News Gist .News

Articles | Politics | Finance | Stocks | Crypto | AI | Technology | Science | Gaming | PC Hardware | Laptops | Smartphones | Archive

Nvidia Unveils High-Performance Bluefield-3 Supernic for Storage Applications

Nvidia has introduced a new iteration of its BlueField-3 Data Processing Unit (DPU), which morphs into a special self-hosted storage powerhouse with an 80GBps memory bandwidth. This upgraded model allows direct integration with NVMe SSDs and GPUs, bypassing the need for an external CPU. The special version, B3220SH, offers advanced capabilities for hardware connections, simplifying data flow and reducing latency in storage-intensive applications.

See Also

Sabrent 30.72TB Enterprise SSD Revolutionizes Data Storage with Unparalleled Performance and Capacity Δ1.77

The Sabrent Rocket Enterprise PCIe 4.0 U.2/U.3 NVMe SSD has set a new benchmark for enterprise storage solutions, offering up to 7,000MB/s read speeds and handling over 56PB of data with one drive write per day durability. This massive 30.72TB model is designed to meet the demands of large-scale operations, including data centers and businesses requiring high-speed, high-endurance storage solutions. With its ultra-low bit error rate and sustained low-latency performance, this SSD is poised to disrupt the enterprise storage market.

Nvidia RTX Pro 6000 Blackwell GPU Spotted with 24,064 CUDA Cores, 96GB GDDR7, and 600W — 11% More Cores than Ada Δ1.74

The upcoming Nvidia RTX Pro 6000 Blackwell GPUs are expected to feature improved performance and higher memory capacities, positioning them as key components in professional workstations. The dual-flavored launch indicates a growing trend of workstation GPUs with enhanced capabilities, catering to specific industry demands. With two variants in the pipeline, Nvidia's strategy for these high-end graphics cards is yet to be fully understood.

AMD RDNA4 Navi 48 Sets New Benchmark for GPU Density Δ1.74

AMD has announced its latest Radeon RX 9000-series GPU, revealing that the Navi 48 die is not only smaller than expected but also holds a record-breaking density of 150 million transistors per square millimeter. This achievement surpasses Nvidia's GB203 die and even outshines the Blackwell consumer peak, setting a new standard for GPU design. The Navi 48's high transistor count is expected to boost performance, making it a formidable competitor in the market.

Boosting PC Performance with Next-Gen Snapdragon X2 Chips Δ1.73

The upcoming Qualcomm Snapdragon X2 processor for Windows PCs may offer up to 18 Oryon V3 cores, increasing core count by 50% compared to the current generation. The new chip's system in package (SiP) will incorporate both RAM and flash storage, featuring 48GB of SK hynix RAM and a 1TB SSD onboard. This next-generation processor is expected to be used in high-end laptops and desktops, potentially revolutionizing PC performance.

Lenovo ThinkBook 16p Gen 6 Specs and Features Unveiled with High-Performance Laptop Δ1.73

The Lenovo ThinkBook 16p Gen 6 laptop features Intel Core Ultra HX processors and Nvidia RTX-series GPUs, making it ideal for professionals who require high-performance computing. The new model boasts a 16-inch 3.2K display, Wi-Fi 7, and enhanced cooling capabilities, providing an optimal user experience for demanding workloads like 3D rendering, video editing, and AI-assisted tasks. Lenovo's latest offering also includes a dedicated Neural Processing Unit (NPU) for AI-based automation and workflow optimization.

Nvidia's GeForce RTX 5070 Falls Short of Hype Δ1.73

The Nvidia GeForce RTX 5070 delivers excellent 1440p gaming performance thanks to its DLSS 4 Multi-Frame Gen technology, but it fails to deliver a significant upgrade over its predecessor. Its tiny two-slot design and cute factor are notable highlights, but the lack of performance increase and skimpy memory capacity limit its appeal for future-proofing. With a price tag that's still relatively high compared to its capabilities, potential buyers should carefully consider their needs before making a purchase.

Micron Shows Off World's Fastest PCIe 6.0 SSD, Hitting 27 GB/S Speeds — Astera Labs PCIe 6.0 Switch Δ1.73

Micron, in collaboration with Astera Labs, has showcased the world's first PCIe 6.0 SSDs at DesignCon 2025, achieving unprecedented sequential read speeds of over 27 GB/s. This remarkable performance, which doubles the speeds of current PCIe 5.0 drives, was made possible through the integration of Astera's Scorpio PCIe 6.0 switch and Nvidia's Magnum IO GPUDirect technology. The advancements in PCIe 6.0 technology signal a significant leap forward for high-performance computing and artificial intelligence applications, emphasizing the industry's need for faster data transfer rates.

