Tech News
Back to home

Articles tagged with "chips"

G.Skill explains how AMD EXPO ULL unlocks additional performance — expanded profiles allow memory makers to include subtiming tweaks for the first time

G.Skill explains how AMD EXPO ULL unlocks additional performance — expanded profiles allow memory makers to include subtiming tweaks for the first time

G.Skill showcased new memory kits at Computex with AMD's EXPO Ultra Low Latency (ULL) support, aiming to reduce memory latencies for improved CPU performance. The EXPO ULL program allows memory makers to adjust sub-timings within primary timings for lower latencies, enhancing memory performance. While faster memory speeds above 6000 MT/s can add latency on AMD platforms, EXPO ULL kits aim to optimize latency without the need for manual adjustments. These kits require stricter binning during production, potentially making them more expensive but appealing to gamers seeking optimal CPU-bound gaming performance.

Tom's Hardware•
Unreleased RTX 3050 Ti engineering sample appears in photos and benchmarks — the RTX 3060 alternative that never happened

Unreleased RTX 3050 Ti engineering sample appears in photos and benchmarks — the RTX 3060 alternative that never happened

An unreleased engineering sample of the RTX 3050 Ti, a desktop version of Nvidia's graphics card, has been spotted in photos and benchmarks. The card, utilizing the GA106 silicon, falls between the RTX 3050 and RTX 3060 in terms of performance. With 26 SMs active, it features 3,328 CUDA cores and offers a balance between performance and power efficiency. Benchmark results show the RTX 3050 Ti potentially being up to 22% faster than the RTX 3050 but up to 13% slower than the RTX 3060. Despite not being released, the RTX 3050 Ti could have been priced around $289, filling a gap in Nvidia's mid-range lineup.

Tom's Hardware•
Intel will reportedly upgrade its Wildcat Lake refresh to an 8-core config next year, leak claims — top-end silicon tipped to feature 4 P-cores and 4 LP-E cores as part of 'Core 400' series

Intel will reportedly upgrade its Wildcat Lake refresh to an 8-core config next year, leak claims — top-end silicon tipped to feature 4 P-cores and 4 LP-E cores as part of 'Core 400' series

Intel is rumored to be upgrading its Wildcat Lake refresh next year to feature an 8-core configuration, with 4 P-cores and 4 LP-E cores as part of the 'Core 400' series. This upgrade would mark a significant improvement over the current 6-core setup. The Wildcat Lake refresh is expected to offer enhanced real-world performance with the addition of more cores. The move to an 8-core configuration aligns with Intel's strategy to differentiate its product lineup and target a more upscale audience. Additionally, the Wildcat Lake refresh is speculated to include 2 Xe3 cores for integrated graphics.

Tom's Hardware•
Nvidia is reportedly still planning fabled RTX 50 Super series for 2026, leak claims — lineup could now include a potential 'RTX 5060 Super' with 12GB of VRAM

Nvidia is reportedly still planning fabled RTX 50 Super series for 2026, leak claims — lineup could now include a potential 'RTX 5060 Super' with 12GB of VRAM

Nvidia is rumored to be working on a Super refresh for its RTX 50-series GPUs, including a potential RTX 5060 Super with 12GB of VRAM. Leaker MEGAsizeGPU claims that this new SKU is in development, alongside previously mentioned models like the RTX 5070 Super, RTX 5070 Ti Super, and RTX 5080 Super. These GPUs are expected to feature varying amounts of VRAM, with the RTX 5060 Super potentially utilizing 4x 3GB modules on a 128-bit bus. Despite previous rumors of cancellation, the leaker suggests that these GPUs are still slated for a 2026 release, although their value may be impacted by high DRAM prices and AI-related GPU cost increases.

Tom's Hardware•
CEO Interview with Daniel Schall of Black Semiconductor

CEO Interview with Daniel Schall of Black Semiconductor

Daniel Schall, CEO of Black Semiconductor, discusses the company's focus on addressing chip-to-chip communication bottlenecks with integrated graphene photonics technology. The company's Black Semiconductor IGP™ technology aims to enable AI scaling by revolutionizing chip communication. By utilizing graphene's unique properties, Black Semiconductor's technology allows for faster and more energy-efficient data transmission between chips. The construction of FabONE, a 300mm facility in Germany, is underway to integrate electronics and photonics using graphene, with plans for pilot production in 2027 and volume production by 2029. The acquisition of Applied Nanolayers in 2025 enhances Black Semiconductor's capabilities in industrial graphene production, ensuring a robust material supply chain for their innovative technology.

SemiWiki•
Microsoft quietly dropped Copilot+ PC branding for Windows 11’s powerful AI laptop, and it won’t tell you why

Microsoft quietly dropped Copilot+ PC branding for Windows 11’s powerful AI laptop, and it won’t tell you why

Microsoft has quietly abandoned the Copilot+ PC branding for its latest powerful AI laptop, the Surface Laptop Ultra, powered by NVIDIA's RTX Spark platform. Despite confirming privately that the laptop still includes an NPU, Microsoft did not mention the Copilot+ branding during the reveal. The absence of Copilot+ branding is speculated to be intentional, possibly due to NVIDIA's preference for associating local AI acceleration with its RTX branding rather than Microsoft's Copilot branding. The move reflects a shift away from the Copilot branding, which had faced controversies and negative perceptions in the past.

