Learn how Korean AI technology company Rebellions uses proprietary dataflow NPUs, chiplets, and HBM to deliver high-efficiency, scalable AI inference for modern…
Learn how Korean AI technology company Rebellions uses proprietary dataflow NPUs, chiplets, and HBM to deliver high-efficiency, scalable AI inference for modern data centers.
Released in June 1978 after a three-month architecture sprint, the 8086 was meant to buy Intel time while it finished its…
Released in June 1978 after a three-month architecture sprint, the 8086 was meant to buy Intel time while it finished its real next-gen processor. But the real one failed, and the stopgap built x86.
Integration and efficiency are reshaping modern system design, from rad-hard ICs in NASA’s Artemis II mission to SoCs…
Integration and efficiency are reshaping modern system design, from rad-hard ICs in NASA’s Artemis II mission to SoCs turning ID badges into wireless systems.
GaN-based power delivery board targets 96.5% efficiency and 2,100 W/in3 for NVIDIA's next-generation 800 VDC data center…
GaN-based power delivery board targets 96.5% efficiency and 2,100 W/in3 for NVIDIA's next-generation 800 VDC data center architecture.
Learn how the EU's premier lending institution is deploying billions in loans, equity, and blended finance to build a…
Learn how the EU's premier lending institution is deploying billions in loans, equity, and blended finance to build a sovereign European chip ecosystem.
In a landmark moment in its 35-year history, Arm has debuted its own silicon product, the AGI CPU, built on the Arm…
In a landmark moment in its 35-year history, Arm has debuted its own silicon product, the AGI CPU, built on the Arm Neoverse platform.
The MAI device from Dimension NXG runs AI models continuously on a 180-mAh battery for up to 14 days, using analog…
The MAI device from Dimension NXG runs AI models continuously on a 180-mAh battery for up to 14 days, using analog in-memory compute to keep power consumption at microwatt levels.
NXP claims that this device represents the first time an applications processor has combined a dedicated AI NPU with…
NXP claims that this device represents the first time an applications processor has combined a dedicated AI NPU with tri-radio wireless connectivity.
Six months after acquiring Arduino, Qualcomm goes big by adding the Dragonwing IQ-8275 AI processor to Arduino's platform.
Six months after acquiring Arduino, Qualcomm goes big by adding the Dragonwing IQ-8275 AI processor to Arduino's platform.
Announced at Embedded World, the new Genio lineup spans value to flagship tiers, with on-device GenAI acceleration…
Announced at Embedded World, the new Genio lineup spans value to flagship tiers, with on-device GenAI acceleration ranging from 6 TOPS to over 50 TOPS.
This article explores the operation of the simplest digital phase detector: the exclusive-OR (XOR) gate.
This article explores the operation of the simplest digital phase detector: the exclusive-OR (XOR) gate.
Qualcomm’s AI-native Wi-Fi 8 platforms, R19-ready 5G modems, and custom Oryon CPUs set the company up for the next…
Qualcomm’s AI-native Wi-Fi 8 platforms, R19-ready 5G modems, and custom Oryon CPUs set the company up for the next generation of mobile and networking infrastructure.
Announced today at Embedded World, the new Core Series 2 may accelerate the industrial edge and real-time applications,…
Announced today at Embedded World, the new Core Series 2 may accelerate the industrial edge and real-time applications, while the healthcare suite may help providers meet the challenges of patient care using edge AI.
In this exclusive interview, the Seattle startup says its software-ML-hardware co-design—announced today—converts…
In this exclusive interview, the Seattle startup says its software-ML-hardware co-design—announced today—converts off-the-shelf FPGA servers into high-efficiency AI inference engines.
New research explores ultra-low-k COF dielectrics, screens topological conductors for nanoscale wires, and reveals…
New research explores ultra-low-k COF dielectrics, screens topological conductors for nanoscale wires, and reveals high-pressure hexagonal GeSn alloys.
Singulation (wafer dicing) is critical for IC wafer manufacturing. Laser/plasma dicing is replacing mechanical methods.…
Singulation (wafer dicing) is critical for IC wafer manufacturing. Laser/plasma dicing is replacing mechanical methods. Lidrotec's laser-liquid technology offers virtually zero-damage cuts.
The company devised the new Maia 200 inference accelerator to improve cost and performance for AI inference processing in…
The company devised the new Maia 200 inference accelerator to improve cost and performance for AI inference processing in Azure Cloud Services.
ST says its new microprocessors balance application-class performance, real-time control, and power management.
ST says its new microprocessors balance application-class performance, real-time control, and power management.
The most recent addition to the MIPS Atlas family of RISC-V processors supports transformer- and agentic-language AI…
The most recent addition to the MIPS Atlas family of RISC-V processors supports transformer- and agentic-language AI models at the edge.
While brain-inspired computing and neural interfaces have long promised transformative advances, scaling those systems…
While brain-inspired computing and neural interfaces have long promised transformative advances, scaling those systems has remained a challenge. New research suggests that the barrier is beginning to fall.