Micron Presents Two Gen6 E3.S SSDs That Are Almost Twice as Fast as the Fastest Gen5 SSDs Currently Δ1.73

Micron has launched a prototype of its PCIe 6.x SSD featuring a remarkable sequential read speed of 27GB/s, marking a significant advancement over its previous models. This breakthrough was demonstrated at DesignCon 2025, where the SSD's performance was enhanced by Astera Labs' Scorpio P-Series Fabric Switch, showcasing the potential of PCIe 6.x technology in high-speed data transfer. While this innovation promises to address the growing demands of AI and cloud computing, widespread availability of PCIe 6.x storage solutions is still years away due to the need for an ecosystem that supports its capabilities.

Apple Unveils M3 Ultra Chip with Industry-Leading Performance and Graphics Capabilities Δ1.73

The new M3 Ultra chip boasts a 32-core CPU, 80-core GPU, and 32-core Neural Engine, making it Apple's most capable processor to date. The chip can pair with up to 16TB of internal storage and up to 512GB of unified memory, offering impressive performance for demanding tasks such as video editing and game development. The updated Mac Studio is set to launch on March 12, starting at $1,999.

Nvidia RTX 6000 Pro: Details of New Workstation Graphics Card Leak Amid Rumours of GTC 2025 Launch Δ1.72

The Nvidia RTX 6000 Pro workstation graphics card is expected to be officially unveiled at GTC 2025, with specifications revealed by Leadtek and VideoCardz. The GPU allegedly boasts 24,064 CUDA cores, 752 Tensor cores, and 188 RT cores, significantly outperforming the current GeForce RTX 5090. Nvidia's forthcoming release promises to revitalize the graphics card market.

Overclocking Performance for Gaming and Creativity Δ1.72

The Lenovo ThinkBook 16p Gen 6 laptop offers exceptional computing power for complex workloads, thanks to its Intel Core Ultra processors and discrete NPU module. Powered by these components, the laptop can handle demanding tasks such as real-time studio-grade acceleration for 3D rendering, modeling, and visualization of complex designs. By overclocking the GPU and CPU to a combined 200W TDP in Geek Mode, the ThinkBook 16p Gen 6 can deliver blistering performance that rivals some of the most powerful gaming laptops on the market.

Nvidia (NVDA) Stock Soars on Oversold Conditions Δ1.72

Nvidia's stock advanced on Friday as buyers rushed in to purchase oversold stocks, driven by the company's stronger-than-expected fourth-quarter results and above-average 2025 sales guidance. The chip maker reported a surge in Q4 sales, with revenue from data centers more than doubling year-over-year, and surpassed its sales guidance by almost $2 billion. Despite some challenges in transitioning to new technology, Nvidia's shares have rallied on optimistic views from analysts.

DeepSeek Brings Disruption to AI-Optimized Parallel File Systems, Releases Powerful New Open-Source 3FS Δ1.72

DeepSeek has made its Fire-Flyer Fire System (3FS) parallel file system fully open-source this week, as part of its Open Source Week event. The disruptive AI company from China brags that 3FS can hit 7.3 TB/s aggregate read throughput in its own server data clusters, where DeepSeek has been using 3FS to organize its servers since at least 2019.3FS is a Linux-based parallel file system designed for use in AI-HPC operations, where many data storage servers are being constantly accessed by GPU nodes for training LLMs.

Lenovo Unveils High-Performance Laptops with RTX 5070 Power Δ1.72

The Lenovo Yoga Pro 9i Aura Edition is the latest laptop to be showcased at Mobile World Congress, featuring an Intel Core Ultra processor paired with Nvidia graphics up to a GeForce RTX 5070. The machine boasts a 16-inch display with a 3200 x 2000 resolution and a 120 Hz tandem OLED panel with up to 1600 nits peak brightness. Lenovo is targeting this laptop at professionals who require on-device AI processing capabilities.

Is RISC-V the Future of Compute? Alibaba Thinks so as It Launches a New Server-Grade Chip that Will Revolutionize High-Performance Computing Δ1.72

Alibaba's latest move with the launch of its C930 server processor demonstrates the company's commitment to developing its own high-performance computing solutions, which could significantly impact the global tech landscape. By leveraging RISC-V's open-source design and avoiding licensing fees and geopolitical restrictions, Alibaba is well-positioned to capitalize on the growing demand for AI and cloud infrastructure. The new chip's development by DAMO Academy reflects the increasing importance of homegrown innovation in China.

The Rise of Edge AI in Rugged Laptops Δ1.72

Getac's new B360 series brings AI, triple batteries, and NVIDIA GPU power to its rugged laptops, enhancing security, speed, and efficiency with AIB360 supports three SSDs, triple batteries, and Thunderbolt 4 connectivity. Built for extremes, the B360 meets MIL-STD-810H and IP66 standards. Getac has launched two new AI-enabled rugged laptops, the B360 and B360 Pro, designed for professionals working in extreme environments and demanding industries.

