Closed TF3RDL closed 7 months ago
I have turned that code inside out and can't find anything wrong with it. Unless you can pinpoint where my logic goes wrong I'm afraid this bug will remain unfixed.
I guess I've leave this open despite being "fixed", as I think a rigorous testing is required to see what this issue is truly fixed or not; if it is not, leave it open, and if it is fixed after very rigorous testing by @stuerp and HA members, close it (but not immediately)
I think I can close this now. There is no bug in the sample averaging; it was the caused by something else.
I think I can close this now. There is no bug in the sample averaging; it was the caused by something else.
Fair enough, unless something pops up again (which I hopefully it doesn't happen again)
Right now, this spectrum visualization has unusually-high sidelobes (equivalent of rectangular window) when the audio data is stereo unless downmix channels to mono DSP is applied, due to its improper way to sum different channels of each sample into mono before FFT