Performance

Measure in the target scene with the target camera and a representative number of moving objects. Compare one setting at a time.

Cost you observe	First controls to inspect
Expensive reflections	Texture Scale, Update Every N Frames, visibility/distance restrictions, capture layer masks.
Many overlapping wakes	Wake Spawn Rate, Min Distance, Max Visible Wakes, Fast Wake Calculation, Fast Wake Foam.
Dense rain or many random ripples	Ripples Per Second, active visual ripple limits, Shader Refresh Rate, and Max Physics Sample Ripples. The current analytical implementation should be profiled on target hardware.
Ripple upload/sample cost	Shader Refresh Rate and Max Physics Sample Ripples. Lower upload frequency does not stop the shader’s time-based animation.
Excess geometry	Base Resolution, distant detail settings, Ripple Patch Smoothing, floating-prop subdivisions.
Many floating objects	Float-point spacing/count and physics ripple sample budget.
Expensive decoration	Debris count, footprint peak sampling, light-ray counts/layers, and prop deformation.
Many automatic current targets	Affect Layers, target mode, and collider scan limits.

Reducing visible ripple limits changes which ripples reach rendering; reducing physics sample limits can also change sampled movement quality. Validate appearance and floating/swimming behavior after tuning. Do not rely on universal frame-rate estimates: the guide provides controls, not hardware-independent performance guarantees.