ESP32 Forum

With the release of the Serial port AT Command Processor ... see:

http://espressif.com/sites/default/file ... les_en.pdf

I got to wondering about it. Are there any compelling use cases for it? If I understand it correctly, it is firmware (an application) that is precompiled and can be loaded into an ESP32. After which, the ESP32 will start communicating via UART to send and receive commands in "AT+" format. The commands allow us to form WiFi connections and from there TCP/IP network connections ... as either a sender or a receiver.

My core question is "When would one consider this pre-built application in a design?". Is there a core story that this would apply to? Is there some industry standard or specification to which these AT commands conform? For example, if the AT commands were a standard, then I could imagine building products that would talk these protocols and then choosing a supplier (eg. Espressif) to supply components that met the standards and hence not locking myself into one vendor.

Most compelling is to update an existing stable design to replace another AT based module and take advantage of price/features of esp32 and avoid hassle of porting whole application. I don't think there is strict standard for AT commands, but there is some overlap. Otherwise to reduce time to market by relying on experience developing for other platforms and not having to learn esp32 development. Also possibly for designs that may have hardware addon option for wifi.

There is a thing known as recommendation V.250 (https://www.itu.int/rec/T-REC-V.250-200307-I/en), which describes, among other things, the overall protocol and some of the standard commands. This doesn't include any application-specific commands (for example, wifi- and tcp- related ones). As far as i know, our AT implementation is partially compliant with V.250 (but also has some quirks).

Kolban, I had those exact same thoughts when I saw this. Your response was a lot more refined than mine. I was thinking perhaps I missed something somewhere and someone had created some kind of modem for WiFi. It would seem strange since everything is configurable through a browser.

There was obviously a lot of work put into coming up with the code but it looks like this is a solution looking for a problem. Perhaps in the future someone will come up with a profitable use case for it.

This is how beginners get started. Go and look on things like Hack-a-Day for the number of projects using the ESP8266 and the AT command set, there's heaps. Now the ESP8266 version wasn't great, had all sorts of issues, but for a place to start with a complicated module it's perfect. It means that a user can use the module just as a WiFi radio rather than a stand alone uC (again, good for beginners who want to hook up a Uno or similar).
Tuanpm even did a SLIP version which was really go to integrate the ESP8266 directly into an Arduino as a radio rather than using it as as standalone module.
Just my 0.02c worth

Lakid74,

Very valid points. I was beginner 20 years ago when things where simpler. You got a chip, looked up the well documented data sheet and started on your learning curve. Today, the level of understanding of these devices is limited. Which isn't such a bad. The more people want Microcontroller based products, the more companies need people that can develop these products from the ground level. It is a win win situation for everyone.

Keep on MCUing!