Omnissiah User Manual

1. 1Set up playback (device or loopback).
2. 2Set your input (mic) and output (speaker) devices.
3. 3Measure the room with Quick Measurement.
4. 4Review the results.
5. 5Apply the correction.

The measuring, reviewing and applying are identical either way. Only the first step differs, and the next section explains which version of it you need.

Omnissiah measures your room the same way whichever speakers you own. The only real choice is where the finished correction gets applied.

On your Tantrum device

If you have a Really Mean Sub (or another Tantrum DSP product), the correction is written into the device itself. Its own DSP chip does the work from then on. That's Path A in step 1 below.

Through loopback

If your speakers aren't Tantrum ones, Omnissiah can correct your computer's audio on its way out instead. You send your Mac's sound into Omnissiah through a virtual audio device (a small free utility such as BlackHole 2ch that acts like a cable between apps), and Omnissiah applies the correction live before passing the audio on to your speakers. That's Path B in step 1 below.

Which one you get, and how they differ

Both are produced by exactly the same measurement and correction process, and both give you the same Before / After comparison and diagnostics at the end. A device correction is the more permanent of the two: once applied it keeps working with the app closed, after a reboot, and for anything else plugged into that device. Loopback corrects only what your computer sends, and only while Omnissiah is open.

Follow either path below, whichever matches your system. Everything from step 2 onwards is the same for both.

Path A — a Tantrum device

1. 1Power on your Really Mean Sub.
2. 2Connect it to your Mac with a USB-C cable.
3. 3In Omnissiah's sidebar, under Devices, your device appears within a second or two. Until it does, the app shows “No devices found…” with “Please connect your device via USB-C cable to your computer.”

Click to zoom

If the app asks you to update the device's firmware first, follow the prompt and wait for it to finish. More detail: Chapter 5 — Managing devices.

Path B — loopback, with any speakers

1. 1Install a virtual audio device if you don't have one. The calibration setup's loopback screen has a Get BlackHole link.
2. 2In Settings → Audio → System Audio, set Loopback Input to that virtual device (BlackHole 2ch, for example).
3. 3Set your Mac's own sound output to the same virtual device (System Settings → Sound → Output), so everything your computer plays goes into Omnissiah.
4. 4Under Speaker Output in the same Audio tab, choose the real output that feeds your speakers. That's where Omnissiah sends the corrected audio.

There's no device to connect or firmware to update on this path. When you reach step 3, pick the Loopback template instead of a hardware speaker layout.

More detail: Chapter 9 — Room correction & calibration (Loopback, system-wide correction) and Chapter 12 — Settings.

Omnissiah needs to know which mic to record from and which output plays the test signals.

1. 1Go to Settings (gear, bottom of the sidebar), then the Audio tab.
2. 2Under Measurement Mic, pick your microphone, then its channel.
3. 3Under Speaker Output, pick the output device that feeds your speakers (if you set this up on Path B, it's already done).
4. 4Raise the mic input until the Mic Level meter shows healthy level (green), without clipping (red).

More detail: Chapter 12 — Settings.

1. 1Click Calibrate in the sidebar.
2. 2On Configure Your Measurement, keep Quick Measurement selected.
3. 3Under System, confirm your speaker layout and that each speaker is assigned to an output. On a Tantrum device that's a hardware layout such as Left + Right + Subwoofer; on a loopback setup, start from the Loopback template.
4. 4Click Start Measurement.
5. 5In the Speaker Playback About to Begin dialog, follow the level-setting instructions, then click I'm Ready 🤘.
6. 6Keep the mic at head height at your main listening position and stay quiet while it plays test noise through each speaker. A quick measurement takes one measurement per speaker by default, so you won't need to move the mic. If you've set more positions, the app pauses and tells you when to move it a few centimetres.

More detail: Chapter 9 — Room correction & calibration.

When measurement finishes, the app calculates correction and shows the Calibration results page.

1. 1Use the Before / After toggle to hear and see the difference.
2. 2Read any Diagnostics cards. They explain anything the correction couldn't fix and what to do about it.
3. 3When you're happy, click Apply to device, then confirm. (On loopback the same button applies the correction to your system audio.)

Click to zoom

On a device, the correction is now stored on the device and active. On loopback it starts working immediately on anything your computer plays. You can check or bypass it later under Settings → Audio → Loopback Correction. More detail: Chapter 10 — Understanding & applying results.

Once you like the result, save it. A backup holds every connected device's settings and your loopback correction, so it's worth doing on either path:

• Click Back up… at the bottom of the Settings screen and choose where to save the file. (With a device connected, the camera button next to Devices in the sidebar does the same thing.)

More detail: Chapter 11 — Backup & restore.

Your system is measured, corrected and applied. From here you can fine-tune levels and EQ by hand (Chapter 7 — Shaping sound), try a different tonal balance (Chapter 14 — Custom target curves), or learn what every screen does (Chapter 4 — Interface tour).

Copyright. © Tantrum Audio. All rights reserved. “Omnissiah”, “Really Mean Sub” and “Tantrum Audio” are names of Tantrum Audio and its products. All other trademarks are the property of their respective owners.

Disclaimer. This guide describes the app as it behaves at the version it was synced with. Features marked beta or coming soon may change. We've done our best to keep every number and menu path accurate, but if the app and this guide disagree, the app is right. Please tell us (see Appendix D — Support).

• Turn your amplifier/interface volume down first, then raise it gradually.
• Never stand close to a speaker while a sweep or test tone is playing.
• Sudden loud sounds can damage hearing and equipment. If something sounds wrong, stop the measurement (there is always a Stop button on screen).

This guide is for owners of a Tantrum Audio Really Mean Sub using the Omnissiah desktop app to set up, measure and correct their system. You don't need to be an audio engineer. Every task is written as plain, numbered steps with a picture. Sections that go deeper are clearly marked Advanced (optional). You can skip them and still get a great result.

The app's name

The app is called Omnissiah. That's the name you'll see on its icon in your Applications folder and in the dock. Some in-app text and older documents call it “DSP Control”, but it's the same app. This guide uses Omnissiah throughout.

How this guide is organised

• Part I – Overview covers what the app does and how to install it.
• Part II – Getting started is the fastest path from unboxing to good sound, plus a tour of the screen.
• Part III – Task reference has one chapter per job: managing devices, shaping sound, measuring, room correction, results, backup, settings.
• Part IV – Advanced covers fine-tuning room correction, custom target curves, diagnostics.
• Part V – Help has troubleshooting, a glossary, and reference appendices.

New users should read Parts I–II in order. If you already know your way around, jump straight to the Part III chapter for the task you want.

Conventions used here

Screenshots show the app in its dark theme (the only theme). Yellow is the app's accent colour. The highlighted control in a picture is usually the one the step refers to.

