Semiconductors news
Subscribe via RSS ↗69 relevant articles · classified by Haiku 4.5 · ingested daily
2d·Launch·positiveIntel Xeon 6+ Computex roundtable interview transcript — Kira Boyko and Tim Wilson on 18A wafer allocation, Clearwater Forest, and dropping hyper-threading
Intel launches Xeon 6+ processors at Computex with 18A node support and hyper-threading removal.
Tom's Hardware ↗
2d·Other·positiveSK hynix to double memory wafer capacity within five years, chairman says — AI-driven shortage will persist until at least 2030
SK Hynix plans to double memory wafer capacity within five years to address AI-driven demand extending through 2030.
Tom's Hardware ↗
2d·Launch·positiveJensen Huang says Nvidia wants to 'reinvent the single most important tool of humanity' with RTX Spark — Nvidia CEO touts support of 'literally every computer maker in the world' for its agentic AI PC platform
Nvidia CEO Jensen Huang announces RTX Spark, an agentic AI PC platform claiming support from major computer makers.
Tom's Hardware ↗
2d·Launch·neutralAMD ‘had to re-engineer’ the Ryzen 7 5800X3D for a re-release — 10th Anniversary Edition chip had ‘a whole body of engineering work’ put into it
AMD re-engineers the Ryzen 7 5800X3D for a 10th Anniversary Edition release after TSMC's original bonding process became unavailable.
Tom's Hardware ↗
2d·Launch·neutralIntel Arc G3 interview transcript — Intel's Senior Product Director talks new handheld chips, Arrow Lake Refresh, and RTX Spark
Intel's Senior Product Director discusses the Arc G3 chip line for handheld gaming, Arrow Lake Refresh, and RTX Spark at Computex 2026.
Tom's Hardware ↗
Subscribe to see the full semiconductors feed
64 more articles, classified against the roster and tagged for source / sentiment / catalyst type.
Already subscribed? Sign in →
2d·Other·neutralIntel says 'something has to give' with memory prices — company says it 'will continue to make sure that there are products which can take care of older memory technologies'
Intel acknowledges persistent memory pricing pressures and commits to supporting legacy DDR4 platforms.
Tom's Hardware ↗
2d·Regulation·negativeChinese military has been acquiring Nvidia chips, even post-Washington export controls, research claims — multiple institutions linked to the PLA asked for Nvidia AI chips, according to publicly available documents
Research alleges Chinese military institutions have acquired Nvidia AI chips in violation of U.S. export controls, citing publicly available procurement documents.
Tom's Hardware ↗
2d·Demo·neutralThe Evolution Of UCIe
Cadence outlines the evolution of UCIe (Universal Chiplet Interconnect Express) as a comprehensive standard for chiplet packaging and multi-foundry integration.
Semiconductor Engineering ↗
2d·Research·neutralChip Industry Technical Paper Roundup: Jun. 2
Micron, Samsung, and AMD contribute to semiconductor research papers on inference scaling, memory reliability, and GPU power management.
Semiconductor Engineering ↗
2d·Demo·neutralIntel Computex 2026 Keynote Live Coverage
Intel CEO Lip-Bu Tan presents the company's AI hardware strategy at Computex 2026 keynote.
ServeTheHome ↗
3d·Demo·neutralMarvell Computex 2026 Keynote Live Coverage
Marvell CEO Matt Murphy addresses Computex 2026 keynote on connectivity solutions for AI data centers.
ServeTheHome ↗
3d·Launch·neutralAMD Radeon RX 9070 GRE review: thoroughly midrange
AMD launches the Radeon RX 9070 GRE at $549, a midrange GPU with strong 1080p and 1440p gaming performance.
Tom's Hardware ↗
3d·Opinion·neutralTSMC Defends Transistor Scaling Amid Huawei’s ‘Her’s Law’ Proposal
TSMC executive defends transistor scaling strategies in response to alternative proposals from competitors.
EE Times ↗
3d·Launch·positiveNVIDIA Computex 2026 News Bytes: Vera Rubin Now In Production, DGX Station Gets Windows
Nvidia announces Vera Rubin platform in full production and brings Windows support to DGX Station systems, available Q4 2026.
ServeTheHome ↗
3d·Launch·positiveComputex 2026 Day Zero Wrap-Up: Nvidia launches RTX Spark Superchip assault on laptop and desktop markets, Intel readies Xeon 6+
Nvidia launches RTX Spark Superchip targeting laptops and desktops; Intel introduces Xeon 6+ at Computex 2026.
Tom's Hardware ↗
3d·Launch·positiveNvidia's RTX Spark could caplitalize where Qualcomm's Arm-based efforts have not — following the expiration of Qualcomm's Windows on Arm deal, Nvidia stands poised to pick up the slack
Nvidia unveils RTX Spark superchip combining a 20-core Grace Arm CPU and Blackwell RTX GPU for Windows PC market.
