All tutorials

Novus Visualizers

Working with layers in Novus Visualizers (and which combinations work best)

Stack up to four reactive layers with blend controls to build depth — a calm background engine, a sharp foreground, accents, and text — and learn the layer combinations that read well instead of turning to mush.

Stacking up to four reactive visualizer layers with blend controls for depth

A single engine reacting to your track is good; a few well-chosen layers reacting together is what makes a visualizer feel designed. Novus Visualizers lets you stack up to four independent layers with blending, and this guide is about using them deliberately — because four busy engines fighting each other looks worse than one good one.

  1. 1

    1. Think in foreground, background, and accent

    The most reliable structure is a quiet background layer (a slow gradient, a soft particle field, a subtle tunnel) with a sharp foreground engine carrying the main motion on top, and at most one accent layer for highlights. That gives the eye a clear focal point and somewhere to rest, which is exactly what a single flat engine cannot do.

    Each layer is fully independent and reacts to the audio on its own, so the background can breathe slowly while the foreground hits hard on the beat.

    • Up to 4 independent, audio-reactive layers.
    • A clear foreground/background hierarchy reads best.
  2. 2

    2. Use blend modes and opacity for depth

    Layers combine through blend controls. Lighten/additive blending stacks glows without muddying them — ideal for an accent layer of sparks or bloom over a darker base. Lowering a layer’s opacity pushes it back in the depth order. The goal is contrast between layers: if two layers have the same brightness and busyness, they compete; if one is clearly subordinate, they combine.

    • Additive/lighten blending for clean glows.
    • Opacity to set depth order.
    • Contrast between layers, not competition.
  3. 3

    3. Combinations that work — and ones that don’t

    Pairs that tend to read well: a 3D tunnel or particle field behind a 2D spectrum or waveform; a slow mandala behind a sharp geometric engine; a dark gradient base behind anything bright. Combinations to avoid: two high-detail engines at full opacity, two engines using the same color theme at the same brightness, or four layers all reacting to the same beat — they pulse as one blob instead of as distinct elements.

    • Good: calm 3D base + sharp 2D foreground.
    • Avoid: two busy engines at equal weight.
    • Avoid: every layer reacting identically.
  4. 4

    4. Assign different reactions to different layers

    Layers do not have to react to the same thing. Let the background respond to the slow RMS loudness envelope so it swells across a section, while the foreground bursts on detected beats and an accent layer flickers on treble onsets. Splitting the audio response across layers is what makes the whole frame feel alive rather than synchronized to one pulse — more on this in the audio-reactivity tutorial.

Less is usually more

Two well-contrasted layers almost always beat four competing ones. Start with a foreground you like, add a calm background for depth, and only add a third or fourth layer if it has a clear, subordinate job. If the frame ever looks busy or muddy, the fix is almost always to drop a layer’s opacity or remove it entirely rather than adding more.

Go deeper