Documents - ambisonicsfrom Norwegian Centre for Technology, Art and MusicZola2026-02-24T00:00:00+00:00https://docs.notam.no/tags/ambisonics/atom.xmlStudio 32026-02-24T00:00:00+00:002026-02-24T00:00:00+00:00
Thom Johansen (thomj@notam.no)
Niklas Adam (na@notam.no)
https://docs.notam.no/manuals/studio-3/<!--  -->
<!-- in case we are using slug -->
<!-- set webp1 = resize_image(path = p, width = w1, height = h1, format = 'webp', quality = 94) %} -->
<!-- set webp2 = resize_image(path = p, width = w2, height = h2, format = 'webp', quality = 96) %} -->
<!-- set webp3 = resize_image(path = p, width = w3, height = h3, format = 'webp', quality = 98) %} -->
<figure>
<!-- <a href="https://docs.notam.no/manuals/studio-3/AmbisonicLogo.png"><img src="https://docs.notam.no/processed_images/AmbisonicLogo.ef9857d2ad0cfc54.png" /></a> -->
<a href="https://docs.notam.no/manuals/studio-3/AmbisonicLogo.png"><img src="https://docs.notam.no/processed_images/AmbisonicLogo.ef9857d2ad0cfc54.png" /></a>
<figcaption>Figure 1.</figcaption>
</figure>
<p><img src="https://docs.notam.no/manuals/studio-3/~/Pictures/ponies.jpg" alt="alt text" /></p>
<p>Overview of Notam's studio 3</p>
<p>This document contains an overview of the setup and use of Notam's studio 3 (studio for spatial audio). It is composed of two parts, with the first detailing the hardware and the second containing a quick first use setup guide.</p>
<h1 id="technical-overview"><a class="zola-anchor" href="#technical-overview" aria-label="Anchor link for: technical-overview"><i class="icon"></i></a>
Technical overview</h1>
<p>The loudspeaker setup is a slightly irregular hemisphere consisting of 24 loudspeakers in three rings. The lower ring of 12 speakers is placed in the horizontal plane at head height.
The two additional rings placed further up consists of 8 and 4 loudspeakers respectively. In addition there is currently one subwoofer. The measured loudspeaker positions can be found in various formats in the ###Coordinates section. It is highly recommended that you utilize these if your software supports it because of the irregular distances to the top four speakers.</p>
<p>The loudspeaker type is KEF LS-50. The subwoofer (Genelec 7070A) is intended to provide the majority of the low frequency content, so the bass has been rolled off somewhat in the loudspeakers.</p>
<p>The sweet spot in the middle of these loudspeakers is at head height above the markings on the floor.</p>
<h2 id="controls"><a class="zola-anchor" href="#controls" aria-label="Anchor link for: controls"><i class="icon"></i></a>
Controls</h2>
<p>The controls you'll find on the desk are quite simple:</p>
<ul>
<li>
<p>There is a mute/panic switch on the metal box.
The 24 loudspeakers are together capable of excessive sound levels,
so keep this switch close if there is a possibility of loud mistakes being made.
This switch will disable all the loudspeakers (except the subwoofer) in all positions but the topmost/forward one.
It's also a handy way of quickly muting all loudspeakers if you do something you know will generate a loud glitch,
for example switching the sample rate.
Please make liberal use of this switch,
the amplifier is powerful enough to damage the speakers if you are supplying a loud enough signal.
Please mute all speakers whenever you leave the studio.
The mute switch currently does not mute the subwoofer,
this is normal.</p>
</li>
<li>
<p>There is a volume controller in the shape of a white round knob. This controls the level of all loudspeakers, including the subwoofer. If you find that you need an even higher level than this can provide, ask the staff. Be advised that this indicates your material is at a rather low level.</p>
</li>
</ul>
<h2 id="audio-connectivity"><a class="zola-anchor" href="#audio-connectivity" aria-label="Anchor link for: audio-connectivity"><i class="icon"></i></a>
Audio connectivity</h2>
<p>On the table you will find an RME Madiface USB. This connects to your Windows or Mac computer via USB. When using this, your computer's output channels 1-24 will be sent to the loudspeakers (see diagram on last page for channel numbering). Output 25 is sent to the subwoofer.</p>
<p>The Madiface USB is supposed to be set up as clock master to allow you to select the sample rate yourself, but thanks to limitations in the hardware, the only easily supported sample rates are 44.1 kHz and 48 kHz. If you have a good reason for requiring either 88.2 kHz or 96 kHz, tell the staff. 192 kHz is not supported.</p>
<p>The audio connections in the studio are MADI based, which makes it non-trivial to connect other hardware to the system. We do however have solutions in place, so feel free to ask. As a quick summary, we can easily accomodate seven analogue line/mic inputs (via the patch panel in the front of the room) and MADI.</p>
<h2 id="gear-list"><a class="zola-anchor" href="#gear-list" aria-label="Anchor link for: gear-list"><i class="icon"></i></a>
Gear list</h2>
<p>Here follows the list of gear making up the signal chain in studio 3 for those who are curious.</p>
<ul>
<li>
<p>RME Madiface USB. Used as an interface to the user's computer inside the studio. For more info, visit the <a href="https://www.rme-audio.de/en/products/madiface_usb.php" target="_blank" rel="noopener noreferrer">RME Madiface USB details page</a>.</p>