Tom's Hardware ↗
3d·Launch·positiveQualcomm aims Snapdragon C laptop chip at the budget laptop segment, as manufacturers feel the DRAM squeeze — analysts warn sub $500 laptop market may disappear before 2028
Qualcomm launches Snapdragon C Platform targeting budget laptops as rising DRAM costs compress the sub-$500 segment.
Tom's Hardware ↗
3d·Other·negativeSeven hospitalized after toxic gas fire at SK hynix advanced memory plant — Cheongju 4th campus incident today led to all 3,600 staff being evacuated
A fire in a fluorine-gas room at SK Hynix's Cheongju factory hospitalized seven workers and evacuated 3,600 staff.
Tom's Hardware ↗
3d·Other·neutralEarly Memory Contention Checks Reduce IC Design Risks
Early memory-contention checks catch design risks in complex SoCs and AI accelerators before silicon tape-out.
EE Times ↗
3d·Launch·positiveIntel Xeon 6+ Clearwater Forest is Out
Intel launches Xeon 6900E+ Clearwater Forest with faster memory, new core designs, and expanded cache for data-center workloads.
ServeTheHome ↗
3d·Demo·neutralSupermicro shows off Vera Rubin NVL72 rack with all-new type of coolant — company claims coolant offers 1,000 times higher electrical impedance over standard cooling
Supermicro demonstrates Vera Rubin NVL72 servers with AMD EPYC CPUs and MI450 accelerators in a novel liquid-cooling system.
Tom's Hardware ↗
3d·Launch·positiveNVIDA Introduces RTX Spark: An Arm SoC for Windows PCs
Nvidia unveils RTX Spark, an Arm-based SoC with 20 CPU cores and Blackwell GPU for premium Windows PCs.
ServeTheHome ↗
3d·Regulation·neutralUS closes loophole that allowed Chinese-owned subsidiaries located outside China to buy AI chips — report claims that hundreds of thousands of advanced AI chips have been acquired through BIS blind spot
The U.S. BIS closes a loophole by clarifying that Chinese-owned subsidiaries outside China are subject to AI chip export controls.
Tom's Hardware ↗
3d·Opinion·neutralIntel warns it has 'a healthy dose of paranoia' over Nvidia entrance into PC market — company says RTX Spark is 'great for the market' while touting the virtues of x86
Intel responds cautiously to Nvidia's RTX Spark consumer SoC announcement, says the move is healthy for the market despite competitive concerns.
Tom's Hardware ↗
3d·Launch·positiveAMD promises 13% uplift with new EXPO ‘Ultra Low Latency’ overclocking on DDR5 DIMMs — automatic memory overclocking delivers 4% improvement over standard EXPO, says AMD
AMD announces EXPO 'Ultra Low Latency' automatic DDR5 memory overclocking, delivering a 13% performance uplift over standard speeds.
Tom's Hardware ↗
3d·Demo·neutralWe went hands-on with Qualcomm's new '$300 and up' ARM laptop platform with mystery eight-core CPU — active-cooled Snapdragon C laptop surfaces in Acer Aspire Go 15
Qualcomm's new Snapdragon C ARM laptop platform with an eight-core CPU demonstrated hands-on at Computex 2026, targeting $300+ price point.
Tom's Hardware ↗
3d·Launch·positiveMicrosoft Surface Laptop Ultra weilds Nvidia's RTX Spark superchip with 128GB of RAM, 20 Arm CPU cores, and a Blackwell GPU — 15-inch mini-LED PixelSense Ultra display rounds out the powerful package
Microsoft announces Surface Laptop Ultra powered by Nvidia's RTX Spark Superchip with 20 Arm CPU cores and 6,144 CUDA cores.
Tom's Hardware ↗
3d·Research·neutralThe Sub-2nm Paradox
Sub-2nm semiconductor scaling faces mounting challenges from process variation, RC delay, and memory limitations that diverge from traditional Moore's Law economics.
Semiconductor Engineering ↗
3d·Launch·positiveNvidia lays out RTX Spark roadmap for laptops and desktop PCs at Computex 2026 — three generations outlined, Rubin with LPDDR6 memory, followed by Rosa Feynman
Nvidia unveils RTX Spark roadmap for consumer GPUs with three planned generations: Rubin, Rosa, and Feynman.
Tom's Hardware ↗- 3d·Launch·positive
Nvidia unveils RTX Spark Superchip for laptops and desktop PCs at Computex 2026 – new platform promises to turn Windows into an agentic AI OS with Arm CPU, Blackwell GPU, and 128GB unified memory
Nvidia unveils RTX Spark Superchip at Computex 2026 — an Arm-based laptop/desktop platform with Blackwell GPU and 128GB unified memory for on-device AI.
