we have, passion we have, and moments like that are such a boost for us, for the stamina, to (keep) it going, to deliver (for) you when the time is right.

Bing users who would rather skip the AI answers now have two official ways to do it, according to a new post on Microsoft's Bing Search Blog. The first is the Microsoft Bing AI Search Choice extension for Chrome and Edge, which adds a single toggle for AI-generated answers that applies to every Bing search. The second needs no install: adding "-ai" to the end of a query removes the AI answer for that one search. Microsoft says its research shows users want "to choose whether or not AI is involved in every session," though it hasn't published that research.

Installing the extension also makes Bing the browser's default search engine, and there doesn't seem to be a way to get one without the other, so it looks like a play for Chrome users as well. It's also still labeled a preview, and it isn't new, since Microsoft's head of search, Jordi Ribas, first announced it back in June. Bing doesn't appear to have a matching switch in its own settings either, so anyone on another browser has to type the operator every time.

What's the News? AMD today announced AMD Versal tamper-resistant ("TR") devices, designed to protect security-critical information against sophisticated physical attacks while delivering the high-performance heterogeneous processing capabilities of AMD Versal AI Edge Series Gen 2. Versal TR devices are the industry's first adaptive SoC or FPGA-based devices providing built-in protection against physical attacks on the silicon backside. A sensor-based protection die is bonded to the main compute die using 3D active-on-active technology, shielding the entire backside of the device, protecting sensitive information, detecting physical attacks and preventing boot if the protection die is tampered with or removed. Combined with embedded CPUs, AI Engines and programmable logic, Versal TR devices help customers strengthen the protection of security-critical information without compromising system performance, size, weight, power or time to market.

Why It Matters? Protecting sensitive information from sophisticated physical attacks is a priority across markets and embedded applications. From data center networking and fintech applications to defense systems and critical industrial infrastructure, customers seeking higher levels of device security today typically build custom solutions in-house using mechanical enclosures, ASICs and other components, adding development cost, specialist resources, verification effort and re-spin risk. Versal TR devices integrate heterogeneous processing with enhanced security features including backside physical attack protection in a single device. This reduces the need for separate compute, security and enclosure-based protection components. This approach can save board area and simplify board design versus custom solutions while giving customers access to AMD tools, technical support and a broad partner ecosystem.

Alphacool is known for its high-end PC case and cooling components, and its latest product, the Apex Benchtable, is no exception. The Benchtable is an open-air case built for flexibility and rapid assembly and disassembly, and it supports everything from Mini-ITX to E-ATX form factors, as well as radiators up to 360 mm and a smorgasbord of water cooling accessories. The Apex Benchtable comes in at €449.98 on the Alphacool online shop.

The Apex Benchtable is a modular frame consisting of CNC corner components connected by carbon fiber tubes. Owing to that modular design, the included 120 mm, 240 mm, and 360 mm fan and radiator mounts can be easily moved and repositioned to accommodate various hardware configurations or setups. It ships with a repositionable front I/O module with 2x USB 3.0 ports, a power button, and separate 3.5 mm mic and headphone jacks.

TechPowerUp partners with Formula V Line to bring our readers from all over the world a chance to win one of dozens of pieces of Formula V One cases, CPU coolers, and case fans! The star attraction here is the Formula V Line G10 case, A spacious mid-tower case with a unique front-panel design that allows you to louver airflow from the three intake fans individually. The case comes in three attractive 2-tone color trims—silvery white, copper black, and gray black. There are a total of ten lucky winners to be picked from the Giveaway:

The first lucky winner gets a Formula V Line G10 case, plus an Ice Nova Pro 3 CPU cooler, and a set of three Air Bridge VT Vision case fans. The second lucky winner gets a Formula V Line G10 case, and an Ice Nova Pro 3 cooler. The third lucky winner gets a Formula V Line G10 case. Winners 4th thru 10th get a selection of Ice Nova Pro 3 coolers and Air Bridge VT Vision 3-fan sets. Once again, the Giveaway is open worldwide (wherever legal); and open from today until November 20.

For more details and to participate, visit this page.

G.SKILL, a leading manufacturer of high-performance memory and PC gaming peripherals, has teamed up with Micro Center to offer a specially priced PC hardware combo deal featuring the Intel Core Ultra 7 270K Plus processor, G.SKILL Ripjaws S5 DDR5 memory, and GIGABYTE Z890 EAGLE WIFI7 PLUS motherboard. This hardware bundle offers PC enthusiasts, gamers, and content creators an opportunity to build or upgrade their systems with high-performance components while enjoying substantial combo savings.

The bundle pairs the Intel Core Ultra 7 270K Plus processor, featuring 24 cores (8 Performance-cores and 16 Efficient-cores) and a maximum turbo frequency of up to 5.5 GHz, with the GIGABYTE Z890 EAGLE WIFI7 PLUS motherboard and a 32 GB (2x16GB) G.SKILL Ripjaws S5 DDR5-6000 CL36 memory kit. Together, these components provide a foundation for a high-performance Intel desktop system.For more information and to explore this combo deal, visit the Intel Core Ultra 7 270K Plus and G.SKILL Ripjaws S5 bundle page at Micro Center.

Penguin Solutions, Inc., the AI Factory Platform Company, today introduced the SMART CXL-8F2W-D4, an eight-DIMM DDR4 Compute Express Link (CXL) memory add-in card that enables organizations to adapt DDR4 memory for use in AI systems. Using DDR4 as an alternative to newer-generation DDR5 memory provides an innovative approach offering performance, sourcing, and economic benefits for the buildout of AI infrastructure.

