Amkor says that Glass Substrates, a packaging technology replacement for CoWoS spearheaded by Intel, is set for commercialization within 3 years. Intel-Partner, Amkor, Says Glass Substrates Will See First Commercialization Within Three Years Advanced Packaging is key to any major foundry business as chips are getting more and more complex to meet growing compute and memory demands. TSMC is the single-most important advanced packaging provider in the world thanks to its CoWoS 2.5D technology. Current chip requirements involve integration of HBM and logic chips in a single package & the number of HBM chips is expanding aggressively. Recently, OpenAI showcased […]

Read full article at https://wccftech.com/intel-backed-glass-substrates-tech-will-be-commercilization-ready-within-three-years/

Intel is not only rolling right now, but also operating smartly, driving its revenue up by salvaging CPU dies and selling them off to hungry AI customers. CPUs Demand Is So High Right Now That Intel Got an Unexpected Margin Lift By Selling Salvaged Chips CPU demand is going off the charts as AI inferencing continues to surge with the arrival of Agentic AI. In countless posts, we have stated why CPUs have become so important for AI, a market that was previously dominated by GPUs. Now, AI inferencing is going after CPUs and Memory, in a big way, and […]

Read full article at https://wccftech.com/intel-selling-cpu-dies-it-used-to-throw-in-the-trash-ai-demand-turns-scrap-into-profts/

Taiwan-based manufacturer is reportedly the country's first to offer LPDDR5X memory for NVIDIA Vera Rubin platforms as the green team diversifies its supply chain to power its Agentic AI powerhouse. NVIDIA Diversifies Its Supply Chain, Adding Taiwan-Based Memory Maker For Vera Rubin's LPDDR5X Solution NVIDIA will require a lot of memory, both low-power and high-bandwidth, to fuel the growing needs of Agentic AI with its Vera Rubin platforms. We know that NVIDIA's Vera Rubin makes use of two types of memory. The Vera CPUs use LPDDR5X DRAM while Rubin GPUs use HBM4 DRAM. Both of these have different purposes. HBM4 […]

Read full article at https://wccftech.com/nvidia-taps-taiwanese-nanya-tech-lpddr5x-memory-for-vera-rubin-platform/

TSMC's production lines are being expanded to accommodate more wafer capacity, with a key focus on the new 2nm and 3nm lines. TSMC Focuses Full Efforts To Expand 2nm and 3nm Capacity As Supply In Shock After Huge Demand From AI Firms Recently, AI firms announced that they will be expanding their compute capabilities big time. This massive expansion means that more chips will be required as these are multi-year projects that will address current & future needs. TSMC, being the leading semiconductor maker, has been enjoying this extra demand, but at the same time, the production lines continue to […]

Read full article at https://wccftech.com/tsmc-3nm-2nm-wafer-output-to-be-boosted-by-the-end-of-2026/

TSMC doesn’t need High-NA EUV to complete its current roadmap – A12, A13, and N2U nodes unveiled TSMC has unveiled its newest technology roadmap, which extends until 2029. On it, TSMC has unveiled three new lithography nodes, none of which require High-NA EUV machines. TSMC believes that standard Low-NA EUV machines will be used for […]

The post Take that Intel – TSMC shuns High-NA EUV with its new foundry roadmap appeared first on OC3D.

Leak season for the next Galaxy S lineup has already begun. A new leak sheds light on specs and features of the Snapdragon 8 Elite Gen 6 and 6 Pro that will potentially power the Galaxy S27 Pro and S27 Ultra globally.

Weibo leaker DCS revealed key specs of Snapdragon 8 Elite Gen 6 and Snapdragon 8 Elite Gen 6 Pro, and the chips powering the Galaxy S27 series may be made by TSMC.

TSMC’s 2nm process node means better efficiency, sustained performance, and improved thermal behavior. Qualcomm is also considering a 2nm Snapdragon chip, but it seems unlikely to be from the 6th generation.

Snapdragon 8 Elite Gen 6 (SM8950) configuration:

• 2 prime cores
• 3 performance cores
• 3 efficiency cores

The Pro variant keeps the same layout but pushes it harder. Higher clocks are expected, but Qualcomm is also stacking more resources around it.

The standard chip reportedly uses 16MB L2 + 6MB SLC cache, while the Pro bumps system-level cache to 8MB SLC. It may sound minor, but SLC plays a big role in reducing memory latency during heavy workloads like gaming and AI tasks.

Graphics processing unit:

• Snapdragon 8 Elite Gen 6 – Adreno 845, 6 slices, 12MB graphics cache
• Snapdragon 8 Elite Gen 6 Pro – Adreno 845, up to 18MB graphics cache

Memory support:

• Gen 6 – LPDDR5X
• Gen 6 Pro – LPDDR6 (quad-channel 24-bit) or LPDDR5X (quad-channel 16-bit)

LPDDR6 is not just about speed; it is about bandwidth efficiency and power scaling. Both chips still share the essentials: integrated 5G with sub-6 and mmWave, Wi-Fi 7, Bluetooth 6.0, UFS 5.0 storage, and 10Gbps USB-C.

The Galaxy S27 and S27+ are expected to use Samsung’s in-house Exynos 2700 in most markets. Meanwhile, regions like the US, Canada, and China could get the Snapdragon 8 Elite Gen 6.

Galaxy S27 Pro and S27 Ultra are tipped to use the Snapdragon 8 Elite Gen 6 Pro globally, which at least brings some consistency at the top end.

Samsung may launch the Galaxy S27 series early next year.

The post Snapdragon 8 Elite Gen 6 and Pro leak hints at Galaxy S27 performance roadmap appeared first on Sammy Fans.