Finish implementing OIT

examples/CMakeLists.txt

@@ -1,7 +1,8 @@
 cmake_minimum_required(VERSION 3.21)
 
 function(add_example name)
-  add_executable("example-${name}" ${name}.cpp)
+  add_executable("example-${name}" ${name}.cpp simplexnoise1234.cpp)
+  target_include_directories("example-${name}" PUBLIC .)
   target_link_libraries("example-${name}" PUBLIC sprstk)
 endfunction()
 

examples/basic.cpp

@@ -1,39 +1,114 @@
 #include <sprstk/sprstk.h>
 
-#include <cstdlib>
+#include <simplexnoise1234.h>
+
+#include <cmath>
 
 namespace
 {
 
+constexpr int SIZE = 256;
+
+double octaves(SimplexNoise1234& simplex, double x, double y, int layers, double persistence, double frequency)
+{
+  double ampl = 1;
+  double maxval = 0;
+  double val = 0;
+
+  for (int i = 0; i < layers; i++)
+  {
+    val += simplex.noise(x * frequency, y * frequency) * ampl;
+
+    maxval += ampl;
+
+    ampl *= persistence;
+    frequency *= 2;
+  }
+
+  return val / maxval;
+}
+
+uint8_t data[SIZE * SIZE];
+
+double ease(double n)
+{
+  return pow(n, 1.5);
+}
+
+uint32_t color(uint8_t r, uint8_t g, uint8_t b, uint8_t a)
+{
+  return (r << 24) | (g << 16) | (b << 8) | a;
+}
+
+int pick_pal(uint8_t height)
+{
+  if (height > 20) { return 0; }
+  if (height > 8) { return 1; }
+  return 2;
+}
+
 void init(sprstk* instance, void* userdata)
 {
   sprstk_palette pal = {};
 
-  for (int i = 0; i < 28; i++)
+  for (int i = 0; i < 24; i++)
   {
-    pal.colors[i] = 0x7F3F0040;
+    uint8_t val = 0x33 / (12 - i / 2.0f) + 0x33;
+    pal.colors[i] = color(val, val, val, 0x60);
   }
 
-  for (int i = 28; i < 32; i++)
+  for (int i = 24; i < 32; i++)
   {
-    pal.colors[i] = 0x00FF0040;
+    uint8_t val = 0x55 / (16 - i / 2.0f) + 0xAA;
+    pal.colors[i] = color(val, val, val, 0x60);
   }
 
   sprstk_set_palette(instance, 0, &pal);
+
+  for (int i = 0; i < 16; i++)
+  {
+    pal.colors[i] = color(0x70 / (5.0f - i / 4.0f), 0x35 / (5.0f - i / 4.0f), 0, 0x60);
+  }
+
+  for (int i = 16; i < 32; i++)
+  {
+    pal.colors[i] = color(0x05 / (8.5f - i / 2.5f) + 0x15, 0x40 / (8.5f - i / 2.5f) + 0x0, 0, 0x60);
+  }
+
+  sprstk_set_palette(instance, 1, &pal);
+
+  for (int i = 0; i < 8; i++)
+  {
+    pal.colors[i] = color(0x20 / (8 - i) + 0x10, 0x40 / (8 - i) + 0x20, 0xB0 / (8 - i) + 0x40, 0x60);
+  }
+
+  sprstk_set_palette(instance, 2, &pal);
+
+  sprstk_set_scale(instance, 0.4f);
+
+  SimplexNoise1234 simplex;
+  for (int i = 0; i < SIZE; i++)
+  {
+    for (int j = 0; j < SIZE; j++)
+    {
+      double value = octaves(simplex, i, j, 6, 0.4, 1.0 / 128.0);
+      data[i + j * SIZE] = 28 * ease((value + 1) / 2) + 3;
+    }
+  }
+
+  for (int i = 0; i < SIZE; i++)
+  {
+    for (int j = 0; j < SIZE; j++)
+    {
+      sprstk_put(instance, i - SIZE / 2, j - SIZE / 2, data[i + SIZE * j], pick_pal(data[i + SIZE * j]));
+    }
+  }
 }
 
 void update(sprstk* instance, float dt, float* userdata)
 {
   *userdata += dt / 2;
   sprstk_set_angle(instance, *userdata);
-
-  for (int i = -512; i < 512; i++)
-  {
-    for (int j = -512; j < 512; j++)
-    {
-      sprstk_put(instance, i, j, 31, 0);
-    }
-  }
 }
 
