From Newsgroup: alt.os.linux


I have onboard HD-Audio + a Xonar sound card but I cannot get
simultaneous output to headset and speakes. This used to be possible
with PulseAudio but that is no longer an option on many distros. The
mobo is an Asus x870e creator (of headaches & diarrhea).


--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: alt.os.linux

On Sun, 14 Sep 2025 21:07:46 -0400, bad sector wrote:

I have onboard HD-Audio + a Xonar sound card but I cannot get
simultaneous output to headset and speakes. This used to be possible
with PulseAudio but that is no longer an option on many distros. The
mobo is an Asus x870e creator (of headaches & diarrhea).

This is possible with a PipeWire patching tool like qpwgraph, but I didnrCOt know of any control-panel-type function that would set it up persistently.
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: alt.os.linux

On Fri, 19 Sep 2025 02:03:10 -0000 (UTC), Lawrence DrCOOliveiro <ldo@nz.invalid> wrote in <10aidku$7uud$1@dont-email.me>:

On Sun, 14 Sep 2025 21:07:46 -0400, bad sector wrote:

I have onboard HD-Audio + a Xonar sound card but I cannot get
simultaneous output to headset and speakes. This used to be possible
with PulseAudio but that is no longer an option on many distros. The
mobo is an Asus x870e creator (of headaches & diarrhea).

This is possible with a PipeWire patching tool like qpwgraph, but I
didnrCOt know of any control-panel-type function that would set it up persistently.

ChatGPT is pretty good at coming up with the systemd unit files to make
this work, as well as the commands.

I use fancy audio routing so chat notification bells and such don't get
routed onto a stream. The trick there is to set up a null sink, route everything you want to stream (or hear) there, then use a loopback to
monitor the null sink.

Two shell scripts for your inspection:

#!/bin/bash

# Create a virtual sink (output)
pactl load-module module-null-sink \
sink_name=virt_out \
sink_properties='device.description=Virtual\ Output'

# See it:
pactl list short sinks
# YourCOll also get a source named: virt_out.monitor

% cut here %<--

Then:

#!/bin/bash
# Get your current default real sink (speakers)
SPEAKERS=$(pactl info | awk -F': ' '/Default Sink:/ {print $2}')

# Loopback the null-sink monitor raA speakers
pactl load-module module-loopback \
source=virt_out.monitor \
sink="$SPEAKERS" \
latency_msec=10

% cut here %<--

Then use pavucontrol to route your audio however you like it. For
your case, you might need another loopback to drive the second set of
speakers, so you'd use the second script again with
'sink="$OTHER_SPEAKERS"'.

Note that I'm using pulse-pipewire, and it works here:

https://imgur.com/a/8MfptB1

...album showing three images, including qpwgraph -- where I
tried to align the blocks to show how the signal gets to my speakers
from vlc.

Incidentally, this null/loopback interface setup is exactly why
I moved all my streaming to Linux. Pulseaudio (and pipewire)
are extremely flexible -- try doing this kind of audio routing
with "The Leading Brand". ;)
--
-v System76 Thelio Mega v1.1 x86_64 NVIDIA RTX 3090Ti 24G
OS: Linux 6.16.7 D: Mint 22.2 DE: Xfce 4.18
NVIDIA: 580.82.09 Mem: 258G
"Should I weed the lawn or say it's a garden?"
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: alt.os.linux

On 9/18/25 10:03 PM, Lawrence DrCOOliveiro wrote:

On Sun, 14 Sep 2025 21:07:46 -0400, bad sector wrote:

I have onboard HD-Audio + a Xonar sound card but I cannot get
simultaneous output to headset and speakes. This used to be possible
with PulseAudio but that is no longer an option on many distros. The
mobo is an Asus x870e creator (of headaches & diarrhea).

This is possible with a PipeWire patching tool like qpwgraph, but I didnrCOt know of any control-panel-type function that would set it up persistently.

Thanks for the help, I've been getting a little more involved than
anticipated :-)

At first I couldn't get qpwgraph for tumbleweed but now 0.8.1-1.2-x86_64
is installed, also asked chatgpt how to do it. It came back with a long
list of options the shortest and simplest one of which was if using
pipewire AND helvum or qjackctl as well as both the onboard sound AND
the Xonar card (surprised that qpwgraph isn't in there but I assume it's
just as good). So I plug the headphones into the motherboard back-panel 'Line-Out' and the speakers into the Xonar card 'Spkrs' port.

Problem is I don't know what to do with these connections and I get
sound out of the speakers only. Parked it for the night. Not knowing
what was going on I next tried only the onboard audio ports but still
NOTHING from the motherboard! I've seen this *before* with this Asus *shitboard*: as you plug the mini audio jack in there is a milisecond of
what sounds like music but that's all, something automatic is instantly killing it. Some have suggested that it is falsely detecting another
port being in use. I have no idea what if anything it is detecting, my headphone is wireless Sennheiser.