Omnissiah is the desktop app for your Really Mean Sub. Your Really Mean Sub is an active subwoofer with a built-in signal processor (DSP) and three outputs:

• one low-passed output that drives the subwoofer, and
• two high-passed outputs that feed your left and right main speakers.

Omnissiah connects to the Really Mean Sub over USB-C and lets you:

• set levels, delay, polarity, a crossover and a parametric EQ for each output;
• measure how your speakers actually sound in your room with a microphone;
• automatically calculate room correction filters that fix what the room is doing to the sound, and load them onto the device;
• back up and restore all of it.

The correction lives on the device, so once you've applied it, your system sounds corrected whether or not the app is running or your computer is connected.

A speaker that measures perfectly flat in a lab will not sound flat in your room. Walls, floor, furniture and the speaker's position create peaks (boomy notes), dips (missing notes) and timing errors. Your ears hear the room + speaker together, not the speaker alone.

Omnissiah's job is to measure that combined result and calculate the opposite (gentle EQ cuts and boosts, small delays, and level matching) so that what reaches your listening position is closer to how the music was meant to sound.

Three concepts, from biggest to smallest:

1. 1Device is your Really Mean Sub. Omnissiah discovers it automatically when you plug it in.
2. 2Outputs are the three channels on the device (the subwoofer output and the two speaker outputs). You select an output to work on it.
3. 3DSP blocks are the processing on each output, in signal order: Gain · Polarity · Mute · Delay · Crossover · Parametric EQ. Chapter 7 — Shaping sound covers each one.

Almost everything in Omnissiah follows the same loop:

1. 1Measure: play test noise through your speakers and capture the room's response with a microphone. (Chapters 8–9.)
2. 2Correct: the app calculates correction filters aimed at a target curve (the tonal balance you want). (Chapter 9.)
3. 3Review & apply: compare before/after, then load the correction onto the device. (Chapter 10.)

You can also skip measurement entirely and adjust the DSP blocks by hand (Chapter 7 — Shaping sound), which is useful for quick tweaks once you're set up.

Computer

• A Mac (macOS). Apple silicon and Intel Macs both work. A Windows version is in development.
• A free USB-C port (or an adapter) to connect the device.

Hardware

• Your Really Mean Sub, powered on, with a USB-C cable to your computer.
• For measurement and room correction: a measurement microphone connected to your Mac (typically a USB measurement mic, or a measurement mic on an audio interface). You don't need a mic to adjust levels and EQ by hand, only to measure and run room correction.

1. 1Get the Omnissiah app from Tantrum Audio (download link or supplied file).
2. 2If it arrives as a disk image or zip, open it and drag Omnissiah into your Applications folder.
3. 3Double-click Omnissiah to launch it.

When Omnissiah starts, it spends a moment getting ready. You'll see the logo and a spinner with “Getting things ready…” while it sets up its audio engine and looks for devices.

This is normal and usually quick. When it finishes, the main window appears.

Room correction needs to record from your microphone, so macOS will ask for permission the first time the app tries to use it.

1. 1When macOS shows the microphone prompt, click Allow (OK).
2. 2If you clicked Don't Allow by mistake, turn it back on in System Settings → Privacy & Security → Microphone and switch Omnissiah on.

If microphone access is blocked, Omnissiah won't start a measurement that can't record anything. It stops with a message telling you to allow the app in System Settings → Privacy & Security → Microphone and try again.

With the app installed and running, connect your device. See Chapter 3 — Quick start or Chapter 5 — Managing devices.

Omnissiah has a sidebar on the left and a content area on the right. What appears on the right depends on what you clicked in the sidebar.

Click to zoom

From top to bottom:

• Logo: click it to return to the Home screen.
• Calibrate: start or review a room measurement and correction (Chapters 9–10).
• Control: open the per-output DSP controls (Gain, Delay, Crossover, EQ) for the currently selected output (Chapter 7).
• Settings: audio devices, calibration options, connected devices, logs (Chapter 12). The active section gets a yellow bar on its left.
• Devices: a heading with a camera button beside it. The camera opens Backup & restore (Chapter 11). Below the heading is the list of outputs on your connected device(s).

The output list

Each output shows an icon, its name, and a status dot:

• a green dot means the output is live;
• a red mute icon means the output is muted.

Click an output to select it (it gets a yellow left border). The Control screen then acts on that output. Click it again to deselect. Linked outputs are stacked together in one card with a small link icon (Chapter 6).

Clicking the logo (or launching the app) shows the Welcome screen: two big shortcut cards (Calibrate and Control), plus a Room Correction section showing the status of any applied correction, and a Devices section listing connected hardware.

Closing the window (the red traffic-light button) hides Omnissiah instead of quitting it, so your device stays connected in the background. To bring it back, click its icon in the Dock or launch it again. Opening Omnissiah while it's already running reopens the existing window rather than starting a second copy. To quit properly, press Cmd-Q or choose Quit from the menu bar.

Omnissiah tells you when something goes wrong with a banner across the bottom of the window:

• ERROR: a red banner with the message and a ⓧ button to dismiss it.
• INITIALIZATION ERROR: shown if the app couldn't start up properly. This one has no dismiss button (restart the app; see Chapter 16 — Troubleshooting).

• “Getting things ready…”: the app is starting (Chapter 2).
• “No devices found…”: no device is connected, so plug one in over USB-C.
• “No Output Selected” on the Control screen: pick an output from the sidebar first.

Now that you know your way around, the chapters that follow cover each task in detail.

Omnissiah looks for Tantrum Audio DSP hardware continuously. Plug your Really Mean Sub into a USB-C port and it appears on its own. There's no “connect” button.

• While nothing is connected, you'll see “No devices found…” and “Please connect your device via USB-C cable to your computer.”
• Once connected, the device shows up under Devices on the Home screen and in Settings → Devices; its outputs appear under Devices in the sidebar.

Click to zoom

Each connected device shows its name, and a line like “DSP · current v1.2.0 · #ABC123”: the device type, its current firmware version, and its ID. On the row you'll find:

• Identify: flashes/identifies the physical device so you can tell which one you're looking at (handy if more than one is connected).
• Reset: factory-resets that device (see below).
• Update to v…: appears only when newer firmware is available (see below).

Click to zoom

Firmware is the software that runs on the device itself. Omnissiah bundles the latest firmware and updates the device over the air (OTA) via USB.

When an update is required

If a connected device's firmware is older than the app expects, the whole app is covered by a FIRMWARE UPDATE REQUIRED overlay. You must update before you can do anything else. This keeps the app and device in step.

Click to zoom

Running the update

1. 1Click Update to v… on the device row.
2. 2A confirmation asks “Are you sure you want to update … to …?” Click Confirm.
3. 3The Update Progress screen shows a bar. You'll see “Getting things ready for the update.” and then a percentage while it works.
4. 4Leave everything alone until it's done. The app reminds you: “This may take a few minutes” and “Please do not turn off the device.”
5. 5At Update Complete, the device restarts shortly on its own.

