"JMartin" said:
The mobo is:
MIC2* *AGND
MICPWR* *+5VA
Line out_R* *BLINE_OUT_R
NC*
Line out_L* *BLINE_OUT_L
The plugs from the front of the case are:
M BIAS
M IN
Speaker R (one wire, two plugs)
Speaker L (one wire, two plugs)
Ground
This is how I connected the wires:
MIC2 -> M IN
MICPWR -> M BIAS
Line out_R* *BLINE_OUT_R -> R Speaker (both plugs)
NC*
Line out_L* *BLINE_OUT_L -> L Speaker (both plugs)
AGND -> Ground
Headphones work front and back. Audio works fine in the back but poor
in the front. That is, using Dragon NaturallySpeaking, the volume
check is good but fails on the sound quality part of the test. It
works fine plugging into the rear panel.
Any thoughts?
Your wiring looks OK, but I wonder if the case wiring is properly
labelled or not ? There are certain computer cases out there,
where there are wiring errors. M_IN should correspond to the
Tip of the mic jack, while M_BIAS should correspond to the Ring
contact. If you plug in a male-male 1/8" stereo cable, you should
be able to measure +5V between the middle contact (Ring) and the
contact closest to the cable (Sleeve-aka-GND), using a voltmeter.
The +5V comes from the bias signal, which is a 2K resistor to the
+5V supply on the motherboard.
To get more information about what is going on, I would want
to use a recording application that has a GUI that draws
an analog waveform of what is being recorded. Then, plug the mic
into first one jack, make a recording, and then repeat for the
second jack. To make a controlled sound source, try to record
the same source material from your stereo, holding the
microphone the same distance from the speaker for both
recordings. Then, compare the waveforms from the two recordings
and see how they differ.
If the only difference in the two recordings is the amplitude,
that implies there is some difference in the gain path between
the two microphone inputs. Changing the gain setting in the
mixer control panel will fix that. Check to see whether the
+20dB microphone boost check box has been selected or not.
If the shape of the waveforms is different, then you'll have
to guess at the cause, based on exactly how the two waveforms
differ. For example, if the waveform is clipped (flat tops on
sine waves for example), then something is limiting the signal.
There could be some kind of noise mixed with the signal. For
some of these, perhaps describing the difference between the
two recordings, or posting the recordings on the web, so
we can look at them, we might be able to give you some
feedback on a possible cause.
A web page on electret microphones claims that the MIC and
MICBIAS signals are shorted together inside an electret
microphone. If this information is in fact correct, then a
wiring error won't account for the problem. If your microphone
is dynamic, I would think reversing the wiring (MIC/MICBIAS)
would give you no sound input from the microphone. So, chances
are, something else is responsible for the difference between
the two paths.
Most likely, the Dragon application is complaining about the
amplitude (and your gain setting). I doubt the application
is clever enough to figure out anything else about the
signals.
HTH,
Paul