According to TechInsights, Inc., AI data centers account for 34% of DRAM demand in 2026, a figure expected to climb to nearly 50% by 2029. This rapid growth is driving up memory costs and putting increasing pressure on infrastructure budgets. For organizations deploying new servers, whether as part of a technology refresh or a new data center buildout, the rising cost of memory presents a significant challenge, underscoring the need for more cost-effective ways to expand memory capacity.

Silicon Motion Technology Corporation, a global leader in designing and marketing NAND flash controllers for solid-state storage devices, today announced that it will showcase its PCIe Gen 5 and PCIe Gen 6 MonTitan enterprise SSD controllers and reference platforms designed to address the evolving storage demands of AI factories, Agentic AI, and large-scale inference deployments at Supercomputing 2026 (SC26), taking place November 16 to 19 in Chicago, Illinois, at Booth #6816. The company will also highlight its enterprise PCIe NVMe boot drive solutions, including a demonstration featuring the NVIDIA DPU platform, alongside its Ferri-eMMC embedded storage solution with the ASPEED AST2600 BMC, showcasing storage technologies for AI servers and next-generation data center infrastructure.

Alex Chou, Senior Vice President of Enterprise Storage & Display Interface Solution Business at Silicon Motion"As AI infrastructure evolves to support multi-agent and increasingly complex agentic workloads, storage performance is becoming a critical factor in overall system performance and efficiency," "Our PCIe Gen 5 and Gen 6 enterprise SSD controllers are designed to address these challenges by delivering high performance, low latency, scalable capacity, and power efficiency, helping customers build the next generation of AI infrastructure."

Microsoft Visual Studio Professional 2026 is available for just $29.97, compared with an MSRP of $499.99. The development environment combines tools for writing, debugging, testing, and collaborating on code, with support for multiple application types and deployment targets.

Built as a 64-bit IDE, Visual Studio Professional 2026 is designed to handle large solutions and complex workloads. Developers can build cross-platform mobile and desktop applications with.NET MAUI, or create responsive web interfaces in C# using Blazor. Development, debugging, and testing tools support.NET and C++ applications targeting Windows, Linux, and containers.

Get lifetime Visual Studio Pro 2026 access for just $29.97.

GIGABYTE, the world's leading technology brand, announced that the AORUS C510 GLASS INFINITY Chassis has received 2027 CES Innovation Award in recognition of its innovative design, practical engineering, and functionality. Featuring a built-in 16-inch display, flexible placement, efficient cooling, and upgradability, the award-winning chassis redefines how a desktop PC can be configured and used.

Designed to support an all-in-one PC setup, the AORUS C510 GLASS INFINITY integrates a 16-inch high refresh rate gaming display directly into the chassis. The display can serve as the system's primary screen, a secondary monitor, or a dedicated display for additional content. It can be mounted on either side of the chassis or detached and placed on its own stand for independent use, allowing users to adapt the display configuration to different environments and workflows.

Axelera AI, the leading provider of purpose-built AI hardware acceleration technology, unveiled the details behind Titania, its third-generation silicon architecture in a chiplet design. Titania is highly programmable and energy efficient and will deliver high performance at interactive speeds, all within the power and cooling limits of existing data centers, rather than the 100-kilowatt-plus liquid-cooled racks that today's highest-density GPU systems require.

Performance expectations for Titania are up to 7x the throughput per watt compared to the best data center grade servers available, and up to a 20x better TCO. Each Titania-based system-in-package (SiP) will contain two Titania chiplets which are then packaged in a range of form factors, including standard PCIe cards as well as OCP-compliant UBB baseboards with eight Titania OAM modules connected in a scale-up network. In support of over $300 million in confirmed Titania commitments, these form factors will be available via OEM partners in an inference server configuration in both air- and liquid-cooled configurations.

Today, the Universal Chiplet Interconnect Express (UCIe) Consortium, the open standards group for interconnects between chiplets within a package, announced the release of the UCIe 4.0 Specification, taking the UCIe architecture beyond chiplet-level interface design to address key requirements for integrating multiple chiplets into a cohesive on-package system.

As chiplet architecture grows more sophisticated, connectivity is only part of the challenge. Systems incorporating compute, accelerator, memory and other chiplets must also address how those components are managed, secured and monitored while efficiently using power, die area and package resources. The UCIe 4.0 Specification addresses these broader system-level requirements with new capabilities spanning manageability and security, efficiency, memory connectivity and reliability.

Corsair is proud to launch the HX1000i SHIFT CRYSTAL, its first power supply with active 12V-2x6 per-pin overcurrent protection via PinProtect+. The power supply also simplifies builds by powering up to 24 iCUE LINK devices directly from the integrated iCUE LINK System Hub, and delivers stunning visuals with the transparent side panels, an RGB-lit fan, and custom-made Japanese capacitors.

The HX1000i SHIFT CRYSTAL began as a prototype shown at COMPUTEX 2025, and the positive response led Corsair to bring it to market. Built on the current HXi SHIFT platform, it is the first Corsair power supply with PinProtect+ per-pin overcurrent protection (OCP) for the native 12V-2x6 GPU connection. Cybenetics Platinum-rated efficiency, ATX 3.1 certification, PCIe 5.1 compliance, and comprehensive Corsair iCUE software control deliver the performance, protection, and flexibility that demanding PCs need.[Editor's Note: Catch the TechPowerUp Review]

NVIDIA's co-founder, chief executive officer, and president has officially signed a letter of resignation, stepping down from the role he has held since 1993. Following a series of controversies, NVIDIA announced on Sunday that the company's board of directors is launching a full-scale independent investigation into recent reports of founder and CEO Jensen Huang having a relationship with a woman the U.S. intelligence community believes is an agent of the Chinese state. Another allegation has since emerged, claiming that Jensen had an inappropriate relationship with one of his direct reports. As a result, Jensen Huang has resigned from all his positions, including President, Chief Executive Officer, and member of the Board, effective immediately.