Click to zoom

A factory reset returns the device to its default settings.

1. 1Click Reset on the device row.
2. 2The reset is sent to the device, and a confirmation appears: “ID:… has been reset.” with “Changes will take effect once you restart the device.”
3. 3Click Close, then restart (power-cycle) the device.

Click to zoom

Settings → Devices → Latest firmware lists the firmware versions bundled with this version of the app, for reference. If none is bundled you'll see “No firmware bundled with this version.”

The Really Mean Sub has three outputs:

• the subwoofer output (low-passed: it only plays bass), and
• two speaker outputs (high-passed: they feed your left and right main speakers, playing everything above the crossover).

Each output has its own full set of DSP controls (Chapter 7). Output names are read from the device, and you can rename an output by clicking its name at the top of the Control page (Chapter 7). The app doesn't force fixed labels like “Ch1”.

1. 1In the sidebar, under Devices, find the output you want.
2. 2Click it. It gets a yellow left border to show it's selected.
3. 3Click Control to open its DSP controls, or click the output again to deselect.

Click to zoom

If you open Control with nothing selected, you'll see “No Output Selected”. Pick one from the sidebar first.

Your two speaker outputs usually want the same EQ, crossover and level. Rather than set each one twice, link them: adjust one and the other follows.

Create a link

1. 1Select an output and open Control.
2. 2Click Link in the header. The Link With Device window lists the other outputs.
3. 3Click the output you want to link with. The two are now grouped (they appear stacked in one card in the sidebar, with a small link icon).

Click to zoom

Choose what the link shares

By default a new link shares everything. You can make individual controls independent. For example, share EQ and crossover but keep Delay separate so you can time-align each speaker.

1. 1On a linked output, click Linked to open Linked Controls.
2. 2For each control (Mute, Gain, Delay, Polarity, EQ, Crossover), click to toggle between Shared (the control changes on every output in the link at once) and Independent (the control changes only the output you're adjusting).
3. 3Click Close.

Remove a link

On a linked output, click Unlink.

If you use a single main speaker, you can still measure and correct it. The mono option lives in the calibration setup rather than the per-output controls here: choose the Mono Speaker layout (Chapter 9).

Select an output in the sidebar (Chapter 6 — Outputs & channel setup), then click Control. You'll see the control surface for that output:

• a header with the output name, a Link button, and Load / Save;
• three cards side by side: Gain, Delay, Crossover;
• a full-width EQ section below.

Click to zoom

Every control is a number you type or step. There are no sliders, so you can dial in an exact value. The Gain and Delay cards also have − / + buttons: click one to step once, or hold it to glide the value in fine steps. A card that isn't doing anything (for example, Gain while the output is muted) dims to show it's inactive.

Changes take effect on the device immediately. There's no separate "apply" step.

The Gain card holds three things: the level itself, plus Polarity and Mute buttons in its header.

Click to zoom

Gain

• Type a level in dB, or use the − / + buttons: one click is 0.1 dB. Hold a button to glide the level smoothly by ear.
• 0 dB is unity (no change), and negative values make the output quieter.
• The available range is set by the device.

Mute

Click the Mute button (speaker icon) to silence the output. It turns red and the header shows a MUTED pill. Click again (Unmute) to restore it.

Polarity

Click the Polarity button to flip the output's polarity (also called phase invert). It turns amber and reads Polarity: inverted. Click again for Polarity: normal.

Click to zoom

Delay holds an output back in time so all your speakers reach your ears together. It's essential when the sub and mains are different distances away.

Click to zoom

• Type a delay value. The unit button cycles through units each time you click it: ms → m → samples → cm → in → ft.
• The − / + buttons step by one step of the unit currently shown (see the table below). Hold one to glide the delay in fine steps.
• The maximum delay is about 8.5 ms (≈ 410 samples, ≈ 2.8 m).
• Reset sets the delay back to 0.

The crossover decides which frequencies each output plays. On the Really Mean Sub, the subwoofer output is low-passed (only bass) and the speaker outputs are high-passed (everything above the crossover). The two meet at the crossover frequency.

Click to zoom

• Type a crossover frequency in Hz. The range is 20–200 Hz, and the Reset button returns it to 90 Hz.
• Choose an alignment from the dropdown: BW (Butterworth) or LR (Linkwitz-Riley).

Parametric EQ (PEQ) lets you cut peaks and lift dips at specific frequencies. Each output has several bands. On the Really Mean Sub the speaker outputs have up to 8 bands and the sub output has 4.

Click to zoom

There are two ways to edit: a graph at the top and control cards below. They stay in sync.

The control cards

Each band card has four fields:

Q is the width of the filter: low Q is wide and gentle, high Q is narrow and surgical.

Type offers seven filter shapes:

1. 1Low Pass: passes lows, rolls off highs.
2. 2High Pass: passes highs, rolls off lows.
3. 3Peaking: a boost or cut around a centre frequency (the everyday EQ shape).
4. 4Notch: a deep, narrow cut (to kill a resonance).
5. 5Low Shelf: lifts or lowers everything below a frequency.
6. 6High Shelf: lifts or lowers everything above a frequency.
7. 7All Pass: changes phase without changing level (advanced).

Each band also has:

• a Bypass (power) button that switches the band off (it dims and turns grey);
• a Reset button that returns that band to its defaults (asks "Are you sure you want to reset this band?").

The graph — dragging nodes

The graph shows a numbered node per band and the combined response curve.

Click to zoom

• Drag a node left/right to change its frequency, up/down to change its gain.
• Double-click a node (or Shift-click it) to reset it.
• Hover a node to see its Band number, Q, Freq and Gain.
• To change a band's Q, use the Q field on its control card.

Bypass all & Flatten

Two buttons sit at the top-right of the graph:

• Bypass all: switches every band off in one click, keeping all your settings. The button becomes Enable all. Click it to switch the bands back on, restoring exactly which bands were on before. Useful for a quick with/without comparison of your EQ.
• Flatten: resets all bands to their defaults after a confirmation ("Are you sure you want to flatten the EQ?"). Use it to start clean.

1. 1Click Calibrate in the sidebar.
2. 2On the setup page, choose Manual Measurement (Open →).

You'll see Real-Time Measurement, with the note "Play pink noise and capture the room response." If the app isn't set up yet, it shows Device Setup Required first. Pick your mic and output, then Apply (Chapter 12 — Settings).

Click to zoom

1. 1Click Start Noise. The app plays pink noise through your selected output and shows the live response. The status pill reads OK — Measuring — playing pink noise.
2. 2Watch the frequency-response chart fill in.
3. 3Click Stop Noise when you're done.