Went into BIOS and verified that the onboard HD-Audio is in fact still 'enabled'. Parked it for another night. Just now as a last resort I
returned into BIOS where the audio feature is STILL enabled but not
working and set the BIOS back to defalts, something I hate doing because
the defaults are so far removed from what I want that doing so each time
needs to be followed by 10 minutes of re-tweaking.

So lo and behold the onboard audio port works again. Parking it for
another night.




--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: alt.os.linux

On Mon, 22 Sep 2025 20:30:37 -0400, bad sector wrote:

At first I couldn't get qpwgraph for tumbleweed but now 0.8.1-1.2-x86_64
is installed, also asked chatgpt how to do it. It came back with a long
list of options ...

I donrCOt know what werCOre supposed to do about this. The AI gives you an answer, but itrCOs not enough? You want us to say whether the AI is right or wrong? Why not just ask the AI itself, while yourCOre at it?
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: alt.os.linux

On 9/23/25 3:44 AM, Lawrence DrCOOliveiro wrote:

On Mon, 22 Sep 2025 20:30:37 -0400, bad sector wrote:

At first I couldn't get qpwgraph for tumbleweed but now 0.8.1-1.2-x86_64
is installed, also asked chatgpt how to do it. It came back with a long
list of options ...

I donrCOt know what werCOre supposed to do about this. The AI gives you an answer, but itrCOs not enough? You want us to say whether the AI is right or wrong? Why not just ask the AI itself, while yourCOre at it?

I just didn't want to give the impression of someone who asks a question
and then never comes back. Anyway, all my related problenms seem to
arise from various hardware/software devs decision to make default the
muting of speaker ports when headphones are plugged in without providing controls suitable for average users to unmute them if they want to use
several output ports. Seems to me that much like a printer when sound
hardware is plugged in it's because the user wants to hear it (or at
least have the controls to select it for simultaneous use).

So on a hunch I tried ChatGPT with this question:

"On an Asus x870e proart creator motherboard using linux with pipewire
and qpwgraph and an Asus Xonar sound card, how to prevent the use of a headphone audio port on the sound-card from blocking simultaneous use of
the 'speakers' port of the sound-card?"

It came back with a multiple step procedure the first two of which don't
work so the rest can't either. I disabled onboard HD-Audio and went at it...


Step 1: Check ALSA auto-mute setting

# alsamixer
Select your Xonar card (press F6 to choose the card).
Look for Auto-Mute, Jack Detection, or Headphone Jack Sense.

There's not one such entry there.



Step 2: Expose both ports to PipeWire

#pw-cli ls Node | grep -A20 Xonar
(this one returns nothing)

#pw-cli
Welcome to PipeWire version 1.4.8. Type 'help' for usage.

remote 0 is named 'pipewire-0'

Error: "unsupported type PipeWire:Interface:SecurityContext"
Error: "unsupported type PipeWire:Interface:Profiler"

#pipewire-0 list sinks | grep -A20 'Xonar'
pipewire-0: command not found


..OR..

#pactl list sinks | grep -A20 'Xonar'
(this one returns nothing)

but using capitalised XONAR does:

https://paste.opensuse.org/pastes/5dd98b9ece6b

still there's nothing about speaker or headphones in that.

Moreover at this stage the sound card isn't producing any sound at all
on speakers OR in headphones together or separately.

That's it for tonight, tomorrow I'll maybe try the ChatGPT bit using
only the on-board audio.



--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: alt.os.linux

On Tue, 23 Sep 2025 22:21:52 -0400, bad sector <forgetski@_INVALID.net>
wrote in <EOScnTgqLsUiyE71nZ2dnZfqnPqdnZ2d@giganews.com>:

On 9/23/25 3:44 AM, Lawrence DrCOOliveiro wrote:

On Mon, 22 Sep 2025 20:30:37 -0400, bad sector wrote:

At first I couldn't get qpwgraph for tumbleweed but now 0.8.1-1.2-x86_64 >>> is installed, also asked chatgpt how to do it. It came back with a long
list of options ...

I donrCOt know what werCOre supposed to do about this. The AI gives you an >> answer, but itrCOs not enough? You want us to say whether the AI is right or >> wrong? Why not just ask the AI itself, while yourCOre at it?

I just didn't want to give the impression of someone who asks a question
and then never comes back. Anyway, all my related problenms seem to
arise from various hardware/software devs decision to make default the muting of speaker ports when headphones are plugged in without providing controls suitable for average users to unmute them if they want to use several output ports. Seems to me that much like a printer when sound hardware is plugged in it's because the user wants to hear it (or at
least have the controls to select it for simultaneous use).

So on a hunch I tried ChatGPT with this question:

"On an Asus x870e proart creator motherboard using linux with pipewire
and qpwgraph and an Asus Xonar sound card, how to prevent the use of a headphone audio port on the sound-card from blocking simultaneous use of
the 'speakers' port of the sound-card?"

