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.
Digital Twin Semiconductor Supply Chain Market to Reach $7.9 Billion by 2033
The Digital Twin Semiconductor Supply Chain market is forecast to reach $7.9 billion by 2033. Growing demand for end‑to‑end visibility, faster time‑to‑market and resilient operations is driving adoption across the industry. Digital twins provide real‑time simulation, predictive analytics and process automation, allowing manufacturers to anticipate bottlenecks, cut costs and improve quality. Major players such as Siemens, IBM and Microsoft are expanding their twin platforms to capture this expanding opportunity.
SmartRay Sensor Unites ECCO X Innovations with High-Precision Glass Inspection Capabilities
SmartRay unveiled the ECCO X 050G sensor, extending its ECCO X family with high‑resolution inline metrology for glass, reflective and transparent surfaces. The sensor delivers up to 40 kHz scan rates, 2.2–2.9 µm vertical and 11–13 µm lateral resolution, and generates 163 million 3‑D points per second...
Quantum Diamonds Expands to Asia
QuantumDiamonds announced the launch of a new regional hub in Taiwan, marking its first major foothold in Asia. The company appointed Peter Lemmens, a veteran with over 25 years in semiconductor leadership, to spearhead commercial and technical operations across the...
CEA-Leti and Fraunhofer IPMS Validate Wafer Exchange for Ferroelectric Memory Materials
CEA‑Leti and Fraunhofer IPMS have successfully demonstrated a wafer‑exchange pilot line for hafnium‑zirconium‑oxide ferroelectric stacks, proving that complex material stacks can be processed across multiple advanced fabs without contamination. The program used 300 mm CMOS cleanrooms, standardized VPD‑ICP‑MS and TXRF checks,...
Imec Names NVIDIA's Jensen Huang as Recipient of the 2026 Imec Lifetime of Innovation Award
Imec announced that Jensen Huang, founder and CEO of NVIDIA, will receive the 2026 imec Lifetime of Innovation Award. The award honors Huang’s pivotal role in creating the programmable GPU, which has evolved from gaming graphics to the core engine...
Keysight Introduces Hands-On Semiconductor Teaching Labs for Universities
Keysight Technologies unveiled three semiconductor teaching lab solutions—Basic Design and Measurement, Parametric Test and On‑Wafer Measurement, and Photonics IC Measurement—to give university students hands‑on experience with professional‑grade tools. The kits replicate real‑world test workflows, letting students set up equipment, perform...
Nordson Electronics Solutions to Demonstrate Automated Fluid Dispensing and Plasma Treatment Systems
Nordson Electronics Solutions will demonstrate its ASYMTEK Vantage fluid dispensing system with a new Class‑100 cleanroom configuration at Productronica China. The company will also showcase the stand‑alone MARCH FlexTRAK plasma system, which provides high‑throughput surface treatment in a compact footprint....
Metallic Oxide Semiconductor Field Effect Transistor Market to Surpass $15.5 Billion by 2033
The global metallic oxide semiconductor field‑effect transistor (MOSFET) market is projected to exceed $15.5 billion by 2033, driven by surging demand for energy‑efficient consumer electronics, electric vehicles, and renewable‑energy systems. Wide‑bandgap silicon‑carbide and gallium‑nitride technologies are accelerating performance gains, while miniaturization...
Broadcom to Build Custom TPUs for Google AI Racks
Broadcom $AVGO just said it has entered into a long term agreement with Google $GOOGL "for Broadcom to develop and supply custom Tensor Processing Units (“TPUs”) for Google’s future generations of TPUs and a Supply Assurance Agreement for Broadcom to...
Automotive Semiconductor Shifts & AI Workloads | TechInsights AWS Seminar
TechInsights hosted a data‑driven seminar for AWS automotive teams, highlighting a decisive shift from distributed electronic modules to centralized compute architectures in vehicles. The briefing detailed how high‑performance automotive SoCs, expanding memory bandwidth, and silicon‑carbide power devices are reshaping AI...
Hong Kong: Secure Edge AI with Compute-in-Memory Innovation
University of Hong Kong researchers unveiled a Co‑Located Authentication and Processing system that merges compute‑in‑memory with hardware‑based security using memristors. The architecture eliminates the von Neumann bottleneck, delivering a 146‑fold boost in energy efficiency and an 18‑fold reduction in chip area...
European Chips Skills Academy Launches the ECSA Learning Platform
The European Chips Skills Academy (ECSA) has launched the ECSA Learning Platform, a free digital hub offering more than 60 introductory to intermediate semiconductor courses. Developed with industry and academic partners, the platform integrates a Knowledge Hub, community collaboration spaces,...