 }

examples/simplexnoise1234.cpp

@@ -0,0 +1,470 @@
+// SimplexNoise1234
+// Copyright © 2003-2011, Stefan Gustavson
+//
+// Contact: stegu@itn.liu.se
+//
+// This library is public domain software, released by the author
+// into the public domain in February 2011. You may do anything
+// you like with it. You may even remove all attributions,
+// but of course I'd appreciate it if you kept my name somewhere.
+//
+// This library is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+// General Public License for more details.
+
+// Modified by the LOVE Development Team to use double precision.
+
+/** \file
+		\brief Implements the SimplexNoise1234 class for producing Perlin simplex noise.
+		\author Stefan Gustavson (stegu@itn.liu.se)
+*/
+
+/*
+ * This implementation is "Simplex Noise" as presented by
+ * Ken Perlin at a relatively obscure and not often cited course
+ * session "Real-Time Shading" at Siggraph 2001 (before real
+ * time shading actually took on), under the title "hardware noise".
+ * The 3D function is numerically equivalent to his Java reference
+ * code available in the PDF course notes, although I re-implemented
+ * it from scratch to get more readable code. The 1D, 2D and 4D cases
+ * were implemented from scratch by me from Ken Perlin's text.
+ *
+ * This is a highly reusable class. It has no dependencies
+ * on any other file, apart from its own header file.
+ */
+
+
+#include	"simplexnoise1234.h"
+
+#define FASTFLOOR(x) ( ((x)>0) ? ((int)x) : (((int)x)-1) )
+
+//---------------------------------------------------------------------
+// Static data
+
+/*
+ * Permutation table. This is just a random jumble of all numbers 0-255,
+ * repeated twice to avoid wrapping the index at 255 for each lookup.
+ * This needs to be exactly the same for all instances on all platforms,
+ * so it's easiest to just keep it as static explicit data.
+ * This also removes the need for any initialisation of this class.
+ *
+ * Note that making this an int[] instead of a char[] might make the
+ * code run faster on platforms with a high penalty for unaligned single
+ * byte addressing. Intel x86 is generally single-byte-friendly, but
+ * some other CPUs are faster with 4-aligned reads.
+ * However, a char[] is smaller, which avoids cache trashing, and that
+ * is probably the most important aspect on most architectures.
+ * This array is accessed a *lot* by the noise functions.
+ * A vector-valued noise over 3D accesses it 96 times, and a
+ * float-valued 4D noise 64 times. We want this to fit in the cache!
+ */
+unsigned char SimplexNoise1234::perm[512] = {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,
+  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
+};
+
+//---------------------------------------------------------------------
+
+/*
+ * Helper functions to compute gradients-dot-residualvectors (1D to 4D)
+ * Note that these generate gradients of more than unit length. To make
+ * a close match with the value range of classic Perlin noise, the final
+ * noise values need to be rescaled to fit nicely within [-1,1].
+ * (The simplex noise functions as such also have different scaling.)
+ * Note also that these noise functions are the most practical and useful
+ * signed version of Perlin noise. To return values according to the
+ * RenderMan specification from the SL noise() and pnoise() functions,
+ * the noise values need to be scaled and offset to [0,1], like this:
+ * float SLnoise = (SimplexNoise1234::noise(x,y,z) + 1.0) * 0.5;
+ */
+
+double  SimplexNoise1234::grad( int hash, double x ) {
+    int h = hash & 15;
+    double grad = 1.0 + (h & 7);   // Gradient value 1.0, 2.0, ..., 8.0
+    if (h&8) grad = -grad;         // Set a random sign for the gradient
+    return ( grad * x );           // Multiply the gradient with the distance
+}
+
+double  SimplexNoise1234::grad( int hash, double x, double y ) {
+    int h = hash & 7;      // Convert low 3 bits of hash code
+    double u = h<4 ? x : y;  // into 8 simple gradient directions,
+    double v = h<4 ? y : x;  // and compute the dot product with (x,y).
+    return ((h&1)? -u : u) + ((h&2)? -2.0*v : 2.0*v);
+}
+
+double  SimplexNoise1234::grad( int hash, double x, double y , double z ) {
+    int h = hash & 15;     // Convert low 4 bits of hash code into 12 simple
+    double u = h<8 ? x : y; // gradient directions, and compute dot product.
+    double v = h<4 ? y : h==12||h==14 ? x : z; // Fix repeats at h = 12 to 15
+    return ((h&1)? -u : u) + ((h&2)? -v : v);
+}
+
+double  SimplexNoise1234::grad( int hash, double x, double y, double z, double t ) {
+    int h = hash & 31;      // Convert low 5 bits of hash code into 32 simple
+    double u = h<24 ? x : y; // gradient directions, and compute dot product.
+    double v = h<16 ? y : z;
+    double w = h<8 ? z : t;
+    return ((h&1)? -u : u) + ((h&2)? -v : v) + ((h&4)? -w : w);
+}
+
+// A lookup table to traverse the simplex around a given point in 4D.
+// Details can be found where this table is used, in the 4D noise method.
+/* TODO: This should not be required, backport it from Bill's GLSL code! */
+static unsigned char simplex[64][4] = {
+    {0,1,2,3},{0,1,3,2},{0,0,0,0},{0,2,3,1},{0,0,0,0},{0,0,0,0},{0,0,0,0},{1,2,3,0},
+    {0,2,1,3},{0,0,0,0},{0,3,1,2},{0,3,2,1},{0,0,0,0},{0,0,0,0},{0,0,0,0},{1,3,2,0},
+    {0,0,0,0},{0,0,0,0},{0,0,0,0},{0,0,0,0},{0,0,0,0},{0,0,0,0},{0,0,0,0},{0,0,0,0},
+    {1,2,0,3},{0,0,0,0},{1,3,0,2},{0,0,0,0},{0,0,0,0},{0,0,0,0},{2,3,0,1},{2,3,1,0},
+    {1,0,2,3},{1,0,3,2},{0,0,0,0},{0,0,0,0},{0,0,0,0},{2,0,3,1},{0,0,0,0},{2,1,3,0},
+    {0,0,0,0},{0,0,0,0},{0,0,0,0},{0,0,0,0},{0,0,0,0},{0,0,0,0},{0,0,0,0},{0,0,0,0},
+    {2,0,1,3},{0,0,0,0},{0,0,0,0},{0,0,0,0},{3,0,1,2},{3,0,2,1},{0,0,0,0},{3,1,2,0},
+    {2,1,0,3},{0,0,0,0},{0,0,0,0},{0,0,0,0},{3,1,0,2},{0,0,0,0},{3,2,0,1},{3,2,1,0}};
+
+// 1D simplex noise
+double SimplexNoise1234::noise(double x) {
+
+  int i0 = FASTFLOOR(x);