It came back with a multiple step procedure the first two of which don't work so the rest can't either. I disabled onboard HD-Audio and went at it...

Step 1: Check ALSA auto-mute setting

# alsamixer
Select your Xonar card (press F6 to choose the card).
Look for Auto-Mute, Jack Detection, or Headphone Jack Sense.

There's not one such entry there.

Step 2: Expose both ports to PipeWire

#pw-cli ls Node | grep -A20 Xonar
(this one returns nothing)

#pw-cli
Welcome to PipeWire version 1.4.8. Type 'help' for usage.

remote 0 is named 'pipewire-0'

Error: "unsupported type PipeWire:Interface:SecurityContext"
Error: "unsupported type PipeWire:Interface:Profiler"

#pipewire-0 list sinks | grep -A20 'Xonar'
pipewire-0: command not found

..OR..

#pactl list sinks | grep -A20 'Xonar'
(this one returns nothing)

but using capitalised XONAR does:

https://paste.opensuse.org/pastes/5dd98b9ece6b

still there's nothing about speaker or headphones in that.

Moreover at this stage the sound card isn't producing any sound at all
on speakers OR in headphones together or separately.

That's it for tonight, tomorrow I'll maybe try the ChatGPT bit using
only the on-board audio.

If you're not getting _any_ sound out of the card, you might have
to configure it to do analog (instead of SPDIF), since the first
mode seems to be 8ch.

pavucontrol is the tool I use to set those things, under the "Configuration" tab. You may see the digital modes listed as "IEC958".

You can also select your default output for playback in the "Output Devices" tab, then try playing something. Because if you can't get sound out of the card _at all_, then it seems to me that that's the first step to
troubleshoot.
--
-v System76 Thelio Mega v1.1 x86_64 NVIDIA RTX 3090Ti 24G
OS: Linux 6.16.8 D: Mint 22.2 DE: Xfce 4.18
NVIDIA: 580.82.09 Mem: 258G
"And it's only ones and zeros."
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: alt.os.linux

On Wed, 9/24/2025 3:50 AM, vallor wrote:

On Tue, 23 Sep 2025 22:21:52 -0400, bad sector <forgetski@_INVALID.net>
wrote in <EOScnTgqLsUiyE71nZ2dnZfqnPqdnZ2d@giganews.com>:

On 9/23/25 3:44 AM, Lawrence DrCOOliveiro wrote:

On Mon, 22 Sep 2025 20:30:37 -0400, bad sector wrote:

At first I couldn't get qpwgraph for tumbleweed but now 0.8.1-1.2-x86_64 >>>> is installed, also asked chatgpt how to do it. It came back with a long >>>> list of options ...

I donrCOt know what werCOre supposed to do about this. The AI gives you an >>> answer, but itrCOs not enough? You want us to say whether the AI is right or
wrong? Why not just ask the AI itself, while yourCOre at it?

I just didn't want to give the impression of someone who asks a question
and then never comes back. Anyway, all my related problenms seem to
arise from various hardware/software devs decision to make default the
muting of speaker ports when headphones are plugged in without providing
controls suitable for average users to unmute them if they want to use
several output ports. Seems to me that much like a printer when sound
hardware is plugged in it's because the user wants to hear it (or at
least have the controls to select it for simultaneous use).

So on a hunch I tried ChatGPT with this question:

"On an Asus x870e proart creator motherboard using linux with pipewire
and qpwgraph and an Asus Xonar sound card, how to prevent the use of a
headphone audio port on the sound-card from blocking simultaneous use of
the 'speakers' port of the sound-card?"

It came back with a multiple step procedure the first two of which don't
work so the rest can't either. I disabled onboard HD-Audio and went at it... >>

Step 1: Check ALSA auto-mute setting

# alsamixer
Select your Xonar card (press F6 to choose the card).
Look for Auto-Mute, Jack Detection, or Headphone Jack Sense.

There's not one such entry there.



Step 2: Expose both ports to PipeWire

#pw-cli ls Node | grep -A20 Xonar
(this one returns nothing)

#pw-cli
Welcome to PipeWire version 1.4.8. Type 'help' for usage.

remote 0 is named 'pipewire-0'

Error: "unsupported type PipeWire:Interface:SecurityContext"
Error: "unsupported type PipeWire:Interface:Profiler"

#pipewire-0 list sinks | grep -A20 'Xonar'
pipewire-0: command not found


..OR..

#pactl list sinks | grep -A20 'Xonar'
(this one returns nothing)

but using capitalised XONAR does:

https://paste.opensuse.org/pastes/5dd98b9ece6b

still there's nothing about speaker or headphones in that.

Moreover at this stage the sound card isn't producing any sound at all
on speakers OR in headphones together or separately.

That's it for tonight, tomorrow I'll maybe try the ChatGPT bit using
only the on-board audio.

