Working on OpenGL explanation

Author: wedesoft

src/volumetric_clouds/main.clj

@@ -29,8 +29,6 @@
              [org.lwjgl.opengl GL GL11 GL12 GL13 GL15 GL20 GL30 GL32 GL42]))
 
 
-;; # Procedural generation of volumetric clouds
-;;
 ;; Volumetric clouds are commonly used in flight simulators and visual effects.
 ;; For a introductory video see [Sebastian Lague's video "Coding Adventure: Clouds](https://www.youtube.com/watch?v=4QOcCGI6xOU).
 ;; Note that this article is about procedural generation and not about simulating real weather.
@@ -39,7 +37,7 @@
 ;;
 ;; [Worley noise](https://en.wikipedia.org/wiki/Worley_noise) is a type of structured noise which is defined for each pixel using the distance to the nearest seed point.
 ;;
-;; ### Noise parameters
+;; #### Noise parameters
 ;;
 ;; First we define a function to create parameters of the noise.
 ;;
@@ -57,7 +55,7 @@
       (make-noise-params 256 8 2) => {:size 256 :divisions 8 :cellsize 32 :dimensions 2})
 
 
-;; ### 2D and 3D vectors
+;; #### 2D and 3D vectors
 ;;
 ;; Next we need a function which allows us to create 2D or 3D vectors depending on the number of input parameters.
 (defn vec-n
@@ -69,7 +67,7 @@
        (vec-n 2 3 1) => (vec3 2 3 1))
 
 
-;; ### Random points
+;; #### Random points
 ;;
 ;; The following method generates a random point in a cell specified by the cell indices.
 (defn random-point-in-cell
@@ -115,7 +113,7 @@
       (plotly/layer-point {:=x :x :=y :y})))
 
 
-;; ### Modular distance
+;; #### Modular distance
 ;;
 ;; In order to get a periodic noise array, we need to component-wise wrap around distance vectors.
 (defn mod-vec
@@ -158,7 +156,7 @@
          0   5   0   0   3.0)
 
 
-;; ### Modular lookup
+;; #### Modular lookup
 ;;
 ;; We also need to lookup elements with wrap around.
 ;; We recursively use `tensor/select` and then finally the tensor as a function to lookup along each axis.
@@ -187,7 +185,7 @@
        (division-index {:cellsize 4} 7.5)  => 1
        (division-index {:cellsize 4} -0.5) => -1)
 
-;; ### Getting indices of Neighbours
+;; #### Getting indices of Neighbours
 ;;
 ;; The following function determines the neighbouring indices of a cell recursing over each dimension.
 (defn neighbours
@@ -204,7 +202,7 @@
        (neighbours 1 10) => [[0 9] [1 9] [2 9] [0 10] [1 10] [2 10] [0 11] [1 11] [2 11]])
 
 
-;; ### Sampling Worley noise
+;; #### Sampling Worley noise
 ;;
 ;; Using above functions one can now implement Worley noise.
 ;; For each pixel the distance to the closest seed point is calculated.
@@ -240,7 +238,7 @@
 ;; [Perlin noise](https://adrianb.io/2014/08/09/perlinnoise.html) is generated by choosing a random gradient vector at each cell corner.
 ;; The noise tensor's intermediate values are interpolated with a continuous function, utilizing the gradient at the corner points.
 
-;; ### Random gradients
+;; #### Random gradients
 ;;
 ;; The 2D or 3D gradients are generated by creating a vector where each component is set to a random number between -1 and 1.
 ;; Random vectors are generated until the vector length is greater 0 and lower or equal to 1.
@@ -305,7 +303,7 @@
       (plotly/base {:=title "Random gradients" :=mode "lines"})
       (plotly/layer-point {:=x :x :=y :y})))
 
-;; ### Corner vectors
+;; #### Corner vectors
 ;;
 ;; The next step is to determine the vectors to the corners of the cell for a given point.
 ;; First we define a function to determine the fractional part of a number.
@@ -346,7 +344,7 @@
          (v2 1 1) => (vec2 -0.25 -0.5)
          (v3 0 0 0) => (vec3 0.75 0.5 0.25)))
 
-;; ### Extract gradients of cell corners
+;; #### Extract gradients of cell corners
 ;;
 ;; The function below retrieves the gradient values at a cell's corners, utilizing `wrap-get` for modular access.
 (defn corner-gradients
@@ -372,7 +370,7 @@
          ((corner-gradients {:cellsize 4 :dimensions 3} gradients3 (vec3 9 6 3)) 0 0 0)
          => (vec3 2 1 0)))
 