The company board has begun the search for a new CEO and, in the meantime, has named Colette Kress, currently serving as vice president and chief financial officer, as the interim CEO. According to the official statement, NVIDIA's board is looking for a permanent CEO and is open to considering both internal and external candidates for the position. Who exactly will take on the role is unknown, as the board must find a suitable successor. During the transition, CFO Colette Kress will continue to perform CEO duties and manage the company.

MONTECH marks its 10th anniversary with the launch of KING 95 3D and KING 95 3D PRO, alongside LINK FAN INF 120 magnetic ARGB fans and LUMI software. The new dual-chamber ATX mid-tower case adds a tempered-glass top to the KING series' signature curved-glass panorama, putting more of the build on display. KING 95 3D retains the series' curved profile while adding a tempered-glass top panel. The 15-degree curved front edge connects the front and side views, while an integrated 118-LED contour ARGB strip follows the chassis, giving the case its own lighting signature.

Noctua today introduced the new NT-CP1 AM5/4 carbon nanotube thermal pad for AMD AM5 and AM4 processors. Leveraging Carbice's novel, space-grade carbon nanotube technology, the NT-CP1 is a premium-quality, easy-to-apply thermal pad that offers superior long-term stability for peace of mind. While thermal pastes and most other thermal pads gradually lose performance over time due to pump-out, delamination, cracking and other degradation effects, Carbice pads keep improving their performance over the first few hundred to a few thousand thermal cycles. This is possible thanks to the pad's robust mechanics, phase-change polymer coating and carbon nanotubes that incrementally conform to surface microstructures, enhancing the heat transfer rate with time and thermal cycling. Once this initial activation phase is completed, the NT-CP1 maintains stable, maintenance-free performance for years and years. As such, it is ideal for everyone who seeks to rule out thermal issues due to degradation and never wants to bother with reapply paste again.

Unlike other carbon or graphite-based thermal pads that can be brittle, slippery and highly electrically conductive, Carbice pads are sturdy, safe and easy to install using a straightforward "peel, press, peel" process. The NT-CP1's aluminium backbone makes it surprisingly robust and easy to remove. The slightly tacky polymer coating helps to keep the pad in place once pressed onto the CPU and ensures that the contact surfaces are electrically non-conductive before usage unless the polymer layer is penetrated.

According to a company called Apex Compute, there is a new player in neural rendering. The company has developed Apex Neural Rendering, which reportedly offers higher quality and performance than NVIDIA's DLSS implementations. It includes three key components: frame generation (ANR FG), super resolution upscaling (ANR SR), and ray reconstruction denoising for ray tracing (ANR RR). Running frame generation on NVIDIA's GeForce RTX 5090 at 4K, the ANR FG achieves a latency of 1.01 ms per generated frame at a 2x setting, while NVIDIA's DLSS Frame Generation records 1.26 ms. We are not sure which version of DLSS FG was tested, but we can assume it was the latest version.

Regarding memory consumption, ANR FG uses only 214 MB of GPU memory, while DLSS FG requires more than double that, at 562 MB. Reportedly, the ANR FG neural network weights fit within just 0.17 MB, which is incredibly efficient. Based on samples on the company's website, the image quality that ANR delivers matches DLSS and even surpasses it in some cases. Apex Compute also managed to get Hogwarts Legacy and The Last of Us Part II Remastered working, showing the difference compared to DLSS, resulting in frame generation with more visual fidelity. ANR FG is a beta injector built around OptiScaler, which works in any game offering NVIDIA DLSS Frame Generation. You can view the results below.

Even while Valve transitioned its Steam client to 64-bit, its Linux client has remained stuck in 32-bit land, but it appears as though that is finally slated to change, with a new 64-bit Steam client for Linux finally launching in the beta branch. This follows on the recent announcement of the beta RT3 client earlier in 2026 marking the start of the transition and promising a more consistent experience across different Linux distributions.

The launch of the 64-bit beta comes alongside Steam's new Achievement Sets update that allows developers to group base game, DLC, and other achievements based on categories. There is also a new achievements page to more easily track game achievements. Ahead of the latest beta launch, Valve also responded to Epic Games's new vibe-coded Epic Games Launcher beta with a mainline Steam client update that resulted in significantly faster cold-start times.

Update Nov 12th: The new Steam Achievements Sets feature has officially been released to the Stable channel as of a November 11 Steam Client update, whose update notes also make mention of several improvements to performance, responsiveness, and stability.

Jacek Rozenek, who has voiced Geralt in Polish since the first Witcher game, is returning for The Witcher 3: Songs of the Past, though his new lines are being run through an AI voice model so they match his original Witcher 3 performance. CD Projekt Red told Game File it worked with Ukrainian AI firm Respeecher on this, with Rozenek's approval. The actor suffered a stroke in 2019 that left him unable to speak for months, and his voice is now lower and gruffer than in the roughly 10,000 lines he recorded for the base game in 2014. CDPR feared the mismatch would stand out, since new battle cries and one-liners can play in older parts of the map too.

Rozenek still performs every line himself. Respeecher's model, trained on his original Witcher 3 recordings, then reshapes the new takes to match how Geralt sounded in 2015. Rozenek himself weighed in on this, saying the point is that this voice should be as close as possible to his voice from 12 years ago. Localization producer Mikołaj Szwed called it a "special case" tied to Rozenek's health and said CDPR's contracts require consent for any AI work. He sees this as a way to use AI "ethically, with consent and also to do something good." Doug Cockle's English Geralt, apparently, needed no such treatment.