+  int i1 = i0 + 1;
+  double x0 = x - i0;
+  double x1 = x0 - 1.0;
+
+  double n0, n1;
+
+  double t0 = 1.0 - x0*x0;
+  t0 *= t0;
+  n0 = t0 * t0 * grad(perm[i0 & 0xff], x0);
+
+  double t1 = 1.0 - x1*x1;
+  t1 *= t1;
+  n1 = t1 * t1 * grad(perm[i1 & 0xff], x1);
+  // The maximum value of this noise is 8*(3/4)^4 = 2.53125
+  // A factor of 0.395 will scale to fit exactly within [-1,1]
+  return 0.395 * (n0 + n1);
+
+}
+
+// 2D simplex noise
+double SimplexNoise1234::noise(double x, double y) {
+
+#define F2 0.366025403 // F2 = 0.5*(sqrt(3.0)-1.0)
+#define G2 0.211324865 // G2 = (3.0-Math.sqrt(3.0))/6.0
+
+    double n0, n1, n2; // Noise contributions from the three corners
+
+    // Skew the input space to determine which simplex cell we're in
+    double s = (x+y)*F2; // Hairy factor for 2D
+    double xs = x + s;
+    double ys = y + s;
+    int i = FASTFLOOR(xs);
+    int j = FASTFLOOR(ys);
+
+    double t = (i+j)*G2;
+    double X0 = i-t; // Unskew the cell origin back to (x,y) space
+    double Y0 = j-t;
+    double x0 = x-X0; // The x,y distances from the cell origin
+    double y0 = y-Y0;
+
+    // For the 2D case, the simplex shape is an equilateral triangle.
+    // Determine which simplex we are in.
+    int i1, j1; // Offsets for second (middle) corner of simplex in (i,j) coords
+    if(x0>y0) {i1=1; j1=0;} // lower triangle, XY order: (0,0)->(1,0)->(1,1)
+    else {i1=0; j1=1;}      // upper triangle, YX order: (0,0)->(0,1)->(1,1)
+
+    // A step of (1,0) in (i,j) means a step of (1-c,-c) in (x,y), and
+    // a step of (0,1) in (i,j) means a step of (-c,1-c) in (x,y), where
+    // c = (3-sqrt(3))/6
+
+    double x1 = x0 - i1 + G2; // Offsets for middle corner in (x,y) unskewed coords
+    double y1 = y0 - j1 + G2;
+    double x2 = x0 - 1.0 + 2.0 * G2; // Offsets for last corner in (x,y) unskewed coords
+    double y2 = y0 - 1.0 + 2.0 * G2;
+
+    // Wrap the integer indices at 256, to avoid indexing perm[] out of bounds
+    int ii = i & 0xff;
+    int jj = j & 0xff;
+
+    // Calculate the contribution from the three corners
+    double t0 = 0.5 - x0*x0-y0*y0;
+    if(t0 < 0.0) n0 = 0.0;
+    else {
+      t0 *= t0;
+      n0 = t0 * t0 * grad(perm[ii+perm[jj]], x0, y0);
+    }
+
+    double t1 = 0.5 - x1*x1-y1*y1;
+    if(t1 < 0.0) n1 = 0.0;
+    else {
+      t1 *= t1;
+      n1 = t1 * t1 * grad(perm[ii+i1+perm[jj+j1]], x1, y1);
+    }
+
+    double t2 = 0.5 - x2*x2-y2*y2;
+    if(t2 < 0.0) n2 = 0.0;
+    else {
+      t2 *= t2;
+      n2 = t2 * t2 * grad(perm[ii+1+perm[jj+1]], x2, y2);
+    }
+
+    // Add contributions from each corner to get the final noise value.
+    // The result is scaled to return values in the interval [-1,1].
+    return 45.23 * (n0 + n1 + n2); // TODO: The scale factor is preliminary!
+  }
+
+// 3D simplex noise
+double SimplexNoise1234::noise(double x, double y, double z) {
+
+    // Simple skewing factors for the 3D case
+#define F3 0.333333333
+#define G3 0.166666667
+
+    double n0, n1, n2, n3; // Noise contributions from the four corners
+
+    // Skew the input space to determine which simplex cell we're in
+    double s = (x+y+z)*F3; // Very nice and simple skew factor for 3D
+    double xs = x+s;
+    double ys = y+s;
+    double zs = z+s;
+    int i = FASTFLOOR(xs);
+    int j = FASTFLOOR(ys);
+    int k = FASTFLOOR(zs);
+
+    double t = (float)(i+j+k)*G3; 
+    double X0 = i-t; // Unskew the cell origin back to (x,y,z) space
+    double Y0 = j-t;
+    double Z0 = k-t;
+    double x0 = x-X0; // The x,y,z distances from the cell origin
+    double y0 = y-Y0;
+    double z0 = z-Z0;
+
+    // For the 3D case, the simplex shape is a slightly irregular tetrahedron.
+    // Determine which simplex we are in.
+    int i1, j1, k1; // Offsets for second corner of simplex in (i,j,k) coords
+    int i2, j2, k2; // Offsets for third corner of simplex in (i,j,k) coords
+
+    /* This code would benefit from a backport from the GLSL version! */
+    if(x0>=y0) {
+        if(y0>=z0)
+        { i1=1; j1=0; k1=0; i2=1; j2=1; k2=0; } // X Y Z order
+        else if(x0>=z0) { i1=1; j1=0; k1=0; i2=1; j2=0; k2=1; } // X Z Y order
+        else { i1=0; j1=0; k1=1; i2=1; j2=0; k2=1; } // Z X Y order
+    }
+    else { // x0<y0
+        if(y0<z0) { i1=0; j1=0; k1=1; i2=0; j2=1; k2=1; } // Z Y X order
+        else if(x0<z0) { i1=0; j1=1; k1=0; i2=0; j2=1; k2=1; } // Y Z X order
+        else { i1=0; j1=1; k1=0; i2=1; j2=1; k2=0; } // Y X Z order
+    }
+
+    // A step of (1,0,0) in (i,j,k) means a step of (1-c,-c,-c) in (x,y,z),
+    // a step of (0,1,0) in (i,j,k) means a step of (-c,1-c,-c) in (x,y,z), and
+    // a step of (0,0,1) in (i,j,k) means a step of (-c,-c,1-c) in (x,y,z), where
+    // c = 1/6.
+
+    double x1 = x0 - i1 + G3; // Offsets for second corner in (x,y,z) coords
+    double y1 = y0 - j1 + G3;
+    double z1 = z0 - k1 + G3;
+    double x2 = x0 - i2 + 2.0f*G3; // Offsets for third corner in (x,y,z) coords
+    double y2 = y0 - j2 + 2.0f*G3;
+    double z2 = z0 - k2 + 2.0f*G3;
+    double x3 = x0 - 1.0f + 3.0f*G3; // Offsets for last corner in (x,y,z) coords
+    double y3 = y0 - 1.0f + 3.0f*G3;
+    double z3 = z0 - 1.0f + 3.0f*G3;
+
+    // Wrap the integer indices at 256, to avoid indexing perm[] out of bounds
+    int ii = i & 0xff;
+    int jj = j & 0xff;
+    int kk = k & 0xff;
+
+    // Calculate the contribution from the four corners
+    double t0 = 0.6f - x0*x0 - y0*y0 - z0*z0;
+    if(t0 < 0.0f) n0 = 0.0f;
+    else {
+        t0 *= t0;
+        n0 = t0 * t0 * grad(perm[ii+perm[jj+perm[kk]]], x0, y0, z0);
+    }
+
+    double t1 = 0.6f - x1*x1 - y1*y1 - z1*z1;
+    if(t1 < 0.0f) n1 = 0.0f;
+    else {
+        t1 *= t1;
+        n1 = t1 * t1 * grad(perm[ii+i1+perm[jj+j1+perm[kk+k1]]], x1, y1, z1);
+    }
+
+    double t2 = 0.6f - x2*x2 - y2*y2 - z2*z2;
+    if(t2 < 0.0f) n2 = 0.0f;
+    else {
+        t2 *= t2;
+        n2 = t2 * t2 * grad(perm[ii+i2+perm[jj+j2+perm[kk+k2]]], x2, y2, z2);
