Nvidia Can Invest $100 Billion Per Year to Grow a Huge AI Ecosystem

Nvidia’s $100 billion commitment to OpenAI over a few years is one of several circular deals Nvidia is making. Upstream chip sales fuel downstream infrastructure bets, creating a self-reinforcing loop that has propelled Nvidia’s market cap to $4.6 trillion. Nvidia Q2 operating cash flow hit $15.4 billion and a growing $56.8 billion war chest. Projections …

Read more

Analog in-memory Computing Attention Mechanism for Fast and Energy-efficient Large Language Models

A Nature paper describes an innovative analog in-memory computing (IMC) architecture tailored for the attention mechanism in large language models (LLMs). They want to drastically reduce latency and energy consumption during inference. The design leverages gain-cell crossbar arrays—capacitor-based memory devices made from oxide semiconductor field-effect transistors (IGZO or ITO)—to store key (K) and value (V) …

Read more

Samsung Versus TSMC Versus Intel

TSMC 3nm process node is the best FinFET technology and TSMC dominates semiconductor chip fabrication with higher transistor density, better yields on a mature technology and broad adoption by clients like Apple and NVIDIA. TSMC’s 2nm Process Scale-Up TSMC’s 2nm (N2) process will introduce nanosheet transistors (a form of Gate-All-Around or GAA architecture). It is …

Read more

TSMC 2025 Technical Symposium Has Roadmap to 2028 and A14 (1.4 nanometer) Node

TSMC’s roadmap has been extended to 2028, with new technologies like N3C and A14 being introduced. TSMC presented the new roadmap at the 2025 Technical Symposium. TSMC is forecasting building 2.5 million chips for humanoid robots and about 2.5 million chips for robotaxi around 2030. TSMC projects that humanoid robots and robotaxi will reach reach …

Read more

Growing Size of AI Chips forces TSMC to Work on Tripling the Size of Wafers

TSMC (Taiwan semiconductor) has a major multi-year plan to create rectangular substrate (510 millimeters by 515 millimeters) with over triple the usable area of current round wafers. The rectangular shape means there would also be less unused area left over at the edges. this means less waste because of the rectangular shape and lower costs …

Read more

TSMC System on Wafer for Over 3.5 Times the Compute by 2027

TSMC introduced its System-on-Wafer (TSMC-SoW™) technology, an innovative solution to bring revolutionary performance to the wafer level in addressing the future AI requirements for hyperscaler datacenters. At the TSMC 2024 North America Technology Symposium, they debuted the TSMC A16™ technology, featuring leading nanosheet transistors with innovative backside power rail solution for production in 2026, bringing …

Read more

Why is There a Shortage of Nvidia AI Chips?

TSMC’s (Taiwan Semiconductor) 2.5D advanced packaging CoWoS (Chip on wafer and wafer on substrate) technology is currently the primary technology used for AI chips. The production capacity of CoWoS packaging technology is a major bottleneck in AI chip output and will stay as a problem for AI chip supply in 2024. Nvidia H100 (AI chips) …

Read more

High Bandwidth Memory Will Stack on AI Chips Starting Around 2026 With HBM4

SK Hynix and Taiwan’s TSMC have established an ‘AI Semiconductor Alliance’. SK Hynix has emerged as a strong player in the high-bandwidth memory (HBM) market due to the generative artificial intelligence (AI) craze. TSMC is the world’s largest semiconductor foundry (consignment production) company. The strategy is to consolidate the victory of the two companies in …

Read more

Samsung Targets Mass Production of 2nm by 2025 and 1.4 nm by 2027

Samsung has a new roadmap that targets mass production of 2nm process technology by 2025 and 1.4nm by 2027. TSMC is planning to start production of 2-nm chips in 2025. Intel Foundry Services (IFS) plans to offer its own GAA-based process at its 18-A node that’s equivalent to 2 nm by 2024. Samsung plans to …

Read more