AMD's Radeon SSD write bug now has a community workaround that doesn't require disabling the External Events Utility. ComputerBase forum member Necandi (necanditool on GitHub) has released Radeon-atieclxx-SSD-Fix, a free, open-source tool that blocks the writes but leaves the service running. The developer found that atieclxx.exe keeps copying its own 660 KB executable out of the driver folder and into the Windows SYSTEM profile every second, or roughly 56 GB of writes per day. User reports put the write rate between 0.3 MB/s and 5 MB/s, and the top end works out to a massive 432 GB a day.

The tool uses Windows file permissions to deny the SYSTEM account write access to that one copied file, so each copy attempt fails and nothing new lands on the drive. Since the service stays active, FreeSync, frame pacing, and Radeon hotkeys should keep working. It also checks whether a PC is affected first, estimates the writes so far, and adds a scheduled task that reapplies the block after driver updates. Apparently, it can even lift the protection on its own once AMD patches the bug. The interface is only in German for now, and it is an unofficial tool, so readers are advised to use it at their own risk.

ROG Xbox Ally X and X20 owners can now use Microsoft's Automatic Super Resolution (Auto SR) upscaler on the handheld's own screen. Microsoft says Auto SR is out of preview on both devices and on select AMD Ryzen Copilot+ PCs. During the Xbox Insider preview that started in April, it only worked with the Ally X docked to an external display. It now works undocked and on battery, and accepts games running at 1080p. The base ROG Xbox Ally misses out, since its chip has no NPU. Auto SR runs its AI model on the NPU, so the upscaling doesn't eat into the GPU's frame time. A new Performance Mode lifts the frame rate ceiling to 120 FPS, from roughly 60 to 75 FPS before.

In Microsoft's Rocket League test on the Ally X at 720p and high settings, Quality Mode averaged around 70 FPS, while Performance Mode hit about 115 FPS. Auto SR can now be paired with Auto HDR as well, though both Performance Mode and Auto HDR support will roll out gradually through Windows updates. Microsoft warns that 1080p on the internal screen may not perform as well as other setups, although older games should still see visual gains. Ally X owners have also had an FSR 4.1 preview driver since last month, which is probably still the better pick in games that support it well, leaving Auto SR for the rest. Enabling it requires the latest Windows 11 and Game Bar updates plus the Auto Super Resolution Package from the Microsoft Store.

It's been known for a while now that Bungie has been toying with the idea of a permanent PvE mode for Marathon, much like Embark has been for Arc Raiders, but it seems as though Bungie is playing things a little more by ear, with the game studio previously announcing that a permanent PvE mode would be coming to Marathon in December. In a recent Marathon developer update, Bungie spilled the beans on what the start of Season 2 will look like and announced that there will be a free week coinciding with the launch of the Symbiosis update.

The Marathon Open Play Week, as it has been dubbed, will start on December 7 and run until December 15, and Bungie is seemingly hopeful that this will allow players to try the game and potentially turn into long-term Marathon players. December 8 marks the introduction of the new Expedition gameplay mode, which is a PvE-exclusive game mode that allows players to "form a team with your friends and explore Tau Ceti IV with a more collaborative mindset."

When Stop Killing Games's Ross Scott previously spoke about the initiative, he mentioned that the upcoming Digital Fairness Act was considering addressing game licensing but ultimately passing up the opportunity. Now, the European Commission has officially published a draft of the Digital Fairness Act, allowing us to see what the proposed legislation is targets and how that might affect video games and their makers. The draft act aims to "tackle unfair commercial practices affecting consumers online, while simplifying and harmonizing parts of the existing legal framework," and it will affect everything from dark patterns and subscription services to loot boxes, in-game purchases, digital contracts, and even influencer marketing.

What will be most pertinent to gamers and gaming companies is that gaming companies will be forced to be more transparent about in-game spending. This means they will need to show the real-world monetary value of virtual items purchased with in-game currency; loot boxes and other paid variable rewards systems will need to have rewards clearly shown as well as the probability of obtaining those rewards, and these systems will be unavailable for minors; and pay-to-win mechanics will be prohibited for minors under certain conditions. Platform designers, like social media companies, will also need to abide by new rules relating to what the Commission is calling "addictive design and attention-maximizing features."

The original Clairvoyance was a groundbreaking tribrid IEM that helped shape the industry and establish Thieaudio's presence in the global audiophile market. Beyond pioneering the tribrid configuration, it introduced a distinctive tuning signature that would go on to influence the industry. After six years of innovation and refinement, Thieaudio proudly presents its modern successor: the Clairvoyance MKII.

An All-New 2DD + 5BA + 2EST Tribrid Configuration
The Clairvoyance MKII features a completely modernized driver platform, combining the latest-generation IMPACT² dual 8 mm isobaric subwoofer system with five balanced armature drivers and dual Sonion electrostatic tweeters. The IMPACT² system delivers tight, impactful sub-bass, while four Knowles BA drivers maximize midrange resolution and preserve the natural timbre of vocals and instruments. A dedicated latest-generation Knowles mid-treble driver enhances vocal presence and imaging, while the dual Sonion EST ultra-tweeters extend the highest frequencies for an airy, lifelike presentation.

SEGA's networked arcade games in Japan have now been offline for more than ten days after unauthorized access to servers behind ALL.Net, SEGA's arcade network, and Aime, its IC card service for player data. SEGA Fave, which runs both, shut the services down at 3 p.m. on November 1 and disclosed the intrusion the next day. Games like Initial D The Arcade, maimai, and CHUNITHM lost online play and card support, and since other manufacturers use ALL.Net as well, Bandai Namco cabinets such as Taiko no Tatsujin and Mario Kart Arcade GP DX went down with them. Some of those Bandai Namco machines reportedly couldn't even run offline.

