Skip to main content
k2k audio logo k2k audio

Back to Crossmorph
Documentation tree

Visual Blend

Visual Blend applies image-processing blend modes (the same ones in Photoshop, GIMP, video compositing tools) to the spectral magnitudes of two inputs.

Parameters

ParameterRangeDefault
Mix0.0 – 1.00.5

Mix — Equal-power blend factor, 0–1. 0 = full Source A; 1 = full Source B; 0.5 = equal blend. Behavior at intermediate values depends on the chosen blend_mode — most modes interpolate proportionally, but some (like Multiply or Screen) produce non-linear blends.

Additional controls

Blend Mode — How A and B are combined, drawn from image-processing blend modes:

  • Additive — A + B (energies sum; can clip on bright sources).
  • Multiply — A × B (only frequencies present in both survive; “intersection” of the two).
  • Maximum — max(A, B) per-bin (the louder of the two at each frequency wins).
  • Minimum — min(A, B) per-bin (the quieter wins; produces sparse, “where both are quiet” output).
  • Screen — inverted multiply; lighter blend that emphasizes loud parts of either source.
  • Difference — |A − B| (highlights spectral differences between the two; bins where A and B are similar become quiet, bins where they differ become loud).

For predictable mixing, Additive with mix is the most musical. The other modes are for sound-design where you want non-linear combinations.

Loop B — When on, Source B loops to match A’s duration; off plays B once.

Phase — Phase reconstruction: Preserve, Blend, Dominant, Complex (see other nodes — same options).

About Visual Blend

Visual Blend applies image-processing blend modes (the same ones in Photoshop, GIMP, video compositing tools) to the spectral magnitudes of two inputs. The result is non-linear combinations that no traditional audio mixing can produce. Difference mode is particularly useful for sound-design: it highlights what’s unique in A vs. B, producing strange “where they disagree” textures. Multiply acts like a spectral mask (only content present in both survives). Maximum is closer to a per-bin OR — anything loud anywhere in either source is preserved. Use it for sound-design transformations where standard crossfading would be too predictable. The conceptual leaning is “spectral image-compositing” rather than “audio mixing.”


Generated 2026-05-05 from K2K_Dev@96730bdc by scripts/gen_lexique.py. Edit _intros/ or _overrides/, not this file.