TechPowerUP NEWS

Socionext Inc., a global leader in System-on-Chip (SoC) design and advanced semiconductor solutions, introduces "Flexlets", a new class of configurable chiplets designed to advance heterogenous integration. As traditional monolithic SoC designs face physical and economic limits—reticle size constraints, yield challenges, and thermal bottlenecks- the industry is turning toward chiplet-based design, where designers can integrate their core features and interface functionalities, provided as chiplets, into a packaged device to improve performance, cost, and accelerate time-to-market.

While chiplet technology opens exciting possibilities for modular designs and scalability, many current solutions are derived from fixed-function ASSPs, limiting flexibility and customization. Socionext's Flexlets overcome this by offering a configurable library of chiplet designs at the RTL level. Unlike traditional approaches, Flexlets empower customers to tailor performance to their unique application needs - whether in high-performance computing, advanced networking, or next-generation automotive systems.[Only registered and activated users can see links. Click Here To Register...]
[Only registered and activated users can see links. Click Here To Register...]

[Only registered and activated users can see links. Click Here To Register...]

Atari and PLAION REPLAI are proud to announce that the Atari 2600+ PAC-MAN Edition, a brand-new version of the legendary console inspired by one of gaming's most recognizable icons, is out now. Launching on Friday 31st October 2025, this special edition brings together two giants of video game history just in time for Halloween, the perfect night to celebrate PAC-MAN's never-ending pursuit of ghosts.

The return of a classic
The Atari 2600+ PAC-MAN Edition is a lovingly reimagined take on the console that defined home gaming for a generation. Now finished in signature PAC-MAN yellow and featuring illuminated character icons across the front panel, this collector's edition combines authentic retro design with modern features including HDMI output, USB-C power, and multiple display modes (4:3/16:9).[Only registered and activated users can see links. Click Here To Register...] [Only registered and activated users can see links. Click Here To Register...] [Only registered and activated users can see links. Click Here To Register...]
[Only registered and activated users can see links. Click Here To Register...]

[Only registered and activated users can see links. Click Here To Register...]

Gaming is no longer just a contest of reflexes and performance. It has evolved into a fusion of technology, vision, and design. Embodying this philosophy, ATTACK SHARK, a gaming peripheral brand specializing in affordable, high-performance mechanical keyboards, gaming mice, and accessories, has recently introduced the R11 ULTRA, its latest flagship carbon fiber gaming mouse. This launch marks a major leap forward in combining advanced material science with top-tier engineering, setting a new benchmark for precision, durability, and responsiveness in professional gaming gear.

The R11 ULTRA's most striking innovation lies in its material: forged dry carbon fiber. Inspired by the full-carbon body of Formula 1 race cars, this material offers an exceptional balance of lightness and strength. With a tensile strength exceeding 3,500 MPa--several times that of steel--dry carbon fiber allows the R11 ULTRA to achieve extreme lightweight construction without compromising structural rigidity. Crafted through a complex multi-stage process that demands artisan-level precision, each shell of the mouse embodies engineering mastery and material artistry.[Only registered and activated users can see links. Click Here To Register...] [Only registered and activated users can see links. Click Here To Register...] [Only registered and activated users can see links. Click Here To Register...] [Only registered and activated users can see links. Click Here To Register...]
[Only registered and activated users can see links. Click Here To Register...]

[Only registered and activated users can see links. Click Here To Register...]

Yesterday, we reported that the latest AMD Software Adrenalin 25.10.2 WHQL drivers [Only registered and activated users can see links. Click Here To Register...] on reference Radeon RX 7900 XTX and RX 7900 XT. However, it appears that this was a mistake, and AMD has contacted TechPowerUp to clarify the situation. To make sure that the functionality is back, we tested USB-C power delivery on our own reference card. Readers might remember that the Made by AMD (MBA) reference Radeon RX 7900 XTX and RX 7900 XT graphics cards include USB Type-C ports. These ports support DisplayPort 2.1 passthrough and provide USB power delivery up to the PD 3.0 standard of 30 W.

While the port does not support USB data transfer, it is wired for DisplayPort and power delivery. This theoretically allows you to connect certain monitors that use a single USB-C connection for both display input and power. We tested the feature to make sure it is supplying the port with sufficient power. Using a USB-C power meter, we recorded 27 W of power output from the AMD Radeon RX 7900 XTX. Impressively, the card can supply full power to the port even without a driver installed, as demonstrated in our test. Installing the latest driver doesn't alter this behavior, indicating that the original driver changelog was just incorrect. To confirm that the 27 W charging is consistent, we charged a laptop with a completely empty battery to full capacity. While charging, we loaded Furmark test, and the load was maintained, proving that the feature works as intended.[Only registered and activated users can see links. Click Here To Register...] [Only registered and activated users can see links. Click Here To Register...]
AMD's statement follows after the break.[Only registered and activated users can see links. Click Here To Register...]

