NeuroLoom 3.1.14 User Guide

Launch price: €39.99
Regular price planned at €59.

Try NeuroLoom free for 14 days. Processing, LOOM and exports are included. The trial starts on first use, with no card or license key required. Download for Windows or Apple silicon Mac.

Secure checkout through Lemon Squeezy. Already purchased?

Watch NeuroLoom demos on YouTube ↗ to hear the sound transformations.

RECENT PRESETS recalls the last ten distinct factory or user presets you auditioned, newest first. Recall respects KEEP MIX / OUTPUT. History is saved with the session; missing user preset files appear disabled.

NeuroLoom is a VST3 audio effect and a Standalone source-processing/LOOM creator. This guide applies to 3.1.14, including Lemon Squeezy activation, visible sliders and source loudness matching.

Windows installation

Windows 10/11 64-bit. The MSI includes the standalone app and VST3 effect. Download the free trial, or use your private receipt link if you already purchased.

Use Neuroloom-3.1.14-Windows-x64.msi for this Windows release. Keep the existing installer scope, close apps using the plug-in, then restart and rescan the DAW. The machine paths are C:\Program Files\EgoB\Neuroloom and C:\Program Files\Common Files\VST3\Neuroloom.vst3.

Insert NeuroLoom on an audio or mixer track. The DAW owns playback and BPM. For FL Studio, use Options > Manage plugins > Find installed plugins, then add NeuroLoom to a Mixer insert.

Starting your free trial

NeuroLoom 3.1.14 starts a free 14-day trial on first use. Download the trial without a card or license key. Standalone and VST3 share the trial in the same OS account. Reinstalling does not restart it. The Apple silicon Mac 3.1.14 package also includes the 14-day trial.

Activating after purchase

To keep using NeuroLoom, buy a license through Lemon Squeezy and enter the receipt key in TRIAL or ACTIVATE while online. The same installation unlocks on Windows or Mac without a reinstall. Your projects and presets stay in place.

Find your key in your Lemon Squeezy receipt or My Orders. Your private installer link is separate from your activation key.

1. Click ACTIVATE in the top bar and paste the key from your Lemon Squeezy receipt. Your private download link is separate from the key.
2. Click ACTIVATE in the dialog while online. The header changes to LICENSE when effects are enabled.
3. Your lifetime license allows two computers. Standalone and VST3 share one activation within the same OS account, including multiple plugin instances.
4. Connect at least once every 30 days. NeuroLoom checks automatically in the background; CHECK LICENSE retries immediately. A temporary outage keeps your remaining offline allowance.
5. To move computers, open LICENSE > DEACTIVATE COMPUTER on the old computer while online, then activate on the new computer. If the old computer is unavailable or both slots are used, contact info@egob.be with your order number.

After the trial expires, input audio passes through unchanged. You can open projects and adjust settings, but effects and processed capture/export require activation. Keys are stored separately from projects and presets: encrypted storage on Windows and login Keychain on Mac. An OS reinstall or a different OS account may need another activation. Do not delete activation data to free a slot; use the deactivation button.

Start from a preset

Enable KEEP MIX / OUTPUT to retain Total Effect (wet/dry mix) and Output gain while loading factory or user presets. It is off by default and separate from the detailed mutation locks. Other preset parameters still load. The option is saved with session state; restoring a project or DAW session restores its actual values. This does not automatically match preset loudness.

1. Choose a factory starting point in the Studio preset browser. The SHOWCASE filter contains ten flagship sounds within the 37-program factory catalog; see the showcase guide.
2. Listen to the source and adjust Total Effect for the dry/wet balance.
3. Raise Damage, Fracture, Space and Motion as needed. Showcase presets recall these four macros at 0%; their detailed effect values still define the sound, so zero macros do not mean a completely dry signal.
4. Lock detailed controls you want protected before mutation. Use A/B and undo/redo to compare changes, then save a .neuroloom preset if required.

Auto Level offers OFF/LIGHT/STRONG compensation. ECO/HQ affects the documented fold stage; it is not a general quality switch for the entire app.

