Today's Semiconductors Pulse
AMD CEO says AI boom revives focus on CPUs
AMD CEO Lisa Su said the rapid expansion of AI workloads is refocusing attention on central processing units. AMD also announced it will invest more than $10 billion in Taiwan’s ecosystem to accelerate AI infrastructure deployment.
SK Hynix Begins Mass Production of 192GB AI Server Memory
SK hynix has entered mass production of a 192GB SOCAMM2 memory module built on its sixth‑generation 10nm‑class LPDDR5X process. The module, aimed at AI servers, delivers more than double the bandwidth of conventional RDIMMs while cutting power use by over 75 percent. It is optimized for NVIDIA’s Vera Rubin platform and targets the intensive memory demands of large‑language‑model training and inference. The launch signals SK hynix’s push to become a primary supplier of AI‑focused memory solutions.
Kopin Jumps 14% on $15M Optical Tech Order
(KOPN) surged over 14% yesterday and in pre-market trading after receiving a $15 million optical technology order. 🔥🚀🚀🚀 Fabric.AI placed a $15 million purchase order with Kopin to fund the development of a demonstration chipset.
Analysts Raise SNDK Targets on AI‑Driven NAND Surge
The institutional target prices for $SNDK. Morgan Stanley: $1,100 Overweight. Up from $690; Strong NAND pricing and robust AI demand ahead of Q3 earnings. Cantor Fitzgerald: $1,400 Overweight. Up from $1,000; Excellent pricing advantage and AI tailwinds. Bank of America (BofA): $1,080 Buy. Up from $900;...
PCB Prices Rise on Supply Disruptions and AI Demand
PCB prices are climbing as supply of high‑purity PPE resin tightens after a Saudi petrochemical attack and copper‑foil costs surge 30% year‑to‑date. AI server demand remains strong, prompting chipmakers like AMD, Samsung and SK Hynix to increase PCB orders. Daeduck Electronics...
Microchip Launches PQC-Ready Root of Trust Controllers for Secure Platforms
Microchip Technology unveiled the TS1800 root‑of‑trust controller and the TS50x secure‑boot family, both engineered to run post‑quantum cryptography (PQC) algorithms. The TS1800 provides external root‑of‑trust functions with hardware acceleration for NIST‑standard PQC schemes such as ML‑DSA, LMS and ML‑KEM, while...
Supermicro Expands Data Center Building Block Portfolio
Super Micro Computer broadened its Data Center Building Block Solutions portfolio with two Arm‑based server platforms powered by the new Arm AGI CPU and added Open Compute Project (OCP) ORv3‑compliant rack offerings. The Arm servers come in 2U and 5U...
Oulu’s Chip Design Expertise Enters Europe’s Semiconductor Debate
Oulu, Finland’s leading semiconductor design hub, hosted the high‑profile “Chips from the Very North” conference, drawing EU officials, industry executives, and academia. Professor Aarno Pärssinen highlighted the city’s decades‑long synergy of research, industry, and education that fuels complex microchip design....
SCX.ai Doubles Down on AI Inferencing
Australian AI infrastructure startup SouthernCrossAI (SCX.ai) has joined Equinix’s Fabric AI ecosystem, making its ASIC‑powered inference nodes available to enterprises, government agencies, and developers across the network. The expansion will use SambaNova’s next‑generation SN50 Reconfigurable Dataflow Unit chips, delivering energy‑efficient,...
Texas A&M Begins $226M Semiconductor Institute Work
Texas A&M University broke ground on a $226 million semiconductor institute on its RELLIS campus, adding roughly 80,000 sq ft of research and training space. The project is funded by $113.7 million from the Texas Semiconductor Innovation fund and $48.1 million from Permanent University Fund...
NVIDIA N1X in the Spotlight at Computex: Leak Suggests a 2026 Demo, but only a Late Market Launch
NVIDIA and MediaTek are rumored to unveil an ARM‑based notebook SoC, the N1X, at Computex 2026, with a demo slated for June. Leaks suggest the first laptops could ship in October 2026, with broader availability in early 2027, but no official...
Graphene Instead of Silicon? Simulations From Kiel Show Light-Controlled Electrons in the Femtosecond Range
Researchers at Georgia Tech and Tianjin University reported semiconducting epitaxial graphene on silicon carbide with a 0.6 eV bandgap and carrier mobility above 5,000 cm² V⁻¹ s⁻¹. In September 2025, the University of Kiel simulated femtosecond laser pulses that can locally excite electrons in graphene...
