# 477: One Thousand New Instructions

- Show: [Embedded.fm](https://hardwarepodcasts.com/podcast/embedded-fm)
- Published: 2024-05-16
- Length: 1h 24m
- Episode: 477
- Page: https://hardwarepodcasts.com/podcast/embedded-fm/2024-05-16-477-one-thousand-new-instructions
- Listen: https://embedded.fm/episodes/477

## Summary

Kwabena Agyeman, founder of OpenMV, explains how SIMD optimization and new Arm Cortex-M55 processors enable complex computer vision and machine learning on low-cost microcontrollers. He details specific algorithmic improvements, such as a 16x speedup in median filters, and discusses the shift from Linux-based systems like NVIDIA Jetson to tinyML for cost and power efficiency. Agyeman also shares his experience scaling OpenMV through the chip shortage and his previous career at Embark Trucks.

## Chapters

- [0:01](https://hardwarepodcasts.com/podcast/embedded-fm/2024-05-16-477-one-thousand-new-instructions#t=1) Introduction and OpenMV Overview
- [4:22](https://hardwarepodcasts.com/podcast/embedded-fm/2024-05-16-477-one-thousand-new-instructions#t=262) SIMD Optimization and Median Filter Case Study
- [20:35](https://hardwarepodcasts.com/podcast/embedded-fm/2024-05-16-477-one-thousand-new-instructions#t=1235) MicroPython, ulab, and Production Use Cases
- [33:45](https://hardwarepodcasts.com/podcast/embedded-fm/2024-05-16-477-one-thousand-new-instructions#t=2025) Wasp Identification Project and Optical Challenges
- [43:26](https://hardwarepodcasts.com/podcast/embedded-fm/2024-05-16-477-one-thousand-new-instructions#t=2606) Event Cameras and Frame Differencing
- [56:25](https://hardwarepodcasts.com/podcast/embedded-fm/2024-05-16-477-one-thousand-new-instructions#t=3385) Scaling OpenMV and Supply Chain Lessons
- [1:02:34](https://hardwarepodcasts.com/podcast/embedded-fm/2024-05-16-477-one-thousand-new-instructions#t=3754) Arm Helium, Cortex-M55, and TinyML Future
- [1:09:38](https://hardwarepodcasts.com/podcast/embedded-fm/2024-05-16-477-one-thousand-new-instructions#t=4178) Reflections on Embedded Engineering and Career

## Topics

- [Embedded firmware](https://hardwarepodcasts.com/topic/embedded-firmware)
- [Microcontrollers & dev platforms](https://hardwarepodcasts.com/topic/microcontrollers-and-platforms)
- [AI & machine learning](https://hardwarepodcasts.com/topic/ai-and-machine-learning)
- [Scaling production](https://hardwarepodcasts.com/topic/scaling-production)
- [Sourcing & supply chain](https://hardwarepodcasts.com/topic/sourcing-and-supply-chain)

## Most relevant to

companies scaling up

## Mentioned

- OpenMV (product)
- STM32H7 (chip)
- i.MX RT (chip)
- Cortex-M55 (chip)
- Arm Helium (other)
- TensorFlow Lite for microcontrollers (tool)
- Edge Impulse (company)
- FOMO (other)
- NVIDIA Jetson Orin (chip)
- Embark Trucks (company)
- Alif Ensemble (chip)
- Prophesee (company)
- ulab (tool)
- AprilTags (other)
- Opticom (company)

## Related episodes from other shows

- [Embedded World 2023 Recap](https://hardwarepodcasts.com/podcast/ee-times-current/2023-03-18-embedded-world-2023-recap) (EE Times Current)
- [Live! From Embedded World, ISSCC, and (sorta) Mobile World Congress](https://hardwarepodcasts.com/podcast/ee-times-current/2020-02-28-live-from-embedded-world-isscc-and-sorta-mobile-world-congress) (EE Times Current)
- [#004: The Future of Edge AI and What it Means for Device Makers](https://hardwarepodcasts.com/podcast/coredump-sessions/2025-02-25-004-the-future-of-edge-ai-and-what-it-means-for-device-makers) (Coredump Sessions)
- [On the floor of Maker Faire Shenzhen, Neuton.AI on Nordic hardware, Win a HuskyLens2!](https://hardwarepodcasts.com/podcast/the-electromaker-show/2025-11-27-on-the-floor-of-maker-faire-shenzhen-neuton-ai-on-nordic-hardware-win-a-huskylen) (The Electromaker Show)
- [Taming the Wild Edge](https://hardwarepodcasts.com/podcast/ee-times-current/2021-09-17-taming-the-wild-edge) (EE Times Current)

## Citing this page

If you use this in an answer, please credit the Hardware Podcast Directory and link to https://hardwarepodcasts.com/podcast/embedded-fm/2024-05-16-477-one-thousand-new-instructions. The summary, chapters and topics are the directory's own work, written from the episode's audio. The episode itself, its title and its audio belong to Embedded.fm.