Auto Capture — one-click measurement

If you want a clean capture without watching the chart, click Auto Capture (wand). It does the whole run for you: starts the noise, waits until the measurement is stable, takes a snapshot, then stops the noise. While it waits, the status pill reads Auto Capture — Waiting for a stable measurement with a progress percentage. To cancel, click Stop Noise. No snapshot is taken.

Reading the charts

• Frequency response (top): level (dB) against frequency (about 25 Hz to 20 kHz, on a log scale). This is the shape of your sound: peaks are too-loud regions, dips are too-quiet ones.
• Coherence overlay: a coloured trace (red → yellow → green) showing how trustworthy the measurement is at each frequency. Green means reliable. Red means noise, reflections or timing problems are making that part of the measurement unreliable.
• Phase response (bottom): the timing/phase of the signal, wrapped between +180° and −180°. Where the measurement is too unreliable to give a meaningful phase reading, the line is left blank rather than drawn as noise.

Hover the chart for a readout of Freq, Gain, Phase and Confidence (the coherence at that point, as a percentage).

Click to zoom

Live diagnostics below the charts

While you measure, a collapsible Diagnostics panel appears below the charts. It runs the same room analysis as the results page, but live: plain-language cards pointing out reflections, room nulls, resonances and measurement-quality problems as the microphone hears them. Move a speaker or quieten the room and watch the cards update.

• Hover a card to shade its frequency band on the chart, so you can see exactly where the finding lives.
• Select a snapshot and the panel describes that capture instead of the live signal (the panel's title shows the snapshot's name).
• The charts and snapshot list stay pinned at the top while the diagnostics scroll beneath them.

For what each card means and what to do about it, see Chapter 15 — Diagnostics.

A snapshot freezes the current measurement so you can keep it, name it, and compare it against others (for example, left vs. right, or before vs. after moving a speaker).

Click to zoom

Capture a snapshot

1. 1While measuring, click Take Snapshot (camera). (It's greyed out until you start measuring.)
2. 2The Edit Snapshot Details window opens straight away. Give it a Name, pick a colour, and optionally set the Speaker ID (which output it belongs to). Click Update.

Or let the app do it: Auto Capture (see above) starts the noise, waits for a stable measurement, and takes the snapshot for you.

The Realtime row at the top of the snapshot list is always the live signal. Each captured snapshot below it has:

• a colour dot: click to show/hide it on the chart;
• a name: click to select it (click again to deselect);
• a red ✕: click to delete it.

Work with the selected snapshot

The Selected Snapshot bar shows the current one, with:

• Edit: rename / recolour / reassign it.
• Manual offset controls (see below).
• Reference: set a snapshot as the reference (Set as Reference) to compare others against it. Remove clears it.

Manual offsets (Delay / Gain / Polarity)

You can nudge a snapshot's timing, level and polarity to see how it would line up, without touching the device:

• Delay: + / − step by 0.5 ms (hold Shift for a fine 0.025× step), and the ↩︎ button resets to 0.
• Gain: + / − step by 0.5 dB (Shift for fine), and ↩︎ resets.
• Polarity: a single button that flips polarity (amber when inverted).

These update the chart instantly. They're a measurement aid, not a device change.

Combine, export and load

The row of buttons under the snapshots:

• Combine: sum two or more snapshots into one response (Combine Snapshots, pick at least two).
• Create Calibration: start room correction (or Align Only) from your snapshots. See Chapter 9 — Room correction & calibration.
• Export CSV / Export All: with a snapshot selected the button is Export CSV (that one snapshot). With nothing selected it's Export All (one CSV per snapshot into a folder you choose). The CSV carries the snapshot's name, colour, sample rate, latency and offsets. See Appendix B — File formats.
• Load: import one or more snapshot CSVs back in (multi-select).

Click the gear (Measurement settings) in the toolbar to open the settings panel.

Click to zoom

The panel also holds the Latency Drift Check diagnostic. See Chapter 15 — Diagnostics.

Averaging modes

Averaging controls how the live curve settles. (Stop the noise first: the dropdown is locked while a measurement runs.)

• Slow / Fast / Ultra-Fast behave like a level meter: the curve continuously eases towards the latest sound, settling over roughly 1.5 s / 500 ms / 250 ms. Slow is the steadiest and the most trustworthy.
• FIFO4 – FIFO32 average exactly the last 4–32 frames of data. The trace settles in a fixed, predictable time and then stops changing if the room does, which is good for careful before/after comparisons.

Custom graph view

Pick Custom… under Graph Scale to set your own window: the Custom Graph View dialog asks for a Scale (the ±dB window), and a From / To frequency range in Hz. Click Save to apply. While Custom is active, an Edit button appears next to the dropdown to reopen the dialog.

1. 1Configure
2. 2Set levels
3. 3Measure (one or more positions)
4. 4Calculate
5. 5Results

Omnissiah walks you through each stage with on-screen instructions. You can press Stop at any point.

Click Calibrate. On Configure Your Measurement:

• Choose your measurement. Pick one of the cards. Quick Measurement (~5 min): a handful of measurements at your listening position. The right choice for most rooms, and the one this chapter follows. Complete Measurement (~20 min) adds many more measurements that also take the rest of the room into account: a main listening area at several positions, plus optional extra areas anywhere people might be (see Complete Measurement below). Manual Measurement opens the free-form tool from Chapter 8 — Measuring your system. Previous Results reopens your last result.
• System: build your speaker layout and assign each speaker to an output.

Build your speaker layout

Under System, pick a Start from template, then adjust:

• Left + Right: two full-range speakers.
• Left + Right + Subwoofer: two speakers crossed over to a subwoofer. (The usual choice for a Really Mean Sub system.)
• Mono Speaker: a single speaker.
• Loopback: a stereo pair on a virtual audio device, for system-wide correction (see Loopback correction below).

Click to zoom

For each speaker row:

1. 1Optionally rename it.
2. 2Choose the output it's connected to (each output can be used once). A speaker with no output shows "Won't be measured."

Use Add speaker / Add subwoofer to add more, and the reset button to remove one. The counter on the right shows "{n} of {n} will be measured."

Mic positions (optional)

With Quick Measurement selected, expand Quick Measurement Settings to choose how many microphone positions to measure per speaker:

• 1: a single fixed point (fastest).
• 3: basic coverage.
• 5: thorough, a wider average.
• 7: extensive coverage.

More positions average out the room and give a more representative correction, but take longer. See Chapter 13 — Tuning room correction for guidance.

(For a Complete Measurement, positions and listening areas are set in the Configure Your Space dialog when you start. See below.)

When at least one speaker is assigned, click Start Measurement. The Speaker Playback About to Begin dialog appears. (If you chose Complete Measurement, the Configure Your Space dialog comes first, described below, and ends on this same step.)

