bluetooth-low-energy-applications VS esp-nimble-cpp

If you need some more background, I wrote a book, Develop your own Bluetooth Low Energy Applications for Raspberry Pi, ESP32 and nRF52 with Python, Arduino and Zephyr, and you can find all code examples in its GitHub repository.

If you want some more background about BLE development, I've written a book with a practical approach: Develop your own Bluetooth Low Energy Applications for Raspberry Pi, ESP32 and nRF52 with Python, Arduino and Zephyr. This also includes a chapter about how to read BLE specifications such as the Health Thermometer Service and create code to support the specification. All code examples from the book are freely available.

Here's an example of a Zephyr application that sends indications for BME280 sensor data in a custom BLE service: https://github.com/koenvervloesem/bluetooth-low-energy-applications/blob/main/4-connections/zephyr/peripheral_bme280/src/main.c

I like NimBLE-Arduino, and this has a BLE client example that you can adapt to your needs, including both regular reads and notifications. Note that this isn't pure C++, this is using Arduino. There's also a version for use with ESP-IDF: esp-nimble-cpp, with a corresponding BLE client example. The examples for both projects are heavily commented and quite easy to follow.

On the ESP32 side, I use this library: https://github.com/h2zero/esp-nimble-cpp

No idea what “nimble” is but the first result on google for “nimble library” points at a esp thing: https://github.com/h2zero/esp-nimble-cpp

I have a project using NimBLE, and I'm trying to understand limitations / best-case / how to improve BLE latency.

Im working on a mesh project and want to use BLE to have communication between a smartphone and the mesh. I worked with this nimble lib before but i cant use it since compiling the MDF project with nimble enabled fails. Basically i need https://github.com/nkolban/ESP32_BLE_Arduino but for IDF.

Have a look at my project https://github.com/koenvervloesem/M5Core2-Heart-Rate-Display. It's using NimBLE-Arduino, but if you prefer to use ESP-IDF instead of the Arduino framework there's also esp-nimble-cpp with almost exactly the same API.