[Only registered and activated users can see links. Click Here To Register...]

SK hynix showcased its full-stack AI memory portfolio for enhancing AI and data center infrastructure at the 2025 OCP Global Summit in San Jose, California from October 13-16. The OCP Global Summit is a conference hosted by the Open Compute Project (OCP), the world's largest open data center technology collaboration community. Held under the theme "Leading the Future of AI", this year's event welcomed experts from leading global companies and developers to share the latest trends in data center and AI infrastructure and discuss developing industry solutions.

Under the slogan "MEMORY, Powering AI and Tomorrow," SK hynix presented a range of innovative technologies which improve the performance and efficiency of AI infrastructure. The company's interactive booth was organized into four sections - HBM, AiM, DRAM, and eSSDs - and featured product-based character designs, 3D models of key technologies, and live demonstrations. The centerpiece of SK hynix's booth was its groundbreaking 12-layer HBM4. In September 2025, the company became the first in the world to complete development of HBM4 and establish the product's mass production system. Featuring 2,048 input/output (I/O) channels, twice that of the previous generation, HBM4 offers increased bandwidth and over 40% greater power efficiency. These specifications make it optimized for ultra-high-performance AI computing systems.[Only registered and activated users can see links. Click Here To Register...] [Only registered and activated users can see links. Click Here To Register...]
[Only registered and activated users can see links. Click Here To Register...]

[Only registered and activated users can see links. Click Here To Register...]

In the quest for peak productivity and immersive creation, we often forget that our body is the primary interface. Traditional keyboards confine our hands, leading to strain and stifled creativity. Enter the era of ergonomic split keyboards. The EPOMAKER Split70, alongside the capable Split65, is designed to break these physical constraints. Combining tri-mode connectivity, hot-swappable customization, and a liberating split design, they promise not just to type, but to transform your interaction with your digital world, offering unparalleled comfort for office warriors, gamers, and DIY enthusiasts alike.

Ergonomics That Set Your Creativity Free
The core of the Split70 is its philosophy: "When a keyboard learns to stretch, creation finds a freer posture." It's not just a tool; it's an extension of your body.

• Natural Posture: The 70% layout is split into two halves, allowing you to position them at a natural shoulder-width distance. This lets your arms and wrists relax, turning every typing session from a cramped effort into a natural, flowing embrace.
• Form Follows Function: You can use the halves together or independently. The left section can even connect solo for lightweight, mobile setups, offering incredible flexibility that adapts to your moment's need.
• Beyond the Basics: While the Split65 offers a solid ergonomic foundation in a 65% form factor, the Split70 introduces a new white color option and independent side light bars, catering to both minimalist and vibrant desk aesthetics.

[Only registered and activated users can see links. Click Here To Register...] [Only registered and activated users can see links. Click Here To Register...] [Only registered and activated users can see links. Click Here To Register...]
[Only registered and activated users can see links. Click Here To Register...]

[Only registered and activated users can see links. Click Here To Register...]

ASUS Republic of Gamers (ROG) today announced the ROG GR70, the first ROG gaming mini PC to feature the AMD Ryzen 9 CPU for desktop-class performance in a compact footprint.

The GR70 Mini PC boasts the NVIDIA GeForce RTX 5070 or 5060 Laptop GPU for exceptional visuals, and an advanced triple-fan cooling system that keeps temperatures and acoustics low. Plus, built-in Wi-Fi 7 and Bluetooth 5.4 allow gamers to engage in fast-paced gameplay without bottlenecks.[Only registered and activated users can see links. Click Here To Register...] [Only registered and activated users can see links. Click Here To Register...] [Only registered and activated users can see links. Click Here To Register...] [Only registered and activated users can see links. Click Here To Register...] [Only registered and activated users can see links. Click Here To Register...]
[Only registered and activated users can see links. Click Here To Register...]

[Only registered and activated users can see links. Click Here To Register...]

NVIDIA today announced that it is working with SK Group to build an AI factory to advance semiconductor research, development and production, as well as cloud infrastructure to support digital twin and AI agent development. SK Group is building an AI factory featuring more than 50,000 NVIDIA GPUs, with the first phase planned for completion by late 2027. Once complete, the system is expected to be one of Korea's largest AI factories.