Semiconductor-Based Quantum Pilot Line ‘SPINS’ Launched with EU Support
The EU has launched the €50 million SPINS pilot line, one of six quantum pilot facilities created under the Chips Act. Coordinated by imec, SPINS unites 25 RTOs, industry players and universities to develop semiconductor‑based spin‑qubit chips. The initiative targets three...
Stitching Precise Patterns – With Lasers
University of Pittsburgh engineers have devised a laser‑induced graphene (LIG) manufacturing method that uses an iron‑oxide ink layer to precisely control graphene formation on polymer films. The technique enables tunable electrode thickness and conductivity, and can create graphene on either...
IVWorks Secures $4.5M Funding to Expand ‘reGaN’ Technology Into RF and AI Power Semiconductor Markets
IVWorks announced a $4.5 million financing round, bringing its total capital raised to $33 million, to accelerate deployment of its proprietary reGaN technology. The company will use the funds to scale mass‑production infrastructure and secure its supply chain for GaN epitaxial wafers...
US Lawmakers Aim to Ban Export of DUV Chipmaking and Etching Tools to Leading Firms in China — Bipartisan Proposal...
A bipartisan group of U.S. senators has introduced the MATCH Act, shifting export controls on advanced wafer‑fabrication equipment from fab‑level to company‑level restrictions. The bill would bar DUV lithography, etching and deposition tools from reaching Chinese firms such as Huawei,...
Why TSMC Grew Four Times Faster than Its Foundry Rivals in 2025 — Price Hikes, Vertical Integration, and Commanding Technology...
The semiconductor foundry market hit a record $320 billion in 2025, with TSMC commanding 38% and growing 36% YoY, far outpacing rivals that collectively rose 8%. Its dominance stems from an unprecedented volume of sub‑5 nm wafers, aggressive price hikes, and a...
Ep 23: Cross-Stack Design and Tooling for Large-Scale Distributed AI Systems with Dr. Tushar Krishna, Georgia Tech
In this episode, Dr. Tushar Krishna discusses his evolution from network‑on‑chip research to designing large‑scale distributed AI systems, emphasizing the importance of cross‑stack co‑design that spans accelerators, memory hierarchies, and interconnect fabrics. He explains how predictable AI data‑flow patterns enable...
Intel "Nova Lake" To Use Xe3 Graphics and Xe3P Display/Media Engine
Intel’s upcoming Nova Lake CPUs, part of the Core Ultra 400 series, will integrate Xe3 graphics and the Xe3P display/media engine rather than the newer Xe4 IP. The platform targets a late‑2026 launch with early‑2027 market availability and will natively support...
IBM, Arm Target Enterprise AI With Mixed-Architecture Approach
IBM and Arm announced a partnership to run Arm‑native applications on IBM Z mainframes and LinuxOne servers through a shared software layer and virtualization. The solution lets enterprises deploy AI workloads across both architectures without rewriting code, preserving the uptime,...
YieldWerx Delivers a Master Class in Co-Packaged Photonics Implementation
YieldWerx announced a forthcoming webinar that will teach semiconductor engineers how to implement co‑packaged photonics (CPO) across the full product lifecycle. The session, led by CEO/CTO Aftkhar Aslam, will detail the 12 cross‑domain challenges—from optical data complexity to test‑flow discontinuities—and...
ACCM Celeritas™ SF1600 Delivers Zero-Skew for PCIe 7, 224 Gbps and Beyond
Advanced Chip & Circuit Materials (ACCM) launched Celeritas SF1600, a laminate and prepreg that eliminates fiber‑weave skew, a root cause of timing errors in 224 Gbps PAM4 and higher data rates. The material delivers zero‑skew performance, a low dielectric constant of 2.80,...
CEA-Leti, CEA-List and PSMC Collaborate to Integrate RISC-V and MicroLED Silicon Photonics Into 3D Stacking and Interposer for...
CEA‑Leti, CEA‑List and Powerchip Semiconductor Manufacturing Corporation (PSMC) announced a partnership to embed CEA‑List’s RISC‑V processor IP and CEA‑Leti’s microLED silicon‑photonic chiplets into PSMC’s 3D‑stacking and interposer platform. The integration replaces traditional copper interconnects with short‑reach, high‑bandwidth optical links, delivering...
Packaging Shift Lets Amazon, Google Embrace Intel Foundry
Pat’s Law of Something: All major vendors and suppliers are always in talks and testing. I do think this will move forward, though. Packaging is an easier commitment than wafers as it doesn’t force a redesign. Not easy but just...