Windows Latest•
AMD's Frank Azor pushes back against claim that FSR 4.1 won't be ported to RDNA 3.5 GPUs — says 'no such decision' has been made

AMD's Frank Azor pushes back against claim that FSR 4.1 won't be ported to RDNA 3.5 GPUs — says 'no such decision' has been made

AMD's Frank Azor refutes claims that FSR 4.1 won't be available for RDNA 3.5 GPUs, stating that no decision has been made to drop support. Azor's response comes after reports suggested this exclusion, allegedly originating from AMD's David McAfee. While Azor's statement doesn't confirm FSR 4.1 support for RDNA 3.5, it hints that it hasn't been ruled out. RDNA 3.5, focused on integrated GPUs, is present in gaming laptops and handheld devices, making continued FSR support crucial. AMD recently announced FSR 4.1 for RDNA 3 and RDNA 2 GPUs, offering enhanced upscaling features.

Tom's Hardware•
Intel: Pushing EMIB Forward: Design Methodology Insights with Synopsys Tools

Intel: Pushing EMIB Forward: Design Methodology Insights with Synopsys Tools

Intel is advancing its Embedded Multi-die Interconnect Bridge (EMIB) technology to facilitate next-gen semiconductor products, emphasizing the importance of robust design methodologies in packaging success. Collaborating with Synopsys, Intel leverages EDA tools to address challenges in advanced package design, analysis, and verification, enabling designers to optimize EMIB benefits while reducing development risks. EMIB technology allows for dense interconnect pathways between dies, offering performance advantages with flexibility and cost efficiency. Synopsys' suite of advanced packaging tools supports Intel's EMIB design ecosystem, facilitating concurrent chip-package co-design to optimize performance and improve product quality. The collaboration underscores the significance of ecosystem readiness in the chiplet era for broader adoption of advanced packaging solutions in AI, high-performance computing, and data center applications.

SemiWiki•
SemiEngineering

Orbital Data Centers Are Souped-Up Satellites – For Now

Orbital data centers, currently referred to as compute centers in space, are being developed to handle AI workloads for people on Earth. These centers are powered by sun-synchronous solar energy but face challenges due to Earth's magnetic field weakening at the poles, leading to radiation effects on advanced node architectures. Companies like SpaceX and Blue Origin are investing in building satellite constellations with increased compute capabilities to meet the growing demand for AI processing power. However, technical challenges such as power generation, thermal management, and space debris need to be addressed before these orbital data centers become mainstream.

SemiEngineering•
Samsung shows first HBM5 mockup with Heat Path Block cooling — thermal race with SK hynix shaping up

Samsung shows first HBM5 mockup with Heat Path Block cooling — thermal race with SK hynix shaping up

Samsung showcased its first physical mockup of HBM5 memory at Computex 2026, featuring a new in-package cooling structure called Heat Path Block (HPB). This move comes in response to SK hynix unveiling its own iHBM thermal design, with both companies addressing the heat bottleneck in the die-to-die interface connecting memory to the processor. Samsung plans to fabricate HBM5's base die on its 2nm process, down from the 4nm node used for previous generations. The HPB design focuses on managing heat within the stack, improving thermal efficiency and performance for next-generation memory solutions.

Tom's Hardware•
Computex 2026 Day Two Wrap-Up: Intel atones for Arrow Lake, Wi-Fi 8 comes into focus

Computex 2026 Day Two Wrap-Up: Intel atones for Arrow Lake, Wi-Fi 8 comes into focus

Intel addresses the disappointment over Arrow Lake's performance at Computex 2026, admitting the need to regain trust with enthusiasts. Wi-Fi 8 is on the horizon with Asus unveiling the ROG Rapture GT-BN98 Pro, promising improved performance over Wi-Fi 7. Meanwhile, MSI showcased high-end Wi-Fi 7 routers like the RadiX BE19000, featuring unique capabilities such as an internal PCIe SSD slot for NAS functionality. The event also saw other tech reveals like Cooler Master's new MWE Gold V4 Power supplies and Microsoft's Surface RTX Spark Dev Box.

Tom's Hardware•
TSMC Expands Use of NVIDIA AI Technologies Across Chip Production Operations

TSMC Expands Use of NVIDIA AI Technologies Across Chip Production Operations

TSMC is expanding its use of NVIDIA AI and accelerated computing technologies in chip design and manufacturing operations, including lithography, process simulation, defect inspection, production scheduling, and factory optimization. By leveraging NVIDIA's CUDA-X libraries and GPU-accelerated computing platforms, TSMC has seen improvements in cycle time and cost effectiveness, particularly in computational lithography. The collaboration aims to address the complexities of modern semiconductor manufacturing and advance towards the angstrom era. AI integration in areas like transistor simulation, factory operations optimization, quality control, and digital twins for manufacturing reflects a broader industry trend towards AI-driven manufacturing for improved yield, energy efficiency, and design cycles. This partnership signifies a significant step towards autonomous, data-driven chip manufacturing and a competitive edge for leading foundries.

SemiWiki•
AMD executives react to Nvidia’s RTX Spark — ‘you’re just wrong if you don’t get a Strix Halo notebook’

AMD executives react to Nvidia’s RTX Spark — ‘you’re just wrong if you don’t get a Strix Halo notebook’

AMD executives are confident in their Strix Halo and upcoming Gorgon Halo products as competition against Nvidia's RTX Spark. They believe their hardware specs, such as 16-core / 32-thread CPU and 128 GB of local memory, position them well. AMD sees Nvidia's entry into the consumer PC market as beneficial for the industry's ecosystem. The battle between AMD and Nvidia in the AI infrastructure space continues, with AMD emphasizing the importance of memory specifications for AI workloads. Both companies are expected to release their respective products, Gorgon Halo and RTX Spark, in the near future.

