Latest News
Geopolitics Is Rewriting Memory Sourcing
SK hynix customers offer to buy its EUV machines and fund new fabs as memory capacity hits zero amid crushing AI-driven shortages — worsening global shortages pry open wallets to the tune of hundreds of millions of dollars
AMD announces MI350P PCIe AI accelerator card with 144GB of HBM3E — roughly 40% faster in FP16 and FP8 theoretical compute compared to Nvidia's H200 NVL competitor
Semiconductor Manufacturing Cost Breakdown | Wafer, Packaging, Test & ASIC Cost Factors
Samsung chip workers reject $340,000 one-time bonus, demand annual payouts like SK hynix's $900,000 — workers want share of AI windfall, impending 18-day strike could cost Samsung up to $11.7 billion
Full Archive
Geopolitics Is Rewriting Memory Sourcing
Geopolitical tensions are rewriting memory sourcing—ditch old habits and future-proof your supply chain to avoid costly disruption. The post Geopolitics Is Rewriting Memory Sourcing appeared first on EE Times.
SK hynix customers offer to buy its EUV machines and fund new fabs as memory capacity hits zero amid crushing AI-driven shortages — worsening global shortages pry open wallets to the tune of hundreds of millions of dollars
The offers go well beyond standard long-term supply agreements.
AMD announces MI350P PCIe AI accelerator card with 144GB of HBM3E — roughly 40% faster in FP16 and FP8 theoretical compute compared to Nvidia's H200 NVL competitor
AMD launches a new MI350P PCIe AI-accelerator card with half the cores and memory of its flagship Instinct MI355X GPU. The new card provides customers with a drop-in upgrade solution for existing air-cooled servers.
Semiconductor Manufacturing Cost Breakdown | Wafer, Packaging, Test & ASIC Cost Factors
Semiconductor Manufacturing Cost Breakdown: Wafer, Packaging, Test, Yield and Production Costs Understanding the semiconductor manufacturing cost breakdown is essential before starting a custom chip, ASIC, or semiconductor production project. The cost of manufacturing a semiconductor device is not only the wafer price. A complete cost model may include design, mask sets, wafer fabrication, wafer
Samsung chip workers reject $340,000 one-time bonus, demand annual payouts like SK hynix's $900,000 — workers want share of AI windfall, impending 18-day strike could cost Samsung up to $11.7 billion
Negotiations between the National Samsung Electronics Union, which represents workers in the company’s chipmaking division, and management have seemingly broken down over a single issue.
Capacity Cuts and Surging Demand for AI Power ICs Set Stage for Mature-Node Foundry Price Increases, Says TrendForce
The global mature-node supply–demand landscape is undergoing a structural shift...
Capacity Cuts and Surging Demand for AI Power ICs Set Stage for Mature-Node Foundry Price Increases, Says TrendForce
The average 8-inch capacity utilization rate among the world’s top 10 foundries is projected to approach 90% in 2026, and remain above 80% through 1H27. Foundries are reallocating 8-inch and 12-inch mature-node capacity toward power-related processes to improve ASP and profitability, benefiting Chinese foundries through order spillover. TSMC’s planned reduction in 12-inch mature-node capacity may further drive order redistribution, creating opportunities for Tier 2 foundries to raise prices.
Apple quietly axes 128GB Mac Studio amid supply constraints and local AI frenzy — highest memory capacity reduced to 96GB, two months after discontinuation of 512GB model
Apple quietly pulls high Unified Memory capacity Mac Studios and Mac minis from the Apple store as the memory crunch continues to affect consumer products.
AMD posts record first-quarter results, driven by skyrocketing data center CPU demand — company expects consumer andgaming revenue to decline in Q2 over rising memory and component costs
AMD sets record data center revenue as demand for CPUs from the AI sector skyrockets and is poised to increase even further as agentic AI gains momentum.
North American AI Data Center Expansion Drives 2026 CapEx of Top Nine CSPs to US$830 Billion, Says TrendForce
TrendForce’s latest findings on the AI industry highlight that several major North American CSPs have recently raised their 2026...
Global semiconductor sales hit nearly $300 billion in Q1 2026 — chips are on track to top $1 trillion for this year, says report
Sales of chips in Q1 2026 hit $298.5 billion and are on track to exceed $1 trillion this year, according to the Semiconductor Industry Association.