New Model Updates Cost‑Per‑Transistor Economics for AI Era
Working on building a new cost per transistor model with updated economics since it's been a few years since I have seen one. So many more dynamics are needed than the ones from a few years ago, pre AI...
![[OFC 2026] Part 2 of 5: CPO and the AI Interconnect Challenge](/cdn-cgi/image/width=1200,quality=75,format=auto,fit=cover/https://substackcdn.com/image/fetch/$s_!FG3y!,w_256,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F3201360a-f898-4593-8267-8cfc0766cc04_1280x1280.png)
[OFC 2026] Part 2 of 5: CPO and the AI Interconnect Challenge
AI compute clusters are doubling annually, pushing Meta's rack designs from 72 GPUs to over 256 nodes and exceeding 1 MW of power per rack. At these scales, copper backplanes face insurmountable limits in power delivery, bandwidth density, and routing complexity....
[OFC 2026] Part 2 of 5: CPO and the AI Interconnect Challenge
AI training racks are scaling rapidly, with Meta's upcoming ORW rack doubling node count and pushing single‑rack power beyond one megawatt. Copper backplanes now face insurmountable limits in bandwidth, power density, and routing complexity. Meta used OFC 2026 to set...
Vishay Launches Automotive PV MOSFET Driver
Vishay Intertechnology has released the VODA1275, an automotive‑grade photovoltaic MOSFET driver built in a compact SMD‑4 package with an 8 mm creepage distance and a CTI‑600 mold compound. The driver offers 20 V open‑circuit voltage, 20 µA short‑circuit current, and an 80 µs turn‑on...
DigiKey Collaborates with STMicroelectronics and Ultra Librarian on eDesignSuite
DigiKey, STMicroelectronics, and Ultra Librarian have launched an enhanced eDesignSuite, a free browser‑based design platform that unifies simulation, component selection, and BOM purchasing. The tool supports power‑management, signal‑conditioning, and NFC/RFID applications, offering multi‑CAD export and direct integration with DigiKey’s catalog....
Parsing the AI and Gaming Future with Nvidia’s Jensen Huang | GTC Q&A
Nvidia CEO Jensen Huang delivered an unscripted two‑hour keynote at GTC 2026, unveiling more than a dozen AI‑focused gaming announcements. The event in San Jose attracted over 30,000 attendees, underscoring the growing appetite for AI‑driven technology in entertainment. Huang highlighted...
Silicon Choices Grow in Importance as Industrial AI Moves Closer to the Factory Floor
Industrial AI is moving from centralized clouds to the factory floor, where real‑time inference must run continuously alongside machines. This shift is driven by latency, data volume, and security concerns that make edge processing essential. As a result, the choice...
Intel Launches Core Ultra 7 251HX Arrow Lake Processors with 18 CPU Cores and 3 Xe GPU Cores
Intel quietly added the Core Ultra 7 251HX to its Arrow Lake HX family, slotting between the Core Ultra 5 245HX and the Core Ultra 7 255HX. The processor features 18 cores – six Performance and twelve Efficient – with a 30 MB Smart Cache and a...
RISC-V Has Momentum. The Real Question Is Who Can Deliver
RISC‑V has moved from a promising ISA to a viable platform as the RVA23 baseline unifies high‑performance compute. Arm’s recent transition to a silicon‑first model reshapes the IP landscape, intensifying competition. Akeana’s Alpine test chip, taped out in a 4 nm...
EPC’s CEO Analyzes GaN’s Move Into Robotics and Data Centers
Over the past few years GaN HEMTs have moved beyond consumer chargers into high‑performance markets such as AI data‑center servers and humanoid robotics. EPC’s CEO Alex Lidow explained that the material’s low on‑resistance, ultra‑fast switching and high power density make it...
Indium Corporation Announces Strategic Agreement for Domestic Critical Metals Recovery
Indium Corporation has signed a long‑term offtake framework with Flash Metals USA, a Metallium subsidiary, to purchase critical metals recovered from electronic scrap using Flash Joule Heating technology. The agreement covers gallium, germanium, copper, tin, gold and indium, with an...
Japan's Rapidus Ramps up 2nm Chip Plans While Eyeing Factories on the Moon
Japan’s state‑backed chipmaker Rapidus has moved its IIM‑1 plant in Hokkaido from construction to an operating pilot line, delivering working two‑nanometer gate‑all‑around prototypes. The company secured a ¥267.6 billion ($1.7 billion) financing round led by the government and over 30 private partners,...
The KV Cache Wars?
A quiet but critical battle is unfolding in agentic AI infrastructure over the key‑value (KV) cache. The KV cache, which stores key and value projections for every token, scales linearly with context length, layer count, batch size, and heads, consuming...