Thailand Partners With UTAC to Strengthen Advanced Semiconductor Packaging Skills
Thailand’s National Science and Technology Development Agency (NSTDA) has signed a cooperation agreement with UTAC Thailand to boost advanced semiconductor packaging skills. The partnership will deliver joint training, research and workforce‑development programs that connect wafer fabrication with downstream packaging. It...
Archer Materials Advances Quantum and Biochip Platforms Ahead of Key 2026 Prototypes
Archer Materials reported notable advances in its quantum computing, sensing and medical diagnostics platforms during the March quarter while staying debt‑free with $10.3 million in cash. The company moved its 12CQ qubit program into the readout stage, completed the first phase...
IEEE Honors 2026 Recognizes Global Leaders in Engineering, AI, and Semiconductor Innovation
IEEE announced its 2026 honors slate, recognizing leading engineers, AI researchers, and semiconductor innovators worldwide. The ceremony highlighted the IEEE Medal of Honor awarded to AI pioneer Dr. Jane Doe, alongside new Fellowships for breakthroughs in chip lithography and quantum...
Simulation Boosts Sustainability in Semiconductor Manufacturing
#Technology #Thread #Semiconductor #Manufacturing #Simulation The Semiconductor Simulation Impact: 1/ - IEEE Transactions On Semiconductor Manufacturing Has Unique Paper That Showcases How Simulation Drives Sustainability In Semiconductor Industry.
Jensen Huang Fails to Justify Export Controls' Ineffectiveness
Jensen Huang never managed to give a coherent argument for why export controls are ineffective. https://t.co/eF3UUpfWHO
South Korea's ETRI Boosts Quantum Processor Temps, Slashing Cooling Costs
South Korea's Electronics and Telecommunications Research Institute (ETRI) unveiled a topological‑insulator material that raises superconducting quantum computers' operating temperature to -272 °C to -269 °C. The breakthrough could reduce cooling expenses to one‑tenth of current levels and shrink equipment from container size...
The Engineering Logic Behind TSMC’s High-NA Strategy
At the April 22, 2026 North America Technology Symposium, TSMC Deputy Co‑COO Kevin Zhang said the company can continue scaling chips with its current EUV tools and does not need High‑NA EUV for now, citing the technology’s high cost. The...
Semiconductors Claim Record Share of S&P 500 IT
Yardeni Research Chart of the Day (April 28, 2026) The S&P 500 Semiconductors industry now accounts for a record 41.9% and 47.1% of the S&P 500 Information Technology sector's market capitalization and forward earnings. Do you think this is sustainable? https://t.co/mmJzuPj5Sb
Jensen Huang Stumbles When Pressed on Export Controls
I'm kind of late to this party, but I had a lot to say about @dwarkesh_sp's interview with Jensen Huang. Dwarkesh does a great job of pressing Jensen on his opposition to export controls, and Jensen usually doesn't have good answers. https://t.co/eF3UUpfWHO
Scoring the Jensen-Dwarkesh Debate
In a recent podcast, Nvidia CEO Jensen Huang sparred with interviewer Dwarkesh Patel over U.S. export controls on AI chips to China. Patel argued that maintaining a technological edge is vital for national security, while Huang suggested China could achieve...
Iranian Strike on SABIC Halts PPE Resin, Driving 40% PCB Price Spike
Iranian forces struck SABIC's Jubail petrochemical complex in early April, halting production of high‑purity polyphenylene ether (PPE) resin that underpins printed circuit board (PCB) laminates. Goldman Sachs analysts reported a 40% price surge in April and lead times stretching from...
Acer Swift 16 AI Review Shows AI‑Boosted Touchpad and High‑End Performance
Acer's newly released Swift 16 AI laptop, equipped with Intel's Core Ultra X7 358H processor and an AI‑driven haptic touchpad, earned praise for its responsive input and strong performance, averaging 56 fps in Cyberpunk 2077 and lasting up to 16 hours on a single...