Tom's Hardware•
Blog Review: Jun. 3

Blog Review: Jun. 3

The article discusses various topics in the tech industry from June 3rd, 2026. Siemens emphasizes the importance of verification IP for design teams to verify interfaces and memories efficiently. Synopsys highlights the need for collaboration in multi-die design from architecture to manufacturing. Cadence mentions the significance of unordered I/O for PCIe 7.0 in AI workloads. Keysight addresses data challenges in AI deployment for semiconductor design. Lastly, Arm introduces an open-source project for edge AI in sign language translation.

SemiEngineering•
AMD says new EXPO ‘Ultra Low Latency’ DDR5 memory should be 'effectively the same price' as current kits — feature will work on existing chipsets, but will require new DIMMs

AMD says new EXPO ‘Ultra Low Latency’ DDR5 memory should be 'effectively the same price' as current kits — feature will work on existing chipsets, but will require new DIMMs

AMD has introduced EXPO Ultra Low Latency DDR5 memory, boasting a 13% uplift over JEDEC standards and a 4% improvement over standard EXPO. The new feature, developed in collaboration with memory partners, enhances sub timings for lower latency. While compatible with existing chipsets, it will require new DIMMs. AMD's David McAfee suggests updating BIOS for compatibility. Despite current inflated memory prices, AMD aims to offer EXPO ULL kits at similar price points to current kits. The feature is expected to deliver improved performance in gaming scenarios, with partners like G.Skill and Kingston already on board to support it.

Tom's Hardware•
The rise of local agentic computing faces a brutal reality: rising DRAM prices — RTX Spark, Gorgon Halo chips subject to 63% DRAM contract price hike this quarter

The rise of local agentic computing faces a brutal reality: rising DRAM prices — RTX Spark, Gorgon Halo chips subject to 63% DRAM contract price hike this quarter

Nvidia unveiled the RTX Spark and AMD introduced the Ryzen AI Max 400 "Gorgon Halo" lineup at Computex 2026, both focusing on local agentic computing. However, the increasing demand for on-device RAM due to local AI computing is facing challenges with rising DRAM prices. AMD's Gorgon Halo SKUs offer increased memory capacity, but the market is impacted by a 58% to 63% DRAM contract price hike this quarter. Despite efforts to expand memory capacity, memory costs are significantly affecting the tech industry, with vendors adjusting prices and making strategic decisions to navigate the memory shortage.

Tom's Hardware•
Using Graph Attention for Virtual Metrology in Semiconductor Manufacturing (Intel Foundry, ASU)

Using Graph Attention for Virtual Metrology in Semiconductor Manufacturing (Intel Foundry, ASU)

Researchers from Arizona State University and Intel Foundry have collaborated on a paper titled "Graph Attention-Based Virtual Metrology for Film Deposition Processes in Semiconductor Manufacturing." The study focuses on the use of artificial intelligence-driven virtual metrology to predict wafer-level characteristics in semiconductor manufacturing. The proposed framework utilizes graph attention mechanisms to model structured dependencies among process variables, improving predictive performance compared to existing models. The approach integrates temporal feature learning and parameter-layer dependency modeling to enhance prediction accuracy and provide meaningful insights into process dynamics, supporting effective monitoring and optimization in semiconductor manufacturing.

SemiEngineering•
Intel’s next-gen LGA1954 socket will support Nova Lake, Razor Lake, and beyond — finally an Intel socket that outlives its CPUs

Intel’s next-gen LGA1954 socket will support Nova Lake, Razor Lake, and beyond — finally an Intel socket that outlives its CPUs

Intel's upcoming LGA1954 socket is rumored to support multiple generations of CPUs, starting with Nova Lake and Razor Lake, following AMD's long-term socket strategy. This move aims to reduce the frustration and cost of frequent motherboard upgrades for Intel users. The 900-series chipset motherboards, especially the Z series, are expected to feature a 64MB BIOS SPI ROM for extended compatibility with future processors. This shift may lead to a distinction between high-end and budget-friendly motherboard options, similar to AMD's AM4 platform. If successful, Intel's LGA1954 socket could revolutionize how consumers view Intel platforms.

Tom's Hardware•
Computex 2026 Day One Wrap-Up: Arm makes a bold play for Windows PCs, PCIe 6.0 SSDs are coming, Asus embraces black and gold for ROG 20th

Computex 2026 Day One Wrap-Up: Arm makes a bold play for Windows PCs, PCIe 6.0 SSDs are coming, Asus embraces black and gold for ROG 20th

Arm is making a bold move for Windows PCs with new chips like Nvidia's RTX Spark and Qualcomm's Snapdragon C. Asus showcased new laptops at Computex 2026, including the Zenbook 14 with an Intel processor and OLED display, along with ROG peripherals featuring black and gold accents. Additionally, Phison unveiled its PCIe 6.0 SSD controller capable of impressive speeds and IOPS, set to sample to customers by the end of 2026. Other highlights from the event include Noctua's thermal pad for AMD chips, Gigabyte's new gaming monitors, Cooler Master's chassis and coolers, and AMD's re-engineered Ryzen 7 5800X3D.

