Re: Low audio output from rear ports on new P5WD2-E Premium board

In article <44c4aa2d$0$69381$ed2619ec@xxxxxxxxxxxxxxxxxxxxxxxxxx>, James
Weston <james.p.weston@xxxxxxxxxxxxxxxxxxxxxxx> wrote:

Whenever the headphones are put into the rear jack the Realtek software
recognises that headphones have been inserted and I am given the
chance to confirm this in a dialogue box, which I then duely do.
For the frontpanel connection I have selected Legacy '97 in the bios so
that headphones are expected in this plug and no response is made from
the Realtek software in this case, which is what one would expect.

Does that enable you to say more?


What happens if you place the front panel header in HDaudio mode,
with the front panel audio cabling disconnected ? Does that change
things in any way on the back of the computer ? Does inserting a
plug into a back connector then result in an amplified signal ?

For power reasons, it would probably not be a good thing to have
the headphone amplifier enabled on all jacks. Maybe the driver
has some limits in that regard. But the specified output impedance
on that chip is so low to begin with, that at least for amplified
speakers, all the jacks should behave the same. On older audio
devices, the expected impedances might have been 32 ohms for
headphones and 600 ohms suitable for line level amplifiers. The
ALC882 is 1 ohm and 100 ohms respectively. The 100 ohms is still
too weak to give full headphone output, but should still be

If you want to be more scientific about what is going on, get
an audio tool to generate a test tone (sine wave). I've done
experiments with Audacity, and it has an option in the menu to
fill a sample window with a sine wave. You can then send that
output to your output jack. You can then read the output
amplitude with a $30 multimeter set to AC volts. Or, using
a male 1/8" stereo to male 1/8" stereo cord, you can route
the output from the jack you want to test, back to a jack
set up as an input.

The only problem I had with the version of Audacity I was
using, was I could not record and playback at the same time.
I ended up using a separate waveform recorder program to
record the signal coming back, for comparison and analysis.

Using a simple loopback plug, won't tell you how the thing
behaves under the load of headphones. You would need a
splitter adapter, that converts one stereo output, to two
stereo outputs, plug headphones into one jack, plug loopback
cable into the second, before routing the signal to an input.

Doing that kind of stuff, should give you some idea as to
whether this is an output amplitude problem, or an impedance
problem. I used this approach to analysing my Soundmax build-in
audio, which had a "muddy" sound. By placing a square wave
pulse in an audacity screen, and recording it on another
input, I was able to see that the driver was applying
reverberation. So the technique of looping the signal
back (or using a second computer, if you don't trust the
first), is a handy way to see what is actually happening.

I have not seen a utility that will dump either AC'97
standard registers or HDaudio standard registers. That
would be an even easier way to determine what is going
on. I suppose the Linux world would make accessing those
registers easier, but then you would not be able to tell
what a Windows driver was messing up by using Windows.

If you are not interested in all that nonsense, changing
driver versions is about the only other option you've
got. Either that, or get a separate sound card.