240MHz

DUAL-CORE LX7

16MB

FLASH STORAGE

8MB

OPI PSRAM

512KB

INTERNAL SRAM

45

GPIO PINS

BT 5.0

LONG RANGE

+20dBm

TX POWER

8µA

DEEP SLEEP

N16R8

16 MB

N8R8

8 MB

N8R2

8 MB

N16R8

~80 MB/s

N8R8

~80 MB/s

N8R2

~40 MB/s ⚠️

ESP32-S3

Xtensa LX7 ✅

• 128-bit SIMD vector unit
• ~1181 CoreMark score
• Native USB OTG built-in
• Bluetooth 5.0 + Long Range
• 45 programmable GPIOs
• 14 capacitive touch channels
• AI/ML acceleration hardware

VS

ORIGINAL ESP32

Xtensa LX6

• No vector unit
• ~994 CoreMark score
• External USB-serial chip needed
• Bluetooth 4.2 only
• 34 programmable GPIOs
• 10 capacitive touch channels
• No AI acceleration

🔀 True Dual-Core FreeRTOS

Run Wi-Fi/BLE on Core 0 and your application on Core 1 simultaneously — no timesharing, true parallelism.

📊 128-bit SIMD Vector Unit

Processes 16 INT8 values in a single clock cycle — the engine that makes edge AI practical.

💡 RISC-V ULP Coprocessor

Wakes at 24 µA to poll sensors in deep sleep — main cores stay off until needed.

🗃️ 8.5 MB Total Usable RAM

512 KB on-chip SRAM + 8 MB external PSRAM = ~8.5 MB available to your code.

🗣️

Voice Wake-Word (ESP-SR)

“Hey Marvin”-style local wake-word detection. No cloud. Latency <100 ms.

👁️

Person / Object Detection

~10 FPS at 96×96 INT8 with TF-Lite Micro on Core 1 while Core 0 handles Wi-Fi.

😀

Face Detection & Recognition

ESP-WHO library — 5–10 FPS face detect with recognition on-device.

🌱

TinyML — Edge Impulse

Train anomaly detection / gesture models in Edge Impulse, deploy to S3 in one click.

🔊

Audio Classification

Classify sounds (glass break, cough, machine noise) locally at <50 ms.

🛡️

Privacy-First AI

All inference on-device — no audio or video ever leaves the hardware.

⌨️ USB HID — Keyboard / Mouse

Board appears as a real keyboard or mouse to any PC/Mac. Build macro pads, stream decks, accessibility tools, gaming controllers.

💿 USB CDC — Serial Programming

Upload sketches and see Serial Monitor over native USB — no CP2102 chip needed at all if you use GPIO 19/20 directly.

💾 USB MSC — Mass Storage

Board appears as a USB flash drive — users drag-drop files without any custom driver.

🎮 USB Host — Connect Devices

Plug a USB keyboard, mouse, or game controller into the S3 and read it as a host device.

📶 Wi-Fi 2.4 GHz (802.11 b/g/n)

📡 Bluetooth 5.0 LE

📷 Camera Frame Buffers

A 640×480 RGB565 frame = 614 KB. With 8 MB PSRAM you hold 3+ frames for double-buffered JPEG encoding at full speed.

🤖 AI Model Weights

MobileNet-V1 INT8 ≈ 2–3 MB. Without PSRAM these models simply cannot load. 8 MB fits most practical TinyML models.

🔊 Audio Buffers

16 kHz 16-bit mono = 32 KB/s. 8 MB PSRAM can buffer ~4 minutes of audio for wake-word and speech processing.

🌐 Web / LVGL UI Caches

Large HTML/JSON responses and full LVGL color frame-buffers for TFT screens without fragmenting the heap.

🔐 Secure Boot V2

Only firmware signed with your private RSA-3072 key will run. Prevents unauthorized firmware from booting.

🔒 Flash Encryption (AES-XTS-256)

All flash contents encrypted at rest. Reading the flash chip directly reveals only ciphertext — your firmware and keys stay secret.

🔑 eFuse OTP Memory

One-time-programmable fuses store root keys, device IDs and security flags permanently — cannot be overwritten.

🧮 Hardware Crypto Engines

AES-128/256, SHA-256/512, RSA, ECC, and True RNG — all accelerated in hardware. TLS/HTTPS at full speed.

🛑 Digital Signature Block

Private key stored in hardware — software can request signatures without ever exposing the raw key bytes.

🔥 HMAC-Based Secure OTA

OTA firmware updates verified by hardware HMAC — prevents downgrade attacks and firmware tampering.

🔢

ADC

2× 12-bit SAR, 20 channels

👆

Touch Sensor

14 capacitive channels

🔁

UART

3× hardware UART

🔗

SPI

4× SPI (2 general)

🔗

I²C

2× I²C (master + slave)

🎵

I²S Audio

2× I²S (mic + speaker)

🌈

LEDC PWM

8 channels, any GPIO

📷

Camera (DVP)

8–16 bit parallel camera

🖥️

LCD (8/16-bit)

Parallel LCD controller

⏱️

Timers

4× 54-bit general timers

📡

RMT (IR/WS2812)

8 channels — IR remote, NeoPixel

🔄

MCPWM

Motor control PWM — servos, BLDC

⚫ BOOT Button (GPIO 0)

• Normal use: Does nothing during normal operation — GPIO 0 is just a GPIO when the board is running.
• During upload: Hold BOOT while pressing RESET (or while connecting USB) — this puts the board into download mode so Arduino IDE / esptool can flash firmware.
• Auto-reset: The CP2102 auto-reset circuit does this automatically on most uploads — you only need to press BOOT manually if auto-reset fails.