• It tells you what's about to happen and how many speakers will be measured.
• Expand Devices & Target to confirm your Measurement Mic, Speaker Output, and Target Curve in one place (target curves are covered in Chapter 14 — Custom target curves).
• Click I'm Ready 🤘.

Now the Set levels stage runs ("First, let's set your speaker volume"):

1. 1The app briefly shows Getting your speakers ready while it mutes the outputs and clears any existing EQ, so the measurement starts from a clean slate.
2. 2It then listens to your room's noise floor. Keep the room quiet.
3. 3Then, for each speaker, raise your interface/amplifier volume until the on-screen meter turns green: clearly above the noise floor but not clipping (red).
4. 4When prompted "Happy with the levels in the room?", click Yes, continue (or No, readjust to try again).

Next the app measures each speaker ("Now we'll measure your room").

Click to zoom

1. 1Place the mic at head height at your main listening position and keep still and quiet.
2. 2The app plays test tones through each speaker back-to-back and shows the live response and its Confidence (coherence). It only pauses when the mic needs to move.
3. 3If you chose more than one position, you'll be prompted to move the mic a small distance (about 5–10 cm) to the next spot. The on-screen ∞ (figure-eight) diagram shows where. Move it, then Continue. (A Complete Measurement uses a different diagram for larger areas, described below.)
4. 4As it measures, a Diagnostics panel below the chart lists what the app is hearing: untreated dips, peaks, measurement-quality issues. Hover a card to shade its frequency band on the chart. These are the same findings you'll see on the results page, and no action is needed here.

Before it calculates, the app shows a Setup assumptions summary: the sub and top roles it detected, the crossover frequency, the measured and corrected frequency range per speaker, and how it will time-align things. Review it and click Calculate.

The Calculating room correction screen optimises EQ filters to match your target curve, one channel at a time.

Click to zoom

• The status card highlights the channel being corrected and its Generation count. The chart shows that channel's corrected curve against the target.
• A progress bar fills as it converges. This usually takes a moment.
• You can expand Best individual details to watch the filters being found.
• If you enabled Phase Correction (Chapter 13 — Tuning room correction), a Fine-tuning phase alignment pass runs after the EQ filters are found.
• If you're impatient, Skip to results stops early and uses the best result so far (letting it finish usually gives a better result).

When it's done, you're taken to the Calibration results page, covered in Chapter 10 — Understanding & applying results.

Choose Complete Measurement when more than one spot matters: a client sofa behind the desk, a whole live room. It measures a main listening area at several positions, plus optional secondary areas, and weights them together. The stages are the same as above. Only the start and the number of mic moves differ.

Clicking Start Measurement opens the Configure Your Space dialog:

1. 1Configure Your Space. Say how large your main listening area is. Single seat: one chair or desk, with a tight ∞ pattern around your head. Large area: a console, several seats or a wide sweet spot, with positions spread across it. Venue: a live room or audience area, with a rough grid over the whole space. The diagram previews the movement pattern, and Positions sets how many mic placements the main area gets (a suggested count follows the size you picked). Click Next.
2. 2Other Listening Areas (optional): add anywhere else people might be (e.g. Client Sofa). Each extra area gets a single quick measurement and counts into the result at its weight: Light — 0.25, Balanced — 0.5, Strong — 0.75, or Equal to main — 1.0. Your main area stays the priority. Remove the suggested area if you don't need one, then Continue.
3. 3Speaker Playback About to Begin: the same final check as the quick flow (speaker count, Devices & Target). Click I'm Ready 🤘.

During the measurement, each prompt is prefixed with the area it belongs to, and the placement diagram matches the area's size (∞ pattern, spread positions, or grid). Your space plan and area weights are remembered for next time.

If you only want your speakers time-, level- and polarity-aligned without EQ correction, you can. From the Manual Measurement screen (Chapter 8 — Measuring your system), select your snapshots, click Create Calibration, and choose Align Only (the button reads Align). This lines the speakers up in time and level without changing their tonal balance.

Advanced setups can route your computer's system audio through the correction in real time using a loopback device. This is set up in Settings → Audio → System Audio (Chapter 12 — Settings). Once a loopback is active, the calibration flow can target it like any other speaker. The Loopback template starts from a stereo pair, but you can Add speaker / Add subwoofer to correct more channels (e.g. left/right/sub) on a multi-channel loopback device.

Click to zoom

• The Before / After toggle at the top switches between your room's measured response and the corrected prediction. It opens on After.
• The faint yellow line is your chosen target curve, the response the correction is aiming for.
• The channel chips below the chart (one per speaker, colour-coded) let you focus a single channel. Click one to see only that channel and its correction filters (a dashed overlay), and to show only that channel's diagnostics. Click again to show all.
• Hover the chart for Freq / Gain / Phase / Confidence at any point.
• The toggle, chart and Apply to device button stay pinned at the top while the rest of the page scrolls beneath them, so the chart is always in view.

Below the chart, the Diagnostics section lists per-channel findings. These are the app being honest about what EQ can and can't do. Deep dips from room nulls, reflections, and low-confidence measurements can't be EQ'd away.

Click to zoom

Each card names the channel and frequency, explains the issue in plain language, and often suggests a fix (usually moving a speaker, the sub, or your seat, not more EQ). Hover a card to shade its frequency band on the chart above, so you can see exactly which part of the curve it's talking about. Clicking a channel chip filters the list to that channel. If everything's clean you'll see “No issues flagged — this measurement looks clean.” The full catalogue of findings is in Chapter 15 — Diagnostics.

• Room correction details: expand to see the phase response chart, each channel's applied Gain, Delay, Polarity and Phase align info, and the individual EQ filters found. (Phase-alignment entries only appear if you've enabled the optional Phase Correction setting in Settings → Calibration, covered in Chapter 13 — Tuning room correction.)
• Setup assumptions: what the app assumed about your speakers (sub/tops, crossover, measured and corrected ranges) when it calculated the correction.

When you're happy:

Click to zoom

1. 1Click Apply to device.
2. 2In Apply calibration results, check the channel-to-output list, then click Apply to device again to confirm. (This may overwrite any correction already on the device.)
3. 3A Correction applied message confirms it (“Your room correction is now active”) and returns you home.

The correction now lives on the device and works with or without the app running.

The applied-correction panel (headed Room Correction on Home and the calibration setup page, Applied Correction in Settings → Calibration) shows whether correction is active and lets you A/B it:

• A status dot reads Room correction active (green), Correction bypassed (amber), or No correction applied (grey).
• The button toggles Bypass (turn correction off) / Enable (turn it back on) so you can hear the difference on real music.