+    }
+
+    double t3 = 0.6f - x3*x3 - y3*y3 - z3*z3;
+    if(t3<0.0f) n3 = 0.0f;
+    else {
+        t3 *= t3;
+        n3 = t3 * t3 * grad(perm[ii+1+perm[jj+1+perm[kk+1]]], x3, y3, z3);
+    }
+
+    // Add contributions from each corner to get the final noise value.
+    // The result is scaled to stay just inside [-1,1]
+    return 32.74 * (n0 + n1 + n2 + n3); // TODO: The scale factor is preliminary!
+}
+
+
+// 4D simplex noise
+double SimplexNoise1234::noise(double x, double y, double z, double w) {
+
+    // The skewing and unskewing factors are hairy again for the 4D case
+#define F4 0.309016994 // F4 = (Math.sqrt(5.0)-1.0)/4.0
+#define G4 0.138196601 // G4 = (5.0-Math.sqrt(5.0))/20.0
+
+    double n0, n1, n2, n3, n4; // Noise contributions from the five corners
+
+    // Skew the (x,y,z,w) space to determine which cell of 24 simplices we're in
+    double s = (x + y + z + w) * F4; // Factor for 4D skewing
+    double xs = x + s;
+    double ys = y + s;
+    double zs = z + s;
+    double ws = w + s;
+    int i = FASTFLOOR(xs);
+    int j = FASTFLOOR(ys);
+    int k = FASTFLOOR(zs);
+    int l = FASTFLOOR(ws);
+
+    double t = (i + j + k + l) * G4; // Factor for 4D unskewing
+    double X0 = i - t; // Unskew the cell origin back to (x,y,z,w) space
+    double Y0 = j - t;
+    double Z0 = k - t;
+    double W0 = l - t;
+
+    double x0 = x - X0;  // The x,y,z,w distances from the cell origin
+    double y0 = y - Y0;
+    double z0 = z - Z0;
+    double w0 = w - W0;
+
+    // For the 4D case, the simplex is a 4D shape I won't even try to describe.
+    // To find out which of the 24 possible simplices we're in, we need to
+    // determine the magnitude ordering of x0, y0, z0 and w0.
+    // The method below is a good way of finding the ordering of x,y,z,w and
+    // then find the correct traversal order for the simplex we’re in.
+    // First, six pair-wise comparisons are performed between each possible pair
+    // of the four coordinates, and the results are used to add up binary bits
+    // for an integer index.
+    int c1 = (x0 > y0) ? 32 : 0;
+    int c2 = (x0 > z0) ? 16 : 0;
+    int c3 = (y0 > z0) ? 8 : 0;
+    int c4 = (x0 > w0) ? 4 : 0;
+    int c5 = (y0 > w0) ? 2 : 0;
+    int c6 = (z0 > w0) ? 1 : 0;
+    int c = c1 + c2 + c3 + c4 + c5 + c6;
+
+    int i1, j1, k1, l1; // The integer offsets for the second simplex corner
+    int i2, j2, k2, l2; // The integer offsets for the third simplex corner
+    int i3, j3, k3, l3; // The integer offsets for the fourth simplex corner
+
+    // simplex[c] is a 4-vector with the numbers 0, 1, 2 and 3 in some order.
+    // Many values of c will never occur, since e.g. x>y>z>w makes x<z, y<w and x<w
+    // impossible. Only the 24 indices which have non-zero entries make any sense.
+    // We use a thresholding to set the coordinates in turn from the largest magnitude.
+    // The number 3 in the "simplex" array is at the position of the largest coordinate.
+    i1 = simplex[c][0]>=3 ? 1 : 0;
+    j1 = simplex[c][1]>=3 ? 1 : 0;
+    k1 = simplex[c][2]>=3 ? 1 : 0;
+    l1 = simplex[c][3]>=3 ? 1 : 0;
+    // The number 2 in the "simplex" array is at the second largest coordinate.
+    i2 = simplex[c][0]>=2 ? 1 : 0;
+    j2 = simplex[c][1]>=2 ? 1 : 0;
+    k2 = simplex[c][2]>=2 ? 1 : 0;
+    l2 = simplex[c][3]>=2 ? 1 : 0;
+    // The number 1 in the "simplex" array is at the second smallest coordinate.
+    i3 = simplex[c][0]>=1 ? 1 : 0;
+    j3 = simplex[c][1]>=1 ? 1 : 0;
+    k3 = simplex[c][2]>=1 ? 1 : 0;
+    l3 = simplex[c][3]>=1 ? 1 : 0;
+    // The fifth corner has all coordinate offsets = 1, so no need to look that up.
+
+    double x1 = x0 - i1 + G4; // Offsets for second corner in (x,y,z,w) coords
+    double y1 = y0 - j1 + G4;
+    double z1 = z0 - k1 + G4;
+    double w1 = w0 - l1 + G4;
+    double x2 = x0 - i2 + 2.0f*G4; // Offsets for third corner in (x,y,z,w) coords
+    double y2 = y0 - j2 + 2.0f*G4;
+    double z2 = z0 - k2 + 2.0f*G4;
+    double w2 = w0 - l2 + 2.0f*G4;
+    double x3 = x0 - i3 + 3.0f*G4; // Offsets for fourth corner in (x,y,z,w) coords
+    double y3 = y0 - j3 + 3.0f*G4;
+    double z3 = z0 - k3 + 3.0f*G4;
+    double w3 = w0 - l3 + 3.0f*G4;
+    double x4 = x0 - 1.0f + 4.0f*G4; // Offsets for last corner in (x,y,z,w) coords
+    double y4 = y0 - 1.0f + 4.0f*G4;
+    double z4 = z0 - 1.0f + 4.0f*G4;
+    double w4 = w0 - 1.0f + 4.0f*G4;
+
+    // Wrap the integer indices at 256, to avoid indexing perm[] out of bounds
+    int ii = i & 0xff;
+    int jj = j & 0xff;
+    int kk = k & 0xff;
+    int ll = l & 0xff;
+
+    // Calculate the contribution from the five corners
+    double t0 = 0.6f - x0*x0 - y0*y0 - z0*z0 - w0*w0;
+    if(t0 < 0.0f) n0 = 0.0f;
+    else {
+        t0 *= t0;
+        n0 = t0 * t0 * grad(perm[ii+perm[jj+perm[kk+perm[ll]]]], x0, y0, z0, w0);
+    }
+
+    double t1 = 0.6f - x1*x1 - y1*y1 - z1*z1 - w1*w1;
+    if(t1 < 0.0f) n1 = 0.0f;
+    else {
+        t1 *= t1;
+        n1 = t1 * t1 * grad(perm[ii+i1+perm[jj+j1+perm[kk+k1+perm[ll+l1]]]], x1, y1, z1, w1);
+    }
+
+    double t2 = 0.6f - x2*x2 - y2*y2 - z2*z2 - w2*w2;
+    if(t2 < 0.0f) n2 = 0.0f;
+    else {
+        t2 *= t2;
+        n2 = t2 * t2 * grad(perm[ii+i2+perm[jj+j2+perm[kk+k2+perm[ll+l2]]]], x2, y2, z2, w2);
+    }
+
+    double t3 = 0.6f - x3*x3 - y3*y3 - z3*z3 - w3*w3;
+    if(t3 < 0.0f) n3 = 0.0f;
+    else {
+        t3 *= t3;
+        n3 = t3 * t3 * grad(perm[ii+i3+perm[jj+j3+perm[kk+k3+perm[ll+l3]]]], x3, y3, z3, w3);
+    }
+
+    double t4 = 0.6f - x4*x4 - y4*y4 - z4*z4 - w4*w4;
+    if(t4 < 0.0f) n4 = 0.0f;
+    else {
+        t4 *= t4;
+        n4 = t4 * t4 * grad(perm[ii+1+perm[jj+1+perm[kk+1+perm[ll+1]]]], x4, y4, z4, w4);
+    }
+
+    // Sum up and scale the result to cover the range [-1,1]
+    return 27.3 * (n0 + n1 + n2 + n3 + n4); // TODO: The scale factor is preliminary!
+}
+//---------------------------------------------------------------------

