UImportanceSamplingLibrary::RandomSobolFloat

function Engine Blueprint Since 5.0

#include "Kismet/ImportanceSamplingLibrary.h"

Access: public Specifiers: staticUFUNCTIONBlueprintPure

Description

Returns a Sobol-distributed quasi-random number between 0 and 1 for the given sample index and dimension. Use for low-discrepancy sampling that fills a range more evenly than uniform random over many draws.

Caveats & Gotchas

• Only 16 independent Sobol dimensions (0-15) are supported — requesting an out-of-range Dimension is not validated against a documented error and should be avoided.
• Seed scrambles the sequence so different objects/instances don't sample identical points, but the same (Index, Dimension, Seed) always produces the same result — it's deterministic, not a fresh random draw each call.
• For a running sequence of samples, NextSobolFloat is cheaper than recomputing RandomSobolFloat from scratch for each Index.

Signature

static float RandomSobolFloat(int32 Index, int32 Dimension, float Seed)

Parameters

Name	Type	Description	Default
Index	int32	Which sequential point in the Sobol sequence.	—
Dimension	int32	Which Sobol dimension to draw from (0 to 15).	—
Seed	float	Random seed in the range 0-1 used to randomize (scramble) across multiple independent sequences.	—

Return Type

float

Example

Sample a low-discrepancy value per iteration C++

for (int32 i = 0; i < 64; ++i)
{
    const float Sample = UImportanceSamplingLibrary::RandomSobolFloat(i, 0, 0.0f);
    // use Sample to jitter a spawn position, etc.
}

Version History

Introduced in: 5.0

Version	Status	Notes
5.6	stable	—

Feedback

Was this helpful?