🔄 RESET Button (EN Pin)

• Press once: Reboots the chip and restarts your sketch from the beginning — like power cycling.
• With BOOT held: Press RESET while holding BOOT → release RESET → release BOOT = enters download mode manually.
• After upload: If the board does not auto-reboot after upload, press RESET once to start the new sketch.

🎓 Beginners

Pre-soldered headers — zero soldering needed. Plug into breadboard and start with Arduino IDE in under 10 minutes.

📚 Students

Perfect for university projects in IoT, AI, embedded systems, wireless comms and final-year research.

⚙️ Engineers

Maximum Flash, PSRAM bandwidth, and hardware security for production commercial IoT and AI products.

🛠️ Hobbyists & Makers

Smart home, wearables, BLE gadgets, AI edge devices, WLED displays, custom keyboards and more.

ARDUINO IDE → TOOLS — EXACT SETTINGS FOR ESP32-S3 N16R8

⬇️ Download .ino

#include <WiFi.h>
const char* ssid     = "Your_WiFi_Name";
const char* password = "Your_WiFi_Password";

void setup() {
  Serial.begin(115200);
  WiFi.mode(WIFI_STA);
  WiFi.begin(ssid, password);
  // Optionally set max TX power for longer range:
  // WiFi.setTxPower(WIFI_POWER_19_5dBm);
  Serial.print("Connecting");
  while (WiFi.status() != WL_CONNECTED) {
    delay(500); Serial.print(".");
  }
  Serial.println("\n✅ WiFi Connected!");
  Serial.print("IP Address : "); Serial.println(WiFi.localIP());
  Serial.print("RSSI (dBm) : "); Serial.println(WiFi.RSSI());
  Serial.print("MAC Address: "); Serial.println(WiFi.macAddress());
}
void loop() {}

Open Serial Monitor at 115200 baud to see IP address and signal strength.

⬇️ Download .ino

#include <WiFi.h>
#include <WebServer.h>
const char* ap_ssid     = "ESP32-S3-AP";
const char* ap_password = "12345678";
WebServer server(80);
void setup() {
  Serial.begin(115200);
  WiFi.softAP(ap_ssid, ap_password);
  Serial.print("AP IP: "); Serial.println(WiFi.softAPIP());
  server.on("/", [](){
    server.send(200, "text/html",
      "<h1>ESP32-S3 N16R8</h1><p>Free PSRAM: "
      + String(ESP.getFreePsram()) + " bytes</p>");
  });
  server.begin();
}
void loop() { server.handleClient(); }

Connect phone to “ESP32-S3-AP” (password: 12345678) → open http://192.168.4.1 in browser.

⬇️ Download .ino

#include <BLEDevice.h>
#include <BLEServer.h>
#define SERVICE_UUID "4fafc201-1fb5-459e-8fcc-c5c9c331914b"
#define CHAR_UUID    "beb5483e-36e1-4688-b7f5-ea07361b26a8"
void setup() {
  Serial.begin(115200);
  BLEDevice::init("ESP32-S3-BLE");
  BLEServer* pServer = BLEDevice::createServer();
  BLEService* pService = pServer->createService(SERVICE_UUID);
  BLECharacteristic* pChar = pService->createCharacteristic(
    CHAR_UUID, BLECharacteristic::PROPERTY_READ);
  pChar->setValue("Hello from ESP32-S3 N16R8!");
  pService->start();
  BLEDevice::getAdvertising()->start();
  Serial.println("✅ BLE Advertising as 'ESP32-S3-BLE'");
}
void loop() { delay(2000); }

Scan with “nRF Connect” app (Android/iOS) → find “ESP32-S3-BLE” → connect → read the characteristic value.

⬇️ Download .ino

#include <WiFi.h>
#include <WebServer.h>
#include <Adafruit_NeoPixel.h>
const char* ssid = "Your_WiFi_Name";
const char* pass = "Your_WiFi_Password";
Adafruit_NeoPixel led(1, 48, NEO_GRB + NEO_KHZ800);
WebServer server(80);
void setup() {
  Serial.begin(115200);
  led.begin(); led.show();
  WiFi.begin(ssid, pass);
  while(WiFi.status() != WL_CONNECTED) { delay(500); }
  Serial.println(WiFi.localIP());
  server.on("/red",   [](){ led.setPixelColor(0,50,0,0);  led.show(); server.send(200,"text/plain","RED");   });
  server.on("/green", [](){ led.setPixelColor(0,0,50,0);  led.show(); server.send(200,"text/plain","GREEN"); });
  server.on("/blue",  [](){ led.setPixelColor(0,0,0,50);  led.show(); server.send(200,"text/plain","BLUE");  });
  server.on("/off",   [](){ led.clear();             led.show(); server.send(200,"text/plain","OFF");  });
  server.begin();
}
void loop() { server.handleClient(); }

Visit http://<board-IP>/red, /green, /blue, or /off in any browser. Install Adafruit NeoPixel library first.

⚡ Ultra-Low Latency

~1 ms delivery time vs ~100 ms for standard Wi-Fi. Perfect for real-time control.

🔋 No Router Needed

Boards communicate directly — great for remote or off-grid deployments.

🌐 Up to 20 Peers

One board can communicate with up to 20 others simultaneously.

🕸️ Mesh with ESP-Mesh

Scale to hundreds of nodes using Espressif’s ESP-WIFI-MESH protocol.

1 × ESP32-S3 WROOM-1 N16R8 Development Board

Pre-soldered (welded) pin headers included. USB cable and accessories not included.

✅ COD Available

Pay after receiving

⚡ Fast Dispatch

Quick processing time

🚚 Inside Dhaka

69 BDT — within 24 hrs

🛵 Outside Dhaka

129 BDT — 24 to 72 hrs