If you're not getting _any_ sound out of the card, you might have
to configure it to do analog (instead of SPDIF), since the first
mode seems to be 8ch.

pavucontrol is the tool I use to set those things, under the "Configuration" tab. You may see the digital modes listed as "IEC958".

You can also select your default output for playback in the "Output Devices" tab, then try playing something. Because if you can't get sound out of the card _at all_, then it seems to me that that's the first step to troubleshoot.

Just a random opinion, but generally on computer sound, there is
the "default but safe" sound subsystem, and then there is the "specialist"
set of options.

On Windows, there is a system mixer and one sound card is coupled to the output.
However, there is also ASIO4ALL which is the FOSS version of ASIO, whereas the owner of ASIO (Steinberg) had output redirection, whereby a stereo stream, the Left channel could come out of one speaker on the HDAudio. The Right channel could come out of the Xonar card and one of its speakers. ASIO was mainly invented
for low latency output, but it did have that redirection capability on the non-free
version.

Windows in fact, does not restrict sound to the system mixer model. I've seen claims in the past, that WinAMP had code written for output to two devices at the same time. You can bypass the system mixer model. And so it should be
on Linux, that any "default" could be bypassed. If you're willing to
write the code on Windows then, you can do just about anything.

On Linux, the best exemplar might be Ubuntu Studio, which has Jack and other stuff.
When I've had that loaded here, I couldn't find a simple enough set of documentation
to have me set something up as a demo. But if I had to guess, in terms of "breadcrumbs",
that would be the kind of orbit I would start with. You should be able to install
Jack on any other debian-related distro.

https://distrowatch.com/table.php?distribution=ubuntustudio # No breadcrumbs in package list

https://en.wikipedia.org/wiki/Ubuntu_Studio

[Jack is the only package that stands out...]

https://www.linux.com/news/ubuntu-studio-supports-serious-audio-adds-little-video-and-graphics/

"Pro audio on Linux (regardless of the distro) always comes down to two components:
the high-end audio server JACK and a low-latency kernel.
Both are available through standard Ubuntu repositories,
but Ubuntu Studio makes them the default and provides sane setups preconfigured."

Again, as a guess, HDAUdio chips are limited to driving two 1/8" jacks with 32 ohm
headphones connected. There never seemed to be a capability to drive 5.1 into pure 32 ohm loads with an HDAudio chip. This is a driver-mediated feature. (Many of the widgets have boost for output, but the chip would overheat if
all were turned on at once. Also, some motherboards only have a 5V 100mA regulator
for the HDAudio chip.)

When an OS uses a "system mixer" centric setup, it is unlikely a configuration of sound output would be set up with "two 2.0 outputs at once" as an option. You would have to check and see if maybe an ALSA modification could do something
like that.

But when splitting loads between two sound cards, that's just not a simple ALSA mod, that's going to take something more. And at a guess, that would be Jack
to do the routing (rather than write something nasty at the ALSA level).
Even if there was a ticky box when in 2.0 mode to drive
two outputs, they would be outputs coming off the same HDAudio chip.
(Two sets of 32 ohm headphones carrying the same music content.)

And if someone felt they didn't have the hardware for this, the NVidia card audio-over-HDMI is a separate audio device from the HDAudio sound chip,
so if you needed test materials for a Jack experiment, you could send a
sound channel to the speakers in your monitor. Not everyone uses the speakers in their monitor, and they may not know they exist.

I even have a stereo output jack on my HDMI to VGA adapter, but that
only counts as a monitor-style audio output. The "destination" in that
case is still NVidia, as nobody knows the chip there is a "sink" for sound. There is no back channel from the chip saying it is doing that.

So you don't need to have a Xonar, to do these experiments. Look carefully
at your collection of goods, for odds and ends like that.

And Bluetooth audio does not count in this case, as we want something
akin to conventional streaming, to see if a Jack scheme might work.
Bluetooth might involve more aggravation to get running as a sink.

It took a long time, for Linux to get the "stereo, 5.1, 7.1" and so on choices, of a default system mixer model. Back before Pulseaudio and Pipewire,
people were doing custom things with ALSA config files, but this is
not for the faint of heart. In my experiments, I was busting things
more often than I was fixing them :-)

Paul
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: alt.os.linux

On Wed, 24 Sep 2025 17:42:55 -0400, Paul wrote:

On Linux, the best exemplar might be Ubuntu Studio, which has Jack
and other stuff.

PipeWire is supposed to subsume all the functionality of JACK and
PulseAudio, and extend it to video as well.

Given upstream Debian has moved to PipeWire, I imagine Ubuntu would
have done so, too.
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: alt.os.linux

On 9/24/25 3:50 AM, vallor wrote:

On Tue, 23 Sep 2025 22:21:52 -0400, bad sector <forgetski@_INVALID.net>
wrote in <EOScnTgqLsUiyE71nZ2dnZfqnPqdnZ2d@giganews.com>:

On 9/23/25 3:44 AM, Lawrence DrCOOliveiro wrote:

On Mon, 22 Sep 2025 20:30:37 -0400, bad sector wrote:

At first I couldn't get qpwgraph for tumbleweed but now 0.8.1-1.2-x86_64 >>>> is installed, also asked chatgpt how to do it. It came back with a long >>>> list of options ...