</li>
<li>
<p>RME Madiface XT. Connected to the Madiface USB inside the studio. Functions as a standalone volume controller and loudspeaker signal processor (bass management, etc.). More info <a href="https://rme-audio.de/madiface-xt.html" target="_blank" rel="noopener noreferrer">here</a>.</p>
</li>
<li>
<p>Ferrofish A32. Converts the MADI signal from the Madiface XT to analogue line signals. Also optionally converts any analogue signals from external equipment to digital MADI signals for use in the studio. More info <a href="https://ferrofish.com/en/products/a32-converter/" target="_blank" rel="noopener noreferrer">here</a>.</p>
</li>
<li>
<p>Sonible D:24. Amplifies the A32 line level signals and drives the 24 loudspeakers inside the studio. The subwoofer is currently active and needs no amplifier. More info <a href="https://www.sonible.com/d24/" target="_blank" rel="noopener noreferrer">here</a>.</p>
</li>
</ul>
<h1 id="first-time-use"><a class="zola-anchor" href="#first-time-use" aria-label="Anchor link for: first-time-use"><i class="icon"></i></a>
First time use</h1>
<h2 id="setting-up"><a class="zola-anchor" href="#setting-up" aria-label="Anchor link for: setting-up"><i class="icon"></i></a>
Setting up</h2>
<ol>
<li>
<p>Set the mute switch to one of the lower positions and the volume controller to a very low level (counter-clockwise).</p>
</li>
<li>
<p>Install the RME Madiface USB drivers at the <a href="https://rme-audio.de/downloads.html" target="_blank" rel="noopener noreferrer">RME download page</a>.</p>
</li>
</ol>
<blockquote class="markdown-alert-note">
<ul>
<li>feel free to do this in advance if you are comfortable doing it. If in doubt ask the staff.</li>
</ul>
</blockquote>
<!-- important=true -->
<blockquote class="markdown-alert-tip">
<ul>
<li>Of the two available drivers for Apple computers, the <strong>Driverkit</strong> one is recommended. Once the driver is properly installed and you've rebooted, the Totalmix FX window should pop up.</li>
</ul>
</blockquote>
<ol start="3">
<li>Some caveats involved with the driver install and Totalmix:
<ul>
<li>If Totalmix shows up with a dialog window saying "detected", featuring "yes" and "no" buttons, press "yes" until it goes away.</li>
</ul>
</li>
</ol>
<blockquote class="markdown-alert-important">
<ul>
<li>If you are using MacOS 10 (10.13 High Sierra or higher), read the special notice contained in the driver package for this operating system, or <em>the driver will possibly not work</em>.<br />
<a href="https://rme-audio.de/macos-usb-drivers.html" target="_blank" rel="noopener noreferrer">How to choose the right RME Driver for your Mac Operating System</a></li>
</ul>
</blockquote>
<ol start="4">
<li>
<p>In the <strong>File</strong> menu of Totalmix, press "Load Workspace" and load the <em>totalmix workspace</em> file located <a href="studio_3_totalmix.tmws" target="_blank" rel="noopener noreferrer">here</a>. Click "yes" if it asks you whether to discard all changes.</p>
</li>
<li>
<p>After switching the mute switch to the top position you should now be good to go, and all the loudspeakers in the studio will be available from your software of choice.</p>
</li>
</ol>
<p>If you have any questions about the Madiface USB setup, please check the troubleshooting section.</p>
<ol start="6">
<li>Notam's decoder files for usage with the IEM SimpleDecoder can be found <a href="https://docs.notam.no/manuals/studio-3/#iem-allrad">here</a></li>
</ol>
<h1 id="reaper-template"><a class="zola-anchor" href="#reaper-template" aria-label="Anchor link for: reaper-template"><i class="icon"></i></a>
Reaper template</h1>
<p><a href="NOTAM-IEM-TEMPLATE.zip" download="">Here</a> you will find a Reaper template for ambisonics work with the IEM Plug-in Suite.</p>
<h1 id="troubleshooting"><a class="zola-anchor" href="#troubleshooting" aria-label="Anchor link for: troubleshooting"><i class="icon"></i></a>
Troubleshooting</h1>
<ul>
<li>
<p>If you have any problems, first bring up the <strong>Fireface USB settings</strong> dialog window (should be found in your dock/task bar/system tray), and make sure the following options are set correctly (top to bottom):</p>
<ul>
<li>Output format set to "64 (32) ch." and "96K Frame"</li>
<li>Sample rate is set to either 44100 Hz or 48000 Hz</li>
<li>Clock source is set to "internal"</li>
</ul>
</li>
<li>
<p>Totalmix should be set up properly after loading the workspace file provided, but as a quick summary, all output channels from 1 to 25 should be set to 0 dB, and channel routing from software to hardware outputs should be 1:1, as it is after performing a "Straight playback" mix reset in the options menu.</p>
</li>
<li>
<p>If the volume controller or mute switch does not work, please contact the staff.</p>
</li>
</ul>
<blockquote class="markdown-alert-note">
<p>When troubleshooting, please remember to set the studio volume knob at a low level and <em>mute all speakers with the mute switch</em> whenever you are doing something drastic like restarting equipment or connecting cables. There might be no sound when you are troubleshooting, but when you find the problem or manage to properly insert that last cable you forgot, you might be surprised at how loud you were really playing...</p>