The new factory will serve SK subsidiaries - including SK hynix and SK Telecom (SKT) - as well as external organizations through a GPU-as-a-service model, accelerating digital transformation and industrial innovation for Korea's industries. Further expanding the NVIDIA and SK Group partnership, the companies are collaborating on the development of SK hynix high-bandwidth memory (HBM) and next-generation advanced memory solutions for NVIDIA GPUs, semiconductor manufacturing and telecommunications infrastructure.[Only registered and activated users can see links. Click Here To Register...]
[Only registered and activated users can see links. Click Here To Register...]

[Only registered and activated users can see links. Click Here To Register...]

NVIDIA today announced plans with Samsung Electronics to build a new AI factory, representing a new era where intelligent computing and chip manufacturing converge. The state-of-the-art AI factory will combine Samsung's semiconductor technologies with NVIDIA platforms to establish the foundation of next-generation, AI-driven production. Powered by more than 50,000 NVIDIA GPUs, Samsung's semiconductor AI factory will be a centerpiece of the company's digital transformation, integrating accelerated computing directly into full-fledged advanced chip manufacturing.

Through this collaboration, Samsung and NVIDIA are setting a global benchmark for AI-driven semiconductor manufacturing at scale, integrating data from physical equipment and production workflows to achieve predictive maintenance, process improvements and increased operational efficiency in autonomous fab environments.[Only registered and activated users can see links. Click Here To Register...]
[Only registered and activated users can see links. Click Here To Register...]

[Only registered and activated users can see links. Click Here To Register...]

Razer, the world's leading lifestyle brand for gamers, has teamed up with Valve to launch the Razer | Counter-Strike 2 Collection, bringing the legendary Dragon Lore skin from in-game prestige to real-world performance across Razer's top-ranked esports gear. A symbol of power, and rarity, Dragon Lore is one of the most recognizable skins in Counter-Strike history. Now, it steps beyond the game, featured on the same gear trusted by top esports pros. The collection fuses Counter-Strike's competitive legacy with Razer's tournament-grade hardware to deliver signature style and pro-level performance.

"Dragon Lore isn't just a skin - it's a symbol of Counter-Strike's legacy," said Addie Tan, Global Head of Lifestyle Division at Razer. "Counter-Strike has defined competitive FPS for over two decades. This collection now brings the iconic design of Dragon Lore into a physical representation of the game as a tribute to the game's history and community, and to the legends still to come. Designed on our award-winning esports gear, it's performance gear built for legends in the making."[Only registered and activated users can see links. Click Here To Register...]
[Only registered and activated users can see links. Click Here To Register...]

[Only registered and activated users can see links. Click Here To Register...]

The WireView Pro II, which was already presented at Computex 2025, is now available for pre-order exclusively in the Thermal Grizzly shop and offers new features for monitoring and data logging as well as new protection functions for graphics cards with a 12VHPWR power connector. The WireView Pro II is a device for measuring the power consumption of graphics cards and introduces precise per-pin monitoring of the 12V-2x6 connector for the first time, providing valuable insights into load distribution on modern high-end graphics cards along with advanced safety features.

A CNC-machined aluminium housing with semi-passive cooling and a high-contrast TFT-IPS display ensures reliable visualization of all critical measurements. Automatic data logging and an extended 2-year warranty for the GPU connector and graphics card provide additional peace of mind in demanding setups.[Only registered and activated users can see links. Click Here To Register...] [Only registered and activated users can see links. Click Here To Register...] [Only registered and activated users can see links. Click Here To Register...] [Only registered and activated users can see links. Click Here To Register...]
[Only registered and activated users can see links. Click Here To Register...]

[Only registered and activated users can see links. Click Here To Register...]

Seemingly as part of its ongoing lawsuit against Palworld's developer, Pocketpair, Nintendo applied for several patents to protect what it saw as intellectual property being ripped from Pokémon games. One of those early patent filings at the Japanese Patent Office (JPO) described effectively the creature capture mechanics you see in most Pokémon and similar games these days. The patent describes the game mechanic as throwing an object at a character, after which the computer determines whether the character in question is successfully captured. It also mentions throwing an object at a character in a game in order to limit the character's movement in the field. According to [Only registered and activated users can see links. Click Here To Register...], this patent application has now been denied, meaning Nintendo may have to find another linchpin in its ongoing lawsuit against Palworld's developer.

The denial of the patent comes after the patent office examined prior art that was submitted by an unnamed third party. As Games Fray goes on to explain, however, the decision to deny Nintendo's patent doesn't necessarily spell victory for Palworld and Pocketpair. Firstly, the patent denial is "non-final," which means Nintendo can contest the ruling and defend its position. Secondly, and this appears to be more important, while Japanese judges regularly do consult with the JPO and respect the decision of the Patent Office, the patent is non-binding, meaning the final decision is still up to the judge's discretion. For Nintendo's part, the patent in question forms a core part of a family of patents, so if it is eventually "finally" denied, Nintendo may lose additional resources to fight against its perceived threats, since that patent would weaken its claim to other patents in the patent family.[Only registered and activated users can see links. Click Here To Register...]