I donrCOt know what werCOre supposed to do about this. The AI gives you an >>> answer, but itrCOs not enough? You want us to say whether the AI is right or
wrong? Why not just ask the AI itself, while yourCOre at it?

I just didn't want to give the impression of someone who asks a question
and then never comes back. Anyway, all my related problenms seem to
arise from various hardware/software devs decision to make default the
muting of speaker ports when headphones are plugged in without providing
controls suitable for average users to unmute them if they want to use
several output ports. Seems to me that much like a printer when sound
hardware is plugged in it's because the user wants to hear it (or at
least have the controls to select it for simultaneous use).

So on a hunch I tried ChatGPT with this question:

"On an Asus x870e proart creator motherboard using linux with pipewire
and qpwgraph and an Asus Xonar sound card, how to prevent the use of a
headphone audio port on the sound-card from blocking simultaneous use of
the 'speakers' port of the sound-card?"

It came back with a multiple step procedure the first two of which don't
work so the rest can't either. I disabled onboard HD-Audio and went at it... >>

Step 1: Check ALSA auto-mute setting

# alsamixer
Select your Xonar card (press F6 to choose the card).
Look for Auto-Mute, Jack Detection, or Headphone Jack Sense.

There's not one such entry there.



Step 2: Expose both ports to PipeWire

#pw-cli ls Node | grep -A20 Xonar
(this one returns nothing)

#pw-cli
Welcome to PipeWire version 1.4.8. Type 'help' for usage.

remote 0 is named 'pipewire-0'

Error: "unsupported type PipeWire:Interface:SecurityContext"
Error: "unsupported type PipeWire:Interface:Profiler"

#pipewire-0 list sinks | grep -A20 'Xonar'
pipewire-0: command not found


..OR..

#pactl list sinks | grep -A20 'Xonar'
(this one returns nothing)

but using capitalised XONAR does:

https://paste.opensuse.org/pastes/5dd98b9ece6b

still there's nothing about speaker or headphones in that.

Moreover at this stage the sound card isn't producing any sound at all
on speakers OR in headphones together or separately.

That's it for tonight, tomorrow I'll maybe try the ChatGPT bit using
only the on-board audio.

If you're not getting _any_ sound out of the card, you might have
to configure it to do analog (instead of SPDIF), since the first
mode seems to be 8ch.

pavucontrol is the tool I use to set those things, under the "Configuration" tab. You may see the digital modes listed as "IEC958".

You can also select your default output for playback in the "Output Devices" tab, then try playing something. Because if you can't get sound out of the card _at all_, then it seems to me that that's the first step to troubleshoot.

Staying on the card at least until it works was a good idea. As with the onboard HD-Audio any screwup into an impossible situation just disables everything. The onboard audio was dead also and, as before, I had to
revert BIOS to defaults and then re-tweak it. Once done the card worked
but only as before i.e. EITHER headphones OR speakers and if I tried Y splitters or any such stunt it just ended up with only the left channel
in the headphones and shitty volumes.

There is NO way to disable Auto-Mute in the card with alsamixer. So I
kocked the card OFF in pavucontrol, reconnected the front panel for the on-board HD-Audio and got to work on that; first positive sign was that alsamixer CAN disable Auto-Mute but then it's pavucontrol that still
only gives you one or the other.

Next I squeezed ChatGPT for maybe half a dozen procedures all but one of
which bombed. The last one of these was in response to question:

"On an Asus x870e proart creator motherboard using linux opensuse
tumbleweed, pipewire and qpwgraph, after having disabled auto-mute in alsamixer, how to make pavu control allow selection of both headphone
and speaker output instead one or the other if there is no
"switch-on-connect =" entry in file analog-output-headphones.conf or in
file analog-output-lineout.conf?"

Only one section worked but it got me what I wanted, it should come up
with the same question by anyone but I'll paste it in here for safety:

--------------------------------
4. If still only one shows up
On some Realtek codecs, you need to override the
ALSA UCM profile so that both ports are considered
valid simultaneously.

Check:
/usr/share/alsa-card-profile/mixer/paths/
for analog-output-headphones.conf and analog-output-lineout.conf.

If those lack switch-on-connect, you can copy them into your home
directory for overrides:

~/.config/alsa-card-profile/mixer/paths/

and then edit them, e.g.:

[Jack Headphone]
required-any = any

Remove or comment out required-any = all or priority entries that force exclusivity.
--------------------------------


This (finally) got me an additional (funneling) *Analog Output Port* in pavucontrol and selecting it drives both the remote-panel headphone and
the backpanel speaker ports of the onboard HD-Audio service. Its single
volume control I can complement with knobs on the headphones and on the speakers. Thanks for the pointers! whaddawhorehouse :-)



--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: alt.os.linux

On 2025-09-24 23:42, Paul wrote:

...

