Flexible 32-Bit Microprocessor Unveiled by Pragmatic Semiconductor
Flexible: British startup Pragmatic Semiconductor has introduced a groundbreaking 32-bit microprocessor known as Flex-RV. This innovative chip is capable of running machine learning models while also being bendable, all at an affordable price point of under a dollar. Built on the open standard RISC-V architecture, the Flex-RV chip utilizes a unique material called indium gallium zinc oxide (IGZO) to achieve its impressive flexibility.
The key advantage of IGZO is its ability to replace traditional silicon, eliminating the need for costly packaging to protect the chip from damage when flexed. By printing IGZO transistors directly onto plastic substrates at low temperatures, Pragmatic has created a highly adaptable microprocessor that can withstand bending without compromising performance.
During a demonstration, the Flex-RV chip was rolled up around a straw and then unfurled to continue functioning seamlessly. While the chip requires careful handling and does not offer high performance with only 12,600 logic gates and a maximum clock speed of 60 kHz, it does feature a low-power machine learning accelerator.
Designed for applications in disposable medical devices, health wearables, soft robotics, and more, the Flex-RV chip consumes less than 6 mW of power. By utilizing the open-source RISC-V instruction set, Pragmatic has also avoided costly licensing fees, making the chip an affordable solution for various industries.
In testing, the Flex-RV chip maintained accuracy even when bent to a 5 mm radius curve, showcasing its resilience and versatility. This advancement in bendable chip technology represents a significant leap forward from previous designs that could only be tested while resting on rigid surfaces.
While previous attempts have been made to create bendable PCBs, the Flex-RV chip sets itself apart by combining flexibility with integrated circuit capabilities that allow it to twist and contort. This innovative approach to microprocessor design was recently published in the prestigious journal Nature on September 25.