-;; ### Influence values
+;; #### Influence values
 ;;
 ;; The influence value is the function value of the function with the selected random gradient at a corner.
 (defn influence-values
@@ -395,7 +393,7 @@
          (influence2 1 1) => 11.0
          (influence3 1 1 1) => 111.0))
 
-;; ### Interpolating the influence values
+;; #### Interpolating the influence values
 ;;
 ;; For interpolation the following "ease curve" is used.
 (defn ease-curve
@@ -437,7 +435,7 @@
          (weights3 0 0 0) => (roughly 0.010430 1e-6)))
 
 
-;; ### Sampling Perlin noise
+;; #### Sampling Perlin noise
 ;;
 ;; A Perlin noise sample is computed by
 ;; * Getting the random gradients for the cell corners.
@@ -478,15 +476,15 @@
 
 ;; ## Mixing noise values
 ;;
-;; ### Combination of Worley and Perlin noise
+;; #### Combination of Worley and Perlin noise
 ;;
 ;; One can mix Worley and Perlin noise by simply doing a linear combination of the two.
 (def perlin-worley-norm (dfn/+ (dfn/* 0.3 perlin-norm) (dfn/* 0.7 worley-norm)))
 
 ;; Here for example is the average of Perlin and Worley noise.
 (bufimg/tensor->image (dfn/+ (dfn/* 0.5 perlin-norm) (dfn/* 0.5 worley-norm)))
 
-;; ### Interpolation
+;; #### Interpolation
 ;;
 ;; One can linearly interpolate tensor values by recursing over the dimensions as follows.
 (defn interpolate
@@ -517,7 +515,7 @@
          (interpolate y3 2.5 3.5 3.0) => 3.0
          (interpolate z3 2.5 3.5 5.5) => 2.0))
 
-;; ### Octaves of noise
+;; #### Octaves of noise
 ;;
 ;; Fractal Brownian Motion is implemented by computing a weighted sum of the same base noise function using different frequencies.
 (defn fractal-brownian-motion
@@ -547,7 +545,7 @@
          (fractal-brownian-motion base1 [0.0 1.0] 0.0) => 0.0
          (fractal-brownian-motion base1 [0.0 1.0] 0.5) => 1.0))
 
-;; ### Remapping and clamping
+;; #### Remapping and clamping
 ;;
 ;; The remap function is used to map a range of values of an input tensor to a different range.
 (defn remap
@@ -580,7 +578,7 @@
        0      2    3      2
        4      2    3      3)
 
-;; ### Generating octaves of noise
+;; #### Generating octaves of noise
 ;;
 ;; The octaves function is to create a series of decreasing weights and normalize them so that they add up to 1.
 (defn octaves
@@ -609,7 +607,7 @@
         low high 0 255)
       0 255)))
 
-;; ### 2D examples
+;; #### 2D examples
 ;;
 ;; Here is an example of 4 octaves of Worley noise.
 (bufimg/tensor->image (noise-octaves worley-norm (octaves 4 0.6) 120 230))
@@ -621,8 +619,11 @@
 (bufimg/tensor->image (noise-octaves perlin-worley-norm (octaves 4 0.6) 120 230))
 
 
-;; ## Testing shaders
+;; ## Volumetric clouds
 
+;; #### OpenGL setup
+;;
+;; In order to render the clouds we create a window and an OpenGL context.
 (GLFW/glfwInit)
 
 (def window-width 640)
@@ -635,7 +636,9 @@
 (GLFW/glfwMakeContextCurrent window)
 (GL/createCapabilities)
 
-
+;; #### Compiling and linking shader programs
+;;
+;; The following method is used compile a shader program.
 (defn make-shader [source shader-type]
   (let [shader (GL20/glCreateShader shader-type)]
     (GL20/glShaderSource shader source)
@@ -664,17 +667,17 @@
     program))
 
 
-(def vertex-test "
-#version 130
+(def vertex-test
+"#version 130
 in vec3 point;
 void main()
 {
   gl_Position = vec4(point, 1);
 }")
 
 
-(def fragment-test "
-#version 130
+(def fragment-test
+"#version 130
 out vec4 fragColor;
 void main()
 {