"Zen 6 Is Done": Intel's Return to Top with Nova Lake Looks Possible with More IPC Uplift vs Zen 6
Intel’s upcoming Nova Lake Core Ultra 400 desktop CPUs, now expected in 2027, are rumored to pack up to 52 cores, higher clock speeds and a notable IPC uplift that could outpace AMD’s Zen 6 chips. Leaker HXL suggests Nova Lake’s P‑core...
14th Gen Core "Raptor Lake" Will Continue to Be "Abundantly Available," Says Intel
Intel confirmed that its 14th‑gen Core “Raptor Lake Refresh” desktop CPUs, the 700‑series chipset, and the LGA1700 socket will stay in production and be “abundantly available.” The company is encouraging motherboard makers to launch boards that support both DDR4 and...
Canadian Listings of Ryzen 9 9950X3D2 Indicate Near-$1000 US Pricing
Canadian retailers ShopRBC and PC‑Canada have posted early listings for AMD's Ryzen 9 9950X3D2 Dual Edition, pricing the chip at CAD $1,374‑$1,375, which converts to roughly USD $990. The listings suggest AMD will set a $999 MSRP when the processor launches on April 22, 2026. The...
$2 WeAct CH32V006F8U6 Mini Core Board Features CH32V006 RISC-V MCU, Supports 3.3V or 5V I/O Voltage
Chinese manufacturer WCH’s CH32V006 RISC‑V microcontroller, launched in 2024, powers the new $2 WeAct CH32V006F8U6 Mini Core development board. The board features a 48 MHz 32‑bit RISC‑V2C core, 8 KB SRAM, 62 KB flash, USB‑C power, and selectable 3.3 V or 5 V I/O via...
Supermicro Owes Its Rapid Rise to $4 Trillion Nvidia—But China Smuggling Allegations and a High-Profile Arrest Could Blow up the...
Supermicro’s rapid growth has been powered by its deep reliance on Nvidia GPUs, which now account for roughly 64% of its component spend and drive about 71% of its revenue. A federal indictment alleges co‑founder Yih‑Shyan “Wally” Liaw smuggled $2.5 billion...
Research Bits: Apr. 6
Researchers at Loughborough University unveiled a nanoporous niobium‑oxide memristor that performs reservoir computing directly in hardware, achieving up to 2,000‑times lower energy consumption than conventional software solutions. The same chip accurately forecasted short‑term Lorenz‑63 chaos, recognized pixelated digits and executed...
NXP Introduces Omlox Starter Kit for Industrial RTLS Deployment
NXP Semiconductors has launched the omlox Starter Kit, a turnkey solution that merges UWB hardware, software and analytics into a single platform built on the omlox open standard. The kit centers on NXP’s Trimension SR048 UWB SoC and MCX W72...
Intel Pulls the Plug on XeSS Support in Unity Game Engine
Intel abruptly discontinued its official XeSS plugin for the Unity game engine, removing support for frame generation, temporal super‑sampling, and antialiasing. This follows the release of Intel's XeSS 3.0 SDK only a month earlier, which introduced multi‑frame generation and shared...
AI Auto-Optimizes CUDA, Outperforms PyTorch and Torch.compile
People are using AI to break CUDA’s moat. Given any pytorch model, it profiles it, ranks bottlenecks by amdahl's law, writes triton or CUDA C++ replacements, and runs 300+ experiments overnight with no human in the loop. - 5.29x over pytorch eager...
Predicting XPUs: Once Mocked, Now Inevitable GPU‑only World
I remember so many people thinking I was nuts to suggest that years ago that XPUs would be successful and that it would be a GPU-only world.
Global 300mm Fab Equipment Spending Set to Rise, AI Drives Semiconductor Investment
Global spending on 300 mm wafer fab equipment is projected to reach $133 billion in 2026, an 18 % rise. The upward trajectory continues to $151 billion in 2027, driven by surging AI chip demand in data centers and edge devices. Logic and micro‑devices...
China’s CXMT IPO Casts Doubt on Memory Upcycle
China’s CXMT IPO clouds memory upcycle outlook Rising supply risks threaten pricing power of Samsung, SK hynix, Micron Clouds? Shortages everywhere you look.... https://t.co/VfzNQDNgtZ
Murata Completes New MLCC Production Facility in Izumo, Japan
Murata Manufacturing has finished a new 10‑story, 69,829 m² multilayer ceramic capacitor (MLCC) plant in Izumo, Japan, after two years of construction. The project cost about 47 billion yen (≈ $313 million) and expands Murata’s production footprint in the region. The facility complements Murata’s...