OpenAI, Oracle Eye Nvidia Chips Worth Billions for Stargate Site Δ1.71

OpenAI and Oracle Corp. are set to equip a new data center in Texas with tens of thousands of Nvidia's powerful AI chips as part of their $100 billion Stargate venture. The facility, located in Abilene, is projected to house 64,000 of Nvidia’s GB200 semiconductors by 2026, marking a significant investment in AI infrastructure. This initiative highlights the escalating competition among tech giants to enhance their capacity for generative AI applications, as seen with other major players making substantial commitments to similar technologies.

The Future of Upscaling Tech in Gaming Δ1.71

AMD FSR 4 has dethroned FSR 3 and Nvidia's DLSS CNN model, according to Digital Foundry, offering significant image quality improvements, especially at long draw distances, with reduced ghosting. The new upscaling method is available exclusively on AMD's RDNA 4 GPUs, but its performance and price make it a strong competitor in the midrange GPU market. FSR 4's current-gen exclusivity may be a limitation, but its image quality capabilities and affordable pricing provide a solid starting point for gamers.

Apple M3 Ultra Makes Geekbench Debut Alongside New Mac Studio Δ1.71

Apple's M3 Ultra chip has debuted on Geekbench, showcasing significant enhancements over its predecessor, the M2 Ultra, with up to 30% better CPU performance and a 13% increase in GPU speed. The new Mac Studio, powered by the M3 Ultra, features advanced specifications, including a remarkable 32 CPU cores and support for 512 GB of unified memory. Despite its impressive capabilities, the pricing of the M3 Ultra-powered Mac Studio raises questions about its market competitiveness against more affordable alternatives.

Zeaus GPU Beats Nvidia's RTX 5090 by 10 Times in Certain Workloads Δ1.71

Bolt Graphics' Zeus GPU platform has been shown to outperform Nvidia's GeForce RTX 5090 in path tracing workloads, with a performance increase of around 10 times. However, the RTX 5090 excels in AI workloads due to its superior FP16 TFLOPS and INT8 TFLOPS capabilities. The Zeus GPU relies on the open-source RISC-V ISA and features a multi-chiplet design, which allows for greater memory size and improved performance in path tracing and compute workloads.

GPU Compute Benchmark Performance Soars for Nvidia RTX 5090 Before Support Patch Is Released Δ1.71

The performance of the Nvidia GeForce RTX 5090 in GPU Compute tests has significantly improved as more samples have passed through PassMark's test site. The release of a patch that should solve problems with the Blackwell card has also contributed to the improvement, allowing the RTX 5090 to reach its true performance potential. With the right support, gamers and PC builders can expect to enjoy most of the benefits of their high-end hardware purchase.

Lenovo Unveils AI Stick with 32 TOPS NPU, A Portable Solution for Local AI Acceleration? Δ1.71

Lenovo has introduced the Lenovo AI Stick, a portable device that connects to PCs via USB-C Thunderbolt and enables local AI acceleration through its 32 TOPS NPU. The stick is designed to provide AI features such as large language models and graphics apps on devices without dedicated hardware, making it an attractive solution for those with powerful processors but no NPUs or slower NPUs. However, the device's power requirements and compatibility with specific systems remain unclear.

Zotac Gaming Geforce Rtx 5090 Solid Review: Benchmarking Nvidia's Flagship Gpu Vs. Founders Edition Δ1.71

The Zotac Gaming GeForce RTX 5090 Solid delivers top-tier performance with Blackwell architecture, outperforming the reference model in terms of speed, cooling, noise, and efficiency, making it ideal for demanding users. Its strong gaming and creative performance is enhanced by features like DLSS 4.0 and multi-frame generation. The card's minimalist design complements subtle lighting, adding a modern touch.

Apple Levels Up the Mac Studio with the M4 Max and Unveils Its Most Powerful Chip Ever, the M3 Ultra Δ1.71

Apple has introduced significant upgrades to the Mac Studio, featuring the new M4 Max chip and the unprecedented M3 Ultra, which offers remarkable performance enhancements for creative professionals. The M4 Max boasts a configurable 14-core CPU and up to 40 GPU cores, while the M3 Ultra, made from dual M3 Max chips, delivers a stunning 32-core CPU and up to an 80-core GPU, supporting extensive memory and storage options. As Apple continues to push the boundaries of its silicon technology, the new Mac Studio models are poised to redefine power and efficiency in the professional computing landscape.

Dynabook Unveils Portégé Z40L-N Laptop with Intel's Latest AI CPU Δ1.71

The Dynabook Portégé Z40L-N is a business-focused laptop powered by Intel’s latest Core Ultra processor (Series 2), featuring a Neural Processing Unit (NPU) optimized for AI tool workloads. The device offers customizable configurations and extended warranties, making it an attractive option for professionals who require both mobility and performance. With its user-replaceable battery and AI power management system, the laptop aims to extend device lifespan and optimize energy consumption.