As market demands continue to escalate, particularly with the rise of edge AI applications, the need for higher processing capabilities in embedded systems grows. The evolution of MCUs from 8-bit to 16-bit to 32-bit highlights this trend, with performance steadily improving and the lines between MCUs and MPUs becoming increasingly blurred to accommodate new applications. Microchip, a leader in embedded solutions, has recently introduced the PIC64GX MPU, marking its foray into the 64-bit MPU market.
The PIC64GX MPU is the first product in Microchip's PIC64 series. It addresses the needs of compute-intensive real-time applications such as intelligent embedded vision and machine learning, while also supporting industries like industrial, automotive, communication, IoT, aerospace, and defense. This 64-bit MPU significantly extends Microchip's product portfolio, offering a comprehensive range of computing solutions from low-end to high-end.
The PIC64GX MPU features a 64-bit RISC-V quad-core processor with asymmetric multiprocessing (AMP) and deterministic latency. This architecture allows Linux, real-time operating systems, and bare-metal applications to run simultaneously on a single processor cluster, meeting the demand for asymmetric processing in intelligent edge applications. As the first RISC-V multi-core solution with AMP capabilities, the PIC64GX MPU is designed to support mixed-criticality systems robustly.
One of the standout features of the PIC64GX MPU is its quad-core design, including a CPU cluster supporting Linux, a fifth microcontroller-level monitor, and 2MB of flexible L2 cache, operating at frequencies up to 625 MHz. These characteristics ensure high processing performance, system reliability, and real-time capabilities.
Notably, the PIC64GX series is pin-compatible with PolarFire SoC FPGA devices, providing significant flexibility for embedded solution development. The PIC64GX series is also supported by the PIC64GX Curiosity evaluation kit and integrates with Microchip's MPLAB Extensions for VS Code, simplifying the development process. Additionally, the product is supported by Linux4Microchip resources, Linux distribution vendors, and the Zephyr® RTOS and related software stacks.
Microchip has not only made strides with the PIC64GX MPU but also plans to introduce the PIC64 High-Performance Spaceflight Computing (PIC64-HPSC) series. These space-grade 64-bit multi-core RISC-V MPUs will offer enhanced computational performance and radiation tolerance for aerospace and defense applications. The HPSC processor development, selected by NASA's Jet Propulsion Laboratory (NASA-JPL), heralds a new era of autonomous space computing.
Traditionally, MCUs (Microcontroller Units) and MPUs (Microprocessor Units) have distinct differences. MCUs are highly integrated, including a CPU, RAM, ROM, I/O interfaces, and other peripherals, making them suitable for low-cost embedded systems running simple RTOS. Conversely, MPUs generally lack integrated memory but boast extensive peripheral resources and require external memory, offering higher computational power for complex tasks and operating systems like Linux.
However, due to advances in manufacturing processes and integration needs, the distinctions between MCUs and MPUs are diminishing. For example, NXP's i.MX RT 1050, introduced in 2017, featured significant RAM and enhanced CPU performance, bridging the gap between MCU and MPU functionalities. Recently, NXP unveiled the S32N55 processor, fabricated using a 5nm process. This processor, though termed a processor, integrates 48MB of SRAM, closely resembling an MCU. It features 16 Arm Cortex-R52 cores, operating at up to 1.2GHz, aimed at high-performance automotive applications.
High-performance MPUs and MCUs are poised to provide more efficient, secure, and reliable support for future intelligent edge computing. With the widespread adoption of 64-bit technology, the embedded field will witness innovative applications and advancements.
Hot News
