Initially i2c_sda and i2c_scl are set to "-1" in wled.h. This can cause a bootloop when trying to initialize I2C with pins (-1, -1).
This fix ensures that usermod initialization will fail when SDA=-1 or SCL=-1.
A few goodies from my development branch, before I continue refactoring so much that I can't extracts nothing anymore.
- GEQ fades a bit slower
- updates to mic profiles
- simple peak detection without FFT (low freqs only)
- added "max used PSRAM" info
- moved all mem infos into one part
- only check PSRAM if BOARD_HAS_PSRAM
- oappend: debug message in case that stack buffer is too small.
Recent changes in mdev back to this branch
- devums to devmax
- Asyncwebserver to 2.0.7
- expressif updates
- new chips updates
- new warnings
esp32mdevmax
- activate 4ld, and rotary and 'hope' there are no conflicts with MPU6050 (sharing i2s bus)
This is a nice little chip info tool that I've created some time ago.
When building with `-D WLED_DEBUG`, it reports some interesting information about your ESP32 on serial monitor.
````
====================================
Chip info for ESP32-D0WDQ5
SDK: v4.4.1-472-gc9140caf8c
------------------------------------
XTAL FREQ: 40 MHz
APB FREQ: 80.0 MHz
CPU FREQ: 240 MHz
ESP32 DEVICE: ESP32-D0WD-V3 rev.3, 2 core(s), 240MHz.
Chip feature flags: 0b110010
* 0b0000010 Chip has 2.4GHz WiFi
* 0b0001000 Chip has Bluetooth LE
* 0b0010000 Chip has Bluetooth Classic
FLASH CHIP FREQ (magic): 80.0 MHz
FLASH SIZE (magic byte): 4.00 MB
FLASH MODE (magic byte): 2 ; 0=QIO, 1=QOUT, 2=DIO, 3=DOUT or other
FLASH CHIP ID: 0x16405E
FLASH CHIP FREQ: 80.0 MHz
FLASH REAL SIZE: 4.00 MB
FLASH REAL MODE: DIO
------------------------------------
RAM HEAP SIZE: 261.79 KB
FREE RAM: 232.57 KB
MAX RAM alloc: 107.99 KB
ESP32 PSRAM: found.
* SPI RAM Chip availeable: 64MBits = 8MBytes
total PSRAM: 4094 KB
FREE PSRAM: 4093.89 KB
MAX PSRAM alloc:4031.99 KB
used PSRAM: 0 Bytes
* PSRAM free after malloc / ps_malloc : 2833.85 KB
* Can allocate big memory with ps_malloc()
* Can allocate big memory with malloc()
CPU #0 - last reset reason = 1 POWERON_RESET => Vbat power on reset
CPU #1 - last reset reason = 14 EXT_CPU_RESET => APP CPU reset by PRO CPU
====================================
````
it's possible that volume samples become negative. In this scenario, our simple noise gate does stupid things, and it looks like "effects temporarily lost the sound".
This fix improves the situation, and makes sure that volume samples are always >= 0.