</blockquote>
<h2 id="speaker-placement-and-channel-order-drawing"><a class="zola-anchor" href="#speaker-placement-and-channel-order-drawing" aria-label="Anchor link for: speaker-placement-and-channel-order-drawing"><i class="icon"></i></a>
Speaker placement and channel order drawing</h2>
<!-- {% set image = resize_image(path="overview.png", width=1280, op="fit_width", format = 'webp', quality = 94) %} -->
<!-- <img src="{{ image.url | safe }}" class=""/> -->
<p><img src="https://docs.notam.no/manuals/studio-3/overview.png" alt="overview" />
Figure 2.</p>
<hr />
<!-- ```python -->
<!-- class ECB(object): -->
<!-- def enc(self, m, k): -->
<!-- cipher = AES.new(key(k), AES.MODE_ECB) -->
<!-- return b64encode(cipher.encrypt(pad(m.encode(encoding), -->
<!-- AES.block_size))) -->
<!-- def dec(self, c, k): -->
<!-- raw = b64decode(c) -->
<!-- cipher = AES.new(key(k), AES.MODE_ECB) -->
<!-- return unpad(cipher.decrypt(raw), AES.block_size) -->
<!-- ``` -->
<!-- ```bash -->
<!-- $ grep 22px bootstrap/css/bootstrap.css -->
<!-- $ cp -r ../phosphor/templates/categories ../phosphor/templates/tags templates -->
<!-- ``` -->
<h1 id="coordinates"><a class="zola-anchor" href="#coordinates" aria-label="Anchor link for: coordinates"><i class="icon"></i></a>
Coordinates</h1>
<p>Loudspeaker coordinates as of 29.11.2024</p>
<h2 id="aed-format"><a class="zola-anchor" href="#aed-format" aria-label="Anchor link for: aed-format"><i class="icon"></i></a>
AED format:</h2>
<blockquote class="markdown-alert-note">
<ul>
<li>Angle, Elevation, Distance</li>
<li>IRCAM Spat Compatible</li>
<li>Positive azimuths CLOCKWISE</li>
<li>0 degree azimuth is forwards (front speaker)</li>
</ul>
</blockquote>
<pre><code data-lang="json">0 0 2
30 0 2
60 0 2
90 0 2
120 0 2
150 0 2
180 0 2
-150 0 2
-120 0 2
-90 0 2
-60 0 2
-30 0 2
22.5 29.2 1.95
67.5 28.7 1.936
112.5 28.8 1.938
157.5 28 1.918
-157.5 28.8 1.931
-112.5 29 1.937
-67.5 28.8 1.939
-22.5 28.5 1.939
45 61.6 1.523
135 59.8 1.538
-135 60 1.545
-45 59.8 1.538
</code></pre>
<h2 id="json-aerc-format"><a class="zola-anchor" href="#json-aerc-format" aria-label="Anchor link for: json-aerc-format"><i class="icon"></i></a>
json AERC format</h2>
<blockquote class="markdown-alert-note">
<ul>
<li>Azimuth, Elevation, Radius, Channel</li>
<li>Negative azimuths is CLOCKWISE</li>
<li>0 degree azimuth is forwards (front speaker)</li>
</ul>
</blockquote>
<p>Created via <code>jq</code> on Notam's decoder file found <a href="https://docs.notam.no/manuals/studio-3/#iem-allrad">here</a></p>
<pre><code data-lang="sh">jq '.LoudspeakerLayout.Loudspeakers \
| sort_by(.Channel) \
| map(del(.IsImaginary, .Gain))' \
s3_iem_decoder_config_2024.json
</code></pre>
<pre><code data-lang="json">[
{
"Azimuth": 0.0,
"Elevation": 0.0,
"Radius": 2.0,
"Channel": 1
},
{
"Azimuth": -30.0,
"Elevation": 0.0,
"Radius": 2.0,
"Channel": 2
},
{
"Azimuth": -60.0,
"Elevation": 0.0,
"Radius": 2.0,
"Channel": 3
},
{
"Azimuth": -90.0,
"Elevation": 0.0,
"Radius": 2.0,
"Channel": 4
},
{
"Azimuth": -120.0,
"Elevation": 0.0,
"Radius": 2.0,
"Channel": 5
},
{
"Azimuth": -150.0,
"Elevation": 0.0,
"Radius": 2.0,
"Channel": 6
},
{
"Azimuth": -180.0,
"Elevation": 0.0,
"Radius": 2.0,
"Channel": 7
},
{
"Azimuth": 150.0,
"Elevation": 0.0,
"Radius": 2.0,
"Channel": 8
},
{
"Azimuth": 120.0,
"Elevation": 0.0,
"Radius": 2.0,
"Channel": 9
},
{
"Azimuth": 90.0,
"Elevation": 0.0,
"Radius": 2.0,
"Channel": 10
},
{
"Azimuth": 60.0,
"Elevation": 0.0,
"Radius": 2.0,
"Channel": 11
},
{
"Azimuth": 30.0,
"Elevation": 0.0,
"Radius": 2.0,
"Channel": 12
},
{
"Azimuth": -22.5,
"Elevation": 29.20000076293945,
"Radius": 1.950000047683716,
"Channel": 13
},
{
"Azimuth": -67.5,
"Elevation": 28.70000076293945,
"Radius": 1.935999989509583,
"Channel": 14
},
{
"Azimuth": -112.5,
"Elevation": 28.79999923706055,
"Radius": 1.937999963760376,
"Channel": 15
},
{
"Azimuth": -157.5,
"Elevation": 28.0,
"Radius": 1.917999982833862,
"Channel": 16
},
{
"Azimuth": 157.5,
"Elevation": 28.79999923706055,
"Radius": 1.930999994277954,
"Channel": 17
},
{
"Azimuth": 112.5,
"Elevation": 29.0,
"Radius": 1.937000036239624,
"Channel": 18
},
{
"Azimuth": 67.5,
"Elevation": 28.79999923706055,
"Radius": 1.939000010490417,
"Channel": 19
},
{
"Azimuth": 22.5,
"Elevation": 28.5,
"Radius": 1.939000010490417,
"Channel": 20
},
{
"Azimuth": -45.0,
"Elevation": 61.20000076293945,
"Radius": 1.523000001907349,
"Channel": 21
},
{
"Azimuth": -135.0,
"Elevation": 59.79999923706055,
"Radius": 1.537999987602234,
"Channel": 22
},
{
"Azimuth": 135.0,
"Elevation": 60.0,
"Radius": 1.544999957084656,
"Channel": 23
},
{
"Azimuth": 45.0,
"Elevation": 59.79999923706055,
"Radius": 1.537999987602234,
"Channel": 24
}
]
</code></pre>
<h1 id="decoder"><a class="zola-anchor" href="#decoder" aria-label="Anchor link for: decoder"><i class="icon"></i></a>
Decoder</h1>
<h2 id="iem-allrad"><a class="zola-anchor" href="#iem-allrad" aria-label="Anchor link for: iem-allrad"><i class="icon"></i></a>
IEM AllRAD</h2>
<p>Notam's decoders made with the AllRAD approach for usage with the <em>SimpleDecoder</em>:</p>
<ul>
<li><a href="studio_3_IEM_decoder_config_5o_maxrE.json" download="">5th order maxrE</a>.</li>