Tom's Hardware ↗ 
6d·Research·neutralBuilding Fixed HW Implementations of Neural Networks (Yale, Cornell et al.)
Researchers propose Physical Foundation Models: fixed-hardware implementations of large neural networks with potential orders-of-magnitude gains in energy efficiency and parameter density.
Semiconductor Engineering ↗
6d·Other·neutralChip Industry Week In Review
Cadence validates third-party IP on Intel Foundry's 18A process node; industry advances in panel packaging, hybrid bonding, and substrates presented at ECTC.
Semiconductor Engineering ↗
6d·Launch·positiveFrom Billions Of Violations To Actionable Insights: Calibre Vision AI
Cadence expands Calibre Vision AI with real-time DRC triage and collaborative closure tools for advanced-node chip design.
Semiconductor Engineering ↗
May 28·Opinion·neutralToward Agentic Verification
Industry analysis: agentic AI is reshaping semiconductor verification workflows, with EDA vendors and design houses facing a pivotal methodological shift.
Semiconductor Engineering ↗
May 27·Opinion·neutralGPUs And RAM Are In Short Supply, But The Real Bottleneck For AI Is Electricians
Infrastructure bottlenecks, particularly electrician shortages, may constrain AI deployment more than hardware availability.
The Next Platform ↗
May 27·Research·positiveJunctionless Transistors Show a New Path to 3D Chips
University of Illinois researchers demonstrate a low-temperature silicon process for monolithic 3D chip fabrication, addressing alignment and density limits in current 3D stacking approaches.
IEEE Spectrum (Semiconductors) ↗
May 25·Launch·neutralPavona Launches Open-Hardware Ecosystem for Secure Chips
GlobalPlatform launches Pavona, an open-source hardware ecosystem with modular designs, reference tooling, and governance for secure chip development.
IEEE Spectrum (Semiconductors) ↗
May 21·Launch·neutralBolt Challenges Nvidia With a Focus on Cutting-Edge Graphics
Bolt Graphics unveils Zeus GPU, a graphics processor targeting creative and scientific workloads where Nvidia's AI-focused designs fall short.
IEEE Spectrum (Semiconductors) ↗
May 19·Research·neutralThe Next 15 Years of Moore’s Law, According to Imec
Imec publishes a 15-year semiconductor roadmap predicting CFET transistor introduction around 2033 and further transitions driven by power reduction.
IEEE Spectrum (Semiconductors) ↗
May 15·Funding·positiveWith Its IPO Done, Cerebras Can Get Back To Pushing The AI Envelope
Cerebras completes its IPO and returns focus to advancing its AI chip architecture and performance.
The Next Platform ↗
May 12·Partnership·positiveOpenAI, Microsoft And Friends Build A Better, More Scalable Ethernet
OpenAI, Microsoft, and others collaborate on next-generation Ethernet networking infrastructure to support large-scale AI cluster scaling.
The Next Platform ↗
May 6·Launch·positiveChip Fab-in-a-Box Could Democratize Semiconductors
InchFab launches a $5M–$15M shipping-container fab system using smaller wafers to lower barriers for chip prototyping and manufacturing.
IEEE Spectrum (Semiconductors) ↗
May 5·Partnership·neutralBroadcom Helps CPU And XPU Makers Go Vertical With Compute
Broadcom helps CPU and XPU makers integrate compute capabilities vertically.
The Next Platform ↗
May 4·Other·neutralMicrosoft Committed To Doubling AI Infrastructure In Two Years
Microsoft commits to doubling its AI infrastructure capacity over two years to support growing demand.
The Next Platform ↗
Apr 27·Other·positiveAI-Driven CPU Shortage Saves Intel’s Financial Cookies
AI-driven CPU demand surge helps Intel shore up financial performance.
The Next Platform ↗
Apr 24·Launch·positiveWith TPU 8, Google Makes GenAI Systems Much Better, Not Just Bigger
Google announces TPU 8, claiming efficiency and performance improvements for generative AI workloads.
The Next Platform ↗
Apr 22·Funding·positiveThe Second Time Will Be The IPO Charm For Cerebras
Cerebras prepares for an IPO, its second attempt at going public after an earlier withdrawal.
The Next Platform ↗
Apr 22·Research·neutralAI Agent Designs a RISC-V CPU Core From Scratch
Verkor.io's agentic AI system designs a RISC-V CPU core autonomously using structured workflows and LLM harnesses.
IEEE Spectrum (Semiconductors) ↗
Apr 20·Other·positiveAI Will Soon Drive A Third Of TSMC’s Business
TSMC's AI-related revenue is expected to grow to roughly one-third of total business as demand for advanced chip manufacturing surges.
The Next Platform ↗
Apr 17·Research·neutralHow HPC And AI Digital Twins Accelerate Quantum Error Correction
HPC and AI digital twins enable simulation-based acceleration of quantum error correction techniques.