Tom's Hardware•
Intel reportedly no longer working on 6-core Nova Lake mobile SKU, alleges new rumor — Wildcat Lake Refresh to become focus for next-gen budget markets instead

Intel reportedly no longer working on 6-core Nova Lake mobile SKU, alleges new rumor — Wildcat Lake Refresh to become focus for next-gen budget markets instead

Intel is rumored to have halted the development of a 6-core Nova Lake mobile SKU, shifting focus to the Wildcat Lake Refresh for next-gen budget markets. The Nova Lake family is expected to offer a significant leap in performance, with the 6-core mobile variant aimed at the budget segment. However, Intel's existing Wildcat Lake lineup, designed for low-end laptops and mini-PCs, already caters to this market. The decision to prioritize Wildcat Lake over Nova Lake's entry-level chip is driven by market demand and cost considerations, especially with the competition from Apple's MacBook Neo.

Tom's Hardware•
AMD is considering a potential Ryzen 5 9600X3D — company says six-core Zen 5 X3D chip 'maybe something we look at doing... later this year'

AMD is considering a potential Ryzen 5 9600X3D — company says six-core Zen 5 X3D chip 'maybe something we look at doing... later this year'

AMD is contemplating the potential release of a Ryzen 5 9600X3D, a six-core Zen 5 X3D chip, later this year. The decision to stick with Zen 4 for the Ryzen 7 7700X3D was explained by AMD's David McAfee at Computex 2026, hinting at a possible Zen 5 chip in the future. McAfee emphasized the company's focus on providing various options for consumers amidst ongoing supply challenges. The Ryzen 5 7600X3D, despite being a limited run product, has shown competitive performance in gaming benchmarks. The introduction of a Ryzen 5 9600X3D could further expand AMD's X3D CPU lineup, offering more affordable options for gamers.

Tom's Hardware•
NVIDIA reveals if RTX Spark (N1X) can run all your Windows 11 apps

NVIDIA reveals if RTX Spark (N1X) can run all your Windows 11 apps

NVIDIA introduces RTX Spark (N1X), an Arm-based processor designed to revolutionize Windows PCs for AI applications. The processor, with up to 128GB of memory, is optimized for AI agents and NVIDIA's graphics stack. While NVIDIA claims it can run all Windows 11 apps, technical benchmarks are yet to be shared. The Surface Laptop Ultra is among the first devices to feature this chip. Windows on Arm has made strides in app compatibility, with popular apps running natively and improved emulation technology like Prism.

Windows Latest•
AMD Radeon RX 9070 GRE review: thoroughly midrange

AMD Radeon RX 9070 GRE review: thoroughly midrange

The AMD Radeon RX 9070 GRE, previously available only in China, is now launching globally at $549. Positioned as a midrange option, it aims to fill the gap between entry-level enthusiast cards and higher-end products. The card features the RDNA 4 architecture with dedicated matrix math accelerators for AI tasks and improved RT units. However, with slower GDDR6 memory and less bandwidth compared to higher-end models, it may be less suited for RT and 4K gaming. AMD faces challenges in software support for features like upscaling and frame generation compared to Nvidia, which could impact its competitiveness in the market.

Tom's Hardware•
Noctua announces new thermal pad for AMD chips in partnership with Carbice — product will work with processors in AM5 and AM4 sockets

Noctua announces new thermal pad for AMD chips in partnership with Carbice — product will work with processors in AM5 and AM4 sockets

Noctua has partnered with Carbice to introduce a new thermal pad, the NT-CP1 AM5/4, optimized for AMD Ryzen CPUs in AM5 and AM4 sockets. The pad is made of carbon nanotubes that enhance thermal conductivity over numerous thermal cycles by conforming to CPU surfaces. Unlike thermal paste, this pad offers a maintenance-free, long-lasting solution for PC builders. Noctua will exclusively distribute Carbice pads for the retail market, with potential future collaborations to develop thermal pads for Intel processors.

Tom's Hardware•
Computex 2026 Day Zero Wrap-Up: Nvidia launches RTX Spark Superchip assault on laptop and desktop markets, Intel readies Xeon 6+

Computex 2026 Day Zero Wrap-Up: Nvidia launches RTX Spark Superchip assault on laptop and desktop markets, Intel readies Xeon 6+

Nvidia made a significant impact at Computex 2026 with the launch of the RTX Spark Superchip, a powerful Arm platform aimed at transforming Windows 11 into an AI operating system. Microsoft has already adopted the chip in its Surface Laptop Ultra, featuring a high-resolution Mini LED display and a variety of ports. Intel introduced the Xeon 6+ processor to compete with AMD in the data center market, boasting impressive core counts and performance improvements. Additionally, AMD unveiled new Ryzen processors and confirmed support for the AM5 socket until at least 2029. Alienware and Asus also showcased new gaming monitors, including OLED options with high refresh rates.

Tom's Hardware•
Qualcomm aims Snapdragon C laptop chip at the budget laptop segment, as manufacturers feel the DRAM squeeze — analysts warn sub $500 laptop market may disappear before 2028

Qualcomm aims Snapdragon C laptop chip at the budget laptop segment, as manufacturers feel the DRAM squeeze — analysts warn sub $500 laptop market may disappear before 2028

Qualcomm introduces the Snapdragon C Platform, a budget-friendly Arm processor for Windows 11 laptops priced at $300, featuring older Kryo cores from Qualcomm's smartphone lineup. Analysts warn that rising memory prices may lead to the disappearance of the sub-$500 laptop market by 2028. The first laptop to feature the Snapdragon C chip is Acer's Aspire Go 15, with specifications including a 15.6-inch display, up to 8GB of memory, and up to 512GB of storage. Manufacturers are facing challenges due to increasing memory costs, impacting the pricing of budget laptops.

