The default stereo panner distributes 2 inputs to 2 outputs. Its
behaviour is controlled by two parameters, width and position. The
default settings for the stereo panner are width=100%,
- position=center (L=50%, R=50%). This panner assumes that you have
- 2 uncorrelated, independent signals that you wish to distribute to
- two outputs.
+ position=center (L=50%, R=50%). This panner assumes that the signals
+ you wish to distribute are either uncorrelated (that means totally
+ independent), or they contain a stereo image which is
+ mono-compatible, such as a co-incident microphone recording, or a
+ stereo image that has been created with pan pots.<sup><a href="#caveat">*</a></sup>
</p>
<div class="well">
<p>
single signal indicator marked "M" (for mono), and whose color will
change to indicate the special state.
</p>
-
<p>
It is possible to invert the outputs (see below) so that whatever
would have gone to the right channel goes to the left and vice
Mouse operations in the upper half of the panner adjust the position
parameter, constrained by the current width setting.
</p>
-
<p>
Mouse operations in the lower half of the panner adjust the width
parameter, constrained by the current position setting.
</p>
-
<p>
To change the position smoothly, press the right button and drag
within the top half of the panner, then release. The position will
be limited by the current width setting. <em>Note: you do not need
to grab the position indicator in order to drag</em>
</p>
-
<p>
To change the width smoothly, press the right button and drag
within the lower half of the panner, then release. The width will be
<dt><key>0</key></dt>
<dd>reset position to center</dd>
<dt><key class="mod2">↑</key></dt>
- <dd>reset width to 180° (100%)</dd>
+ <dd>reset width to full (100%)</dd>
</dl>
<h4>Using the scroll wheel/touch scroll</h4>
<dt>down / <key class="mod1">down</key></dt>
<dd>move position 1° / 5°to the right</dd>
</dl>
+
+<h2><a name="caveat"></a>Panning caveats</h2>
+
+<div class="well">
+Note that the stereo panner will introduce unwanted side effects on
+material that includes a time difference between the channels, such
+as AB, ORTF or NOS microphone recordings, or delay-panned mixes.<br />
+With such signals, when you reduce the with, you are summing two signals
+with different delays, which will introduce comb filtering.
+</div>
+<p>
+Let's take a look at what happens when you record a source at 45° to the
+right side with an ORTF array and then manipulate the width.
+</p>
+<p>
+For testing, we apply a pink noise signal to both inputs of an Ardour stereo
+bus with the stereo panner, and feed the bus output to a two-channel analyser.
+Since pink noise contains equal energy per octave, the readout is a straight line:
+</p>
+<img src="/images/stereo-panner-with-ORTF-fullwidth.png" />
+<p>
+To simulate an ORTF, we use Robin Gareus' stereo balance
+control LV2 to set the level difference and time delay. Ignore the Trim/Gain
+— its purpose is just to align the test signal with the 0dB line of the
+analyser.
+</p>
+<p>
+Recall that an ORTF microphone pair consists of two cardioids spaced 17 cm
+apart, with an opening angle of 110°.<br />
+For a source at 45° to the right, the time difference between the capsules
+is 350 usecs or approximately 15 samples at 44.1 kHz. The level difference
+due to the directivity of the microphones is about 7.5dB.
+</p>
+<p>
+Now for the interesting part: if we reduce the width of the signal to 50%,
+the time-delayed signals will be combined in the panner. Observe what
+happens to the frequency response of the left and right outputs:
+</p>
+<img src="/images/stereo-panner-with-ORTF-halfwidth.png" />
+<p>
+You may argue that all spaced microphone recordings will get comb filters
+later, when the two channels recombine in the air between the speakers. But
+perceptually, this is a world of difference, since our hearing system is
+very good at eliminating comb filters in the real world, if their component
+signals are spatially separated. But once you combine two delayed signals
+inside your signal chain, this spatial separation is lost. As usual, you
+get to keep the pieces.
+</p>
+<div class="well">
+Depending on your material and on how much you need to manipulate the width,
+the comb filter may be acceptable. Then again, it may not. Listen
+carefully for artefacts if you manipulate unknown stereo signals — many
+orchestra sample libraries for example do contain time-delay components.
+</div>
+