The actions on the row below the chart, to the right of the channel chips:

Click to zoom

• Calculate Again: re-run the correction (it re-confirms the setup assumptions first). Useful after you change the target curve or tuning options.
• Export results: opens a chooser with three formats (below). Pick one, then click Export. (Other destinations are coming.)
• Send Feedback: if a result doesn't look or sound right, this saves the results file and opens a pre-filled email to Tantrum Audio. Add a short note about what you noticed, then attach the saved file yourself before sending (the app tells you where it saved it). This genuinely helps us improve the correction. See Appendix D — Support.

Start from the top discards the current results and restarts calibration from the beginning (it confirms first).

One backup file holds:

• every connected device's stored settings (levels, delay, polarity, crossover, EQ);
• the app's calibration results and the target curve;
• your loopback correction, if you use one.

It does not include your measurement snapshots (export those separately, see Chapter 8 — Measuring your system).

The file is a .json file named like dsp-control-backup_2026-07-02_15-30-00.json.

There are three ways in. All do the same thing.

From the sidebar (quickest):

1. 1Click the camera button next to the Devices heading in the sidebar.
2. 2In Backup & restore, click Back Up….
3. 3Choose where to save. You'll see “Backup saved to …”.

Click to zoom

From the Settings footer: open Settings. The Back up… and Restore… buttons sit at the bottom of the window, beside Reset to factory defaults…, whichever tab you're on.

From Settings → Devices: Backup & restore → Back up → Back Up….

1. 1Open Backup & restore (sidebar camera), the Settings footer, or Settings → Devices.
2. 2Click Restore… (in Settings → Devices it's Restore → Choose File…) and pick your backup file.
3. 3The app applies it and shows a Backup restored report listing what happened to each device and to the calibration.

What restore does:

• For each device in the file that is currently connected (matched by serial number), its settings are pushed to the device.
• Devices in the file that aren't plugged in are skipped (you'll see them listed as “skipped (not connected)”). A device mid-firmware-update is skipped too.
• The app's calibration results and target curve are loaded back in.
• Loopback correction is restored.

When restore won't run

• You can't restore while a calibration is running. Stop it first.
• The app only accepts its own backup files, and can't read a backup made by a newer version of the app than you're running (update the app first).

This is where you tell the app which microphone and speakers to use, and how the audio engine runs.

Click to zoom

Devices

• Measurement Mic: “Used during calibration.” Pick your microphone, then its channel. The Mic Level meter beside it shows the incoming level: green is healthy, red is clipping.
• Speaker Output: “Plays test signals to your speakers.” Pick the audio output that feeds your speakers, then a channel option: All Channels, a pair (e.g. Ch1+Ch2), or a single channel. The 🔊 Test Speakers button plays test noise while you hold it, so you can confirm sound is coming out.

Audio Engine

• Sample Rate: 44.1 kHz up to 192 kHz. Higher captures more detail.
• Buffer Size: Auto (diagnostic), or a fixed size from 64 up to 2048 samples. Smaller means lower latency, and larger is more forgiving on slower machines. The default is 512 samples.

System Audio (loopback) — advanced

Route your computer's own audio (music, films) through room correction in real time:

• Loopback Input: the input carrying your system audio (choose -- Disable Loopback -- to turn it off). This needs a virtual audio device such as BlackHole 2ch. The calibration setup's loopback screen has a download link if you don't have one (Chapter 9 — Room correction).
• Loopback Correction: shows whether a correction is applied to that system audio, lists the active EQ bands, and gives you Bypass / Enable and Reset Correction buttons.

Click to zoom

Diagnostics

Holds the Latency Drift Check, covered in Chapter 15 — Diagnostics.

The Calibration tab holds the room-correction tuning options (Boost/Cut Trust, measurement confidence, alignment, and more) and the Target Curve picker. These are covered in Chapter 13 — Tuning room correction and Chapter 14 — Custom target curves.

Click to zoom

• Connected devices: the same device list as Home, where you identify, reset, and update firmware (Chapter 5 — Managing devices).
• Latest firmware: the firmware bundled with this app version.
• Backup & restore: save/load your settings and calibration (Chapter 11 — Backup & restore).

The Logs tab shows Activity: the 100 most recent messages reported by connected devices, newest first, each with a status dot (green, or red for an error). If there's nothing yet you'll see “No activity yet.” This is mainly useful when troubleshooting with Tantrum Audio support.

The footer's Reset to factory defaults… clears the app's saved settings (audio choices, calibration and snapshots) and restarts with defaults.

1. 1Click Reset to factory defaults….
2. 2Confirm in Reset to factory defaults?.
3. 3At Reset complete, click Close app, then reopen it to start fresh.

• Quick Positions / Full Positions: how many microphone positions to measure per speaker. More positions average out the room better but take longer. The How many positions should I use? panel explains the trade-off, and 5 is the recommended balance for most rooms.
• Clock Drift Correction: Enable only when your mic input and speaker output are on different audio interfaces. Default Disable. (Same setting as on the measurement screen: see Chapter 8 — Measuring your system.)

How the app matches level and timing between your speakers.

• Gain Estimation sets how playback level is matched: Target Level Match (default-style, matches to the target), Minimum Summed Error (Least Squares), or Matched Amplitude (RMS).
• Delay Estimation sets how the alignment delay is found: Minimum Phase Error (Least Squares) or Cross Correlation Peak (GCC-PHAT).
• Time-Align Tops: whether your left and right speakers are time-aligned to each other, or only the subwoofer is aligned to the tops. Default Subwoofer only. Choose All speakers if your left and right are at noticeably different distances.
• Phase Correction: after alignment, an extra pass fine-tunes the phase still left between each speaker and the reference, which can tighten the stereo image. Default Off. It makes calibration take noticeably longer, and it only helps where the remaining phase error is large enough to hear.

How hard and how accurately the optimiser works.

• Max Generations: Overkill (80) / Best (50) / Balanced (30) / Fastest (15). More generations = more accurate, slower.
• Fitness Threshold: Best (10%) / Balanced (20%) / Fastest (30%). Lower = more accurate, slower.
• Confidence Weighting decides which frequencies set the bar for a "trustworthy" measurement: 95% Strict (clearest bins only), 70% Balanced (the default), 50% Relaxed, 10% Permissive. Tighten to Strict if correction seems to chase measurement noise. Only relax below Balanced if correction is leaving obvious problems untreated.
• Correction Filters: the maximum number of EQ filters the correction may use per channel (a number from 1 to 48; default 24). The device itself always caps it too: a channel never gets more filters than its hardware has slots for, so the default means "use every slot the device offers". Fewer filters give a simpler, easier-to-review correction and leave slots free for manual EQ, at the cost of finer detail.
• Boost Trust sets how noisy a frequency's measurement can be before a boost stops counting fully: Off (clearest frequencies only), Slight (80%, default), Moderate (65%), Maximum (50%). Higher settings let the app boost into less-certain regions, so use them with care.
• Cut Trust applies the same idea to cuts: Slight (80%), Moderate (65%), Standard (50%, default), Widest (30%). Cutting is safer than boosting, so this can be more relaxed.
• Boost Restraint sets how strongly the optimiser is discouraged from boosting into deep, unfixable dips: Off, Gentle (recommended), Strong.
• Target Curve: the tonal balance to aim for (see Chapter 14 — Custom target curves).