And if someone felt they didn't have the hardware for this, the NVidia card audio-over-HDMI is a separate audio device from the HDAudio sound chip,
so if you needed test materials for a Jack experiment, you could send a
sound channel to the speakers in your monitor. Not everyone uses the speakers in their monitor, and they may not know they exist.

You know when you send a movie from a laptop (or media computer) to the
big display at the sitting room.
--
Cheers, Carlos.
ESEfc-Efc+, EUEfc-Efc|;
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: alt.os.linux

On 2025-09-25 23:39, bad sector wrote:

This (finally) got me an additional (funneling) *Analog Output Port* in pavucontrol and selecting it drives both the remote-panel headphone and
the backpanel speaker ports of the onboard HD-Audio service. Its single volume control I can complement with knobs on the headphones and on the speakers.-a-a Thanks for the pointers!-a whaddawhorehouse EfOe

Not trivial at all! Wow. Efai
--
Cheers, Carlos.
ESEfc-Efc+, EUEfc-Efc|;
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: alt.os.linux

On 9/25/25 6:59 AM, Carlos E.R. wrote:

On 2025-09-25 23:39, bad sector wrote:

This (finally) got me an additional (funneling) *Analog Output Port*
in pavucontrol and selecting it drives both the remote-panel headphone
and the backpanel speaker ports of the onboard HD-Audio service. Its
single volume control I can complement with knobs on the headphones
and on the speakers.-a-a Thanks for the pointers!-a whaddawhorehouse EfOe

Not trivial at all! Wow. Efai

I had previously interrogated the Asus CEO 'facility' and got NOWHERE
with them... did that like a dozen times and all I really got out of the 'facility' was that they only support winblows and then mostly w11 only.
It's peculiar that they never even mentioned chatgpt to help me out,
me who like all their customers put food on their table. Could it be
that microcancer's long claws even forbid them to do that much? Whaaaaaaatever, this sound issue was just one of many others.


--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: alt.os.linux

On 9/24/25 5:42 PM, Paul wrote:

On Wed, 9/24/2025 3:50 AM, vallor wrote:

On Tue, 23 Sep 2025 22:21:52 -0400, bad sector <forgetski@_INVALID.net>
wrote in <EOScnTgqLsUiyE71nZ2dnZfqnPqdnZ2d@giganews.com>:

On 9/23/25 3:44 AM, Lawrence DrCOOliveiro wrote:

On Mon, 22 Sep 2025 20:30:37 -0400, bad sector wrote:

At first I couldn't get qpwgraph for tumbleweed but now 0.8.1-1.2-x86_64 >>>>> is installed, also asked chatgpt how to do it. It came back with a long >>>>> list of options ...

I donrCOt know what werCOre supposed to do about this. The AI gives you an >>>> answer, but itrCOs not enough? You want us to say whether the AI is right or
wrong? Why not just ask the AI itself, while yourCOre at it?

I just didn't want to give the impression of someone who asks a question >>> and then never comes back. Anyway, all my related problenms seem to
arise from various hardware/software devs decision to make default the
muting of speaker ports when headphones are plugged in without providing >>> controls suitable for average users to unmute them if they want to use
several output ports. Seems to me that much like a printer when sound
hardware is plugged in it's because the user wants to hear it (or at
least have the controls to select it for simultaneous use).

So on a hunch I tried ChatGPT with this question:

"On an Asus x870e proart creator motherboard using linux with pipewire
and qpwgraph and an Asus Xonar sound card, how to prevent the use of a
headphone audio port on the sound-card from blocking simultaneous use of >>> the 'speakers' port of the sound-card?"

It came back with a multiple step procedure the first two of which don't >>> work so the rest can't either. I disabled onboard HD-Audio and went at it...


Step 1: Check ALSA auto-mute setting

# alsamixer
Select your Xonar card (press F6 to choose the card).
Look for Auto-Mute, Jack Detection, or Headphone Jack Sense.

There's not one such entry there.



Step 2: Expose both ports to PipeWire

#pw-cli ls Node | grep -A20 Xonar
(this one returns nothing)

#pw-cli
Welcome to PipeWire version 1.4.8. Type 'help' for usage.

remote 0 is named 'pipewire-0'

Error: "unsupported type PipeWire:Interface:SecurityContext"
Error: "unsupported type PipeWire:Interface:Profiler"

#pipewire-0 list sinks | grep -A20 'Xonar'
pipewire-0: command not found


..OR..

#pactl list sinks | grep -A20 'Xonar'
(this one returns nothing)

but using capitalised XONAR does:

https://paste.opensuse.org/pastes/5dd98b9ece6b

still there's nothing about speaker or headphones in that.