Use the Standalone LOOM workflow

1. Load a supported audio file and let the decoded-memory preflight finish.
2. Set START/END on the source timeline and choose LOOP or ONE-SHOT. Play, pause, stop and seek use the selected 5/12/20/30 ms fade.
3. Switch to Loom and select a shape: variation A is heard immediately. Named shapes have distinct effect recipes; ??? keeps the free mutation behavior. Adjust Sanity/Chaos, press LOOM to apply that intensity, then audition A-D. Shape comparisons retain the same starting sound and respect locked controls and Total Effect.
4. Use Keep to retain a result or Again for another variation. Suggested moments are measured source regions; audition them before deciding to use one.
5. Save a .neuroloomproject to retain exact LOOM variants, source reference, range and export choices. If a source moves, use relink when reopening it.

The current project format writes schema 3 and reads schemas 1–3. The two autosave generations support crash recovery; explicit project saves remain the way to keep named work. A second Standalone launch focuses the existing window rather than starting another recovery owner.

Capture and export

Use the capture/Keep controls for a bar-aligned result, or export/drag the selected range as WAV. Keep the range and LOOP/ONE-SHOT choice intentional: audition, capture, suggested moments and export share that range.

For creator video, choose 9:16, 1:1 or 16:9 and a Loom Rings, Spectral Flux or Vector Bloom profile. Enter an optional Producer / artist name to show it in the video, cover and preview. Check Effected sound only to export just the processed sound; leave it unchecked for before followed by after. The name and checkbox are saved in projects and host state.

15 SEC and 30 SEC are exact total lengths for both video and WAV. Availability follows the loaded sample, or the selected moment when a range is selected. Below 15 seconds, only FULL is enabled; from 15 to under 30 seconds, 15 SEC and FULL are enabled; at 30 seconds or longer, all three are enabled. A 15-second moment therefore enables 15 SEC and FULL, even inside an eight-minute song. Choosing a suggested moment selects its range for playback and export. Choose FULL SAMPLE / ALL MOMENTS in the moment menu to restore the whole sample and browse all suggestions again. An unavailable saved duration switches to FULL.

Fixed durations render the loaded audio from the start of the selected moment using the current effect settings. No kept capture is required. A short kept capture cannot restrict a loaded sample. Without a loaded sample, export uses the available kept audio instead. Audio never repeats to fill an export; a before/after video gives each section half the selected total duration.

FULL renders the whole loaded song with the current effect settings, regardless of the selected moment or timeline range. With before/after enabled, it includes the whole original song followed by the whole processed song. Without a loaded song, FULL exports the complete kept capture. It has no three-minute cutoff; exports that exceed memory or disk limits report an error.

During export, the moving EXPORTING - PLEASE WAIT bar stays visible until the job finishes. Use its CANCEL button to stop; the indicator remains visible while cancellation completes. Wait for the completion message before opening the result files.

Open preview.html alongside its exported WAV and press Play. Local-file playback uses the browser's native audio controls; the visuals follow analysis saved with the export and stay synchronized when seeking. Re-export older packages to receive the corrected preview.

Export is an owned background job. The setup advisor can report missing encoder support, insufficient destination space or source-size limits.

Troubleshooting and performance

• No sound: check host/Standalone playback, the loaded source/range and output device.
• Unexpected preset result: select the intended preset, inspect locks, and compare A/B before changing several macros at once.
• Choppy graphics: lower visual quality. AUTO/HIGH/MEDIUM/LOW changes painting, not audio quality or DSP parameters.
• Source/export refused: read the preflight/job message and advisor guidance. The advisor does not change your DAW or audio settings automatically.

If playback crackles, increase your audio buffer and reduce the number of active effects or plug-in instances. For help, send your OS, DAW, NeuroLoom version and audio-buffer settings to info@egob.be.

Enter values and understand the bottom controls

Drag the Sanity/Chaos slider or click its percentage box and type a value. Arrow keys make small changes; double-click the slider to return to 35%. The four Studio macros have their own aligned row with room for 100.0%.

The bottom START and END sliders choose the part of the source to play and capture. Their values are percentages of the entire file. PLAYHEAD shows the current position; drag it or enter a value to seek. These fields do not change the sound character. LOOP repeats the selected region.

