My speakers don't play as loud as they used to.

Is this a LM problem? Or should I invest in a new pc sound system?

https://www.youtube.com/watch?v=qYZKT1BpQpY

--
Pity the fool who followed his GPS over the cliff

On 1/29/2024 4:40 PM, Nic wrote:
> My speakers don't play as loud as they used to.
>
> Is this a LM problem? Or should I invest in a new pc sound system?
>
> https://www.youtube.com/watch?v=qYZKT1BpQpY
>
>

Solution is simple :-)

Make a sine wave in Audacity.

Play it through the speaker.

Using a male to male 1/8" extension cable and a multimeter,
put the multimeter on the 2V AC full scale range. With all gain
controls known to you, adjust the sine wave until it reads 1.1V RMS.
On a 2000 count meter, it would be 1.100V .

This is a typical value for an HDAudio chip on LineOut. You can
track down the datasheet for your chip, and look up the full scale
output value. Compare to the multimeter.

I've had typical errors of a factor of 10. I would be wondering why
my speakers were not working right. Instead of 1100 on the meter, it
would read around 100 or ten times too small.

When this happens, then it is your job to search for a gain control
which is not properly adjusted. At one time, ALSAMixer had a couple
controls that needed to be pushed up.

And this is not using the "200% control in VLC", this is just passing
an undistorted sine wave through the plumbing, and having 1.1V come out
on LineOut. As proof you have not accidentally set a hidden control
to a too low value.

I *have* had too-low sound.

I fixed it.

Just, not a problem on LM213 right now. LM213 is fine.

*******

OK, tested mine.

LM213 fresh install (FreeFileSync project flushed).

Audacity. Sine wave 440Hz. 5 minutes worth. Amplitude set
to 1.0 instead of 0.9 or whatever the default was.

Played to LineOut. Got 1.25V RMS. A new record for the computer room.
Most chips would read 1.1V.

Plugged the audio signal into the AMD machine (LineOut to LineIn). Took a bit of
faffing about to get the recorder working. Recorded the signal from
the other machine. Carried the .m4a file back to the LM213 machine.
Loaded into Audacity. Waveform is a sine wave, with no clipping.
Which means when the 1.25V RMS signal was created, it was a sine wave
and was not a clipped sine (leading to an exaggerated true-RMS level).
If there was artificial amplification going on (the VLC "200%" kind),
then that could clip, and I saw no clipping. The 1.25V signal
is a real live sine wave.

test=pass

Note that volume controls can be logarithmic or linear, and sometimes,
it is necessary to push the slider all the way up, to get a level
you want.

Summary: I just wanted to do a quick test to indicate no panic is required.
Get your multimeter out, 2V AC full scale, and work on your LineOut level first.
They never read exactly 1 volt. The 1.1V is not a mistake, it is design intent.
I'm really surprised I got 1.25V and without distortion either. Anything
above 1.0V is a pass. They leave a little margin to ensure they can cleanly
pass the 1.0V value OK.

The three contacts on the 1/8" plug are Tip, Ring, Sleeve.
Measure from Tip to Sleeve. Put the red lead on Tip, put the
black lead on Sleeve if you want, measure the AC volts developed.

LineOut is 32 ohms nominal (it's intended to drive 32 ohm headphones
or it can drive a 10Kohm input amplifier). The actual output impedance
(the engineering quantity), is around 1 ohm. But that does not mean you
can connect a 1 ohm speaker to it. The chip really does not have the
power to drive like that. But it can push 1 volt into 32 ohms. None of
these values will cause any kind of problem with the multimeter taking a
reading. Plenty of power output for the multimeter test. Even a Simpson from 60 years
ago, will read 1.1 volts for the AC sine wave. Using your fingers to hold
the probes onto the 1/8" plug, will not affect the reading, and neither
will you be "electrocuted". I would not tell you to test this way, if there
was any danger factor involved. There are no vacuum tube voltages involved here.

Paul

On 1/29/2024 7:45 PM, Paul wrote:
> On 1/29/2024 4:40 PM, Nic wrote:
>> My speakers don't play as loud as they used to.
>>
>> Is this a LM problem? Or should I invest in a new pc sound system?
>>
>> https://www.youtube.com/watch?v=qYZKT1BpQpY
>>
>>
>

One other thing. Hollywood DVDs are recorded at -15 dB,
so yes, the level will seem too low on those. That might
be the reason VLC has a bit of gain, to try to bring that
level back up a bit.

If you rip a Hollywood title, and find the level is too low,
you can use FFMPEG to re-record the sound and amplify it a bit.
You can also load the sound track from a video into Audacity
and "normalize" it, to figure out just how much amplification
it can accept without clipping and distortion. Your Audacity
experiments in that case, tell you how much gain you can
apply in your FFMPEG command.

Paul

On Mon, 29 Jan 2024 16:40:58 -0500, Nic wrote:

> My speakers don't play as loud as they used to.
>
First, is your hearing getting worse for some reason? Do you have
something else to compare your hearing with?

What is the output sound setting in the settings applet?

Finally, if you dual boot in and out of Windows. After using windows for
my one windoze program, I have to do a Windows shutdown, or I have no
sound in _any_ Linux.

On 1/29/24 10:10 PM, azigni wrote:
> On Mon, 29 Jan 2024 16:40:58 -0500, Nic wrote:
>
>> My speakers don't play as loud as they used to.
>>
> First, is your hearing getting worse for some reason? Do you have
> something else to compare your hearing with?
>
> What is the output sound setting in the settings applet?
>
> Finally, if you dual boot in and out of Windows. After using windows for
> my one windoze program, I have to do a Windows shutdown, or I have no
> sound in _any_ Linux.
I would not discount hearing loss at my age 77. But I solved my problem
in the Alsa Mixer.