Moreover at this stage the sound card isn't producing any sound at all
on speakers OR in headphones together or separately.

That's it for tonight, tomorrow I'll maybe try the ChatGPT bit using
only the on-board audio.

If you're not getting _any_ sound out of the card, you might have
to configure it to do analog (instead of SPDIF), since the first
mode seems to be 8ch.

pavucontrol is the tool I use to set those things, under the "Configuration" >> tab. You may see the digital modes listed as "IEC958".

You can also select your default output for playback in the "Output Devices" >> tab, then try playing something. Because if you can't get sound out of the >> card _at all_, then it seems to me that that's the first step to
troubleshoot.

Just a random opinion, but generally on computer sound, there is
the "default but safe" sound subsystem, and then there is the "specialist" set of options.

On Windows, there is a system mixer and one sound card is coupled to the output.
However, there is also ASIO4ALL which is the FOSS version of ASIO, whereas the
owner of ASIO (Steinberg) had output redirection, whereby a stereo stream, the
Left channel could come out of one speaker on the HDAudio. The Right channel could come out of the Xonar card and one of its speakers. ASIO was mainly invented
for low latency output, but it did have that redirection capability on the non-free
version.

Windows in fact, does not restrict sound to the system mixer model. I've seen claims in the past, that WinAMP had code written for output to two devices at the same time. You can bypass the system mixer model. And so it should be
on Linux, that any "default" could be bypassed. If you're willing to
write the code on Windows then, you can do just about anything.

On Linux, the best exemplar might be Ubuntu Studio, which has Jack and other stuff.
When I've had that loaded here, I couldn't find a simple enough set of documentation
to have me set something up as a demo. But if I had to guess, in terms of "breadcrumbs",
that would be the kind of orbit I would start with. You should be able to install
Jack on any other debian-related distro.

https://distrowatch.com/table.php?distribution=ubuntustudio # No breadcrumbs in package list

https://en.wikipedia.org/wiki/Ubuntu_Studio

[Jack is the only package that stands out...]

https://www.linux.com/news/ubuntu-studio-supports-serious-audio-adds-little-video-and-graphics/

"Pro audio on Linux (regardless of the distro) always comes down to two components:
the high-end audio server JACK and a low-latency kernel.
Both are available through standard Ubuntu repositories,
but Ubuntu Studio makes them the default and provides sane setups preconfigured."

Again, as a guess, HDAUdio chips are limited to driving two 1/8" jacks with 32 ohm
headphones connected. There never seemed to be a capability to drive 5.1 into pure 32 ohm loads with an HDAudio chip. This is a driver-mediated feature. (Many of the widgets have boost for output, but the chip would overheat if all were turned on at once. Also, some motherboards only have a 5V 100mA regulator
for the HDAudio chip.)

When an OS uses a "system mixer" centric setup, it is unlikely a configuration
of sound output would be set up with "two 2.0 outputs at once" as an option. You would have to check and see if maybe an ALSA modification could do something
like that.

But when splitting loads between two sound cards, that's just not a simple ALSA
mod, that's going to take something more. And at a guess, that would be Jack to do the routing (rather than write something nasty at the ALSA level).
Even if there was a ticky box when in 2.0 mode to drive
two outputs, they would be outputs coming off the same HDAudio chip.
(Two sets of 32 ohm headphones carrying the same music content.)

And if someone felt they didn't have the hardware for this, the NVidia card audio-over-HDMI is a separate audio device from the HDAudio sound chip,
so if you needed test materials for a Jack experiment, you could send a
sound channel to the speakers in your monitor. Not everyone uses the speakers in their monitor, and they may not know they exist.

My video card also sends HDMI audio to the monitor but since I use my
desktop for music the monitor speakers are useless. I don't even use the
HDMI ports, only the DisplayPort ones. All I need is front stereo and a
wide spectrum (mostly a speaker limit). Monitor speakers might be ok for speech on youtube or narrow-spectrum music but that's about all. I am
very impressed though by the quality of my onboard HD-Audio which, with
less than 10% of the hardware on a sound card, delivers just about the
same (except for 1/4" jacks on some older Xonar models).

What I would like the industry to come up with is a MUCH better 1/8"
jack design. When you sweat a gallon of struggle and strife recording a 10-minute piece (and you know you may never in your life be able to do
as well again, all depending) it's no fun to have to recommence because
of one freakin' intermittent contact break! Every time the topic hits my
mind I see those female battery terminal cleaner brushes with hundreds
of stiff steel whiskers. Someting like that, maybe with less than
hundreds, but a rig that makes it impossible for less than three to be
always in solid contact unless YOU want to pull it.

I even have a stereo output jack on my HDMI to VGA adapter, but that
only counts as a monitor-style audio output. The "destination" in that
case is still NVidia, as nobody knows the chip there is a "sink" for sound. There is no back channel from the chip saying it is doing that.