Match the source level

MATCH SOURCE LOUDNESS starts checked. It follows the original input's average signal level during audition, independently of a preset's input and output gain. Matching settles smoothly over the first few seconds. Capture then matches the measured source range RMS more precisely, without counting the appended effect tail as extra silence to normalize. Matching is bounded to prevent extreme boosts. Very quiet or very dynamic material may not reach exactly the same perceived loudness. This is RMS matching, not LUFS analysis.

Use RESULT LEVEL for the final manual adjustment: 0 dB is the matched level, -6 dB is quieter, and positive values are louder. The adjustment stays after matching and is saved in presets/projects. Capture again after changing it. Clear the matching checkbox to use the effect's own level plus your trim. The older Studio LEVEL mode is disabled while source matching takes precedence.

WAV and drag exports retain the captured level instead of applying a fixed -14 dB RMS target. Export reduces gain only when needed for the -1 dBFS sample peak ceiling, so peak protection can make very dynamic results quieter.

Interface appearance

Use the Theme selector near the top of the window to choose Original, Graphite (charcoal panels and teal accents), or Copper (warm brown panels and amber accents). The choice is remembered for this app on this computer. It changes only the interface; audio, saved projects, and rendered visual colours keep their existing settings.

Showcase preset bank

The SHOWCASE filter contains ten flagship factory sounds. They are appended as host programs 15–24 (zero-based); the original fifteen program numbers and favourite IDs stay stable.

Select SHOWCASE in the Studio preset browser, then load a sound. All 24 detailed controls, Auto Level, HQ quality, and all four performance macros recall explicitly. Damage, Fracture, Space, and Motion start at 0%. The parent values are the supplied neuroloom.txt recommendations; the acceptance fixture in Tests/ShowcasePresetChecks.h checks all of them after both minimum and maximum previous parameter states.

The browser supports the new Tonal/Vocal categories and every supplied tag. Search also finds best source, intensity, bank name, and related variation names. Searching GHOST BREAK, for example, finds BREAK MY DRUMS. The preset tooltip shows the complete metadata and related names. Intensity is descriptive metadata on a 0–100 scale, not an audio control.

Named LOOM variations

Loading any parent below prepares four exact LOOM choices while keeping the parent audible. Switch to Loom and choose the labelled A/B button, then KEEP. BEFORE restores the supplied parent values. Remaining slots hold bounded variations generated from a fixed seed; all four macros remain zero. The same choices are also available through Studio's existing four variation buttons. Recalling the parent rebuilds the same choices.

These ten names are related variations, not extra factory programs. The attachment specified their parent/name relationships but no parameter values. The following authored overrides define their sounds; every omitted value inherits the exact parent. Values use the processor's plain units: percent for mix/amount controls, Hz for frequencies/rate, milliseconds for delay time, dB for output, and zero-based division choices (0 = 1/4, 1 = 1/8, 2 = 1/16, 3 = 1/32, 4 = 1/64, 5 = 1/128).

Exact variant payloads and names survive Keep, offline DNA, host state, project save/open, and project recovery through the existing LOOM state serialization. No LOOM/project/DNA schema change is needed: the optional loomVariantName property travels with each exact parameter tree. DNA import retains the existing parent-lineage behavior.

LOOM, AGAIN, and Daily Loom continue to generate new choices from the current sound and selected personality/seed/locks. Generated choices clear authored names so a newly generated sound cannot be mislabelled GHOST BREAK. Saving a user preset retains the current sound, zero macros, category, tags, best-source and intensity metadata; projects and DNA retain the full A–D set.

These settings implement the proposed sound designs. Automated recall and audio safety checks do not establish subjective listening quality.

Validation

Windows x64 Release standalone and VST3 builds succeeded. The full 20-test CTest suite passed; all seven checks affected by the final footer layout change also passed after rebuilding. The UI fixture checks named variation selection, Keep, source/intensity metadata, category/tag discovery, and non-overlapping footer controls at the 1100 px layout.

The callback report covers all 25 factory programs across five sample rates: 135 release-gated scenarios pass with zero callback allocations. Its separate 16-instance diagnostic remains outside the certified eight-instance ceiling and fails, as explicitly reported by the existing gate. The allocation matrix also exercises every prepared showcase A–D state across five rates, six block sizes, and mono/stereo layouts.