Ambi Decoder Speaker

s3g Ambi Decoder Speaker 64 decodes ambisonic ACN/SN3D input to a 64-channel speaker output bus. Use it when the REAPER session's speaker layout is part of the composition, installation, or encoded-field chain.

The decoder supports first through seventh order. The output bus stays 64 channels so routing is predictable; the active speaker count decides which output lanes are used.

Ambi Decoder Speaker interface showing the 25-speaker Dome 25 layout from the top with its concentric speaker layers visible — Ambi Decoder Speaker showing the available 25-speaker `DOME 25` layout in top view.

Workflow

Insert the appropriate Speaker, Object, Adaptive, or Sub decoder after the encoded source and any Ambisonic processing on a track wide enough for its input and output buses. Match ORDER to the source, then choose the speaker layout and active COUNT, entering measured positions in CUSTOM when needed, or set SUBS for Ambi Decoder Sub.
Route each active output lane to its matching physical speaker or subwoofer feed.
Set the decode mode and weighting or the sub cutoff and width, verify the output map, and trim individual outputs while audio is playing.

Signal Flow

Read the active ambisonic channels from the input bus.
Build a speaker matrix from the selected order and layout.
Write one output lane per active speaker and silence unused lanes.

Diagram of ACN/SN3D input, decoder matrix, speaker layout, and 64-channel output

Decoder Variants

s3g Ambi Decoder Speaker 64: the main curated/custom speaker decoder.
s3g Ambi Decoder Object 64: blends the decoded field with a directional object-panning path using VBAP, LBAP, or DBAP.
s3g Ambi Decoder Adaptive 64: blends diffuse and focused decode paths using confidence and transient cues.
s3g Ambi Decoder Sub: derives up to eight low-frequency sub feeds from ACN/SN3D input, with cutoff, directional width, sub count, and output trim controls.

Layouts

The layout menu provides common speaker shapes plus custom editing.

For room-style presets, speaker numbering starts at the stereo-right position and moves clockwise around each elevation layer, then continues upward. DOME24 and DOME25 follow the s3g-mc dome panner structure: 12 lower speakers, 8 middle speakers, 4 upper speakers, and a top point for DOME25. SPHERE24 uses the fixed 24-direction ordering expected by the Ambi Effect workflows.

Layout	Use
`QUAD`	Four speakers at zero elevation.
`CUBE8`	Eight corner speakers around a cube.
`CUBE17`	The 17-point cube layout used by the s3g-mc cube panner.
`DOME24`	Twenty-four speaker hemisphere dome: 12 lower, 8 middle, 4 upper.
`DOME25`	Dome24 plus a top speaker, matching the s3g-mc 25-channel dome panner shape.
`QUAD+OH`	Quad plus two overhead speakers at `-90` and `+90` azimuth.
`SPHERE24`	The 24-channel virtual sphere used by Ambi Effects and encoded-field routing.
`CUSTOM`	Generate a full-sphere or hemisphere speaker field, then edit speakers directly.

Controls

Control	Meaning
ORDER	Selects the ambisonic order from `1OA` to `7OA`.
MODE	Chooses the decoder matrix style: `BASIC`, `EPAD`, or `MMD`.
COUNT	Sets how many output lanes are used, from 2 to 64. In `CUSTOM`, this also generates a new speaker field.
FIELD	Chooses how `CUSTOM` layouts are generated: `SPHERE` distributes speakers over a full sphere, and `HEMI` keeps them in the upper hemisphere.
WGT	Chooses ambisonic order weighting: `NONE`, `MAXRE`, or `INPHASE`.
WIDTH	Broad decoder matrix trim.
MIXER	Switches the large left view to per-speaker `GAIN` trims, mute/solo buttons, and the final `OUT` gain. Speaker gains are shown 16 at a time, with more pages as needed.
MAP	Switches the large left view to a decoder map. It uses the current selected-speaker `AZ`/`EL` direction as a probe and shows which speakers receive energy from the current order, weighting, mode, gains, and layout.

Changing COUNT or FIELD creates a new CUSTOM layout automatically. SELECTED SPEAKER, AZIMUTH, ELEVATION, and DISTANCE edit one speaker at a time. AZIMUTH, ELEVATION, and DISTANCE have sliders for quick movement and text boxes for exact entry. Any position edit keeps the layout in CUSTOM.

Field View

The speaker field follows the same camera convention as the Ambi Encoder Point. TOP draws 0 degrees at the top, 180 at the bottom, -90 to the right, and +90 to the left. SIDE draws positive elevation upward. 3/4 is for inspecting layers and custom layouts.

Use - / + to zoom the camera view. Drag blank space in the field to rotate the view. Camera movement changes only the visual inspection angle, not speaker AED positions. The selected camera view, manual rotation, and zoom are saved with the host project.

Connection lines show the intended layout mesh, not nearest-neighbor analysis and not speaker list order. A cube draws as a cube. A dome draws as rings with vertical layer connections.

The ordinary field uses shared AED color, while MAP replaces it with a normalized low-to-high response scale. See Interpreting Color before comparing the two views.

Choosing a Decode Mode

Choose BASIC for direct speaker-basis weighting and a clear reference decode. Choose EPAD for a regularized, energy-preserving matrix decode. Choose MMD to blend direct and energy-weighted behavior across multiple modes.

WGT is separate from MODE. It tapers ambisonic orders before the signal reaches the speaker outputs. MAXRE is the default because it usually gives a useful balance of localization and stability. INPHASE is more conservative and can help with irregular layouts. NONE leaves the orders unweighted for comparison.

REAPER Setup

Use a track with enough channels for the selected ambisonic input and speaker output. The plugin has a fixed 64-channel input bus and fixed 64-channel output bus so REAPER pin routing remains stable while layout and order change inside the plugin.

For Ambi Effect reference work, use SPHERE24 as the virtual speaker layout. For real speaker playback, start with the layout closest to the room, then switch to CUSTOM for measured speaker positions.