The officially official Devuan Forum!

aluma

Member

Registered: 2022-10-26

Posts: 646  

Re: Resampling and the Gibbs phenomenon with Audacity

My post #4

What our posts have in common is that they have nothing to do with this forum itself, the forum of the Devuan distribution, and your links and advice on proprietary formats and paid resamples in no way correspond to Freedom Hacks, as I understand it.

igorzwx

Member

Registered: 2024-05-06

Posts: 103  

Re: Resampling and the Gibbs phenomenon with Audacity

With pure alsa, the sound was everywhere, except for the browser, which “did not see” the sound system. Either pipewire or jack was required.

If it works with pulseaudio, it does not mean that it cannot work with "pure alsa".

But I can agree that it might be difficult to create a correct ~/.asoundrc
That is why, I am using arateconf.

There is no sound device defined in the FF Daedalus settings.

I also do not have sound devices defined in the Firefox settings (about:support).
And, nevertheless, sound works with "pure alsa".

Last edited by igorzwx (2024-07-21 17:06:44)

aluma

Member

Registered: 2022-10-26

Posts: 646  

Re: Resampling and the Gibbs phenomenon with Audacity

I also do not have sound devices defined in the Firefox settings (about:support).
And, nevertheless, sound works with "pure alsa".

These are again just words. Screenshot, as in my thread at the link above, details (output of the corresponding bash commands) of the operating system in the studio.

igorzwx

Member

Registered: 2024-05-06

Posts: 103  

Re: Resampling and the Gibbs phenomenon with Audacity

Firefox is playing video in youtube:

$ firefox 2>&1
[Child 3342, MediaDecoderStateMachine #1] WARNING: 7fa31a74d4c0 Could not set cubeb stream name.: file ./dom/media/AudioStream.cpp:321
Input:  44100 Hz, 2 ch, 's32_le' (0xa): dummy = 0, period = 1764
Output: 48000 Hz, 2 ch, 's16_le' (0x2): dummy = 0, period = 1920
Rates:  44100 --> 48000 (J: 0.00%, T: FFT, W: Vorbis)
Ok. 

FFT is the fftrate resampler. It is GPL3, the source code is here:
_https://github.com/PetrovSE/fftrate

$ fuser -av $(find /dev/snd -type c 2>/dev/null)
                     USER        PID ACCESS COMMAND
/dev/snd/controlC0:  igor       2229 F.... mate-settings-d
                     igor       2324 F.... mate-volume-con
/dev/snd/pcmC0D0c:
/dev/snd/pcmC0D0p:   igor       3175 F...m firefox-esr
/dev/snd/seq:
/dev/snd/timer:      igor       3175 f.... firefox-esr

It works without apulse, and it works with apulse as well

$ apulse firefox 2>&1
Input:  48000 Hz, 2 ch, 's32_le' (0xa): dummy = 0, period = 1920
Output: 48000 Hz, 2 ch, 's16_le' (0x2): dummy = 0, period = 1920
Rates:  48000 --> 48000 (J: 0.00%, T: None, W: Planar)
Ok. 

Notice that the resampler is not active (T: None).
It seems that it makes sense to run browsers with apulse.
It helps to improve sound quality with Min, Brave, Chromium and Chrome.
It might be a bug in ALSA backend.

The same video with Chromium:

$ apulse chromium 2>&1
Input:  44100 Hz, 2 ch, 's32_le' (0xa): dummy = 0, period = 1764
Output: 48000 Hz, 2 ch, 's16_le' (0x2): dummy = 0, period = 1920
Rates:  44100 --> 48000 (J: 0.00%, T: FFT, W: Vorbis)
Ok. 

$ fuser -av $(find /dev/snd -type c 2>/dev/null)
                     USER        PID ACCESS COMMAND
/dev/snd/controlC0:  igor       2229 F.... mate-settings-d
                     igor       2324 F.... mate-volume-con
/dev/snd/pcmC0D0c:
/dev/snd/pcmC0D0p:   igor       7968 F...m firefox-esr
                     igor       8428 F...m chromium
/dev/snd/seq:
/dev/snd/timer:      igor       7968 f.... firefox-esr
                     igor       8428 f.... chromium 

If you do not believe my words, why should I believe yours?
It might be very difficult to believe that you really want to configure ALSA.

Last edited by igorzwx (2024-07-22 01:31:26)

aluma

Member

Registered: 2022-10-26

Posts: 646  

Re: Resampling and the Gibbs phenomenon with Audacity

If you do not believe my words, why should I believe yours?

I asked you to post information that I had repeatedly posted myself for my system.
But I don’t care at all, I’ll leave it at that.

igorzwx

Member

Registered: 2024-05-06

Posts: 103  

Re: Resampling and the Gibbs phenomenon with Audacity

I asked you to post information that I had repeatedly posted myself for my system.

I hope that I am mistaken, but it looks like you are simulating "problems with ALSA", in order to convince people to use pulseaudio.