<li><a href="studio_3_IEM_decoder_config_5o_inphase.json" download="">5th order inphase</a>.</li>
<li><a href="studio_3_IEM_decoder_config_7o_maxrE.json" download="">7th order maxrE</a>.</li>
<li><a href="studio_3_IEM_decoder_config_7o_inphase.json" download="">7th order inphase</a>.<br />
For more information on the approach check IEM's <a href="https://plugins.iem.at/docs/allradecoder/" target="_blank" rel="noopener noreferrer">AllRADecoder Guide</a></li>
</ul>
<blockquote class="markdown-alert-important">
<p>Use the same json file with the <em>DistanceCompensator</em>. Dont forget to enable the 25th channel for the sub. And same with the <em>SimpleDecoder</em>. Set the sub channel to <strong>discrete</strong> and choose channel 25.</p>
</blockquote>
<h2 id="weights"><a class="zola-anchor" href="#weights" aria-label="Anchor link for: weights"><i class="icon"></i></a>
Weights</h2>
<p>These are two different <strong>decoder optimization strategies</strong> (also called "shelf filters" or "weighting schemes") used when decoding Ambisonics to loudspeaker arrays. They address a fundamental trade-off between <strong>spatial accuracy</strong> and <strong>perceptual quality</strong>.</p>
<hr />
<p>A basic ("sampling") Ambisonics decoder reconstructs the sound field accurately in theory, but our ears don't perceive it that way in practice. Higher-order components contribute energy that can cause tonal coloration, diffuse imaging, or even <strong>unstable/phasey</strong> perception — especially with limited loudspeaker arrays. Weightings are applied to the Ambisonic orders to compensate.</p>
<hr />
<h3 id="maxre-maximum-energy-vector"><a class="zola-anchor" href="#maxre-maximum-energy-vector" aria-label="Anchor link for: maxre-maximum-energy-vector"><i class="icon"></i></a>
MaxRE (Maximum Energy Vector)</h3>
<p><strong>MaxRE</strong> maximizes the <strong>energy vector (rE)</strong>, which is a measure of the direction from which the listener perceives the majority of sound energy. It's derived from psychoacoustics — specifically, the idea that at high frequencies, the auditory system localizes based on <strong>energy</strong> rather than pressure.</p>
<ul>
<li>Applies a cosine-based window across orders: weights are <code>cos(π·n / 2(N+1))</code> for order <em>n</em> in an order-<em>N</em> system</li>
<li>Tends to <strong>narrow the perceived source width</strong> and improve localization sharpness</li>
<li>At the cost of a slight reduction in the accuracy of the reconstructed wave field</li>
<li>Particularly important for <strong>high frequencies</strong> (above ~700 Hz), where the energy vector dominates perception</li>
<li>Gives a <strong>longer rE vector</strong> — meaning perceived sound is more directionally focused</li>
</ul>
<blockquote>
<p><strong>Think of it as:</strong> optimized for <em>where</em> the sound seems to come from.</p>
</blockquote>
<hr />
<h3 id="inphase"><a class="zola-anchor" href="#inphase" aria-label="Anchor link for: inphase"><i class="icon"></i></a>
InPhase</h3>
<p><strong>InPhase</strong> ensures that all loudspeakers in the array reproduce signals that are <strong>in phase with each other</strong> for a source at any direction. This prevents cancellation artifacts and phase conflicts between speakers.</p>
<ul>
<li>Applies heavier attenuation to higher orders than MaxRE</li>
<li>Produces a <strong>wider, more diffuse</strong> image, but with robust, stable mono-compatible output</li>
<li>Avoids negative loudspeaker contributions (no speaker ever "pulls" the image in the wrong direction)</li>
<li>Important for <strong>low frequencies</strong> and for systems where listener position varies (e.g., large audiences)</li>
</ul>
<blockquote>
<p><strong>Think of it as:</strong> optimized for <em>stability and cohesion</em> of the sound image.</p>
</blockquote>
<hr />
<h3 id="side-by-side-comparison"><a class="zola-anchor" href="#side-by-side-comparison" aria-label="Anchor link for: side-by-side-comparison"><i class="icon"></i></a>
Side-by-Side Comparison</h3>
<table><thead><tr><th>Property</th><th>Basic (no weighting)</th><th>MaxRE</th><th>InPhase</th></tr></thead><tbody>
<tr><td>Goal</td><td>Accurate wave field</td><td>Max energy vector</td><td>All speakers in phase</td></tr>
<tr><td>High-freq localization</td><td>Poor</td><td><strong>Best</strong></td><td>Moderate</td></tr>
<tr><td>Image stability</td><td>Unstable</td><td>Good</td><td><strong>Best</strong></td></tr>
<tr><td>Source width</td><td>Narrow but phasey</td><td>Moderate</td><td>Widest</td></tr>
<tr><td>Order attenuation</td><td>None</td><td>Moderate</td><td><strong>Heaviest</strong></td></tr>
<tr><td>Negative speaker gains?</td><td>Yes</td><td>Sometimes</td><td><strong>Never</strong></td></tr>
<tr><td>Best for</td><td>Analysis</td><td>Playback (HF)</td><td>Large venues / LF</td></tr>
</tbody></table>
<hr />
<!-- ### Order Attenuation -->
<!-- In Ambisonics, a sound field is encoded as a series of **spherical harmonic components**, each belonging to an **order** (0, 1, 2, 3, … up to order N for an N th-order system): -->
<!-- - **Order 0** (W) — the omnidirectional component -->
<!-- - **Order 1** (X, Y, Z) — first-order directional components -->
