ESP32-S3: The Low-Cost Revolution in Graphing Calculators

Published on May 09, 2026 | Translated from Spanish

The NumOS project, created by a 15-year-old, demonstrates that the ESP32-S3 microcontroller can replace 150-euro graphing calculators for just 20 euros worth of components. This chip, common in 3D hardware and embedded projects, offers enough power to run complex mathematical visualizations, breaking the oligopoly of Texas Instruments and Casio. We analyze its performance as a key piece for democratizing scientific visualization tools.

[ESP32-S3 microcontroller with touchscreen running 3D mathematical graphics on a blue background]

Technical analysis of the ESP32-S3 in graphical environments 🔧

The ESP32-S3 integrates a dual-core Xtensa LX7 processor at 240 MHz with 512 KB of SRAM and WiFi and BLE connectivity. For embedded graphics, its integrated LCD controller can handle displays up to 800x480 pixels with 16-bit color, sufficient for rendering mathematical functions in real time. Compared to the Z80 or ARM Cortex-M4 processors of traditional calculators, the ESP32-S3 offers up to 10 times higher performance in floating-point calculations. Its cost, under 5 euros per unit, allows building a complete device with a touchscreen, battery, and 3D-printed case for less than 20 euros, compared to 150 euros for equivalent models from Texas Instruments.

Implications for educational hardware and 3D design 🖨️

This project highlights how optimizing low-cost components can transform stagnant sectors. The ESP32-S3, common in 3D hardware prototypes, demonstrates that specialized chips are not needed for scientific visualization. 3D printing of custom cases and open programming allow the device to be adapted for any classroom, eliminating the internet dependency required by tools like GeoGebra. El-EnderJ's initiative forces a rethinking of the real value of educational hardware, where price no longer justifies planned obsolescence.

What performance or functionality limitations does the ESP32-S3 face when running NumOS compared to a traditional high-end graphing calculator, and how does the project overcome them?

(PS: If your computer is smoking when you open Blender, you might need more than a fan and faith)