The Next Platform ↗
Apr 16·Other·positiveThe Embrace Of AI In Design Transforms Cadence And Its Customers
Cadence reports AI adoption in chip design tools is reshaping the company and its customer base.
The Next Platform ↗
Apr 12·Research·neutralTerahertz Waves Spy on a Chip’s Internal Activity
Researchers at Adelaide University demonstrate terahertz waves can remotely observe transistor activity inside packaged chips without disruption, potentially replacing X-ray and electronic probing methods.
IEEE Spectrum (Semiconductors) ↗
Apr 10·Opinion·neutralFor Enterprises, GPUs Need Virtualization As Much As CPUs Ever Did
Enterprise GPUs require virtualization layers similar to CPU infrastructure to optimize utilization and workload management.
The Next Platform ↗
Apr 7·Other·positiveBroadcom And Google Benefit Mightily From Anthropic’s Meteoric Growth
Broadcom and Google benefit from increased semiconductor and infrastructure demand driven by Anthropic's growth.
The Next Platform ↗
Apr 2·Research·positiveNvidia Software Pushes MLPerf Inference Benchmarks To New Highs
Nvidia's software optimizations achieve top scores in MLPerf inference benchmarks.
The Next Platform ↗
Apr 2·Demo·neutralBroadcom Makes Its Pitch To Run Kubernetes On VMware VCF
Broadcom demonstrates Kubernetes integration capabilities on VMware VCF, positioning its infrastructure software for cloud-native workloads.
The Next Platform ↗
Mar 31·Partnership·positiveThe $2 Billion Nvidia Deal With Marvell Is About A Lot More Than NVLink Fusion
Nvidia and Marvell announce a $2B partnership to integrate NVLink technology and advance interconnect solutions.
The Next Platform ↗
Mar 25·Launch·positiveArm Comes Full Circle With Homegrown, AI-Tuned Server CPU
Arm launches a new server-class CPU architecture optimized for AI workloads and data center deployment.
The Next Platform ↗
Mar 19·Other·neutralStartups Bring Optical Metamaterials to AI Data Centers
Startups including Neurophos use optical metamaterials to build photonic data-center switches for AI workloads, offering improved bandwidth and power efficiency over electronic alternatives.
IEEE Spectrum (Semiconductors) ↗
Mar 18·Research·positiveNvidia’s Always-On Chip Detects Faces in Less Than a Millisecond
Nvidia researchers develop an always-on face-detection SoC consuming under 5 milliwatts with 99% accuracy, demonstrated at ISSCC.
IEEE Spectrum (Semiconductors) ↗
Mar 16·Launch·positiveWith Nvidia Groq 3, the Era of AI Inference Is (Probably) Here
Nvidia announces Vera Rubin inference chips incorporating $20B IP license from Groq, marking the company's entry into inference-optimized processors.
IEEE Spectrum (Semiconductors) ↗
Mar 10·Demo·positiveIntel Demos Chip to Compute With Encrypted Data
Intel demonstrates Heracles chip, accelerating fully homomorphic encryption tasks 5,000x versus standard server CPUs.
IEEE Spectrum (Semiconductors) ↗
Mar 6·Other·positiveCustom Compute Engine Biz Growing More Than Marvell Ever Hoped
Marvell's custom compute engine business is exceeding internal growth projections.
The Next Platform ↗
Mar 5·Opinion·neutralBroadcom May Become The Biggest Counterbalance To Nvidia
Analysis of Broadcom's potential to emerge as a competitive counterweight to Nvidia in AI chip markets.
The Next Platform ↗
Feb 26·Opinion·positiveSo Far, Nobody Turns Tokens Into Money Like Nvidia
Commentary on Nvidia's dominance in converting AI compute demand into revenue, with no competitors matching its market position.
The Next Platform ↗
Feb 25·Demo·positiveSambaNova Pits Its Engineering Against Nvidia For Agentic AI
SambaNova demonstrates engineering optimizations for agentic AI workloads positioned as an alternative to Nvidia's platform.
The Next Platform ↗
Feb 23·Launch·positiveAMD Says “Helios” Racks And MI400 Series GPUs On Track For 2H 2026
AMD confirms Helios rack systems and MI400 series GPUs are on track for delivery in H2 2026.
The Next Platform ↗
Feb 19·Demo·positiveTaalas Etches AI Models Onto Transistors To Rocket Boost Inference
Etched demonstrates a technique to embed AI models directly into transistor hardware to accelerate inference performance.
The Next Platform ↗
Oct 14·Partnership·positiveOracle First In Line For AMD “Altair” MI450 GPUs, “Helios” Racks
Oracle secures early access to AMD's Altair MI450 GPUs and Helios rack systems for AI infrastructure.
The Next Platform ↗