Bandai Namco's titles began a trial comeback on November 8 with online play and card use restored. SEGA's own lineup is still waiting. As of 9 a.m. on November 10, the official ALL.Net service account said there was no estimate for when service would resume, with another update planned for around 9 a.m. on November 12, Japan time. SEGA Fave says it hasn't found any leaked customer or business partner data so far, and payment information like credit card details wasn't stored on the affected servers. The company hasn't said how the attackers got in.

Grand Theft Auto VI is reportedly incorporating AMD's FSR Upscaling 4.1.2 technology on Xbox consoles, according to preload game files. Both amd_fidelityfx_loader_xbox.dll and amd_fidelityfx_upscaler_xbox.dll are present in the Xbox version of GTA VI, indicating that the game will eventually support FSR 4.1.2 technology, although it hasn't been advertised at launch. Interestingly, only Sony's PlayStation Spectral Super Resolution (PSSR) is supported at launch, and it requires a PlayStation 5 Pro console to enable, which is not available on the regular PlayStation 5.

As a reminder, last week, Xbox Series X and Xbox Series S consoles finally received machine learning-based upscaling technology with the latest FSR Upscaling 4.1.2. Microsoft's Game Development Kit (GDK) now officially includes FSR Upscaling 4.1.2 in two models. One is the full-scale upscaling model based on the INT8 data type, and the other is a new and efficient Lite INT8 model. Both models are compatible with both consoles, but the key point is that game developers have the flexibility to choose which model to implement. Depending on the demands of a game, the appropriate model can be applied to achieve optimal upscaling performance.

HAVN is proud to announce the launch of the HS 420 Mercedes-AMG PETRONAS F1 Team Edition, an officially licensed collaboration that brings together HAVN's performance-centric Hybrid Structure design with the team's iconic Black and Teal livery, Silver Star badge and distinctive design language.

From the hexagonal pattern to the individually numbered Special Edition Plaque, every detail has been carefully considered to celebrate the team's pursuit of speed, precision, and engineering excellence.

The Legend of Zelda: Ocarina of Time remake sold over 5.2 million units worldwide in its first four days on Switch 2, Nintendo announced, with 810,000 of those coming from Japan. According to Nintendo, that's the highest four-day global figure for any Nintendo-published Switch 2 game so far, and it counts both physical and digital copies. The remake, a full rebuild of the 1998 Nintendo 64 classic, launched on November 5. For a sense of scale, the Switch 2 console itself sold over 3.5 million units in its first four days back in June 2025.

Nintendo's latest earnings put the Switch 2 at 26.35 million units shipped as of September 30, so the remake has sold the equivalent of roughly a fifth of that base. It's a rough comparison, of course, since more consoles have shipped since then and shipped units don't equal owners. In just four days, Ocarina of Time has also passed the 5.05 million lifetime units Donkey Kong Bananza had reached by the end of September. The timing works out well for Nintendo too, as Switch 2 hardware shipments fell 41% last quarter while software sales kept climbing. For comparison, The Legend of Zelda: Tears of the Kingdom sold over 10 million units in its first three days back in 2023, albeit on the much larger installed base of the original Switch.

Developer Chokri Hammedi (blue0x1) has released EmuFive, an experimental emulator that runs PS5 games on a jailbroken PS4. The best result so far is God of War: Sons of Sparta, which apparently runs at around 40 FPS and above on a PS4 Pro. The PS5 version of Hollow Knight: Silksong boots too, though that game already has a native PS4 release. Both are 2D games, of course, and the developer notes that plenty of other games are still sluggish. EmuFive doesn't run on top of the PS4's own operating system. It's a small Linux system, installed through PS4 Multiboot and started from a GoldHEN payload.

At its core is KytyPS5, the same PS5 emulator that recently got Marvel's Wolverine to run on PC, and showed good progress in many other PS exclusive titles. Rather than emulating the PS5's processor, KytyPS5 hands the game's x86-64 code straight to the host CPU. That's an issue on the PS4, as its Jaguar cores lack the AVX2, FMA, and BMI2 extensions that Zen 2-targeted PS5 games use all over the place. Hammedi's patch gets around it by translating those instructions into older SSE4.2, AVX, and BMI1 equivalents, mostly at load time. That works, but it piles extra overhead onto a CPU that is already far slower.

Zalman has silently listed on its website two new 360 mm AIO coolers and a series of 120 mm fans. The AIOs are the ALPHA3 LX A36 and ALPHA3 SE A36, with similar specs but different pump-cover designs and features. The fan lineup includes the ZM-HF120 and ZM-HF120R. Those are available in black and white, with standard and reverse blade configurations. Let's look first at the AIO coolers. Both use a pump with a pure copper base. It operates at 800-3,400 RPM with a water flow of 1.3 L/min. The aluminium radiator measures 394 x 120 x 29 mm and comes with three 120 mm ARGB fans integrated into a single frame. Each fan uses a nine-blade design and rubber anti-vibration dampers. Fan speeds range from 600 to 2,000 RPM, with 32.9 dBA, 71.14 CFM airflow, and 2.06 mmH₂0 static pressure.

The ALPHA3 LX A36 stands out with two magnetic pump covers. One is an LCD that shows in real time CPU temperature, utilization, and clock speed. When the PC is powered off, the display switches to an always-on clock powered by motherboard standby power. The time is synchronized when the system starts. The second is a plain plastic cover for users who prefer a simpler appearance. The ALPHA3 SE A36 model instead features a layered 3D-pattern pump cover with ARGB lighting. The AIO coolers are compatible with Intel LGA 1851, 1700, 1200, and 115x sockets, as well as AMD AM5 and AM4 platforms.

