Chris Veigl

hi, after a few experiments I found a work-around which perfomes quite nicely: * removed the calls to `i2s_zero_dma_buffer` as suggested * added a call to `i2s_set_dac_mode(I2S_DAC_CHANNEL_DISABLE)` when playback is...

thanks! :-) great to hear that the patch is useful!

i didn't connect an oscilloscope by now - so i've no idea if the signal behaves the same on the M5Stack where I have deployed my application... I suppose this...

sounds as it looks ;-) strange! did you use this fork? https://github.com/ChrisVeigl/ESP8266Audio as I find the time I'll hook up an oscilloscope and have a look! BTW: why do you...

the fork has no other changes than the patch, though ..

i might add that the patch ChrisVeigl@1ce70dc works flawlessly in our application with the M5Stack, where we play many short wav/mp3 files sequentially in order to create sentences. (The M5Stack...

yes - but i think this would just be the line https://github.com/ChrisVeigl/ESP8266Audio/blob/1ce70dcd14a70cfe5ebe899e761cfc4f9e3ec278/src/AudioOutputI2S.cpp#L207 I did not try the patch for other configurations. I suppose @Koxx3 is using the internal DAC -...

> not sure if i2s_zero_dma_buffer() can be made into filling the buffers with anything different than 0 - it's part of the ESP-IDF, so we can't change it. we could...

I have now a setup available which uses the internal DAC (an ODroid-Go device, see: https://github.com/hardkernel/ODROID-GO/blob/master/Documents/ODROID-GO_REV0.1_20180518.pdf) I can confirm that my patch creates noise in this setup; the patched code...

uups ! that bug was in there for a log time! actually the median value was added by a contributor and i did not reallly check it's functionality well enough...