Category: Processors
-
Rambus has launched new HBM4E controller IP for accelerator designs. The block targets AI and HPC devices that need much higher memory bandwidth, with early-access licensing now open.
-
STMicroelectronics has started shipping China-made STM32 microcontrollers to customers now. Production begins with the STM32H7 family and is set to expand to STM32H5 and STM32C5 devices.
-
Infineon is packing more current into the AI power stage. Its latest TLVR module targets the growing board-level constraints inside AI servers, where power density, transient response, and thermal headroom are becoming central design limits.
-
RANsemi has secured a tactical 5G baseband design win now. Its RNS802 SoC sits at the heart of Apeiroon’s deployable 4G and 5G systems for defence, public safety, and temporary private networks.
-
AI hardware efficiency is pushing materials science back into focus. Cambridge researchers have developed a hafnium oxide memristor that could cut energy use in analogue in-memory computing.
-
Programmable photonics is edging closer to practical semiconductor deployment. Lumotive’s new 2D beamforming chip points to flatter optical architectures for sensing, communications, and AI data-centre networking.
-
Camera intelligence is moving on-device as edge silicon gets leaner. Hailo is using ISC West 2026 to push low-memory, low-power AI vision deeper into mass-market security designs.
-
Mythic is hardening analog AI around foundry-ready embedded memory technology. Its next APU generation will use SST’s memBrain and SuperFlash to push compute-in-memory efficiency further into deployable silicon.
-
SolidRun has pushed edge AI deeper into embedded compute. Its new COM Express Type 6 modules pair AMD’s Ryzen AI Embedded P100 with LP-CAMM2 memory and up to 59 TOPS for harsh-environment deployments.
-
Texas Instruments has launched two MCU families with integrated TinyEngine neural processing units, targeting low-cost edge AI inference and intelligent real-time motor-control applications.