Week #4 — Hardware

Sachin C -

This week, as I continued developing the artificial intelligence model’s dataset, I took a small break and began investigating the hardware I would be using for the experiment.

  1. Raspberry Pi 4 Model B
    • Processor: Quad-core Cortex-A72
    • RAM: 2GB/4GB/8GB LPDDR4
    • Why Test It? This is a higher-end microcontroller with more memory and processing power than typical embedded systems. It will serve as a benchmark to compare how well more constrained devices perform.
  2. ESP32
    • Processor: Dual-core Xtensa LX6
    • RAM: 520KB SRAM
    • Why Test It? The ESP32 is popular for IoT applications due to its exceptionally low power consumption and built-in Wi-Fi/Bluetooth capabilities. Testing it will help determine how well MobileNet can function on a truly lightweight microcontroller.
  3. Arduino Portenta H7
    • Processor: Dual-core Arm Cortex-M7 and Cortex-M4
    • RAM: 8MB SDRAM
    • Why Test It? This high-performance microcontroller bridges the gap between traditional Arduino boards and more powerful embedded systems, offering an interesting middle ground for AI deployment.
  4. Texas Instruments AM69 Series Board
    • Processor: Arm Cortex-A72 + AI-specific accelerators
    • RAM: 8GB LPDDR4
    • Why Test It? This board is optimized for AI applications with built-in accelerators, making it ideal for comparing how dedicated AI hardware enhances performance.

 

Next week, I’ll check back in and update you all on the progress of my dataset. I’ve thoroughly enjoyed creating this (albeit disk space-consuming) mess inside my computer, so I look forward to seeing you soon!

More Posts

Comments:

All viewpoints are welcome but profane, threatening, disrespectful, or harassing comments will not be tolerated and are subject to moderation up to, and including, full deletion.

    adam_b
    Hi Sachin, I'm so glad to hear you are enjoying your project! Will you be using all of these tools in tandem, or testing them separately?
    March 6, 2025 at 2:58 pm - Reply
    rohan_va
    Hi Sachin, the technology you're putting together for this project is fascinating! You mention that the Raspberry Pi model you plan to use has more memory and processing power than other typical systems. Is there a name/example for these other systems, or are they just the industry standard?
    March 7, 2025 at 10:12 am - Reply
    camille_bennett
    Hi Sachin, sounds interesting to compare the different hardware. Do you have access to all of them? Is price point a consideration in your comparison?
    March 10, 2025 at 10:26 am - Reply
    sachin_c
    Hey Adam, I intend on testing each piece of hardware separately. They will all run the same model and be fed the same prompts, however.
    March 12, 2025 at 10:48 am - Reply
    sachin_c
    Hi Rohan! The other systems are actually the other ones I'm testing! Per the general processing power of most microcontrollers on the market, the Raspberry Pi model I'm using is a little bit more powerful.
    March 12, 2025 at 10:50 am - Reply
    sachin_c
    Hi Ms. Bennett, I have access to hardware through my site placement. Price points tend to be relatively similar and moderate for microcontrollers, as they do not have the processing power of a desktop computer, for example.
    March 12, 2025 at 10:51 am - Reply

Leave a Reply

Your email address will not be published. Required fields are marked *