Tom's Hardware•
Seven hospitalized after toxic gas fire at SK hynix advanced memory plant — Cheongju 4th campus incident today led to all 3,600 staff being evacuated

Seven hospitalized after toxic gas fire at SK hynix advanced memory plant — Cheongju 4th campus incident today led to all 3,600 staff being evacuated

A fire at SK hynix's Cheongju 4th campus led to the evacuation of 3,600 employees, with seven workers hospitalized due to exposure to toxic fluorine gas. The fire, which started in a gas room, was quickly extinguished, but fluorine leakage occurred. Despite the incident, the company assures no disruption to production. Fluorine gas is highly reactive and dangerous to humans, with significant health risks at certain concentrations. Cleanup and safety inspections are ongoing at the facility.

Tom's Hardware•
Nvidia says RTX Spark chip will support all major anti-cheat and DRM technologies — Fortnite, Valorant, Denuvo, and more to work natively with Windows on Arm

Nvidia says RTX Spark chip will support all major anti-cheat and DRM technologies — Fortnite, Valorant, Denuvo, and more to work natively with Windows on Arm

Nvidia has introduced the RTX Spark Superchip, enabling Windows on Arm devices to support major anti-cheat and DRM technologies like Easy Anti-Cheat, BattlEye, and Denuvo. While games will still be emulated on Arm-based devices, the native support for anti-cheat software will allow popular titles like Fortnite and Valorant to run smoothly. This collaboration between Nvidia, Microsoft, and game developers aims to enhance gaming experiences on Arm platforms, bridging the gap that previously hindered gaming on non-x86 systems. The RTX Spark Superchip signifies a significant advancement in making Windows PC games accessible on a wider range of devices.

Tom's Hardware•
Computex 2026 Live: Every update and announcement from day one in Taipei

Computex 2026 Live: Every update and announcement from day one in Taipei

Computex 2026 is currently underway in Taipei, with Tom's Hardware reporters providing live coverage of the latest PC hardware announcements. Some key highlights include the Microsoft Surface Laptop Ultra featuring Nvidia's RTX Spark superchip with 128GB of RAM, Nvidia's roadmap for RTX Spark in laptops and desktop PCs, and Intel unveiling the Crescent Island AI GPU with up to 480GB of LPDDR5X. AMD is also making waves with the global release of the Radeon RX 9070 GRE and confirming support for AM5 through 2029. Additionally, Dell, Alienware, and Intel are showcasing new products aimed at challenging competitors in the market.

Tom's Hardware•
Intel warns it has 'a healthy dose of paranoia' over Nvidia entrance into PC market — company says RTX Spark is 'great for the market' while touting the virtues of x86

Intel warns it has 'a healthy dose of paranoia' over Nvidia entrance into PC market — company says RTX Spark is 'great for the market' while touting the virtues of x86

Intel is closely monitoring Nvidia's entrance into the PC market with its RTX Spark SoC range, expressing a mix of caution and confidence in response to the competition. Intel acknowledges Nvidia's expertise in gaming and AI but highlights potential compatibility and pricing concerns with the new platform. While Intel has made strides in consumer graphics, it has yet to directly challenge Nvidia's dominance. Despite the competition, Intel plans to continue collaborating with Nvidia, particularly in the data center space. The article emphasizes the importance of compatibility and pricing as key factors to watch as Nvidia's RTX Spark designs are set to launch in the fall.

Tom's Hardware•
AMD promises 13% uplift with new EXPO ‘Ultra Low Latency’ overclocking on DDR5 DIMMs — automatic memory overclocking delivers 4% improvement over standard EXPO, says AMD

AMD promises 13% uplift with new EXPO ‘Ultra Low Latency’ overclocking on DDR5 DIMMs — automatic memory overclocking delivers 4% improvement over standard EXPO, says AMD

AMD introduces EXPO Ultra Low Latency (ULL), an automatic memory overclocking feature that promises a 13% performance boost in over 30 games compared to JEDEC standard DDR5 speeds. This new tool also delivers a 4% improvement over the standard EXPO, according to AMD's internal benchmarks. The company has not specified the release date for ULL but has lined up memory partners to support the feature. AMD's benchmarks show significant performance gains with EXPO ULL on a Ryzen 7 9700X CPU, but details on how ULL achieves the extra performance compared to standard EXPO remain unclear.

Tom's Hardware•
We went hands-on with Qualcomm's new '$300 and up' ARM laptop platform with mystery eight-core CPU — active-cooled Snapdragon C laptop surfaces in Acer Aspire Go 15

We went hands-on with Qualcomm's new '$300 and up' ARM laptop platform with mystery eight-core CPU — active-cooled Snapdragon C laptop surfaces in Acer Aspire Go 15

Qualcomm's new Snapdragon C platform, aimed at affordable laptops, features a mystery eight-core CPU running at 1.67 GHz with a base speed of 3.01 GHz. The chip is speculated to be a customized version of mobile-focused Kryo chips. The Acer Aspire Go 15, showcasing this platform, boasts a 16.5" FHD display, up to 8 GB of memory, a 53 Wh battery, and up to 512GB of storage. The laptop includes features like a FHD webcam, USB-C ports, HDMI port, USB-A port, and Wi-Fi 6 support. Pricing details are yet to be revealed, but the system is expected to fall within the "$300 and up" range.