examples/simplexnoise1234.h

@@ -0,0 +1,48 @@
+// SimplexNoise1234
+// Copyright © 2003-2011, Stefan Gustavson
+//
+// Contact: stegu@itn.liu.se
+//
+// This library is public domain software, released by the author
+// into the public domain in February 2011. You may do anything
+// you like with it. You may even remove all attributions,
+// but of course I'd appreciate it if you kept my name somewhere.
+//
+// This library is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+// General Public License for more details.
+
+// Modified by the LOVE Development Team to use double precision.
+
+/** \file
+		\brief Declares the SimplexNoise1234 class for producing Perlin simplex noise.
+		\author Stefan Gustavson (stegu@itn.liu.se)
+*/
+
+/*
+ * This is a clean, fast, modern and free Perlin Simplex noise class in C++.
+ * Being a stand-alone class with no external dependencies, it is
+ * highly reusable without source code modifications.
+ */
+
+class SimplexNoise1234 {
+
+  public:
+    SimplexNoise1234() {}
+    ~SimplexNoise1234() {}
+
+/** 1D and 2D float Perlin noise
+ */
+    static double noise( double x );
+    static double noise( double x, double y );
+    static double noise( double x, double y, double z );
+    static double noise( double x, double y, double z, double w);
+
+  private:
+    static unsigned char perm[];
+    static double  grad( int hash, double x );
+    static double  grad( int hash, double x, double y );
+    static double  grad( int hash, double x, double y, double z );
+    static double  grad( int hash, double x, double y, double z, double t );
+};

include/sprstk/sprstk.h

@@ -30,11 +30,14 @@ void sprstk_del(sprstk* instance);
 void sprstk_run(sprstk* instance);
 void sprstk_stop(sprstk* instance);
 
+void sprstk_clear(sprstk* instance);
+
 void sprstk_put(sprstk* instance, int x, int y, unsigned int layers, unsigned int palette_lookup);
 void sprstk_putz(sprstk* instance, int x, int y, unsigned int layers, unsigned int palette_lookup, unsigned int z_offset);
 
 void sprstk_set_palette(sprstk* instance, unsigned int index, const sprstk_palette* palette);
 
+void sprstk_set_scale(sprstk* instance, float scale);
 void sprstk_set_angle(sprstk* instance, float angle);
 
 #ifdef __cplusplus

src/sprstk.cpp

@@ -7,7 +7,11 @@
 
 #include <cmath>
 
-const char* MESH_SHADER_CODE = R"(
+#define OIT_LAYERS 32
+#define _STRINGIFY(x) #x
+#define STRINGIFY(x) _STRINGIFY(x)
+
+const char* AZ_PASS_MESH_SHADER_CODE = R"(
 #version 460
 
 #extension GL_NV_mesh_shader : require
@@ -17,11 +21,13 @@ layout (triangles, max_vertices = 128, max_primitives = 64) out;
 
 layout (location = 0) out PerVertexData
 {
-  vec4 color;
+  flat uint layer;
+  flat uint color;
 } v_out[];
 
 layout (location = 1) uniform vec3 screen_size_and_pixel_scale;
-layout (location = 2) uniform mat2 rotation_matrix;
+layout (location = 2) uniform float scale;
+layout (location = 3) uniform mat2 rotation_matrix;
 
 struct TileInfo
 {
@@ -54,31 +60,32 @@ void main()
 
   uint layer_count = bitfieldExtract(t_info.position, 20, 5);
 
-  vec2 positions[4] = { vec2(0, 0), vec2(1, 0), vec2(0, 1), vec2(1, 1) };
+  float minsize = min(screen_size_and_pixel_scale.x, screen_size_and_pixel_scale.y);
+
+  vec2 positions[4] = { vec2(-0.5, -0.5), vec2(0.5, -0.5), vec2(-0.5, 0.5), vec2(0.5, 0.5) };
   for (uint i = 0; i < 4; i++)
   {
     positions[i] += stack_position;
     positions[i] *= screen_size_and_pixel_scale.zz;
-    positions[i] /= vec2(min(screen_size_and_pixel_scale.x, screen_size_and_pixel_scale.y));
   }
 
   uint z_offset = bitfieldExtract(t_info.position, 25, 2);
 
   uint palette_lookup = bitfieldExtract(t_info.position, 27, 5);
   ColorInfo c_info = color_infos[palette_lookup];
-  float a = bitfieldExtract(c_info.color[gl_LocalInvocationID.x], 0, 8);
+  uint c = c_info.color[gl_LocalInvocationID.x];
-  float b = bitfieldExtract(c_info.color[gl_LocalInvocationID.x], 8, 8);
-  float g = bitfieldExtract(c_info.color[gl_LocalInvocationID.x], 16, 8);
-  float r = bitfieldExtract(c_info.color[gl_LocalInvocationID.x], 24, 8);
 
   for (uint i = 4 * gl_LocalInvocationID.x; i < 4 * gl_LocalInvocationID.x + 4; i++)
   {
     vec4 position = vec4(rotation_matrix * positions[i % 4], float(4 * gl_LocalInvocationID.x + z_offset) / 128, 1);
-    position.y += 24 * position.z / screen_size_and_pixel_scale.z;
+    position.xy /= screen_size_and_pixel_scale.xy;
-    position.xy *= vec2(0.05);
+    position.xy *= scale;
+    position.y += 20 * gl_LocalInvocationID.x * scale / screen_size_and_pixel_scale.y;
     gl_MeshVerticesNV[i].gl_Position = position;
 
