You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
malformed (wrongly scaled?) waveform with input_alsa #4
Did you compare this with PulseAudio input? The DFT is a linear transform, so DFT-based visualizations should also exhibit the same problem if input waveforms are incorrectly scaled.
Did you compare this with PulseAudio input? The DFT is a linear
transform, so DFT-based visualizations should also exhibit the same
problem if input waveforms are incorrectly scaled.
Reply to this email directly or view it on GitHub: #4 (comment)
I really can't see anything wrong with the input signal from ALSA. If there is any incorrect scaling, it has to be elsewhere in lv_audio.c where the samples are converted.
If you're using lv_scope to check the waveforms, be aware that peak amplitudes will show up at 1/4 and 3/4 of the display height. They will not reach the top or bottom of the display.
I really can't see anything wrong with the input signal from ALSA. If
there is any incorrect scaling, it has to be elsewhere in lv_audio.c
where the samples are converted.
If you're using lv_scope to check the waveforms, be aware that peak
amplitudes will show up at 1/4 and 3/4 of the display height. They will
not reach the top or bottom of the display.
Reply to this email directly or view it on GitHub: #4 (comment)
i suppose it's not even directly linked to lv but to my snd-hw driver. It offers various alsa devices (hw:, plugdev:, etc) and iirc i read something about software processing on one of them. I can't just hardcode another device into input_alsa.c, this doesn't work. I merely wanted to know if anyone else is experiencing this...
Activity
kaixiong commentedon Apr 11, 2012
Did you compare this with PulseAudio input? The DFT is a linear transform, so DFT-based visualizations should also exhibit the same problem if input waveforms are incorrectly scaled.
rigid commentedon Apr 11, 2012
Chong Kai Xiong reply@reply.github.com schrieb:
no, only input_alsa and input_debug work.
kaixiong commentedon Apr 11, 2012
PulseAudio works for me, but I have to use pavucontrol to be able to redirect and capture output audio streams.
rigid commentedon Apr 23, 2012
it seems input_alsa needs a complete facelift and VisParam support selection of different hardwares. https://github.com/mumble-voip/mumble/blob/master/src/mumble/ALSAAudio.cpp could serve as example, it looks quite clean.
kaixiong commentedon Apr 24, 2012
I really can't see anything wrong with the input signal from ALSA. If there is any incorrect scaling, it has to be elsewhere in lv_audio.c where the samples are converted.
If you're using lv_scope to check the waveforms, be aware that peak amplitudes will show up at 1/4 and 3/4 of the display height. They will not reach the top or bottom of the display.
rigid commentedon Apr 24, 2012
Chong Kai Xiong reply@reply.github.com schrieb:
i suppose it's not even directly linked to lv but to my snd-hw driver. It offers various alsa devices (hw:, plugdev:, etc) and iirc i read something about software processing on one of them. I can't just hardcode another device into input_alsa.c, this doesn't work. I merely wanted to know if anyone else is experiencing this...
kaixiong commentedon Apr 24, 2012
Hmm.. I have other issues with the ALSA input plugin though. The most serious one is that It hangs after a while.
rigid commentedon Jan 5, 2013
issue seems to be fixed in cxx branch
Plugins (gforce): Fix new-delete-type-mismatch
Plugins (gforce): Fix new-delete-type-mismatch
Plugins (corona) : Fix allocation size to fix heap-buffer-overflow
Plugins (corona) : Fix allocation size to fix heap-buffer-overflow
Plugins (corona) : Fix allocation size to fix heap-buffer-overflow
Plugins (corona): Fix allocation size to fix heap-buffer-overflow