RESEARCH — RENDERING ALGORITHM
ISE
Immersive Sound Engine
ISE unifies gain-based and delay-based spatialization through acoustic physics. One algorithm covers every behavior: a single Focus parameter morphs the rendering from diffuse WFS-like energy to precise point localization — up to 128 sources on 128 speakers.
THE PROBLEM
Too many algorithms, not enough engine
The spatial audio industry has fragmented into incompatible camps. VBAP positions discrete sources well but leaves holes in the field between speakers. Ambisonics is mathematically elegant but needs high orders — and channel counts most systems cannot carry — for precise localization. Wave Field Synthesis promises physical reconstruction but demands hundreds of speakers and milliseconds of latency.
For practitioners this means constant friction: the theatre designer learns VBAP for one venue and faces an Ambisonic rig at the next; the installation artist masters a tool that the exhibition hardware refuses; and a source placed at 45° in VBAP simply does not sound like the same source at 45° in Ambisonics. Creative intent gets lost in translation.
What practitioners need is not more algorithms — it is fewer. One tool that adapts. One set of parameters that make sense. One behavior that travels across systems.
HOW IT WORKS
How it works
One algorithm, every behavior
No catalog of incompatible methods: gain and delay spatialization are unified through real acoustic physics — inverse square law, delay compensation, intelligent normalization.
The Focus parameter
Low focus spreads energy like WFS; medium behaves like natural panning; high focus locks sources to precise positions. One knob, a continuum of behaviors.
Massive scalability
Up to 128 sources on 128 speakers, cartesian or spherical coordinates, adjustable speed of sound — at a few samples of latency.
UNDER THE HOOD
Going deeper
ISE unifies gain-based and delay-based spatialization into one engine built on acoustic physics: amplitude follows the inverse square law — the same law sound obeys in air — and timing differences between speakers follow the speed of sound, producing interaural cues your brain interprets instinctively.
Physics alone would be unusable: raw inverse-square gains would make a source at 1 meter a hundred times louder than one at 10 meters. ISE's normalization subtracts the smallest attenuation from all channels — relative distance relationships survive, headroom stays constant. Likewise, a 'tight' delay mode subtracts the minimum delay from every channel: the spatial impression of relative timing is preserved while absolute latency is minimized.
The Focus parameter is not a blend between presets — it is continuous control over the rendering mathematics. A logarithmic curve spreads energy across many speakers (diffuse, WFS-like, ideal for ambience); a linear curve concentrates on the nearest while keeping neighbors present (natural panning, VBAP-like, with smoother transitions and wider sweet spot); an exponential curve locks a source to the closest speaker (pinpoint, for spotlit performers and picture-locked effects). One project, one session, one tool — sources that need precision get high Focus, enveloping elements get low.
AT A GLANCE
At a glance
| Input | Mono per source — up to 128 sources |
| Output | Up to 128 speakers |
| Latency | A few samples |
| Implementation | C/C++ · 44.1 to 192 kHz · embedded-ready |
| Availability | Ships in Ripl · OEM licensing |
PARAMETERS
Hands on the algorithm
| Focus | Continuous morph: diffuse (log) → natural (linear) → precise (exp) |
| Coordinates | Cartesian (x, y, z) or spherical |
| Speed of sound | Adjustable, 343 m/s default |
| Capacity | Up to 128 sources × 128 speakers |
| Formats | 44.1 – 192 kHz · 16 / 24 / 32-bit float |
POSITIONING
Compared to the alternatives
vs algorithm catalogs
VBAP, DBAP, HOA, WFS… competitors make you choose per source and live with the discontinuities. ISE is one continuum — no choice, no rupture.
vs format lock-in
No proprietary renderer or certified hardware required — ISE drives any speaker set you can describe.
APPLICATIONS
Where it fits
Live sound & theatre
Create the spatial score once; it travels with the production and adapts to each venue's speakers — a thrust stage with wraparound becomes a proscenium frontal array from the same session. Rain and ambience at low Focus, dialogue at medium, spot effects tracking performers at high — and cues fired from show control spatialize instantly.
Immersive installations
Position sources, set Focus, done — curators can adjust the spatial behavior without an audio specialist on call. The same project serves the 8-speaker temporary gallery and the 64-speaker permanent hall, and the few-sample latency lets visitor tracking and gestures drive sources in real time.
Spatial music production
Position elements with coordinates or angles — stereo panning extended to 3D — and let Focus carry the musical intent: lead vocal medium-high, synth pads low, kick and snare high while the overheads spread. Automate it all from the DAW and render to 5.1, 7.1.4 or any custom array.
VR / AR & gaming
Engine position updates become spatial changes within microseconds, keeping audio locked to picture and head motion. And because the distance modeling is real physics, a sound at 3 meters sounds like it looks — the brain accepts the illusion because the laws are consistent.
INTEGRATION
Built to live inside your product
| Delivery | C · C++ · MATLAB · .dsp — full source code |
| Platforms | macOS · Windows · Linux · embedded ARM · iOS · Android |
| DSP platforms | Flow DSP · Audio Weaver — solutions in preparation |
| Documentation | Whitepaper — every algorithm explained, in the clear |
OEM LICENSING
- ■ One-time payment per brand
- ■ Full source code — C, C++, MATLAB, .dsp
- ■ Whitepaper — all algorithms explained
- ■ Integration support included
- ■ Free updates
- ■ Volume discounts on multiple licenses
SHIPS IN