-    v_out[i].color = vec4(r, g, b, a) / vec4(256, 256, 256, 256);
+    v_out[i].layer = 4 * gl_LocalInvocationID.x + z_offset;
+
+    v_out[i].color = c;
   }
 
   for (uint i = 6 * gl_LocalInvocationID.x; i < 6 * gl_LocalInvocationID.x + 6; i++)
@@ -90,19 +97,86 @@ void main()
 }
 )";
 
-const char* FRAGMENT_SHADER_CODE = R"(
+const char* AZ_PASS_FRAGMENT_SHADER_CODE = R"(
 #version 460
 
-layout(location = 0) out vec4 FragColor;
+layout (location = 0) out vec4 FragColor;
+
+layout (binding = 0, rg32ui) uniform restrict writeonly uimage3D AZBuffer;
+layout (binding = 1, r32ui) uniform restrict uimage2D AZNextBuffer;
 
 in PerVertexData
 {
-  vec4 color;
+  flat uint layer;
+  flat uint color;
 } fragIn;
 
 void main()
 {
-  FragColor = fragIn.color;
+  const uint position = imageAtomicAdd(AZNextBuffer, ivec2(gl_FragCoord.xy), 1);
+
+  if (position < )" STRINGIFY(OIT_LAYERS) R"()
+  {
+    imageStore(AZBuffer, ivec3(gl_FragCoord.xy, position), uvec4(fragIn.layer, fragIn.color, 0, 0));
+  }
+
+  FragColor = vec4(0);
+}
+)";
+
+const char* COMPOSITE_PASS_VERTEX_SHADER = R"(
+#version 460
+
+vec2 positions[3] = { vec2(-2.1, -1.1), vec2(0, 3.1), vec2(2.1, -1.1) };
+
+void main()
+{
+  gl_Position = vec4(positions[gl_VertexID], 0, 1);
+}
+)";
+
+const char* COMPOSITE_PASS_FRAGMENT_SHADER = R"(
+#version 460
+
+layout (location = 0) out vec4 FragColor;
+
+layout (binding = 0, rg32ui) uniform restrict readonly uimage3D AZBuffer;
+layout (binding = 1, r32ui) uniform restrict readonly uimage2D AZNextBuffer;
+
+void main()
+{
+  uvec2 data[ )" STRINGIFY(OIT_LAYERS) R"(];
+
+  const uint layer_count = imageLoad(AZNextBuffer, ivec2(gl_FragCoord.xy)).x;
+
+  for (uint i = 0; i < layer_count; i++)
+  {
+    data[i] = imageLoad(AZBuffer, ivec3(gl_FragCoord.xy, i)).xy;
+  }
+
+  for (int i = 0; i < layer_count; i++)
+  {
+    for (int j = i; j > 0 && data[j - 1].x > data[j].x; j--)
+    {
+      const uvec2 temp = data[j];
+      data[j] = data[j - 1];
+      data[j - 1] = temp;
+    }
+  }
+
+  vec3 color = vec3(0);
+  for (int i = 0; i < layer_count; i++)
+  {
+    uint a = bitfieldExtract(data[i].y, 0, 8);
+    uint b = bitfieldExtract(data[i].y, 8, 8);
+    uint g = bitfieldExtract(data[i].y, 16, 8);
+    uint r = bitfieldExtract(data[i].y, 24, 8);
+    vec4 temp_color = vec4(r, g, b, a) / vec4(255);
+
+    color = vec3(temp_color.rgb * temp_color.a + color.rgb * (1 - temp_color.a));
+  }
+
+  FragColor = vec4(color, 1);
 }
 )";
 
@@ -124,8 +198,13 @@ public:
     callbacks(callbacks),
     userdata(userdata),
     should_stop(false),
-    prev_ticks(0)
+    prev_ticks(0),
+    prev_resized_ticks(0),
+    resized(false)
   {
+    gl.az_buffer = 0;
+    gl.az_next_buffer = 0;
+
     if (!callbacks.update)
     {
       throw std::runtime_error("No update callback");
@@ -158,10 +237,8 @@ public:
 
         if (e.type == SDL_EVENT_WINDOW_RESIZED)
         {
-          int width, height;
+          prev_resized_ticks = prev_ticks;
-          SDL_GetWindowSizeInPixels(sdl.window, &width, &height);
+          resized = true;
-          glViewport(0, 0, width, height);
-          glProgramUniform3f(gl.program, 1, width, height, 8);
         }
       }
 
@@ -169,17 +246,46 @@ public:
       float dt = (current_ticks - prev_ticks) / 1000.0f;
       prev_ticks = current_ticks;
 
-      gl.tile_count = 0;
+      if (resized && (current_ticks - prev_resized_ticks) / 1000.0f > 0.5f)
+      {
+        prev_resized_ticks = 0;
+        resized = false;
+
+        int width, height;
+        SDL_GetWindowSizeInPixels(sdl.window, &width, &height);
+        glViewport(0, 0, width, height);
+        glProgramUniform3f(gl.az_pass_program, 1, width, height, 8);
+
+        update_buffers();
+      }
+
       callbacks.update(this, dt, userdata);
 
       glClearColor(0, 0, 0, 1);
       glClear(GL_COLOR_BUFFER_BIT);
-      int i;
+
-      for (i = 0; i < gl.tile_count; i += 65535)
+      constexpr uint32_t color = 0;
+      glClearTexImage(gl.az_buffer, 0, GL_RG_INTEGER, GL_UNSIGNED_INT, &color);
+
+      constexpr uint8_t z = 0;
+      glClearTexImage(gl.az_next_buffer, 0, GL_RED_INTEGER, GL_UNSIGNED_BYTE, &z);
+
+      glUseProgram(gl.az_pass_program);
+      for (int i = 0; i < gl.tile_count; i += 65535)
       {
-        glDrawMeshTasksNV(i, 65535);
+        int count = 65535;
+        if (i + 65535 > gl.tile_count)
+        {
+          count = gl.tile_count - i;
+        }
+        glDrawMeshTasksNV(i, count);
       }
 
