In another exciting “Ask an ipXpert” segment, Eamon catches up with Henrik Flodell from Alif Semiconductor for a deep dive into the Ensemble family of microcontrollers and “fusion processors” – the latter is an exciting combination of MCU and application processor cores, as explained in our previous article and interview about/with Alif Semiconductor. But as a design engineer, what are the benefits of these multiple cores in a single device and Alif’s implementation of such an architecture?
As Henrik explains, Alif has built their Ensemble devices from the ground up for low power consumption, with an emphasis on bringing mid- to high-level AI workloads to battery-operated devices, far beyond the realms of the typical TinyML workloads expected on many general-purpose MCUs. With Alif’s Ensemble family, you can expect to be doing, for example, facial/image recognition on a 32-bit general-purpose microcontroller using development tools and environments that you will already be familiar with. Thanks to the clever implementation of running AI workloads efficiently on Alif’s devices, you’ll also see new areas where AI can be used to improve a product design and user experience, squeezing the most life out of smaller batteries or greatly extending battery life in existing product form factors.
Enjoy this in-depth discussion on the complete range of Ensemble devices, from the single Cortex-M55 + Ethos-U55 E1 MCU, all the way to the E7 fusion processor that features two sets of Cortex-M55 + Ethos-U55 pairs and two Cortex-A32 cores on top of that for running Linux, for example. Most importantly, pay close attention to Henrik’s explanations of how you can use these cores effectively to make great AI-enabled products, from wearables to home security systems, vending machines, and more.
For a full technical writeup of Alif’s Ensemble family, head to the board page by following the link below, where you can learn more and apply for a development kit to evaluate this technology for your next commercial project.