Digital & Communication Technology
Taking 64-bit processing into the mass market
By Venki Narayanan, director of marketing for Microchip’s FPGA business unit, Microchip Technology T
here is considerable demand for more performance for the Internet of Things (IoT), but without increasing cost, either in development or deployment. Designers are also looking for a consistent development environment and access to a wider ecosystem of software to speed up the time to market.
The global edge computing market is expected to grow by more than 30 per cent in the next five years to serve an ever-increasing range of mission-critical applications. More data processing, machine learning and edge AI inference is requiring ever higher performance, secure processors. This is another major challenge for developers. The RISC-V open instruction architecture allows for 64-bit innovation with cost, performance and process enhancements to address a wide range of IoT applications. The open nature of the instruction set allows extensions to boost the performance of these applications, especially for real-time computer vision, without adding to the cost. Using RISC-V allows more cost-effective 64-bit silicon and designs that can run full operating systems, such as Linux, and Microchip has leveraged RISC-V technology for the PIC64GX microprocessor (MPU) to support the development of edge AI
The awareness of AMP for mixed criticality systems has been integrated into the MPLAB Extensions to provide a robust development environment tailored for embedded system developers.
A hardware development kit allows engineers to test the capabilities of the 64-bit MPU. This enables engineers to run high-level operating systems such as Linux with their own custom code and a full tool chain for developing rugged edge applications.
systems that need mixed criticality. The PIC64GX provides four high- performance RISC-V cores with peripherals for AI, computer vision and a wide range of applications at a price point that is aimed at the embedded processing market. The quad core processing architecture helps simplify the development process for engineers developing AI applications at the edge with Linux, and also gives access to a whole ecosystem of established tools to speed up development times. A fifth core handles the monitor functions for the chip but is also available for running real-time or bare metal workloads directly.
The first versions of these 64-bit cores
have been extensively tested in silicon in Microchip’s PolarFire system on chip (SoC), and this technology is now being extended to mass market applications with the PIC64GX microprocessors. The 600 MHz clock is not pushing the limits of chip design or requiring a leading-edge process technology, enabling the affordable price point. Instead, having four cores provides the performance and flexibility that developers need with power consumption that allows the chip to be used for all kinds of edge applications.
The cores in the cluster can run a standard Linux distribution or the Zephyr real-time operating system, all with a secure root of trust. This combination of operating systems has been thoroughly tested in various real- world applications. While the four cores enable asymmetric multicore processing (AMP) beyond single or dual core use, it is the development environment and high-level operating systems that really open up edge AI for developers.
AMP allows developers to assign dedicated processing units for time-sensitive tasks, ensuring predictable performance and maintaining low latency. Linux process scheduling is usually non-real-time, with unpredictable interrupt latencies. AMP enables workloads with hard real-time needs to run an RTOS for critical control, while maintaining application-level functions running on Linux or other software on another core.
24 December/January 2026 Components in Electronics
The software tools also support a continuum within the Microchip family, from the 8-bit AVR through the 32-bit PIC32 microcontrollers to the PIC64GX MPU, all are compatible with a single development environment. This continuum allows developers to choose the right device for their application and easily scale up in performance as required.
Over time though, a standard device may not be enough for some IoT applications. If a developer starts out using the standard PIC64GX microprocessor and wants more custom logic, they can migrate to the cost- reduced PolarFire Core SoC to provide for more flexibility and functionality through the programmable FPGA fabric.
For designs that need to be in orbit, the high-performance spaceflight compute PIC64-HSPC has 8 RISC-V cores with time sensitive networking (TSN) and machine learning. A 10-core version, the PIC64HX, supports TSN in industrial and mission- critical intelligent edge with the same development environment. This is another advantage of the PIC64 architecture, with the ability to add a wide range of peripherals around the core cluster in the same way as the PIC16 and PIC32 MCU families. And of course there is support from Microchip with reference designs, tools and software to help with the development of embedded systems. With all of this, the PIC64GX is bringing 64-bit multicore AMP into the embedded arena for the next generation of edge IoT devices.
https://www.microchip.com/
www.cieonline.co.uk
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52