<!-- - **Order 2** — second-order, finer directional detail -->
<!-- - …and so on -->
<!-- Higher orders encode increasingly fine-grained spatial detail, but they also contribute more **energy and phase complexity** to the loudspeaker signals. -->
<!-- **Order attenuation** simply means **multiplying each order's components by a gain factor less than 1** before decoding — progressively reducing (or "rolling off") the contribution of higher orders. -->
<!-- It's the mechanism by which MaxRE and InPhase actually do their job. -->
<!-- For example: -->
<!-- | Order | Example Weight | -->
<!-- |---|---| -->
<!-- | 0 | 1.0 (unchanged) | -->
<!-- | 1 | 0.8 | -->
<!-- | 2 | 0.5 | -->
<!-- | 3 | 0.2 | -->
<!-- | … | … | -->
<!-- #### MaxRE vs InPhase Attenuation -->
<!-- The difference between MaxRE and InPhase is **how aggressively** this is applied: -->
<!-- - **MaxRE** uses a moderate cosine-shaped roll-off — higher orders are reduced but still meaningfully present -->
<!-- - **InPhase** uses a much steeper roll-off — higher orders are nearly zeroed out, leaving mostly the low-order terms -->
<!-- #### The Trade-off -->
<!-- More attenuation = more stable and phase-coherent playback, but less spatial resolution. You're essentially trading away the fine directional detail that higher orders provide in exchange for a cleaner, more robust output. -->
<h3 id="usage"><a class="zola-anchor" href="#usage" aria-label="Anchor link for: usage"><i class="icon"></i></a>
Usage</h3>
<p>In practice, a well-designed Ambisonics decoder uses <strong>both</strong>, applied in different frequency bands via shelf filters:</p>
<ul>
<li><strong>Low frequencies (~below 700 Hz):</strong> InPhase weighting (or basic), because phase coherence matters more perceptually</li>
<li><strong>High frequencies (~above 700 Hz):</strong> MaxRE weighting, because energy-based localization dominates</li>
</ul>
<p>This is the basis of the <strong>dual-band or "psychoacoustic" decoder</strong> described by Gerzon and later formalized in tools like the <strong>AllRAD</strong> and <strong>EPAD</strong> decoder designs.</p>
<h1 id="order-overview-0th-7th"><a class="zola-anchor" href="#order-overview-0th-7th" aria-label="Anchor link for: order-overview-0th-7th"><i class="icon"></i></a>
Order Overview (0th – 7th)</h1>
<p>In 3D (full-sphere) Ambisonics, each order N adds <strong>(2N+1)</strong> new spherical harmonic components.
The total number of channels for a complete N th-order system is <strong>(N+1)²</strong>.</p>
<h2 id="order-table"><a class="zola-anchor" href="#order-table" aria-label="Anchor link for: order-table"><i class="icon"></i></a>
Order Table</h2>
<table><thead><tr><th>Order (N)</th><th>New Components (2N+1)</th><th>Total Channels (N+1)²</th><th>ACN Channel Indices</th><th>Common Label</th><th>Spatial Detail</th></tr></thead><tbody>
<tr><td>0</td><td>1</td><td>1</td><td>0</td><td>W</td><td>Omnidirectional — no directional information</td></tr>
<tr><td>1</td><td>3</td><td>4</td><td>1–3</td><td>Y, Z, X</td><td>Front/back, up/down, left/right — basic directionality</td></tr>
<tr><td>2</td><td>5</td><td>9</td><td>4–8</td><td>V, T, R, S, U</td><td>Quadrupole patterns — moderate spatial detail</td></tr>
<tr><td>3</td><td>7</td><td>16</td><td>9–15</td><td>Q, O, M, K, L, N, P</td><td>Octupole patterns — good localization, typical VR/spatial audio target</td></tr>
<tr><td>4</td><td>9</td><td>25</td><td>16–24</td><td>—</td><td>Fine directional resolution — high-quality immersive audio</td></tr>
<tr><td>5</td><td>11</td><td>36</td><td>25–35</td><td>—</td><td>Very fine resolution — approaching perceptual limits for most listeners</td></tr>
<tr><td>6</td><td>13</td><td>49</td><td>36–48</td><td>—</td><td>Extremely fine resolution — research / high-end reproduction systems</td></tr>
<tr><td>7</td><td>15</td><td>64</td><td>49–63</td><td>—</td><td>Near-theoretical maximum perceptual resolution for human hearing</td></tr>
</tbody></table>
<h2 id="channel-labels-fuma-notation"><a class="zola-anchor" href="#channel-labels-fuma-notation" aria-label="Anchor link for: channel-labels-fuma-notation"><i class="icon"></i></a>
Channel Labels (FuMa Notation)</h2>
<p>The letters come from <strong>FuMa (Furse-Malham) notation</strong>, the older naming convention for Ambisonic channels. Each letter maps to a specific <strong>spherical harmonic shape</strong> — essentially a 3D pattern describing which directions that component is sensitive to.</p>
<hr />
<h3 id="order-0"><a class="zola-anchor" href="#order-0" aria-label="Anchor link for: order-0"><i class="icon"></i></a>
Order 0</h3>
<table><thead><tr><th>Label</th><th>Description</th></tr></thead><tbody>
<tr><td><strong>W</strong></td><td>The omnidirectional component. Picks up sound equally from all directions — like a perfect omni microphone. It's the "base" of the entire sound field.</td></tr>
</tbody></table>
<hr />
<h3 id="order-1"><a class="zola-anchor" href="#order-1" aria-label="Anchor link for: order-1"><i class="icon"></i></a>
Order 1</h3>
<table><thead><tr><th>Label</th><th>Axis</th><th>Description</th></tr></thead><tbody>
<tr><td><strong>X</strong></td><td>Front/Back</td><td>Figure-of-eight pattern pointing forward and backward</td></tr>