Tom's Hardware•
Microsoft builds its ultimate MacBook Pro rival with the NVIDIA powered Surface Laptop Ultra

Microsoft builds its ultimate MacBook Pro rival with the NVIDIA powered Surface Laptop Ultra

Microsoft has unveiled the Surface Laptop Ultra, a powerful MacBook Pro competitor powered by NVIDIA RTX SPARK. The laptop runs Windows on Arm and boasts impressive hardware specifications, including up to 128GB unified memory and a 15-inch mini-LED PixelSense Ultra touchscreen. Microsoft optimized Windows 11 for the NVIDIA RTX Spark platform to maximize performance and power efficiency. The NVIDIA partnership brings native CUDA support and enhanced GPU capabilities, promising a new level of performance for professional users. Despite the premium price tag, the Surface Laptop Ultra aims to redefine professional computing with its cutting-edge features and design.

Windows Latest•
The Sub-2nm Paradox

The Sub-2nm Paradox

The article discusses the challenges faced in semiconductor manufacturing at 2nm and below. It highlights the impact of process variation and physics on design, manufacturing, and economics. The focus is on faster data movement and more efficient computing rather than just increasing transistor density. The article also touches on the complexities of achieving performance, power consumption, and transistor density promises at these advanced nodes. Additionally, it mentions the shift towards multi-die assemblies of chiplets to address the limitations of single-die designs.

SemiEngineering•
Nvidia lays out RTX Spark roadmap for laptops and desktop PCs at Computex 2026 — three generations outlined, Rubin with LPDDR6 memory, followed by Rosa Feynman

Nvidia lays out RTX Spark roadmap for laptops and desktop PCs at Computex 2026 — three generations outlined, Rubin with LPDDR6 memory, followed by Rosa Feynman

Nvidia unveils its RTX Spark roadmap for laptops and desktop PCs at Computex 2026, promising at least two more generations of Spark platforms beyond the Grace Blackwell RTX Spark chips. Future generations will include Vera Rubin Sparks with LPDDR6 memory and a Rosa Feynman Spark with faster memory. The company aims to build a full product and partner ecosystem, emphasizing its commitment to delivering future generations of Sparks. Nvidia also plans to release Windows on Arm-compatible versions of its high-end DGX Station desktop PC, showcasing its dedication to creating a robust ecosystem for partners and consumers. More details on Nvidia's RTX Spark systems and ecosystem are expected to be revealed at Computex 2026.

Tom's Hardware•
Nvidia unveils RTX Spark Superchip for laptops and desktop PCs at Computex 2026 – new platform promises to turn Windows into an agentic AI OS with Arm CPU, Blackwell GPU, and 128GB unified memory

Nvidia unveils RTX Spark Superchip for laptops and desktop PCs at Computex 2026 – new platform promises to turn Windows into an agentic AI OS with Arm CPU, Blackwell GPU, and 128GB unified memory

Nvidia introduced the RTX Spark Superchip at Computex 2026, aiming to revolutionize Windows into an agentic AI platform for laptops and desktops. The platform features an Arm CPU, Blackwell GPU, and 128GB unified memory, promising efficiency and power for AI interactions. The RTX Spark Superchip offers up to 20 Arm CPU cores, a Blackwell GPU with 6,144 CUDA cores, and 128GB of LPDDR5X RAM, supporting AI agents and 120-billion-parameter models. Nvidia collaborates with partners like Dell, HP, Lenovo, Asus, MSI, and Microsoft to deliver high-performance laptops with OLED displays, extended battery life, and premium design. The RTX Spark platform also enhances creative and gaming capabilities, with promises of 100 FPS 1440p gaming and Adobe Photoshop and Premiere optimizations for GPU acceleration and AI workflows.

Tom's Hardware•
Intel details long-awaited Crescent Island AI GPU at Computex, boasts up to 480 GB of LPDDR5X to combat memory shortages — company shares more details of its Xe3P inference accelerator at Computex

Intel details long-awaited Crescent Island AI GPU at Computex, boasts up to 480 GB of LPDDR5X to combat memory shortages — company shares more details of its Xe3P inference accelerator at Computex

At Computex 2026, Intel unveiled details about its upcoming Data Center GPU, Crescent Island, based on the Xe3P GPU architecture. The GPU will utilize LPDDR5X memory, offering up to 480GB of memory capacity. This design choice aims to maximize memory capacity while maintaining high bandwidth, potentially making Crescent Island more efficient for AI inference tasks. The air-cooled card with modest power requirements is expected to fit into traditional GPU servers, appealing to companies seeking on-premise inferencing solutions. Intel plans to launch Crescent Island in the second half of 2026, with more information expected as the year progresses.