Logitech employees are getting a day off just to play video games. The company has announced "Logi Game Day," which it calls the first company-wide gaming day in its history, pitched as a way to recognize the cognitive and professional benefits of gaming. Logitech G will host a streaming event at its Americas headquarters in San Jose on November 19, which happens to be the day GTA 6 launches on PS5 and Xbox Series X|S. Logitech G general manager Robin Piispanen will host it.

Employees aren't locked into GTA 6, though Piispanen does name-drop it in the press release alongside Call of Duty: Modern Warfare 4 and Forza. CEO Hanneke Faber framed the whole thing as a push against an old stigma, saying that it's time to retire the stereotype that gaming is a waste of time. Logitech also points to its own Ctrl Tomorrow study (coming from a gaming peripheral maker, of course), where 81% of gamers said gaming builds real-world confidence, reduces anxiety, and helps their day-to-day work performance.

Intel's Xeon 7 "Diamond Rapids" seems to have made its first public appearance as actual silicon. Intel had two packages on display at the Tencent Global Digital Ecosystem Conference in Shenzhen on October 29, one showing the heat spreader and the other flipped to show the contact side, though photos of them are only now making the rounds. These chips are meant for the massive LGA 9324 socket on Intel's Oak Stream platform, and the bottom side shows a huge pad area to match.

The specs themselves aren't new, since Intel detailed the package design at HOT CHIPS back in August. The flagship packs up to 256 Panther Cove P-cores without SMT, spread across 16 compute chiplets built on Intel 18A-P, along with 1.28 GB of last-level cache. It also gets 16 memory channels with MRDIMM support up to 12,800 MT/s and 128 PCIe Gen 6 lanes. At the event, Intel pitched Diamond Rapids as a CPU designed for agentic AI infrastructure and reportedly announced joint development with Tencent Cloud for its CVM instances. The new Tencent Cloud instances launched at the same event, however, run on Xeon 6+ "Clearwater Forest" chips.

GitHub has disabled 144 repositories hosting Nintendo Switch modchip firmware following a DMCA notice from Nintendo of America. The notice, dated November 6 and posted to GitHub's public DMCA archive this week, names a single repository called Ansem-SoD/Picofly. Because its fork network had more than 100 repositories and Nintendo claimed that all or most of the forks infringed to the same extent, GitHub took the entire network down. The targeted firmware covers SX Core, SX Lite, Picofly, and HWFLY chips, among others.

Patched consoles like the Switch Lite and Switch OLED need a soldered modchip that glitches the boot process to run custom software. Nintendo says this firmware lets modchips "illegally circumvent Nintendo's TPMs" by bypassing the boot sequence, which makes distributing it trafficking in circumvention technology under the DMCA. The notice also cites the U.S. criminal case against Team-Xecuter, whose member Gary Bowser pleaded guilty in 2021. A guilty plea isn't a ruling that modchips are illegal, of course, but Nintendo doesn't have to prove that for GitHub to act on a DMCA notice. Chips that are already installed keep working, while owners lose an easy source for firmware updates and flashing new chips.

MSI's GeForce RTX 5080 cards could cost up to 8,000 yuan (about $1,200) more in China this month, according to a ChannelGate post shared on X by harukaze5719. The post outlines MSI's November distributor pricing compared to October, with RTX 5080 models going up by 1,300 to 8,000 yuan depending on the card. That's roughly $195 at the low end, while the top figure alone is more than the RTX 5080's entire $999 launch MSRP. Yes, that's apparently a $1,200 increase planned for the RTX 5080 at the top end.

The rest of the RTX 50 stack goes up too. RTX 5070 Ti models rise by 800 to 4,000 yuan (about $120 to $600), and the RTX 5070 by 500 to 1,000 yuan (75 to $150). The RTX 5060 Ti 16 GB picks up 400 yuan, its 8 GB sibling 200 yuan, and the RTX 5060 between 200 and 500 yuan. Even the RTX 3050 gets a 100 yuan bump. These are distributor prices in China, not new MSRPs, and the post doesn't say which RTX 5080 models sit at the top of that range. Colorful and Galax reportedly started sharing similar hikes with their partners earlier this month, so MSI seems to be following the same path. We can certainly expect other AIB partners to follow suit soon.

In the six years since the PlayStation 5 launched, Sony says that the gaming console has officially sold more than 100 million units, as of a November 2026 announcement by Sony. As per the console maker, it has sold as many as 4.2 million units between April and September 2026, which is actually a significant decline compared to the previous year's 6.5 million units in the same period in 2025—despite the expected increase in sales from the imminent launch of GTA VI.

So far, this puts the PS5 as Sony's fourth most successful console platform to date, behind the PlayStation 1 (102.4 million sales), PlayStation 4 (117 million), and the PlayStation 2 (over 160 million units), but sales are likely to continue until and even after the expected 2028 launch of the PS6. At the time of writing, only the PSP and the PlayStation 3 have sold less than the PlayStation 5. Sony also provided some insights into software sales for its PlayStation platforms, with the PS4 and PS5 combined having sold a claimed 1.795 billion units (including physical and download-only titles, PS VR titles as well as software bundled with PS4, PS5).

Razer, the leading global lifestyle brand for gamers, today announced the Razer | Honkai: Star Rail Collection, developed in collaboration with HoYoverse. Following previous collaborations with Genshin Impact and Zenless Zone Zero, the collection brings the world of Honkai: Star Rail to life through officially licensed gaming peripherals.

Inspired by Silver Wolf LV.999, the invincible player from Honkai: Star Rail, the collection reimagines Razer's award-winning gaming hardware with her unmistakable aesthetic and competitive spirit, delivering premium gaming gear built for players who are always one move ahead.

More than a year of rising memory prices has placed mounting cost pressure on the consumer electronics industry, according to TrendForce's latest memory industry research. Notebook shipments, in particular, are expected to decline sharply quarter-over-quarter beginning in 3Q26, leaving PC OEMs grappling with sluggish sales and swelling finished-goods inventories.