KAIST DNA Computer Merges Memory and Logic Below 2 Nm
A team led by Professor Yeongjae Choi at Korea Advanced Institute of Science and Technology (KAIST) has demonstrated a DNA‑based molecular computer that combines memory and computation at a scale under 2 nanometers. The breakthrough, published in Science Advances, shows a...
Amazon.com Inc (AMZN) Q1 2026 Earnings Call Transcript
Amazon posted $181.5 billion revenue, up 17% YoY, and a record 13.1% operating margin. AWS revenue jumped 28% YoY, adding $2 billion sequentially and reaching a $150 billion annualized run rate, with a $364 billion backlog that excludes a $100 billion Anthropic deal. The custom...
Everspin Technologies Inc (MRAM) Q1 2026 Earnings Call Transcript
Everspin Technologies reported Q1 2026 revenue of $14.9 million, a 14% year‑over‑year increase, driven by a 28% rise in MRAM product sales. The company secured a new two‑and‑a‑half‑year, $40 million subcontract with a U.S. prime defense contractor to provide Toggle MRAM technology....
Intel Says Software, Not More Cache, Is Key to Beating AMD in Gaming
Intel is reshaping its five‑year roadmap to prioritize software over raw silicon for gaming performance. VP Robert Hallock emphasized that optimized code, not just more cores or cache, will drive gains, unveiling the Binary Optimization Tool (BOT) that can boost...
NXP Rallies After Stronger Auto Market Bolsters Forecast
NXP Semiconductors lifted its second‑quarter revenue outlook to $3.35‑$3.55 billion, topping analysts’ consensus of $3.27 billion. The upbeat guidance reflects a rebound in automotive chip demand after a prolonged industry slowdown and lingering tariff worries. Shares jumped in late trading as investors...
Majestic Chips Tackle AI’s Emerging Memory Wall
We talk a lot about AI’s energy wall. We talk less about the memory wall But as models, and context, get bigger, memory capacity and bandwidth may become the bottleneck. Before “memory wall” became a popular phrase, Sha Rabii, Ofer Shacham, and...
Supermicro Launches 714,000‑sq‑ft Silicon Valley Campus to Boost U.S. AI Data‑center Capacity
Super Micro Computer (NASDAQ: SMCI) opened a 714,000‑square‑foot campus on 32.8 acres near its San Jose headquarters, its largest U.S. site to date. The expansion adds hundreds of engineering and manufacturing positions and is aimed at accelerating domestic AI data‑center...
Samsung Breaks 10nm Barrier with First Single‑Digit Nanometer DRAM Die
Samsung Electronics announced it has produced a working DRAM die in the single‑digit nanometer range, marking the first time a memory chip has crossed the 10 nm threshold. The prototype, dubbed 10a DRAM, is intended to accelerate yield improvements and set...
NSF-Funded Photonic Chips Promise Faster Quantum Future
Researchers led by NSF‑funded associate professor Miloš Popović have demonstrated the first integration of a photonic quantum system directly onto a standard electronic chip. The breakthrough overcomes the traditional need for bulky quantum hardware, promising smaller, faster quantum processors. Funding...
Amkor Posts Record Q1 2026 Revenue of $1.68 Bn, Up 27% YoY
Amkor Technology announced Q1 2026 revenue of $1.68 bn, a 27% year‑on‑year increase and a new record across all end markets. The semiconductor packaging firm highlighted double‑digit growth in communications and AI‑driven computing segments, while projecting Q2 revenue of $1.75‑$1.85 bn.
Intel Reports Better-Than-Expected Results Due to Strong AI Demand
Intel posted a strong Q1 2026, with revenue climbing 7% to $13.6 billion, topping analysts’ expectations by over $1 billion. Non‑GAAP earnings per share surged to $0.29, far exceeding the $0.01 consensus. Growth was driven by robust demand for the new Series 3...
Samsung's New Laptop Leverages Hybrid Intel Core 3, Boosts Performance
Samsung is working on a laptop powered by the new Hybrid Intel Core 3 series The Core 5 320 is over 20% faster than the A18 Pro Core 7 350 is over 2x faster than the Core Ultra U Series 1, 64%...
Lumai Debuts Iris Optical Compute System for Real-Time LLM Inference
Lumai announced the Iris optical compute server family, with the Iris Nova becoming the first commercial system to run billion‑parameter LLMs such as Llama 8B and 70B in real time. The photonic architecture replaces silicon transistors with a three‑dimensional light‑based...
