GoKawiil
Tech News
Microsoft has launched the Surface Laptop Ultra, a new flagship Windows laptop built around Nvidia's new RTX Spark N1X 'superchip,' aimed at handling on-device AI, gaming and content creation in one machine. At 4.4 pounds, it's lighter than the 16-inch MacBook Pro but is the thickest among the new wave of N1X-powered laptops, including HP's OmniBook Ultra.
Microsoft's new Surface Laptop Ultra, built with an Nvidia RTX Spark N1X chip, starts at $2,599 and can be configured up to $5,899 depending on processor and memory. The base 24GB unified memory model ships with Windows 11 Home, while higher-tier configurations come with Windows 11 Pro and up to 64GB of memory. During a review briefing, Microsoft gave only a vague answer when asked whether the 24GB base model has enough memory for tasks like running AI models locally.
The Verge's staff spent a week testing Nvidia's new Arm-based RTX Spark N1X processor across five laptops, including Microsoft's Surface Laptop Ultra, Asus's ProArt P16, Lenovo's Yoga 9n and HP's OmniBook Ultra. All units share the same 20-core CPU with 6,144 CUDA cores, though a cheaper 18-core variant with fewer CUDA cores also exists, with prices ranging from $2,599.99 to $6,999.99. The chip combines RTX Blackwell graphics with large unified memory pools of up to 128GB, aimed at running local AI models.
Microsoft has released the Surface Laptop Ultra, a premium Windows-on-Arm laptop built with Nvidia to showcase its new RTX Spark chips, which are designed to run AI models locally alongside demanding games and creative software. The top-spec model costs $5,900 and features a 15-inch 3:2 MicroLED touchscreen reaching 2,000 nits, an all-aluminum design, and a new Magnetic Connect charging port built into a USB-C port.
A Washington Post investigation citing an internal fundraising spreadsheet found Apple and Nvidia each committed $25 million to fund Trump's planned $600 million White House ballroom, tying them as the largest donors alongside the Adelson Family Foundation and Charlie Ergen's household and donor-advised fund. Together these four donations make up roughly 30% of the $333 million in contributions documented. Other tech firms named include Google (via a $22 million settlement-linked trust), Microsoft and Palantir ($10 million each), Amazon and Meta ($5 million each), and smaller sums from HP, Micron, Salesforce, Cisco, Coinbase, Ripple, ServiceNow and T-Mobile.
Nvidia and Cisco have confirmed making multimillion-dollar donations to the renovation of a White House ballroom championed by President Trump, joining a donor list reported by The Washington Post to exceed $330 million. Nvidia and Apple each reportedly gave $25 million, while Cisco confirmed a $5 million contribution; Amazon, Palantir, ServiceNow, Meta and Microsoft were also named as major donors, funneled through a nonprofit trust.
The Washington Post obtained a donor list showing tech, telecom and crypto companies as the largest contributors to President Trump's White House ballroom project, totaling more than $183 million. NVIDIA, Apple and Dish Network co-founder Charles Ergen each gave $25 million, with Microsoft, Meta, Palantir, Google, Amazon, Cisco, Blackrock, Lockheed Martin, Gemini and Coinbase among other donors. About $77 million of the funds remain unaccounted for since the donations aren't subject to public disclosure or campaign finance rules.
Volvo and Nvidia announced an expanded version of their near decade-long partnership, under which Volvo will integrate three Nvidia technologies: the Hyperion sensor and compute hardware, the Alpamayo driving software stack, and the Halos safety evaluation layer. The system is designed to handle both urban streets and highways, combining AI-based and classical software approaches with multiple sensor types for redundancy. Volvo did not disclose financial terms or specify which model will launch first, though executives said a new vehicle using the stack will arrive within the decade.
TSMC posted its sixth consecutive quarter of record profit and forecast fourth-quarter revenue up 58% year-over-year, driven by demand for advanced chipmaking. Executives say wafer capacity is stretched thin as clients including Nvidia, AMD, Apple, Google, Amazon, Microsoft, Meta and OpenAI compete for production slots.
Micro Center has changed the listing status of Nvidia's GeForce RTX 5090 to 'Not Carried,' a classification indicating the retailer does not plan to restock the card, rather than a temporary out-of-stock label. Custom RTX 5090 models from partners like Asus, MSI, and Gigabyte remain listed on the site but are sold out at all physical stores, with the sole exception of a $11,000 Asus ROG 20th Anniversary bundle. The change follows recent rumors that Nvidia has halted production of the card.
Nvidia has launched Nemotron 3.5 Super VL, a 120-billion-parameter open-weight model aimed at agentic enterprise tasks. The company says benchmark testing by Artificial Analysis shows it is five times faster than the next-best open model in its class, outperforming InclusionAI's Ling 3.0 Flash and DeepSeek's V4.1 Flash. Nvidia also cited a 44% accuracy gain in a medical insurance claims task after post-training, surpassing Claude Opus 5 on that specific benchmark.
Nvidia and Cisco have jointly unveiled Ultra-Dense Small Form Factor Pluggable (USFP), a new optical transceiver standard supporting 400Gb/s per lane, the first interface to do so. Cisco says the module is designed to eventually reach total transfer rates of 12.8Tb/s, offering four times the bandwidth of existing 200Gb/s optics and 1.3x better power efficiency. Initial USFP transceivers will have 16 full-duplex lanes and work with both switches and Nvidia's SuperNIC adapters, which move data between GPUs using the RoCE protocol.