Tom's Hardware•
AMD’s formerly China-exclusive Radeon RX 9070 GRE goes global for $549 on June 2 — RDNA 4 GPU will bridge the gap between RX 9060 XT and RX 9070

AMD’s formerly China-exclusive Radeon RX 9070 GRE goes global for $549 on June 2 — RDNA 4 GPU will bridge the gap between RX 9060 XT and RX 9070

AMD is making the formerly China-exclusive Radeon RX 9070 GRE available globally on June 2 for $549. The GPU features 48 CUs and 12GB of GDDR6 memory, offering 21% higher gaming performance than the GeForce RTX 5060 Ti 16GB. Despite lower specs than the RX 9070 and RX 9070 XT, the RX 9070 GRE maintains a 220W TDP. The move aims to address the market gap for a sub-$550 model amidst rising prices due to the ongoing memory shortage.

Tom's Hardware•
Watch Nvidia's Computex 2026 keynote here — Jensen Huang takes the stage for Computex and GTC Taipei at 8pm PT / 11pm ET on May 31

Watch Nvidia's Computex 2026 keynote here — Jensen Huang takes the stage for Computex and GTC Taipei at 8pm PT / 11pm ET on May 31

Nvidia CEO Jensen Huang will kick off Computex 2026 with a keynote address on May 31 at 8pm PT / 11pm ET, focusing on AI breakthroughs. The event will likely cover enterprise AI and potentially reveal consumer announcements, such as the N1X chip. The keynote is expected to run for about two hours, with a possibility of consumer-related updates but unlikely to include GeForce news. Nvidia is also hosting GTC Live with pregame events leading up to the keynote.

Tom's Hardware•
AMD confirms AM5 support through 2029 — Zen 4 and 5 platform will likely see two more generations, at least

AMD confirms AM5 support through 2029 — Zen 4 and 5 platform will likely see two more generations, at least

AMD has confirmed support for its AM5 socket through 2029, indicating at least two more generations beyond the current Zen 4 and Zen 5 platforms. The extension of support was announced at Computex 2026, with previous commitments only reaching up to 2027. Despite AMD's silence on specific future products, the likelihood of Zen 6 CPUs and potentially Zen 7 on the AM5 socket by 2029 is high. Additionally, AMD teased a new version of its EXPO memory overclocking feature, EXPO Ultra Low Latency, promising a 4% performance boost compared to its predecessor.

Tom's Hardware•
AMD brings back Ryzen 7 5800X3D, launches Ryzen 7 7700X3D to combat rising component prices — eight-core X3D CPUs arrive under $350 for AM4 or AM5, DDR4 or DDR5

AMD brings back Ryzen 7 5800X3D, launches Ryzen 7 7700X3D to combat rising component prices — eight-core X3D CPUs arrive under $350 for AM4 or AM5, DDR4 or DDR5

AMD is releasing the Ryzen 7 5800X3D and Ryzen 7 7700X3D CPUs to address the increasing component prices. The Ryzen 7 5800X3D, priced at $349, features eight Zen 3 cores, 4.5 GHz boost clock, and 100 MB of cache. It will be available on June 25. The Ryzen 7 7700X3D, a downclocked version of the Ryzen 7 7800X3D with eight Zen 4 cores, launches at $329. AMD aims to offer competitive options for both AM4 and AM5 platforms with DDR4 or DDR5 support.

Tom's Hardware•
Nvidia's long-awaited N1/N1X SoC specs leak ahead of Computex launch — N1 to feature up to 20 Arm-based cores, standard N1 equipped with 12- and 10-core configs

Nvidia's long-awaited N1/N1X SoC specs leak ahead of Computex launch — N1 to feature up to 20 Arm-based cores, standard N1 equipped with 12- and 10-core configs

Nvidia is finally set to launch the N1 family of SoCs at Computex, with leaked specs revealing two SKUs: N1 and N1X. The standard N1 comes in 12-core and 10-core configurations with 8x PCIe 5.0 lanes and up to 64GB of memory support. The more powerful N1X features a 20-core and 18-core variant with higher CUDA cores and PCIe lanes, targeting high-end markets. Nvidia's re-entry into the laptop market with these chips could bring competition to AMD, Intel, and Apple, but pricing may be a concern, especially during a RAM shortage. N1X is expected to compete in the $2,000+ market, while the N1 could offer a midrange option under $1,500.

Tom's Hardware•
Lenovo Yoga Slim 7x review: Snapdragon X2 Elite makes its case

Lenovo Yoga Slim 7x review: Snapdragon X2 Elite makes its case

The Lenovo Yoga Slim 7x is a premium ARM ultraportable featuring Qualcomm's Snapdragon X2 Elite for enhanced performance and battery life. With a sleek design, comfortable keyboard, OLED touch panel, and sharp webcam, it offers a compelling package for those not reliant on x86 apps. The laptop's port selection is limited to USB-C, requiring adapters for other connections. Performance-wise, the Slim 7x excels in productivity tasks and offers impressive battery life, outlasting competitors like the MacBook Air. Overall, it presents a strong option for users seeking a balance of performance, battery life, and affordability in an ARM laptop.

Tom's Hardware•
Microsoft veteran recalls the last time Nvidia and Arm was the future of Windows — shares a video of ‘the first time Windows ran on Nvidia Tegra Arm’ from 2010

Microsoft veteran recalls the last time Nvidia and Arm was the future of Windows — shares a video of ‘the first time Windows ran on Nvidia Tegra Arm’ from 2010

Microsoft veteran Steven Sinofsky reflects on the past excitement around Windows running on Nvidia Tegra Arm chips in 2010/11, sharing a video of the milestone. While the initial collaboration with Nvidia was promising, Windows on Arm also supported Qualcomm and TI processors. Despite the enthusiasm, the partnership with Qualcomm didn't achieve significant breakthroughs. As Nvidia and Arm gear up for a new Windows era, the success of this latest endeavor remains to be seen, with potential challenges like pricing to address.