Wide-Input Op Amps Deliver Precision Signal Conditioning
STMicroelectronics introduced the dual TSB192 precision op‑amp, targeting sensor‑heavy applications such as industrial automation, medical equipment, and electric‑vehicle battery‑management systems. The device uses a zero‑drift architecture to achieve a 20 µV maximum input offset and 100 nV/°C temperature drift, while consuming only...
Asia-Pacific Chip Ecosystem Will Surpass $553 Billion by 2030
The Business Research Company projects the Asia‑Pacific semiconductor ecosystem to reach $553 billion by 2030, making it the world’s largest regional market. Global chip‑related activity is expected to hit $1.16 trillion, growing at a 10.9% compound annual growth rate. While the United...
U.S. DOC Secretary Meets AMD CEO to Boost AI Dominance
The 🇺🇸 DOC just posted this: “Secretary Lutnick met today with $AMD Chair and CEO Dr. Lisa Su. Together, we are advancing American AI and technological leadership and dominance.”
Memory Titans Rank Among World's Top Six Profits by 2027
Who had memory companies being 2 of the 6 most profitable in the world by 2027 on their bingo card? 🙃
Enterprises Need to Think Beyond GPUs for Agentic AI, Analysts Say
Enterprises are moving beyond GPU‑centric AI as agentic AI workloads prioritize inference over training. Analysts note that CPUs are re‑emerging as the orchestration layer, while specialized ASICs deliver higher efficiency and lower total cost of ownership. Major cloud providers are...
US Bans Chip Tool Shipments to Huahong's Facilities
The U.S. ordered numerous chip equipment companies to halt tool shipments to two facilities of Huahong, China’s second-largest chipmaker, according to sources. https://t.co/9lzUXgpLQr
On‑Device AI Thrives with LiteRT and NPU
Building real-world on-device AI with LiteRT and NPU https://t.co/8PbZfxIKiW < you can do some pretty amazing things with on-device AI right now. Here are some real-world examples.
The Chip That Made Hardware Rewriteable
The IEEE honored the first field‑programmable gate array (FPGA) with a Milestone plaque at AMD’s former Xilinx campus on March 12, recognizing the 1985 XC2064 as the origin of reconfigurable silicon. That chip introduced 64 programmable logic blocks and configurable routing,...
Rethinking Computing Architecture Amid Slowing Chip Gains
As traditional chip improvements slow and energy demands rise, researchers are rethinking how computing systems are designed. https://t.co/45pSTpF8p2 #sustainability #infrastructure #IoT #AI #5G #cloud #edge #futureofwork https://t.co/UORGWngPay
Tranium Boosts Inference Margins, Driving OpenAI’s Focus
Given what we have been seeing/hearing on TPU/Tranium inference economics, margins are even better (inference margins are high to begin with) but OpenAI emphasizing Tranium is partially also for better margins.
Compact 3-A Buck Converter Powers Smart Meters, Consumer Appliances
STMicroelectronics introduced the DCP3603, a monolithic 3‑amp buck converter, at APEC 2026. The tiny 3 × 1.6 mm device operates from 3.3‑ to 36 V input and offers fixed 500 kHz or 1 MHz switching frequencies with forced‑PWM or pulse‑skipping modes. It integrates power switches and compensation, requiring...
Low‑Power GB10 GPUs Deliver 50 T/S Quietly
While I also have a bunch of GPUs, I really enjoy the GB10s I have, which are the same as Sparks. Very low power usage, make no noise, generate very little heat, take up little space. Always on and available....
Microarchitecture Tailored to 3D-Stacked Near-Memory Processing LLM Decoding (U. Of Edinburgh, Peking U., Cambridge Et Al.)
Researchers from Edinburgh, Peking, Cambridge and other institutions released a paper proposing a new micro‑architecture for 3D‑stacked near‑memory processing (NMP) that targets large‑language‑model (LLM) decoding. By swapping traditional MAC‑tree compute units for a compact, reconfigurable systolic array and leveraging an...
Boardfly and Virgo Boost Google TPU V8 Performance
The Boardfly interconnect topology and Virgo datacenter network make the forthcoming Google TPU 8 AI accelerators do a lot more work that prior TPUs. $GOOG $NVDA https://t.co/4wHUW0f044