+      glMemoryBarrier(GL_SHADER_IMAGE_ACCESS_BARRIER_BIT);
+
+      glUseProgram(gl.composite_pass_program);
+      glDrawArrays(GL_TRIANGLES, 0, 3);
+
       SDL_GL_SwapWindow(sdl.window);
     }
 
@@ -191,6 +297,11 @@ public:
     should_stop = true;
   }
 
+  void clear()
+  {
+    gl.tile_count = 0;
+  }
+
   void put(int x, int y, unsigned int layers, unsigned int palette_lookup, unsigned int z_offset = 0)
   {
     x += 512;
@@ -219,13 +330,18 @@ public:
     gl.color_info_map[index] = *palette;
   }
 
+  void set_scale(float scale)
+  {
+    glProgramUniform1f(gl.az_pass_program, 2, scale);
+  }
+
   void set_angle(float angle)
   {
     const float arr[4] = {
       cosf(angle), -sinf(angle),
       sinf(angle),  cosf(angle)
     };
-    glProgramUniformMatrix2fv(gl.program, 2, 1, false, arr);
+    glProgramUniformMatrix2fv(gl.az_pass_program, 3, 1, false, arr);
   }
 
 private:
@@ -234,6 +350,8 @@ private:
   bool should_stop;
 
   uint64_t prev_ticks;
+  uint64_t prev_resized_ticks;
+  bool resized;
 
   struct
   {
@@ -242,12 +360,16 @@ private:
   } sdl;
   struct
   {
-    unsigned int program;
+    unsigned int vao;
+    unsigned int az_pass_program;
+    unsigned int composite_pass_program;
     unsigned int tile_buffer;
     TileInfo* tile_buffer_map;
     unsigned int tile_count;
     unsigned int color_buffer;
     sprstk_palette* color_info_map;
+    unsigned int az_buffer;
+    unsigned int az_next_buffer;
   } gl;
 
   void init_sdl()
@@ -262,7 +384,7 @@ private:
     SDL_GL_SetAttribute(SDL_GL_CONTEXT_PROFILE_MASK, SDL_GL_CONTEXT_PROFILE_CORE);
     SDL_GL_SetAttribute(SDL_GL_DOUBLEBUFFER, 1);
     SDL_GL_SetAttribute(SDL_GL_DEPTH_SIZE, 24);
-    SDL_GL_SetAttribute(SDL_GL_CONTEXT_FLAGS, SDL_GL_CONTEXT_DEBUG_FLAG);
+    //SDL_GL_SetAttribute(SDL_GL_CONTEXT_FLAGS, SDL_GL_CONTEXT_DEBUG_FLAG);
 
     sdl.window = SDL_CreateWindow("sprstk", 640, 480, SDL_WINDOW_OPENGL | SDL_WINDOW_RESIZABLE);
     if (!sdl.window)
@@ -294,19 +416,19 @@ private:
       throw application_error("Mesh shaders not supported");
     }
 
-    glEnable(GL_BLEND);
-    glBlendFuncSeparate(GL_SRC_ALPHA, GL_ONE_MINUS_SRC_ALPHA, GL_ONE, GL_ZERO);
-
     glDebugMessageCallback([](GLenum source, GLenum type, GLuint id, GLenum severity, GLsizei length, const GLchar* message, const void* userdata)
     {
       SDL_LogWarn(SDL_LOG_CATEGORY_APPLICATION, "%s", message);
     }, nullptr);
 
+    glGenVertexArrays(1, &gl.vao);
+    glBindVertexArray(gl.vao);
+
     int success;
     char info_log[512];
 
     unsigned int mesh = glCreateShader(GL_MESH_SHADER_NV);
-    glShaderSource(mesh, 1, &MESH_SHADER_CODE, nullptr);
+    glShaderSource(mesh, 1, &AZ_PASS_MESH_SHADER_CODE, nullptr);
     glCompileShader(mesh);
     glGetShaderiv(mesh, GL_COMPILE_STATUS, &success);
     if (!success)
@@ -315,41 +437,73 @@ private:
       SDL_LogError(SDL_LOG_CATEGORY_APPLICATION, "Mesh shader: %s", info_log);
     }
 
-    unsigned int fragment = glCreateShader(GL_FRAGMENT_SHADER);
+    unsigned int az_fragment = glCreateShader(GL_FRAGMENT_SHADER);
-    glShaderSource(fragment, 1, &FRAGMENT_SHADER_CODE, nullptr);
+    glShaderSource(az_fragment, 1, &AZ_PASS_FRAGMENT_SHADER_CODE, nullptr);
-    glCompileShader(fragment);
+    glCompileShader(az_fragment);
-    glGetShaderiv(fragment, GL_COMPILE_STATUS, &success);
+    glGetShaderiv(az_fragment, GL_COMPILE_STATUS, &success);
     if (!success)
     {
-      glGetShaderInfoLog(fragment, sizeof(info_log), nullptr, info_log);
+      glGetShaderInfoLog(az_fragment, sizeof(info_log), nullptr, info_log);
-      SDL_LogError(SDL_LOG_CATEGORY_APPLICATION, "Fragment shader: %s", info_log);
+      SDL_LogError(SDL_LOG_CATEGORY_APPLICATION, "AZ Pass fragment shader: %s", info_log);
     }
 
