I'm trying to find the theoretical output range of improved Perlin noise for 1, 2 and 3 dimensions. I'm aware of existing answers to this question, but they don't seem to accord with my practical findings.
If n is the number of dimensions then according to [1] it should be [-sqrt(n/4), sqrt(n/4)]. According to [2] (which refers to [3]) it should be [-0.5·sqrt(n), 0.5·sqrt(n)] (which amounts to the same thing).
This means that the ranges should be approximately:
| Dimensions | Range |
|---|---|
| 1 | [-0.5, 0.5] |
| 2 | [-0.707, 0.707] |
| 3 | [-0.866, 0.866] |
However when I run the following code (which uses Ken Perlin's own reference implementation of improved noise from his website), I get higher values for 2 and 3 dimensions, namely approximately:
| Dimensions | Range |
|---|---|
| 1 | [-0.5, 0.5] |
| 2 | [-0.891, 0.999] |
| 3 | [-0.997, 0.999] |
With different permutations I even sometimes get values slightly over 1.0 for 3 dimensions, and for some strange reason one of the bounds for two dimension always seems to be about 0.89 while the other is about 1.00.
I can't figure out whether this is due to a bug in my code (I don't see how since this is Ken Perlin's own code) or due to those discussions not being correct or not being applicable somehow, in which case I would like to know what the theoretical ranges are for improved Perlin noise.
Can you replicate this? Are the results wrong, or can you point me to a discussion of the theoretical values that accords with this outcome?
The code:
public class PerlinTest {
public static void main(String[] args) {
double lowest1DValue = Double.MAX_VALUE, highest1DValue = -Double.MAX_VALUE;
double lowest2DValue = Double.MAX_VALUE, highest2DValue = -Double.MAX_VALUE;
double lowest3DValue = Double.MAX_VALUE, highest3DValue = -Double.MAX_VALUE;
final Random random = new SecureRandom();
for (int i = 0; i < 10000000; i++) {
double value = noise(random.nextDouble() * 256.0, 0.0, 0.0);
if (value < lowest1DValue) {
lowest1DValue = value;
}
if (value > highest1DValue) {
highest1DValue = value;
}
value = noise(random.nextDouble() * 256.0, random.nextDouble() * 256.0, 0.0);
if (value < lowest2DValue) {
lowest2DValue = value;
}
if (value > highest2DValue) {
highest2DValue = value;
}
value = noise(random.nextDouble() * 256.0, random.nextDouble() * 256.0, random.nextDouble() * 256.0);
if (value < lowest3DValue) {
lowest3DValue = value;
}
if (value > highest3DValue) {
highest3DValue = value;
}
}
System.out.println("Lowest 1D value: " + lowest1DValue);
System.out.println("Highest 1D value: " + highest1DValue);
System.out.println("Lowest 2D value: " + lowest2DValue);
System.out.println("Highest 2D value: " + highest2DValue);
System.out.println("Lowest 3D value: " + lowest3DValue);
System.out.println("Highest 3D value: " + highest3DValue);
}
static public double noise(double x, double y, double z) {
int X = (int)Math.floor(x) & 255, // FIND UNIT CUBE THAT
Y = (int)Math.floor(y) & 255, // CONTAINS POINT.
Z = (int)Math.floor(z) & 255;
x -= Math.floor(x); // FIND RELATIVE X,Y,Z
y -= Math.floor(y); // OF POINT IN CUBE.
z -= Math.floor(z);
double u = fade(x), // COMPUTE FADE CURVES
v = fade(y), // FOR EACH OF X,Y,Z.
w = fade(z);
int A = p[X ]+Y, AA = p[A]+Z, AB = p[A+1]+Z, // HASH COORDINATES OF
B = p[X+1]+Y, BA = p[B]+Z, BB = p[B+1]+Z; // THE 8 CUBE CORNERS,
return lerp(w, lerp(v, lerp(u, grad(p[AA ], x , y , z ), // AND ADD
grad(p[BA ], x-1, y , z )), // BLENDED
lerp(u, grad(p[AB ], x , y-1, z ), // RESULTS
grad(p[BB ], x-1, y-1, z ))),// FROM 8
lerp(v, lerp(u, grad(p[AA+1], x , y , z-1 ), // CORNERS
grad(p[BA+1], x-1, y , z-1 )), // OF CUBE
lerp(u, grad(p[AB+1], x , y-1, z-1 ),
grad(p[BB+1], x-1, y-1, z-1 ))));
}
static double fade(double t) { return t * t * t * (t * (t * 6 - 15) + 10); }
static double lerp(double t, double a, double b) { return a + t * (b - a); }
static double grad(int hash, double x, double y, double z) {
int h = hash & 15; // CONVERT LO 4 BITS OF HASH CODE
double u = h<8 ? x : y, // INTO 12 GRADIENT DIRECTIONS.
v = h<4 ? y : h==12||h==14 ? x : z;
return ((h&1) == 0 ? u : -u) + ((h&2) == 0 ? v : -v);
}
static final int p[] = new int[512], permutation[] = { 151,160,137,91,90,15,
131,13,201,95,96,53,194,233,7,225,140,36,103,30,69,142,8,99,37,240,21,10,23,
190, 6,148,247,120,234,75,0,26,197,62,94,252,219,203,117,35,11,32,57,177,33,
88,237,149,56,87,174,20,125,136,171,168, 68,175,74,165,71,134,139,48,27,166,
77,146,158,231,83,111,229,122,60,211,133,230,220,105,92,41,55,46,245,40,244,
102,143,54, 65,25,63,161, 1,216,80,73,209,76,132,187,208, 89,18,169,200,196,
135,130,116,188,159,86,164,100,109,198,173,186, 3,64,52,217,226,250,124,123,
5,202,38,147,118,126,255,82,85,212,207,206,59,227,47,16,58,17,182,189,28,42,
223,183,170,213,119,248,152, 2,44,154,163, 70,221,153,101,155,167, 43,172,9,
129,22,39,253, 19,98,108,110,79,113,224,232,178,185, 112,104,218,246,97,228,
251,34,242,193,238,210,144,12,191,179,162,241, 81,51,145,235,249,14,239,107,
49,192,214, 31,181,199,106,157,184, 84,204,176,115,121,50,45,127, 4,150,254,
138,236,205,93,222,114,67,29,24,72,243,141,128,195,78,66,215,61,156,180
};
static { for (int i=0; i < 256 ; i++) p[256+i] = p[i] = permutation[i]; }
}
Ken’s not using unit vectors. As [1] says, with my emphasis:
Third, there are many different ways to select the random vectors at the grid cell corners. In Improved Perlin noise, instead of selecting any random vector, one of 12 vectors pointing to the edges of a cube are used instead. Here, I will talk strictly about a continuous range of angles since it is easier – however, the range of value of an implementation of Perlin noise using a restricted set of vectors will never be larger. Finally, the script in this repository assumes the vectors are of unit length. If they not, the range of value should be scaled according to the maximum vector length. Note that the vectors in Improved Perlin noise are not unit length.
For Ken’s improved noise, the maximum vector length is 1 in 1D and √2 in 2D, so the theoretical bounds are [−0.5, 0.5] in 1D and [−1, 1] in 2D. I don’t know why you’re not seeing the full range in 2D; if you shuffled the permutation I bet you would sometimes.
For 3D, the maximum vector length is still √2, but the extreme case identified by [1] isn’t a possible output, so the theoretical range of [−√(3/2), √(3/2)] is an overestimate. These folks tried to work it out exactly, and yes, the maximum absolute value does seem to be strictly greater than 1.