正文 · AI 翻译
Thanks to IT Home reader 华南吴彦祖 for the tip!
IT Home reported on November 11 that Arm issued an announcement yesterday (November 10) introducing the Cortex-M42 processor and adding it to its Flexible Access and Total Access technology licensing programs.
In terms of positioning, the Cortex-M42 processor is aimed primarily at smart devices sensitive to cost and chip area, with the goal of bringing stronger machine learning and digital signal processing capabilities to products such as wearables and industrial controllers.
In terms of architecture, the Cortex-M42 processor uses the Armv8.1-M architecture, combining Helium technology with a compact AHB bus design to make it easier for chipmakers to upgrade existing Cortex-M4 and Cortex-M33 microcontroller designs.
Helium is a set of vector extensions for Cortex-M processors that lets the same CPU run control code and process data operations suited to vectorization more efficiently.
In terms of performance, compared with the Cortex-M33, the new Cortex-M42 processor offers up to 3 times the vector computing performance and up to 2 times the vector computing energy efficiency. IT Home notes: Vector computing means using one processor instruction to perform the same operation on a group of data, primarily improving data processing efficiency through SIMD (single instruction, multiple data). SIMD places multiple data elements in a larger register; each element’s position is called a “lane,” and the same operation is then performed on multiple lanes.
In practical applications, the new processor can be used for audio processing and keyword recognition in wearables, as well as sensor analysis and condition monitoring in industrial equipment. Arm also said, by way of example, that it can process signals near robot motors and joints to help detect anomalies or signs of wear.
The launch also expands Arm’s technology licensing portfolio. As of November 2026, Flexible Access had 319 active members and had enabled more than 500 independent tape-out projects; Total Access had 60 partners and had supported more than 600 projects in total. Arm hopes to encourage more devices to perform intelligent processing locally by expanding access to its technology.