Tom's Hardware•
You can still run the original Nvidia Control Panel by grabbing it from the Microsoft Store today — app remains useful to adjust a handful of RTX Pro and Quadro features, and may be handy for troubleshooting

You can still run the original Nvidia Control Panel by grabbing it from the Microsoft Store today — app remains useful to adjust a handful of RTX Pro and Quadro features, and may be handy for troubleshooting

Nvidia has retired the old Nvidia Control Panel (NVCP) in favor of the Nvidia App, but the original NVCP is still available for download from the Microsoft Store. While most settings have been migrated to the new app, the NVCP remains useful for adjusting RTX Pro and Quadro features, especially those not yet available in the Nvidia App. Users may find it handy for troubleshooting or if upcoming Nvidia software updates cause issues with the new app's controls.

Tom's Hardware•
ATX12VO V3 standard shrinks the connector and maximizes power efficiency — new 8-pin connector also brings smarter power supply monitoring

ATX12VO V3 standard shrinks the connector and maximizes power efficiency — new 8-pin connector also brings smarter power supply monitoring

Intel is set to release the ATX12VO V3 standard, enhancing power efficiency and introducing a smaller 8-pin connector for improved power supply monitoring. This revision aims to simplify power supply design, increase efficiency during low-power states, and offer new Low Power and High Power modes. The new standard will reduce connector sizes by up to 83%, making system layouts more compact and cost-effective. Additionally, support for PMBus communication and real-time power utilization data will enhance system monitoring and prevent power supply overload. Intel may announce the launch at the upcoming Computex 2026 expo.

Tom's Hardware•
Huawei chairman thanks the US for export restrictions on chips, says it supercharged China’s semiconductor industry — Washington’s export controls encouraged Chinese firms to invest in R&D and build their own tech stack competing with American tech

Huawei chairman thanks the US for export restrictions on chips, says it supercharged China’s semiconductor industry — Washington’s export controls encouraged Chinese firms to invest in R&D and build their own tech stack competing with American tech

Huawei's Chairman expressed gratitude for U.S. export restrictions on chips, stating they spurred China's semiconductor industry growth. The ban led Chinese firms to invest in R&D and develop their own tech stack to compete with American counterparts. Despite challenges like smuggling and black market activities for chips, Chinese companies are now releasing competitive alternatives to U.S. chipmakers. The ban, while initially delaying Chinese AI development, ultimately accelerated the country's progress in creating domestic semiconductor solutions. This shift has significantly impacted the market share of American tech companies like Nvidia in China.

Tom's Hardware•
Nvidia and Microsoft tease "a new era of PC" ahead of Computex 2026 — coordinated social media posts could indicate that rumored N1X laptops will be Windows on Arm systems

Nvidia and Microsoft tease "a new era of PC" ahead of Computex 2026 — coordinated social media posts could indicate that rumored N1X laptops will be Windows on Arm systems

Nvidia and Microsoft are hinting at a significant development in the PC world ahead of Computex 2026, with social media posts suggesting the possible launch of N1X laptops running Windows on Arm. The N1X platform is rumored to feature a powerful GB10 Superchip with an RTX 5070-class GPU and a Mediatek-designed 20-core Arm CPU complex. If Microsoft supports N1X, it could expand the platform's appeal by bringing the full Windows app ecosystem to these devices. However, N1X PCs may be expensive due to their advanced features and the current silicon shortage. More details are expected to be revealed at Computex 2026.

Tom's Hardware•
AMD's flagship RX 9070 XT GPU hits an all-time low $629 in Amazon lightning sale — Gigabyte's Gaming OC graphics card now $110 off

AMD's flagship RX 9070 XT GPU hits an all-time low $629 in Amazon lightning sale — Gigabyte's Gaming OC graphics card now $110 off

AMD's flagship RX 9070 XT GPU, the Gigabyte Gaming OC Radeon, is currently on sale for an all-time low price of $629 on Amazon, $110.99 below its usual price. The graphics card features 64 RDNA 4 compute units, 4096 stream processors, and 16GB of GDDR6 VRAM, making it ideal for 1440p gaming. While it may not excel in ray tracing compared to Nvidia, it offers strong performance in rasterization and frame throughput. This deal offers a significant discount and is a competitive option for gamers seeking high-quality graphics at a lower price point.

Tom's Hardware•
Redditor buys a Ryzen 9 9950X3D2 Dual Edition, gets a normal 9950X3D instead

Redditor buys a Ryzen 9 9950X3D2 Dual Edition, gets a normal 9950X3D instead

A Redditor purchased what was supposed to be an AMD Ryzen 9 9950X3D2 Dual Edition processor but received a regular 9950X3D instead. The discrepancy was discovered when the packaging revealed the incorrect CPU's heatspreader label. The situation raises questions about how the mix-up occurred, with possibilities ranging from a packaging error at AMD to tampering with the product after purchase. The Redditor's experience highlights the importance of vigilance during unboxing and checking for intact factory seals. Amazon is reportedly addressing the issue, emphasizing the need for thorough quality checks to prevent such incidents in the future.

TweakTown•

We use cookies

We use cookies to ensure you get the best experience on our website. For more information on how we use cookies, please see our cookie policy.