So you don't need to have a Xonar, to do these experiments. Look carefully
at your collection of goods, for odds and ends like that.

And Bluetooth audio does not count in this case, as we want something
akin to conventional streaming, to see if a Jack scheme might work.
Bluetooth might involve more aggravation to get running as a sink.

It took a long time, for Linux to get the "stereo, 5.1, 7.1" and so on choices,
of a default system mixer model. Back before Pulseaudio and Pipewire,
people were doing custom things with ALSA config files, but this is
not for the faint of heart. In my experiments, I was busting things
more often than I was fixing them :-)

Paul

--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: alt.os.linux

On Fri, 9/26/2025 9:03 AM, bad sector wrote:

On 9/25/25 6:59 AM, Carlos E.R. wrote:

On 2025-09-25 23:39, bad sector wrote:

This (finally) got me an additional (funneling) *Analog Output Port* in pavucontrol and selecting it drives both the remote-panel headphone and the backpanel speaker ports of the onboard HD-Audio service. Its single volume control I can complement with knobs on the headphones and on the speakers.-a-a Thanks for the pointers!-a whaddawhorehouse EfOe

Not trivial at all! Wow. Efai

I had previously interrogated the Asus CEO 'facility' and got NOWHERE with them... did that like a dozen times and all I really got out of the 'facility' was that they only support winblows and then mostly w11 only. -aIt's peculiar that they never even mentioned chatgpt to help me out, me who like all their customers put food on their table. Could it be that microcancer's long claws even forbid them to do that much? Whaaaaaaatever, this sound issue was just one of many others.

The Asus Tech Support have helped out the occasional user.

One USENETTER had the left channel of his Front Stereo, not work.
As soon as he contacted Support, he was told that if there
were nine standoffs on the motherboard tray, actually you were
only supposed to install eight of them. The ninth metal post
would touch the Left Audio output signal and short it out.
the user removed the excess post, and bingo, working left channel.

the person answering that, didn't have to query engineering,
they already knew the design had an issue. (Modern motherboards
use paper stickers adhered to the board, to "point at" things
you should know.) A motherboard with a dis-allowed mounting post
location, would have an arrow where the post was going to line up.
One of my boards here, had four arrows on the bottom, an arrow-fest.

But other times, the Asus Support mailbox was full (and no indication
they cared one way or another). Like most things in life, quality is
variable.

Paul
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: alt.os.linux

On Fri, 26 Sep 2025 09:10:15 -0400, bad sector wrote:

What I would like the industry to come up with is a MUCH better 1/8"
jack design.

Pro-audio setups have been using XLR for decades.
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: alt.os.linux

On 9/26/25 1:24 PM, Paul wrote:

On Fri, 9/26/2025 9:03 AM, bad sector wrote:

On 9/25/25 6:59 AM, Carlos E.R. wrote:

On 2025-09-25 23:39, bad sector wrote:

This (finally) got me an additional (funneling) *Analog Output Port* in pavucontrol and selecting it drives both the remote-panel headphone and the backpanel speaker ports of the onboard HD-Audio service. Its single volume control I can complement with knobs on the headphones and on the speakers.-a-a Thanks for the pointers!-a whaddawhorehouse EfOe

Not trivial at all! Wow. Efai

I had previously interrogated the Asus CEO 'facility' and got NOWHERE with them... did that like a dozen times and all I really got out of the 'facility' was that they only support winblows and then mostly w11 only. -aIt's peculiar that they never even mentioned chatgpt to help me out, me who like all their customers put food on their table. Could it be that microcancer's long claws even forbid them to do that much? Whaaaaaaatever, this sound issue was just one of many others.

The Asus Tech Support have helped out the occasional user.

One USENETTER had the left channel of his Front Stereo, not work.
As soon as he contacted Support, he was told that if there
were nine standoffs on the motherboard tray, actually you were
only supposed to install eight of them. The ninth metal post
would touch the Left Audio output signal and short it out.
the user removed the excess post, and bingo, working left channel.

the person answering that, didn't have to query engineering,
they already knew the design had an issue. (Modern motherboards
use paper stickers adhered to the board, to "point at" things
you should know.) A motherboard with a dis-allowed mounting post
location, would have an arrow where the post was going to line up.
One of my boards here, had four arrows on the bottom, an arrow-fest.

But other times, the Asus Support mailbox was full (and no indication
they cared one way or another). Like most things in life, quality is variable.

Paul

It would be unpleasant but not a huge deal to run into fuckups on a $5
chinese printed board for a garage opener but when you're paying close
to a grand it's called a low ball ripoff. At 81 I'm left with no more
patience with such crap. To aggravate the insult, in an age when online documentation really is a zero-cost service, what they call
documentation and what could under ideal circumstances salvage
situations, doesn't even measure up to decorated asswipe. They KNOW they
have a HUGE problem, what they don't seem to have is the resolve to fix it.


--- Synchronet 3.21a-Linux NewsLink 1.2