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.
Source
Published
The article discusses various tech news updates, including educational grants, the Bambu X1, and more. Unfortunately, due to the technical code included in the content, it is not possible to provide a concise summary of the key points.
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.
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.
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.
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.