On 1/29/24 7:45 PM, Paul wrote:
> On 1/29/2024 4:40 PM, Nic wrote:
>> My speakers don't play as loud as they used to.
>>
>> Is this a LM problem? Or should I invest in a new pc sound system?
>>
>> https://www.youtube.com/watch?v=qYZKT1BpQpY
>>
>>
> Solution is simple :-)
>
> Make a sine wave in Audacity.
>
> Play it through the speaker.
>
> Using a male to male 1/8" extension cable and a multimeter,
> put the multimeter on the 2V AC full scale range. With all gain
> controls known to you, adjust the sine wave until it reads 1.1V RMS.
> On a 2000 count meter, it would be 1.100V .
>
> This is a typical value for an HDAudio chip on LineOut. You can
> track down the datasheet for your chip, and look up the full scale
> output value. Compare to the multimeter.
>
> I've had typical errors of a factor of 10. I would be wondering why
> my speakers were not working right. Instead of 1100 on the meter, it
> would read around 100 or ten times too small.
>
> When this happens, then it is your job to search for a gain control
> which is not properly adjusted. At one time, ALSAMixer had a couple
> controls that needed to be pushed up.
>
> And this is not using the "200% control in VLC", this is just passing
> an undistorted sine wave through the plumbing, and having 1.1V come out
> on LineOut. As proof you have not accidentally set a hidden control
> to a too low value.
>
> I *have* had too-low sound.
>
> I fixed it.
>
> Just, not a problem on LM213 right now. LM213 is fine.
>
> *******
>
> OK, tested mine.
>
> LM213 fresh install (FreeFileSync project flushed).
>
> Audacity. Sine wave 440Hz. 5 minutes worth. Amplitude set
> to 1.0 instead of 0.9 or whatever the default was.
>
> Played to LineOut. Got 1.25V RMS. A new record for the computer room.
> Most chips would read 1.1V.
>
> Plugged the audio signal into the AMD machine (LineOut to LineIn). Took a bit of
> faffing about to get the recorder working. Recorded the signal from
> the other machine. Carried the .m4a file back to the LM213 machine.
> Loaded into Audacity. Waveform is a sine wave, with no clipping.
> Which means when the 1.25V RMS signal was created, it was a sine wave
> and was not a clipped sine (leading to an exaggerated true-RMS level).
> If there was artificial amplification going on (the VLC "200%" kind),
> then that could clip, and I saw no clipping. The 1.25V signal
> is a real live sine wave.
>
> test=pass
>
> Note that volume controls can be logarithmic or linear, and sometimes,
> it is necessary to push the slider all the way up, to get a level
> you want.
>
> Summary: I just wanted to do a quick test to indicate no panic is required.
> Get your multimeter out, 2V AC full scale, and work on your LineOut level first.
> They never read exactly 1 volt. The 1.1V is not a mistake, it is design intent.
> I'm really surprised I got 1.25V and without distortion either. Anything
> above 1.0V is a pass. They leave a little margin to ensure they can cleanly
> pass the 1.0V value OK.
>
> The three contacts on the 1/8" plug are Tip, Ring, Sleeve.
> Measure from Tip to Sleeve. Put the red lead on Tip, put the
> black lead on Sleeve if you want, measure the AC volts developed.
>
> LineOut is 32 ohms nominal (it's intended to drive 32 ohm headphones
> or it can drive a 10Kohm input amplifier). The actual output impedance
> (the engineering quantity), is around 1 ohm. But that does not mean you
> can connect a 1 ohm speaker to it. The chip really does not have the
> power to drive like that. But it can push 1 volt into 32 ohms. None of
> these values will cause any kind of problem with the multimeter taking a
> reading. Plenty of power output for the multimeter test. Even a Simpson from 60 years
> ago, will read 1.1 volts for the AC sine wave. Using your fingers to hold
> the probes onto the 1/8" plug, will not affect the reading, and neither
> will you be "electrocuted". I would not tell you to test this way, if there
> was any danger factor involved. There are no vacuum tube voltages involved here.
>
> Paul
>

Thanks Paul, I loaded Alsa Mixer and solved my problem, there was a
slider that was not up to full. Just curious, how would a Decibel meter
hold up against the RMS Voltage method?

On 1/30/2024 6:15 AM, Nic wrote:

>
> Thanks Paul, I loaded Alsa Mixer and solved my problem,
> there was a slider that was not up to full. Just curious,
> how would a Decibel meter hold up against the RMS Voltage method?

Using a multimeter is to prove "something is working".

The voltage output of the computer is real and tangible.
It's not just colored splotches on a computer screen.

You will notice another test in the post, was this:

Computer #1 -------------------> Computer #2
LineOut LineIn

Then, see if there is a captured signal or not.

[Picture]

https://i.postimg.cc/qqdnhpG7/waveform-seen.gif

That looks like a pretty good level on LineIn. And
if you don't have a multimeter, it's the next best thing.
It assumes the second computer isn't up to any shenanigans
(gain settings nominal).

I *know* my multimeter works. My computers, I'm not quite as sure :-)

[Picture] Has an RS-232 optoisolated interface for readings...

https://i.postimg.cc/Jh9FYd8k/multimeter.gif

About all you can expect from my multimeter collection,
is an answer good to around 2-3% or so.

Paul

Glad you got it working. Remember when Linux prided itself that it could
run on any old junk, while Windoze was the picky OS? How times change.