North American AI Data Center Expansion Drives 2026 CapEx of Top Nine CSPs to US$830 Billion, Says TrendForce
TrendForce’s latest findings on the AI industry highlight that several major North American CSPs have recently raised their 2026 capital expenditure (CapEx) guidance in response to strong AI demand. As a result, TrendForce has revised its forecast for the combined CapEx of the world’s top nine CSPs—Google, AWS, Meta, Microsoft, Oracle, ByteDance, Tencent, Alibaba, and Baidu—up to approximately US$830 billion in 2026, with the annual growth rate raised from 61% to 79%.
Latest Apex Legends update smooths out stutter on Ryzen X3D CPUs — physics algos apparently struggled at high frame rates unlocked by V-Cache
The Apex Legends Season 29 update reduces stuttering by addressing physics calculations that struggled to keep up at high frame rates enabled by modern high-end CPUs.
A Detailed Evaluation of A Production Server With High-End MRDIMM Main Memory (BSC, Micron, Intel, UPC)
A new technical paper, “Performance and Energy Benefits of MRDIMMs,” was published by researchers at Barcelona Supercomputing Center, Universitat Politecnica de Catalunya, Micron and Intel Corporation. Abstract “Multiplexed Rank DIMMs (MRDIMMs) have recently emerged as memory devices that enable higher bandwidth without increasing DRAM chip frequencies. This paper presents a detailed performance, power and energy... » read more The post A Detailed Evaluation of A Production Server With High-End MRDIMM Main Memory (BSC, Micron, Intel, UPC) appeared first on Semiconductor Engineering.
Replacing GPU Compute Dies With PNM-Enabled HBM Cubes For Long-Context Decode Attention (UCSD, Columbia, Yonsei U., NVIDIA, Samsung)
A new technical paper, “AMMA: A Multi-Chiplet Memory-Centric Architecture for Low-Latency 1M Context Attention Serving,” was published by researchers at UC San Diego, Columbia University, Yonsei University, NVIDIA, and Samsung. Abstract “All current LLM serving systems place the GPU at the center, from production-level attention-FFN disaggregation to NVIDIA’s Rubin GPU-LPU heterogeneous platform. Even academic PIM/PNM... » read more The post Replacing GPU Compute Dies With PNM-Enabled HBM Cubes For Long-Context Decode Attention (UCSD, Columbia, Yonsei U., NVIDIA, Samsung) appeared first on Semiconductor Engineering.
End-to-End ASIC Manufacturing Solutions | From Design to Production
End-to-End ASIC Manufacturing Solutions: From Concept to Production Silicon Developing a custom ASIC is not only a design project. It is a complete semiconductor supply chain project. A successful ASIC program requires architecture, circuit design, verification, layout, tape-out, wafer fabrication, packaging, testing, qualification, yield management, logistics, and long-term production support. This is why
China pushes for 70% homegrown silicon wafer use as domestic firm ramps up 12-inch production — government seeking to localize critical chip supply chain amid AI boom and export restrictions
China is targeting 70% local wafer sourcing as firms like Eswin scale 12-inch production, aiming to reduce reliance on foreign suppliers and support growing AI chip demand.
AI Optical Interconnect Boom Drives U.S. Firms to Expand Southeast Asia Outsourcing, Opening the Door for Cross-Industry Entrants, Says TrendForce
Global shipment volume of optical transceivers is projected to more than triple from 26.5 million units in 2023 to over 92 million units by 2026...
Chip Industry Technical Paper Roundup: May 5
3D-stacked near-memory LLM decoding; In-SoIC ESD protection for chiplets; thermal management for adv. packaging; quantum circuits on chiplets; agentic AI for formal verification; multilayer nanostructure characterization; liquid cooling for adv. packages; NV photonic coupler arrays. The post Chip Industry Technical Paper Roundup: May 5 appeared first on Semiconductor Engineering.
AI Optical Interconnect Boom Drives U.S. Firms to Expand Southeast Asia Outsourcing, Opening the Door for Cross-Industry Entrants, Says TrendForce
Global shipment volume of optical transceivers is projected to more than triple from 26.5 million units in 2023 to over 92 million units by 2026, according to TrendForce’s latest AI infrastructure research. This massive market opportunity—combined with geopolitical factors—is driving a restructuring of the global optical communications supply chain. It is also accelerating outsourcing strategies among U.S. vendors in Southeast Asia, creating an entry point for non-traditional technology players to move into AI optical communications.
News auto-refreshes every 2 hours from TrendForce, Tom's Hardware, SemiWiki, Semiconductor Engineering, and AnySilicon. All articles are permanently archived and searchable.