A smartwatch based on ESP32 from LilyGo. Currently under development.
Telegram chatgroup is here: https://t.me/TTGO_Watch
Clone this repository and open it with platformIO. Build and upload. On a terminal in vscode you can do it with
pio run -t upload
or simple press "build and upload" in platformIO.
- the webserver crashes the ESP32 really often
- the battery indicator is not accurate, rather a problem with the power management unit ( axp202 )
- from time to time the esp32 crashes accidentally
- and some other small things
On startup you see the main screen (time tile). It show the time and the current weather (if correct configure). Now you can swipe with you fingers up, down, left and right between the four main screens. The four screens are organized in time, apps, note and setup tile. For the weather app you need an openweather.com api-id. http://openweathermap.org/appid is a good starting point.
For customise your ir-codes, use WConfigurator.
The bluetooth notification work with gadgetbridge very well. But keep in mind, bluetooth in standby reduces the battery runtime. In connection with OsmAnd the watch can also be used for navigation. Please use the osmand app, otherwise a lot of messages will be displayed.
For quick clock application development use the new QuickGLUI - high level API. See here.
Internal RAM is very limited, use PSRAM as much as possible. When you work with ArduinoJson, include this
#include "hardware/json_psram_allocator.h"
and create your json with
SpiRamJsonDocument doc( 1000 );
to move your json into PSRAM, here is enough RAM for all the crazy stuff you will do. And use
#include "hardware/alloc.h"
with
MALLOC( ... ); // repleace malloc or ps_malloc
CALLOC( ... ); // repleace calloc or ps_calloc
REALLOC( ... ); // repleace realloc or ps_realloc
as often as possible. It managed internal or PSRAM for you. And one very important thing: Do not talk directly to the hardware!
To play sounds from the inbuild speakers use hardware/sound.h
:
#include "hardware/sound.h"
[...]
// MP3 from SPIFFS:
// void sound_play_spiffs_mp3( const char *filename );
// example:
sound_play_spiffs_mp3( "/sound.mp3" )
// or WAV from PROGMEM via
//void sound_play_progmem_wav( const void *data, uint32_t len );
There is a configuration tile to enable/disable all sound output and set the global volume.
In the display settings page 2 you can set what happens when the button is pressed for 2 seconds. This will normally takes you back to the main tile. If the option for a screenshot is activated, a screenshot is taken instead. This can be downloaded via the built-in FTP server (binary and passive mode, username: TTWatch and password: passord), if activated. The file name is screen.data.
Or the other way:
The firmware has an integrated webserver. Over this a screenshot can be triggered. The image has the format RGB565 and can be read with gimp. From bash it look like this
wget x.x.x.x/shot ; wget x.x.x.x/screen.565
Special thanks to the following people for their help:
5tormChild
bwagstaff
chrismcna
datacute
jakub-vesely
joshvito
JoanMCD
NorthernDIY
rnisthal
paulstueber
ssspeq
and the following projects:
ArduinoJson
AsyncTCP
ESP32SSDP
ESPAsyncWebServer
LVGL
TFT_eSPI
TTGO_TWatch_Library
ESP8266Audio
pubsubclient
Every Contribution to this repository is highly welcome! Don't fear to create pull requests which enhance or fix the project, you are going to help everybody.
If you want to donate to the author then you can buy me a coffee.