Against this backdrop, some manufacturers are planning to adjust memory procurement for 4Q26—and even 1Q27—while considering a greater reliance on private-label models in their shipment mix.

XGIMI, a global leader in premium smart projection and home entertainment, today announced that the HORIZON 30 Pro and HORIZON 30 Ultra Max are officially available for purchase at XGIMI.com. Following its October debut, the new generation of high-performance home projectors combines premium picture quality with flexible setup, making big-screen entertainment a natural fit for everyday living spaces.

The three models - HORIZON 30 Pro, HORIZON 30 Max and HORIZON 30 Ultra Max - combine XGIMI's RGB Triple Laser technology with up to 6,000 ISO lumens of brightness, native contrast reaching 10,000:1, and 4K 120 Hz performance. These combined capabilities deliver brighter pictures, greater depth in darker scenes, and smoother motion for movies, sports, streaming and gaming.

IBM (NYSE: IBM) today announced IBM Quantum Nighthawk r3, its next-generation processor, will be built to accurately execute 10,000 quantum operations across 520 qubits to tackle larger, more complex problems while maintaining the speed and quality demonstrated by the current generation Nighthawk available today. In 2027, IBM plans to link two Nighthawk r3 processors into a single system with more than 1,000 qubits.

To be at the frontier of quantum computing, industry-relevant applications demand both computational quality and speed. Speed matters because a quantum computer delivers its answer in samples: each run of a quantum circuit returns one result. Applications increasingly require circuits executing thousands of quantum operations on approximately 100 qubits or more, generating millions to billions of samples. Scale and quality enable increasingly complex computations, but speed determines whether these demanding workloads can be completed in practical timeframes and at a reasonable cost.

BIOSTAR, a leading manufacturer of edge AI solutions, industrial computers, motherboards, graphics cards, and storage devices, today announces the BIPTL-IHT, a Thin Mini-ITX industrial motherboard engineered for intelligent edge systems, AIoT deployments, automation platforms, and advanced system integration applications. Powered by Intel Core Ultra series 3 processors based on the Panther Lake-H platform with integrated AI capabilities, the BIPTL-IHT enables businesses to turn data into intelligent action through efficient edge AI computing, extensive connectivity, and flexible expansion within a compact 170 x 170 mm form factor.

Designed for system integrators building modern business infrastructure, the BIPTL-IHT delivers the performance and adaptability required for applications that rely on stable data processing, connected devices, and responsive system control. It is well-suited for deployments such as customer data management systems, industrial automation platforms, AIoT gateways, and smart edge devices, helping streamline system development while maintaining consistent performance across demanding workloads.

Chieftec is excited to introduce the CHIEFTEC ATLAS, a professional E-ATX workstation chassis designed for high-performance systems that demand extensive expansion capabilities and efficient cooling. With its modern design and spacious interior, the ATLAS provides a versatile foundation for powerful workstations and demanding hardware configurations. With its spacious interior and extensive expansion capabilities, the ATLAS is designed to accommodate powerful hardware configurations for professional applications and high-performance builds.

NVIDIA's rumored GeForce RTX 5080 24 GB model has reportedly been put on hold. The company is struggling to secure enough 3 GB GDDR7 memory modules and is instead opting for the less expensive 2 GB modules for the standard 16 GB version. According to a report from Board Channels, NVIDIA has already informed its add-in card (AIC) partners to officially cancel the January 2027 launch plan for the 24 GB model, citing economic challenges. As we are currently experiencing one of the worst supply chain shortages for memory modules, using the more expensive 3 GB GDDR7 modules would result in a significantly higher MSRP and, consequently, a higher price for customers. Therefore, the 2 GB GDDR7 memory modules will continue to be used, and the RTX 5080 16 GB model will remain available.

Readers may recall that NVIDIA did not plan any other changes to the model besides the memory configuration, as the CUDA core count was to remain the same. The new 24 GB model would potentially have slightly higher TDP and clock speeds, but the main improvements would come from increased memory capacity and bandwidth. Since it seems we won't be getting the revised model, and there's no rumor of a "SUPER" overhaul anytime soon, we will have to stick with the standard lineup a bit longer until the next-generation GeForce RTX 60-Series "Rubin" arrives.

ProArt, the ASUS sub-brand created by creators, for creators, today announced its participation at Adobe MAX 2026, where it will showcase a complete ecosystem of AI-powered creator solutions designed to accelerate content creation, streamline workflows, and empower the next generation of creative professionals. Event attendees can experience hands-on demonstrations of the ProArt GR1X, a new agentic AI mini PC, and ProArt P16 & P14 laptops, bringing powerful on-device AI capabilities directly to creators, as well as ProArt OLED monitors featuring exceptional color accuracy and HDR performance. The demos will span AI content creation, visual effects production, professional photo editing, and color-critical workflows.

RTX Spark Creator Solutions
ProArt GR1X mini PC: Create without limits. Your AI agent, always on
The ProArt GR1X is a mini PC that brings on-device AI to creative workflows, from the Adobe Creative Suite, to generative AI tools like ComfyUI. The RTX Spark carries the full NVIDIA Stack, CUDA, TensorRT, NVFP4, RTX Ray Tracing, DLSS, and Optix. With up to 128 GB of unified memory, users can run agents alongside their largest creative projects, and iterate without per-token or per-generation fees. Because the GR1X has a 6144-core NVIDIA Blackwell RTX GPU, users can keep client footage and project files on their own machine, and feel confident when they utilize AI on their most sensitive client projects. Apps, tools, and workflows are connected into one seamless experience, private by default. RTX Spark brings NVIDIA's full-stack AI platform, ready for the AI workloads of today and whatever comes next. Delivered on the familiar, trusted Windows platform.