[Only registered and activated users can see links. Click Here To Register...]

AMD has confirmed a hardware defect in the RDSEED instruction on its new Zen 5 processors, cataloged as [Only registered and activated users can see links. Click Here To Register...] and assigned CVE-2025-62626. The company reports that the 16-bit and 32-bit forms of RDSEED can return zero values, which may be mistakenly interpreted as a successful entropy read. The 64-bit form is unaffected. AMD rates the issue as high severity and suggests short-term software workarounds: prioritize using the 64-bit RDSEED path where possible, mask RDSEED from software discovery, or treat a zero return as a failure and retry. Firmware and microcode updates are being rolled out to OEMs. AMD's advisory provides product-family target dates, such as AGESA TurinPI 1.0.0.8 for EPYC 9005, with a target date of November 14, 2025, and consumer families targeted for late November 2025. Users should install vendor BIOS/microcode updates when available and consider regenerating any sensitive keys created on unpatched systems.

RDSEED is a low-level instruction that supplies entropy used for cryptographic keys and other security functions. When it returns predictable values without detection, newly generated keys or nonces may be weaker than intended. According to AMD, the problem affects the 16-bit and 32-bit forms of RDSEED, while the 64-bit form does not appear to be impacted. The company is distributing firmware and platform updates to correct the issue, but until those updates are applied, the recommended interim measures are straightforward: prefer the 64-bit RDSEED path where possible, prevent software from discovering or using the affected RDSEED forms, or treat a zero result as a failure and retry. Once updates become available, users should install them promptly and consider regenerating any sensitive keys created on unpatched systems.[Only registered and activated users can see links. Click Here To Register...]

[Only registered and activated users can see links. Click Here To Register...]

Amazon Web Services (AWS) and OpenAI have signed a $38 billion deal spanning several years, seven to be more precise, to boost OpenAI's AI operations using AWS's high-end cloud setup. This agreement lets OpenAI use clusters with hundreds of thousands of NVIDIA GB200 and GB300 GPUs, the complete planned capacity is expected to be deployed for use by the end of 2026. These GPUs are set up across Amazon EC2 UltraServers linked by a fast, high-capacity network built to handle both inference and large-scale model training. AWS's previous experience managing clusters exceeding 500,000 chips has been a key factor in OpenAI's decision.

The new infrastructure is expected to serve a range of OpenAI applications, from ChatGPT inference workloads to training next-generation foundation models. OpenAI's move to AWS comes after a series of recent partnerships with NVIDIA, [Only registered and activated users can see links. Click Here To Register...], [Only registered and activated users can see links. Click Here To Register...], [Only registered and activated users can see links. Click Here To Register...], and Oracle making sure they can tap into a mix of high-end chips and cloud systems. OpenAI CEO Sam Altman stated that the company plans to invest $1.4 trillion to build 30 gigawatts of computing capacity, equivalent to the power consumption of roughly 25 million U.S. households. So far, OpenAI has partnered with [Only registered and activated users can see links. Click Here To Register...] for 10 GW, AMD for 6 GW, and [Only registered and activated users can see links. Click Here To Register...] for 4.5 GW of compute capacity, moving steadily toward its stated goal of building a 30 GW AI infrastructure network.[Only registered and activated users can see links. Click Here To Register...] [Only registered and activated users can see links. Click Here To Register...] [Only registered and activated users can see links. Click Here To Register...]

[Only registered and activated users can see links. Click Here To Register...]

Adeia Inc., the technology company known for developing foundational innovations in the semiconductor industry, announced today that it has filed patent infringement lawsuits against Advanced Micro Devices, Inc. (AMD) in the U.S. District Court for the Western District of Texas.

"Earlier today, Adeia took legal action against AMD to protect our intellectual property rights and the investments we have made in our foundational semiconductor technologies," said Paul E. Davis, chief executive officer of Adeia. "For years, AMD's products have incorporated and made extensive use of Adeia's patented semiconductor innovations, which have greatly contributed to their success as a market leader. After prolonged efforts to reach a mutually agreeable resolution without litigation, we believe this step was necessary to defend our intellectual property from AMD's continued unauthorized use."[Only registered and activated users can see links. Click Here To Register...]
[Only registered and activated users can see links. Click Here To Register...]

[Only registered and activated users can see links. Click Here To Register...]