<tr><td><strong>Y</strong></td><td>Left/Right</td><td>Figure-of-eight pattern pointing left and right</td></tr>
<tr><td><strong>Z</strong></td><td>Up/Down</td><td>Figure-of-eight pattern pointing up and down</td></tr>
</tbody></table>
<p>These three together with W give you a complete <strong>first-order B-format</strong> signal — the classic output of a Soundfield microphone, and the foundation of all Ambisonics.</p>
<hr />
<h3 id="order-2"><a class="zola-anchor" href="#order-2" aria-label="Anchor link for: order-2"><i class="icon"></i></a>
Order 2</h3>
<table><thead><tr><th>Label</th><th>Description</th></tr></thead><tbody>
<tr><td><strong>R</strong></td><td>Vertical "axial" component — a zonal harmonic aligned to the Z axis</td></tr>
<tr><td><strong>S</strong></td><td>Diagonal pattern in the XZ plane</td></tr>
<tr><td><strong>T</strong></td><td>Diagonal pattern in the YZ plane</td></tr>
<tr><td><strong>U</strong></td><td>Horizontal "saddle" pattern — differentiates front/back from left/right in the XY plane</td></tr>
<tr><td><strong>V</strong></td><td>Diagonal pattern in the XY plane, rotated 45° from U</td></tr>
</tbody></table>
<hr />
<h3 id="order-3"><a class="zola-anchor" href="#order-3" aria-label="Anchor link for: order-3"><i class="icon"></i></a>
Order 3</h3>
<table><thead><tr><th>Label</th><th>Description</th></tr></thead><tbody>
<tr><td><strong>K</strong></td><td>Zonal harmonic along Z axis (vertical, narrow lobe)</td></tr>
<tr><td><strong>L</strong></td><td>Tesseral pattern in the XZ plane</td></tr>
<tr><td><strong>M</strong></td><td>Tesseral pattern in the YZ plane</td></tr>
<tr><td><strong>N</strong></td><td>Sectoral pattern mixing X and Y axes</td></tr>
<tr><td><strong>O</strong></td><td>Sectoral pattern mixing X and Y axes, rotated</td></tr>
<tr><td><strong>P</strong></td><td>Diagonal pattern in the XZ plane</td></tr>
<tr><td><strong>Q</strong></td><td>Diagonal pattern in the YZ plane</td></tr>
</tbody></table>
<hr />
<h3 id="useful-intuition"><a class="zola-anchor" href="#useful-intuition" aria-label="Anchor link for: useful-intuition"><i class="icon"></i></a>
Useful intuition</h3>
<p>The letters themselves are somewhat <strong>arbitrary</strong> — they're just shorthand tags assigned by Furse and Malham in the 1990s. What actually matters is the <strong>spherical harmonic function</strong> each one represents, which determines its polar pattern in 3D space.</p>
<p>A useful intuition is that as you go up in order:</p>
<ul>
<li><strong>Order 0</strong> — one blob, no directionality</li>
<li><strong>Order 1</strong> — adds three figure-of-eight lobes along the three axes</li>
<li><strong>Order 2</strong> — adds five more complex "clover-leaf" and saddle shapes</li>
<li><strong>Order 3</strong> — adds seven even more intricate patterns</li>
</ul>
<p>Each new set of patterns fills in finer and finer angular detail, which is why higher orders give better spatial resolution.</p>
<p>Modern systems (AmbiX / ACN+SN3D) dropped the letter names entirely above order 1 and just use the numeric <strong>(N, M)</strong> degree/order index, which scales cleanly to any order without needing to invent new letters.</p>
<h2 id="notes"><a class="zola-anchor" href="#notes" aria-label="Anchor link for: notes"><i class="icon"></i></a>
Notes</h2>
<ul>
<li><strong>ACN</strong> (Ambisonic Channel Numbering) is the standard modern ordering, defined as <code>N² + N + M</code> for degree N, order M.</li>
<li><strong>Channel count grows quadratically</strong> — doubling the order more than quadruples the channels.</li>
<li><strong>3rd order (16ch)</strong> is the most common target for practical spatial audio (e.g. game engines, VR headsets, streaming).</li>
<li><strong>HOA (Higher Order Ambisonics)</strong> typically refers to order 2 and above.</li>
<li>Component letter names (W, X, Y, Z, etc.) are used in <strong>FuMa</strong> notation and only standardized up to order 3; higher orders are typically referred to by their ACN index or (N, M) degree/order pair.</li>
<li>The <strong>SN3D</strong> normalization scheme (used in AmbiX format) is the current standard for encoding these components.</li>
</ul>
<h1 id="recommendations"><a class="zola-anchor" href="#recommendations" aria-label="Anchor link for: recommendations"><i class="icon"></i></a>
Recommendations</h1>
<blockquote class="markdown-alert-tip">
<p>Many workflows are possible in our Spatial Audio Studio.<br />
The staff is mostly using <a href="https://www.reaper.fm/" target="_blank" rel="noopener noreferrer">Reaper</a> with the <a href="https://plugins.iem.at/" target="_blank" rel="noopener noreferrer">IEM Plug-in Suite</a> and <a href="https://supercollider.github.io/" target="_blank" rel="noopener noreferrer">SuperCollider</a> with <a href="https://www.ambisonictoolkit.net/documentation/supercollider/" target="_blank" rel="noopener noreferrer">ATK</a>.</p>
<p>In general we work with <a href="https://en.wikipedia.org/wiki/Ambisonics" target="_blank" rel="noopener noreferrer">Ambisonics</a> and <a href="https://link.springer.com/content/pdf/10.1007/978-3-030-17207-7_3.pdf" target="_blank" rel="noopener noreferrer">VBAP</a>. But if needed Dolby Atmos is possible.