These affect only the chart view, never the captured data or the correction:

• Smoothing: fractional-octave smoothing of the displayed curve (default 1/6 Oct).
• Coherence: hide low-confidence frequencies (default Low = show all).
• Graph Scale: vertical zoom, either Full Scale, ±24dB Norm, or Custom…. Custom lets you set your own ±dB window (Scale) and frequency range (From / To), and an Edit button reopens the custom view while it's active.

The full range and default tables are in Appendix A — Reference tables.

The Calibration Target picker appears wherever you set a target:

• Settings → Calibration → Target Curve → Edit…, or
• inside the Speaker Playback About to Begin dialog (Devices & Target → Target Curve → Edit…) when you start a measurement.

Click to zoom

Pick a preset card (or Custom), see it drawn on the chart, then confirm to make it the target. The chart's vertical scale is ±12 dB.

Choose the Custom card to open the point editor ("Custom target — your own control points"). It starts from whichever curve was selected, so you can use a preset as your starting point and tweak it. (A saved custom target reopens straight into the editor.)

Click to zoom

A curve is a list of control points, each a frequency and a gain. The app draws a smooth line through them.

1. 1Click + Add point to add a point (it starts at 1000 Hz / 0 dB).
2. 2For each point, set Hz (the frequency, 20 – 20,000) and dB (the gain there, −12 – +12, in 0.5 dB steps).
3. 3Remove a point with its ✕ (you must keep at least two points).
4. 4Enter points in any order. The rows stay where they are while you edit, and the app sorts the curve by frequency behind the scenes.
5. 5Confirm to make it your target.

Example — a subtle "house curve"

A mild, natural balance similar to Tantrum Flat:

A little lift under ~200 Hz, flat through the mids and top. Keep boosts gentle: the correction still has to fit them within its headroom limits.

Diagnostic cards are plain-language notes about your room and your measurement: reflections, room nulls, resonances, and measurement-quality problems. They're the app being straight with you. Some room problems can't be fixed with EQ, and it's better to know than to be misled by a "flat" graph. You'll meet them in three places:

• On the results page, after a correction: per-channel notes about anything the correction left untreated, and why. Selecting a channel on the chart filters the list to that channel.
• On the measurement screen (see Chapter 8 — Measuring your system): the same findings appear live below the charts as you measure (or for whichever snapshot you select), phrased as what the microphone is hearing right now. Move a speaker and watch them update.
• During a guided calibration's measuring step, beneath its chart.

Click to zoom

Each card shows the channel, an approximate frequency, a short title, an explanation, and often a suggested action. Hover a card to shade its frequency band on the chart above. Their colour shows severity:

• amber: a warning about measurement quality
• yellow (accent): an advisory about the room
• hairline: information (usually "left alone on purpose")

What the findings mean

The table shows the results-page wording. The live cards on the measurement screen describe the same findings as observations: "Peak at ~250 Hz" rather than "Peak at ~250 Hz — left uncorrected", since no correction has run yet.

Coherence and SNR

Two quality numbers underpin the cards and charts:

• Coherence (shown as Confidence): how much of what the mic heard is genuinely the test signal versus noise/reflections, per frequency. High is good.
• SNR (signal-to-noise ratio): how far your measurement rose above the room's noise floor overall. A low SNR is why you'll be asked to raise the level and quieten the room.

This diagnostic checks whether the round-trip measurement latency stays stable between runs. It matters when your microphone and speakers are on separate audio interfaces: their clocks can drift, which stops two separately-taken measurements lining up to a common time reference. That in turn breaks time-alignment between speakers.

You'll find it in the Measurement settings gear (see Chapter 8 — Measuring your system) and in Settings → Audio → Diagnostics.

Click to zoom

Running it

1. 1Stop any measurement first (the diagnostic drives the measurement itself).
2. 2Keep the mic and speaker fixed and the room quiet.
3. 3Choose the number of Runs (3 / 5 / 10 / 20; default 5).
4. 4Click Baseline, which runs with the audio streams kept open (your measurement noise floor), then Re-open streams, which forces a full stream re-open before each run, exposing the timing shift that separate interfaces introduce.

Reading the result

Each block reports the Latency range (and its spread) and the Datum range across runs, plus the path to a CSV written to the app's diagnostics folder.

• If Re-open streams shows a range much larger than Baseline, drift is confirmed: your input and output are on different clocks.
• The fix is to use one interface for both mic and speakers if you can, or to enable Clock Drift Correction (see Chapter 8 — Measuring your system).

The app is stuck on "Getting things ready…".

It's still initialising. Give it a few seconds. If it never finishes, quit and reopen. If it still hangs, unplug the device, reopen, then reconnect.

There's an "INITIALIZATION ERROR" banner at the bottom.

The app couldn't start its audio engine. Quit and reopen it. Check no other app is holding your audio interface exclusively, and that your interface is connected and powered before launch.

"No devices found…" — my device isn't showing up.

1. 1Check the USB-C cable is connected and the device is powered on.
2. 2Try a different cable/port (some cables are charge-only, so use a data cable).
3. 3Remember the app looks for Tantrum Audio hardware, not your audio interface. An interface won't appear here.
4. 4Reconnect. The list refreshes automatically within a second or two.

"FIRMWARE UPDATE REQUIRED" is blocking everything.

Your device's firmware is older than this app expects. Click Update to v…, confirm, and let it finish. Don't unplug or quit during the update. See Chapter 5 — Managing devices.

The firmware update showed an error.

Click Close, reconnect the device, and try again. If it keeps failing, try a different USB port/cable and make sure the device stays powered throughout.

I clicked Reset by accident.

Device Reset acts immediately (there's no confirmation before it runs) and takes effect when you restart the device. If you had a backup, restore it (see Chapter 11 — Backup & restore). Otherwise, re-run room correction.

No sound during a test or measurement.

1. 1Settings → Audio → Speaker Output: check the right output is selected. Use 🔊 Test Speakers to confirm.
2. 2Check the device/amp is on and turned up (start low).
3. 3Check the output isn't Muted on the Control screen (red MUTED pill).

The mic level meter shows nothing / stays grey.

1. 1Settings → Audio → Measurement Mic: pick your mic and the correct channel.
2. 2Grant microphone permission: System Settings → Privacy & Security → Microphone → Omnissiah on.
3. 3Raise the mic's input/preamp gain.