-    gl.program = glCreateProgram();
+    gl.az_pass_program = glCreateProgram();
-    glAttachShader(gl.program, mesh);
+    glAttachShader(gl.az_pass_program, mesh);
-    glAttachShader(gl.program, fragment);
+    glAttachShader(gl.az_pass_program, az_fragment);
-    glLinkProgram(gl.program);
+    glLinkProgram(gl.az_pass_program);
-    glGetProgramiv(gl.program, GL_LINK_STATUS, &success);
+    glGetProgramiv(gl.az_pass_program, GL_LINK_STATUS, &success);
     if (!success)
     {
-      glGetProgramInfoLog(gl.program, sizeof(info_log), nullptr, info_log);
+      glGetProgramInfoLog(gl.az_pass_program, sizeof(info_log), nullptr, info_log);
-      SDL_LogError(SDL_LOG_CATEGORY_APPLICATION, "Program: %s", info_log);
+      SDL_LogError(SDL_LOG_CATEGORY_APPLICATION, "AZ pass program: %s", info_log);
     }
 
     glDeleteShader(mesh);
-    glDeleteShader(fragment);
+    glDeleteShader(az_fragment);
+
+    unsigned int vertex = glCreateShader(GL_VERTEX_SHADER);
+    glShaderSource(vertex, 1, &COMPOSITE_PASS_VERTEX_SHADER, nullptr);
+    glCompileShader(vertex);
+    glGetShaderiv(vertex, GL_COMPILE_STATUS, &success);
+    if (!success)
+    {
+      glGetShaderInfoLog(vertex, sizeof(info_log), nullptr, info_log);
+      SDL_LogError(SDL_LOG_CATEGORY_APPLICATION, "Vertex shader: %s", info_log);
+    }
+
+    unsigned int composite_pass_fragment = glCreateShader(GL_FRAGMENT_SHADER);
+    glShaderSource(composite_pass_fragment, 1, &COMPOSITE_PASS_FRAGMENT_SHADER, nullptr);
+    glCompileShader(composite_pass_fragment);
+    glGetShaderiv(composite_pass_fragment, GL_COMPILE_STATUS, &success);
+    if (!success)
+    {
+      glGetShaderInfoLog(composite_pass_fragment, sizeof(info_log), nullptr, info_log);
+      SDL_LogError(SDL_LOG_CATEGORY_APPLICATION, "Composite pass fragment shader: %s", info_log);
+    }
+
+    gl.composite_pass_program = glCreateProgram();
+    glAttachShader(gl.composite_pass_program, vertex);
+    glAttachShader(gl.composite_pass_program, composite_pass_fragment);
+    glLinkProgram(gl.composite_pass_program);
+    glGetProgramiv(gl.composite_pass_program, GL_LINK_STATUS, &success);
+    if (!success)
+    {
+      glGetProgramInfoLog(gl.az_pass_program, sizeof(info_log), nullptr, info_log);
+      SDL_LogError(SDL_LOG_CATEGORY_APPLICATION, "Composite pass program: %s", info_log);
+    }
+
+    glDeleteShader(vertex);
+    glDeleteShader(composite_pass_fragment);
 
     if (!success)
     {
       throw application_error("Failed to compile and link shader program");
     }
 
-    glUseProgram(gl.program);
-
     int width, height;
     SDL_GetWindowSizeInPixels(sdl.window, &width, &height);
     glViewport(0, 0, width, height);
-    glProgramUniform3f(gl.program, 1, width, height, 8);
+    glProgramUniform3f(gl.az_pass_program, 1, width, height, 8);
 
     glGenBuffers(1, &gl.tile_buffer);
     glBindBufferBase(GL_SHADER_STORAGE_BUFFER, 0, gl.tile_buffer);
@@ -365,15 +519,49 @@ private:
     gl.color_info_map = (sprstk_palette*)glMapNamedBufferRange(gl.color_buffer, 0, sizeof(sprstk_palette) * (1 << 5), GL_MAP_WRITE_BIT | GL_MAP_COHERENT_BIT);
 
     const float arr[4] = {1, 0, 0, 1};
-    glProgramUniformMatrix2fv(gl.program, 2, 1, false, arr);
+    glProgramUniformMatrix2fv(gl.az_pass_program, 3, 1, false, arr);
+    glProgramUniform1f(gl.az_pass_program, 2, 1);
+
+    update_buffers();
   }
 
   void destroy_gl()
   {
-    glDeleteProgram(gl.program);
+    glDeleteTextures(1, &gl.az_buffer);
+    glDeleteTextures(1, &gl.az_next_buffer);
+    glDeleteProgram(gl.az_pass_program);
+    glDeleteProgram(gl.composite_pass_program);
+    glDeleteBuffers(1, &gl.color_buffer);
     glDeleteBuffers(1, &gl.tile_buffer);
     SDL_GL_DestroyContext(sdl.context);
   }
+
+  void update_buffers()
+  {
+    if (gl.az_buffer)
+    {
+      glDeleteTextures(1, &gl.az_buffer);
+      glDeleteTextures(1, &gl.az_next_buffer);
+    }
+
+    int width, height;
+    SDL_GetWindowSizeInPixels(sdl.window, &width, &height);
+
+    glGenTextures(1, &gl.az_buffer);
+    glBindTexture(GL_TEXTURE_3D, gl.az_buffer);
+    glTexStorage3D(GL_TEXTURE_3D, 1, GL_RG32UI, width, height, OIT_LAYERS);
+    glTexParameteri(GL_TEXTURE_3D, GL_TEXTURE_MIN_FILTER, GL_NEAREST);
+    glTexParameteri(GL_TEXTURE_3D, GL_TEXTURE_MAG_FILTER, GL_NEAREST);
+
+    glGenTextures(1, &gl.az_next_buffer);
+    glBindTexture(GL_TEXTURE_2D, gl.az_next_buffer);
+    glTexStorage2D(GL_TEXTURE_2D, 1, GL_R32UI, width, height);
+    glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST);
+    glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_NEAREST);
+
+    glBindImageTexture(0, gl.az_buffer, 0, true, 0, GL_READ_WRITE, GL_RG32UI);
+    glBindImageTexture(1, gl.az_next_buffer, 0, false, 0, GL_READ_WRITE, GL_R32UI);
+  }
 };
 
 extern "C"
@@ -420,6 +608,11 @@ void sprstk_stop(sprstk* instance)
   instance->stop();
 }
 
+void sprstk_clear(sprstk* instance)
+{
+  instance->clear();
+}
+
 void sprstk_put(sprstk* instance, int x, int y, unsigned int layers, unsigned int palette_lookup)
 {
   instance->put(x, y, layers, palette_lookup);
@@ -435,6 +628,11 @@ void sprstk_set_palette(sprstk* instance, unsigned int index, const sprstk_palet
   instance->set_palette(index, palette);
 }
 
+void sprstk_set_scale(sprstk* instance, float scale)
+{
+  instance->set_scale(scale);
+}
+
 void sprstk_set_angle(sprstk* instance, float angle)
 {
   instance->set_angle(angle);