^^.</p>
</blockquote>
Ambisonics Archive2025-12-03T00:00:00+00:002025-12-03T00:00:00+00:00
Niklas Adam (na@notam.no)
https://docs.notam.no/archive/ambisonics/<p>There are two work categories as of today 270226.
<em>Historical</em> (H) and <em>Contemporary</em> (C).</p>
<!-- Der er to verk kategorier per dags dato 05.06.2025. -->
<!-- *Historisk* og *Samtid*. -->
<!-- Alle verker der er notert som ***Arkivert*** de finnes i stien `Studioarkiv/Ambisonics Arkiv` på `smb://nasc.i.notam02.no` -->
<h2 id="list-of-works"><a class="zola-anchor" href="#list-of-works" aria-label="Anchor link for: list-of-works"><i class="icon"></i></a>
List of works</h2>
<table><thead><tr><th>Code</th><th>Composer</th><th>Work</th><th>Format</th><th>Year</th><th>Status</th></tr></thead><tbody>
<tr><td>H</td><td>Gunnar Sønstevold</td><td>Stormen</td><td>?</td><td>?</td><td>Archived</td></tr>
<tr><td>H</td><td>Arne Nordheim</td><td>Solitaire</td><td>5OA</td><td>?</td><td>Archived</td></tr>
<tr><td>H</td><td>Arne Nordheim</td><td>PolyPoly</td><td>?</td><td>?</td><td>Archived</td></tr>
<tr><td>H</td><td>Arne Nordheim</td><td>Vevnad</td><td>5OA</td><td>?</td><td>Archived</td></tr>
<tr><td>H</td><td>Knut Wiggen</td><td>Sommarmorgon</td><td>5OA</td><td>?</td><td>Archived</td></tr>
<tr><td>H</td><td>Knut Wiggen</td><td>Etyd</td><td>5OA</td><td>?</td><td>Archived</td></tr>
<tr><td>H</td><td>Knut Wiggen</td><td>Resa</td><td>5OA</td><td>?</td><td>Archived</td></tr>
<tr><td>H</td><td>Knut Wiggen</td><td>Massa</td><td>5OA</td><td>?</td><td>Archived</td></tr>
<tr><td>H</td><td>Knut Wiggen</td><td>EMS för sig själva</td><td>5OA</td><td>?</td><td>Archived</td></tr>
<tr><td>H</td><td>Bjørn Fongaard</td><td>Galaxe</td><td>5OA</td><td>?</td><td>Archived</td></tr>
<tr><td>H</td><td>Bjørn Fongaard</td><td>Sinfonia Miceotonalis no 1</td><td>5OA</td><td>?</td><td>Archived</td></tr>
<tr><td>H</td><td>Irma Salo Jæger/Sigurd Berge/Jan Erik Vold</td><td>Blikk</td><td>?</td><td>?</td><td>FRIN</td></tr>
<tr><td>H</td><td>Karin Krog</td><td>Santa Monica</td><td>?</td><td>?</td><td>FRIN</td></tr>
<tr><td>H</td><td>Karin Krog</td><td>For my two JB's</td><td>?</td><td>?</td><td>FRIN</td></tr>
<tr><td>H</td><td>Karin Krog</td><td>As a wife has a cow</td><td>?</td><td>?</td><td>FRIN</td></tr>
<tr><td>H</td><td>Karin Krog</td><td>Aria with Fontana Mix</td><td>?</td><td>?</td><td>FRIN</td></tr>
<tr><td>H</td><td>Jonathan Harvey</td><td>Arditti Quartet – String Quartet No. 4 (2003) With Live Electronics</td><td>5OA</td><td>?</td><td>Archived</td></tr>
<tr><td>?</td><td>Jøran Rudi</td><td>Bruitsderaison</td><td>5OA</td><td>?</td><td>FRIN</td></tr>
<tr><td>?</td><td>Jøran Rudi</td><td>Babel Study</td><td>5OA</td><td>?</td><td>FRIN</td></tr>
<tr><td>?</td><td>Geir Jonsson</td><td>Cor Mundum Crea</td><td>5OA</td><td>?</td><td>FRIN</td></tr>
<tr><td>C</td><td>Gyrid Kaldestad</td><td>Fragmented Voices</td><td>5OA</td><td>?</td><td>Archived</td></tr>
<tr><td>C</td><td>Risto Holopainen</td><td>In Vito</td><td>5OA</td><td>?</td><td>Archived</td></tr>
<tr><td>C</td><td>Bethan Kellough</td><td>Aven</td><td>TOA</td><td>?</td><td>Archived</td></tr>
<tr><td>C</td><td>Niilas</td><td>Vinter</td><td>TOA</td><td>?</td><td>Archived</td></tr>
<tr><td>C</td><td>Cristian Vogel</td><td>Mermans Lament</td><td>TOA</td><td>?</td><td>Archived</td></tr>