"Microphone access is denied" when starting a measurement or calibration.

macOS is blocking the app's microphone access, so nothing can be recorded. Turn Omnissiah on under System Settings → Privacy & Security → Microphone, then try again. Note that after updating the app, macOS may ask for microphone permission once more. That's expected, and your settings and calibrations are unaffected.

The sound stutters, crackles or clicks.

1. 1Quit and reopen the app.
2. 2Try a different Buffer Size (Settings → Audio). The default is 512 samples. Leave Auto (diagnostic) alone unless we ask you to use it.
3. 3If it's the loopback monitor clicking, make sure you're on the latest app version.

My measurement is red (low coherence / "Weak measurement").

Low coherence means noise or reflections are drowning the test signal:

1. 1Raise the playback level (safely) so the signal is well above the room noise.
2. 2Quieten the room: close windows, turn off fans/AC, don't talk.
3. 3Use a proper measurement microphone at ear height at your seat.
4. 4Re-measure. See Chapter 15 — Diagnostics.

Correction sounds over-processed / too boosty.

Set Boost Restraint → Strong and/or Boost Trust → Off (Settings → Calibration), then Calculate Again (see Chapter 13 — Tuning room correction). Lowering Correction Filters there also gives a simpler, gentler correction.

The results graph still shows big dips after correction.

Deep dips are usually room nulls or reflections, and EQ can't fix them. Check the Diagnostics cards for the specific advice (usually move the speaker, sub or your seat). See Chapter 15 — Diagnostics.

Two speakers won't line up in time / the latency seems to wander.

If your mic and speakers are on different audio interfaces, their clocks drift. Use one interface for both if you can, or enable Clock Drift Correction. Confirm with the Latency Drift Check (see Chapter 15 — Diagnostics).

I applied a correction but it doesn't sound different.

Check the Applied Correction panel says Room correction active (not bypassed). Use the Bypass / Enable toggle to A/B it (see Chapter 10 — Results).

I restored a backup but nothing changed on the device.

Restore loads calibration results into the app but doesn't auto-apply them. Open the results and click Apply to device. Also: devices that weren't plugged in when you restored are skipped ("not connected"). Reconnect and restore again, or apply manually.

Restore won't run.

• You can't restore while a calibration is running. Stop it first.
• A backup made by a newer app version can't be read. Update the app.

I closed the window but the app is still running.

That's by design: closing the window hides Omnissiah instead of quitting it, so a connected device keeps working. Reopen it from the Dock or by launching it again (that reopens the existing window rather than starting a second copy). To quit properly, use Cmd-Q or File → Quit.

Does the app run on Intel Macs?

Yes. Omnissiah ships as a universal binary and runs on both Apple silicon and Intel Macs.

Does the correction keep working when the app is closed?

Yes. Applied correction is stored on the device and works without the app or computer connected.

Do I need the microphone for everyday use?

No, only for measuring and room correction. Manual level/EQ tweaks (Chapter 7 — Shaping sound) don't need a mic.

Will updating the app reset my tuning?

No. Your saved calibration tuning is preserved and migrated across app updates.

Where's the app called "DSP Control"?

That's the same app: Omnissiah. Some in-app text still uses the older name.

Delay units

Unit cycle order: ms → m → samples → cm → in → ft. Engine sample rate for conversions: 48,000 Hz. Speed of sound: 343 m/s.

The Gain and Delay cards also have − / + stepper buttons: a click moves one step, holding glides the value in steps (see Appendix C — Mouse & keyboard interactions).

Saved by Back Up… (see Chapter 11 — Backup & restore).

Default name: dsp-control-backup_YYYY-MM-DD_HH-MM-SS.json

Contents:

• app version and the time it was created;
• each connected device: its serial number (used to match on restore), name, type, firmware version, and stored settings (levels, delay, polarity, crossover, EQ);
• the app's calibration results and target curve;
• loopback correction and its bypass state (if any).

Not included: measurement snapshots.

Exported by Export CSV (one snapshot) or Export All (one CSV per snapshot into a folder), and imported with Load (see Chapter 8 — Measuring your system).

• A header comment carries the snapshot's name, colour, sample rate, latency and offsets, so a re-imported CSV keeps its identity.
• The body is the measured response data: one row per frequency point, with the level (dB), phase and coherence.
• Because it's plain CSV, you can also open it in a spreadsheet or another measurement tool.

Saved by Save and loaded by Load on an output's page (the device page has the same pair for a whole device). See Chapter 6 — Outputs and channels.

• Holds that output's settings (gain, mute, polarity, delay, crossover and EQ) so you can move a setup between outputs or keep favourite configurations.
• Loading is checked first: a file with more EQ bands than the output has slots, or with out-of-range or corrupt values, is refused with an explanation and nothing is changed.

Export results on the results page (see Chapter 10 — Results) offers three destinations:

• Omnissiah file — the full calibration results, filters and responses together. Re-open it in Omnissiah later, or send it to Tantrum Audio with Send Feedback. This is the file to attach when you email feedback.
• REW filter file — the correction EQ bands as Room EQ Wizard “Filter Settings” files (Generic), one per channel. In REW, import via Filter Tasks → Import.
• Corrected response CSVs — each channel's corrected frequency and phase response, one CSV per channel in the snapshot CSV format above, so they load back into the Measurements page to compare.

To change a band's Q (width), use the band's control block below the graph. The EQ graph is covered in Chapter 7 — Shaping sound.

Measurement offsets are covered in Chapter 8 — Measuring your system.

Email tom@tantrumaudio.co.uk. Room correction is in beta, and real-world results help us improve it, so please do get in touch, especially if something doesn't look or sound right.

The results page can package everything we need for you:

1. 1On the Calibration results page, click Send Feedback.
2. 2Write a short note in “What looked or sounded wrong?” This is the most useful part. For example: “the left channel is much quieter after correction” or “bass sounds boomy.”
3. 3Click Save & open email. The app saves your results file and opens a pre-filled email.
4. 4Attach the saved results file yourself before sending. The email tells you where it was saved (drag that file into the email).

Click to zoom

The email automatically includes your app version, operating system, and connected device/firmware info, so we can reproduce what you saw.

If you email us directly rather than via the app, please include:

• What happened and what you expected instead.
• Your app version (shown in Settings, bottom of the page) and macOS version.
• Your device and its firmware version (from the device row).
• If it's about a measurement or correction, the exported results file (see Chapter 10 — Results) and, if relevant, your snapshot CSVs (see Chapter 8 — Measuring your system).
• Anything from Settings → Logs (Activity) that looks relevant.

Many issues are covered in Chapter 16 — Troubleshooting & FAQ. It's worth a look first: device-not-found, no-audio, low-coherence and restore-didn't-apply are the usual suspects and all have quick fixes.