Epomaker previously revealed the Galaxy75 in an official livestream video, but the peripheral maker has now officially launched the Galaxy75 alongside a new keyboard, the Bamboo75. As we covered previously, the Galaxy75 features a relatively standard 75% layout, an aluminium case, polycarbonate plate, a mix of PBT and PC/ABS keycaps, wireless connectivity with an 8,000 mAh battery, and Epomaker's Creamy Jade linear switches. The official launch price, which was previously unknown, is $109 on Epomaker's site, and the Galaxy75 is available in three colorways: black, purple, and pink.

The Epomaker Bamboo75, on the other hand, is a slightly more interesting affair. It takes a similar, but notably not identical, layout to the Galaxy75 but swaps out the cutesy pastel and minimalist aesthetics of the Galaxy75 for a more nature-inspired design, thanks to a case made of bamboo and keycaps made of a mix of ABS and PBT with wood grain print on the top. Those Cherry profile keycaps also feature a side-printed shine-through legend, which allows the south-facing per-key RGB backlighting to shine through while keeping the consistent wood aesthetic when looking at the keyboard from the top. Like the Galaxy75, the Bamboo75 also packs wireless connectivity, with both Bluetooth and 2.4 GHz connectivity, and it also has an 8,000 mAh battery, although Epomaker doesn't provide any claims of battery life.

Epomaker has officially unveiled the Alice40 as its latest foray into the world of 40% mechanical keyboards—after the TH40 we reviewed previously and the Luma40 after that—and this time, the brand has gone for yet another unique layout option. As the name suggests, the Alice40 is based on the Alice layout that is commonly recommended to first-time ergonomic keyboard buyers, for its similarity to a regular unibody keyboard, although the Alice40 has a few features that set it apart from other mechanical keyboards in the same $69.99 price bracket.

For starters, the Alice40 modifies the traditional Alice layout with columnar stagger, meaning the columns are vertically offset from one another instead of the rows being horizontally offset. This is meant to make the keys align more appropriately to the natural movement of your fingers while you're typing. Another interesting quirk of the Alice40 is the knob between the two halves of the alpha cluster. As expected from a $69.99 mechanical keyboard, it has a plastic case, but it also has flip-out feet that allow users to adjust the typing angle between 5° and 8.5°. The case also features a gasket mount design, much like the Epomaker TH40, and it is stuffed with sound-damping materials to cut down on echoes and reverberation inside the case.

AMD's Madhu Rangarajan, corporate VP of Compute and Enterprise AI, says the company never set out to build its Zen 6 EPYC 9006 "Venice" chips around agentic AI, a not-so-subtle dig at NVIDIA and Arm, which both pitch their latest server CPUs as made for that exact job. Rangarajan still called EPYC 9006 the best CPU for agentic workloads during a media session covered by Tom's Hardware. Chief Zen architect Mike Clark said much the same in a separate talk, admitting AMD didn't see agentic computing coming, though he added that the company always expected the CPU to keep a role in AI software. Here's what Rangarajan said:

We think that the EPYC 9006 CPU is the best CPU for the agentic era, and spoiler alert, we didn't magically know four years ago that agentic was going to be a thing and purpose-built something for agentic. The reality is an agentic workflow is a diverse set of tasks… and in order to execute those tasks well, you need compute that is able to handle a diverse set of workloads with a full portfolio.

GTA 6 is now the most pre-ordered game ever on both the PlayStation Store and the Microsoft Store, according to Sony and Microsoft. Sony announced the record in a PlayStation Blog post that opened PS5 pre-loading nine days before the November 19 launch, pointing to "unprecedented demand" for the early start. Xbox then called it "the most pre-ordered title of all time on the Microsoft Store" while confirming that GTA 6 will stream to Xbox One through Cloud Gaming. Neither company shared actual numbers, however.

Sensor Tower's tracker fills some of that gap. As of November 10, the firm estimates 9.69 million pre-orders worth about $942 million net of returns, up from 7.71 million a week earlier and 6.45 million in late October. PlayStation holds roughly 75% of that with 7.23 million copies and $703 million, while Xbox sits at 2.46 million copies and $239 million. That averages out to around $97 per copy, which fits with most buyers going for the $100 Ultimate Edition. If accurate, GTA 6 has passed the 8 million pre-orders CD Projekt reported for Cyberpunk 2077 in 2020. The comparison isn't quite like-for-like, though, since Cyberpunk also launched on PC and likely counted physical copies, while Sensor Tower seems to track digital console orders only.

The Philips Evnia 27M2D5901A is a 27-inch (68.58 cm) Dual Mode gaming monitor that lets users switch between detailed 5K visuals and high-speed QHD performance. The Philips Evnia 27M2D5901A features a Fast IPS panel with Dual Mode technology. Users can choose a 5K resolution for detailed visuals or switch to QHD for refresh rates of up to 350 Hz. For a more immersive gaming experience, the monitor combines built-in RGB backlighting with AI-enhanced Ambiglow, and AmbiScape to extend lighting effects to Matter-compatible smart lights.

Different games call for different screen settings. Dual Mode lets users switch between 5K for greater detail and QHD for a higher refresh rate. In 5K mode, the monitor delivers 5120 x 2880 pixels at 165 Hz and can be overclocked to 180 Hz. On a 27-inch panel, the high pixel density keeps textures, text, and interface elements clear, making this mode suitable for visually detailed games, content creation, and everyday work. When speed matters more, the QHD mode runs at 330 Hz and can be overclocked to 350 Hz at 2560 x 1440 pixels. The higher refresh rate provides smoother motion during fast-paced and competitive games.