<tr><td>C</td><td>Idin Samimi Mofakham</td><td>...of struggle and hope</td><td>5OA</td><td>?</td><td>Archived</td></tr>
<tr><td>?</td><td>Nikolai Winge Sildre</td><td>Tanker</td><td>5OA</td><td>?</td><td>Archived</td></tr>
<tr><td>C</td><td>Ernst van der Loo</td><td>Hypnagogia</td><td>5OA</td><td>2024</td><td>Archived</td></tr>
<tr><td>?</td><td>Gisle Kverdok</td><td>Marble Surpluss</td><td>5OA</td><td>?</td><td>Archived</td></tr>
<tr><td>C</td><td>Øystein Wyller Odden</td><td>Stone Piece (Martin Ødegaard Stykekvintett)</td><td>5OA</td><td>?</td><td>Archived</td></tr>
<tr><td>C</td><td>Kevin Volans Chakra</td><td>Sisu</td><td>5OA</td><td>?</td><td>Archived</td></tr>
<tr><td>C</td><td>Stian Westerhus</td><td>Amputation</td><td>5OA</td><td>2016</td><td>Archived</td></tr>
<tr><td>C</td><td>Ola Moen</td><td>Prologue</td><td>5OA</td><td>?</td><td>Archived</td></tr>
<tr><td>C</td><td>Ruben Sverre Gjertsen</td><td>Silje Aker, Aria. Espeviken</td><td>5OA</td><td>?</td><td>Archived</td></tr>
<tr><td>C</td><td>Mariam Gviniashvili</td><td>Revelations</td><td>?</td><td>?</td><td>FRIN</td></tr>
<tr><td>C</td><td>Sara Hamming</td><td>English Speakers</td><td>5OA</td><td>?</td><td>Archived</td></tr>
<tr><td>C</td><td>Anders Vinjar</td><td>The Music is a Wave</td><td>?</td><td>?</td><td>FRIN</td></tr>
<tr><td>C</td><td>Per Martinsen (Mental Overdrive)</td><td>Waveform</td><td>?</td><td>?</td><td>FRIN</td></tr>
<tr><td>C</td><td>Tulle Ruth</td><td>Spor (rein-symfoni)</td><td>7OA</td><td>2024</td><td>Archived</td></tr>
<tr><td>C</td><td>Ruth Wilhelmine Meyer</td><td>Lokeslottet / The Loki Castle</td><td>7OA</td><td>2024</td><td>Archived</td></tr>
<tr><td>C</td><td>Natasha Barrett</td><td>Glass Eye</td><td>?</td><td>?</td><td>FRIN</td></tr>
<tr><td>C</td><td>Balint Lazcko</td><td>Burst</td><td>?</td><td>?</td><td>FRIN</td></tr>
<tr><td>C</td><td>Vilbjørg Broch</td><td>ES QC – using mathematical quasicrystals to inform spatial audio</td><td>HOA</td><td>?</td><td>Archived</td></tr>
<tr><td>C</td><td>Vilbjørg Broch</td><td>Phe noType</td><td>HOA</td><td>2024/2025</td><td>Archived</td></tr>
<tr><td>C</td><td>Vilbjørg Broch</td><td>To the wind</td><td>HOA</td><td>2024</td><td>Archived</td></tr>
<tr><td>C</td><td>Vilbjørg Broch</td><td>Netop som Digter</td><td>HOA</td><td>2025</td><td>Archived</td></tr>
<tr><td>C</td><td>Vilbjørg Broch</td><td>Alloys of the Vast 2</td><td>HOA</td><td>2025</td><td>Archived</td></tr>
<tr><td>C</td><td>Giuseppe Pisano</td><td>Anecdote_#1: a chuva, talvez __ la pluie, peut-être</td><td>7OA</td><td>?</td><td>Archived</td></tr>
<tr><td>C</td><td>Giuseppe Pisano</td><td>Scie Luminosissime</td><td>7OA</td><td>?</td><td>Archived</td></tr>
<tr><td>C</td><td>Totaleee</td><td>Non è un atlante di traiettorie algo-siderali</td><td>7OA</td><td>?</td><td>Archived</td></tr>
<tr><td>C</td><td>Totaleee</td><td>Non è un compendio di etologia numerico-digitale</td><td>6OA</td><td>?</td><td>Archived</td></tr>
<tr><td>C</td><td>Thorvald Bugge Helle</td><td>Tempting Soil</td><td>5OA</td><td>?</td><td>Archived</td></tr>
<tr><td>C</td><td>Signe Heinfelt</td><td>Fairies</td><td>5OA</td><td>?</td><td>Archived</td></tr>
<tr><td>C</td><td>Mads Kjeldgaard</td><td>I Solens Flint</td><td>7OA</td><td>?</td><td>FRIN</td></tr>
</tbody></table>
<!-- | C | Biosphere | Substrata (utvalg) | ? | ? | FRIN | -->
<!-- | C | Niklas Adam | Litho Graf | 7OA | 2026 | Archived | -->