Faster random implementation

.gitignore

@@ -8,3 +8,6 @@ raytracer
 
 # Actual output image
 image.ppm
+
+# Profiler data
+perf.*

Makefile

@@ -5,6 +5,8 @@ FLAGS=-Ofast -march=native -g -Wall -Wextra -Wpedantic
 raytracer: camera.hpp color.hpp hittable.hpp hittable_list.hpp main.cpp material.hpp ray.hpp rtweekend.hpp sphere.hpp vec3.hpp $(INCLUDE)/Remotery.c $(INCLUDE)/Remotery.h
 	g++ $(FLAGS) -I$(INCLUDE) $(LIBS) main.cpp -o raytracer
 
+make debug:
+
 image: raytracer
 	@./raytracer > image.ppm
 	@if [ $$TERM = "xterm-kitty" ]; then\

hittable_list.hpp

@@ -34,8 +34,8 @@ bool hittable_list<T>::hit(const ray& r, float t_min, float t_max, hit_record& r
     bool hit_anything = false;
     float closest_so_far = t_max;
 
-    
-    for (uint32_t i = 0; i < objects.size(); ++i)
+    uint32_t s = objects.size();
+    for (uint32_t i = 0; i < s; ++i)
     {
         T *object = &objects[i];
         if (object->hit(r, t_min, closest_so_far, temp_rec))

main.cpp

@@ -2,9 +2,7 @@
 #include <stdint.h>
 #include <stdlib.h>
 
-
-// Disable profiling
-#define RMT_ENABLED 1
+#define RMT_ENABLED 0
 
 // Lib includes
 #pragma GCC diagnostic push
@@ -21,9 +19,15 @@
 #include "sphere.hpp"
 #include "camera.hpp"
 
+#ifdef DEBUG
+#define print_timers() print_timers_()
+#else
+#define print_timers()
+#endif
+
 color ray_color(const ray& r, const hittable& world, int32_t depth);
 float hit_sphere(const point3& center, float radius, const ray& r);
-void print_timers();
+
 hittable_list<sphere> random_scene();
 
 hittable_list<sphere> random_scene() {
@@ -188,10 +192,10 @@ int32_t main()
     rmt_DestroyGlobalInstance(rmt);
 }
 
+#ifdef DEBUG
 debug_record debug_record_array[__COUNTER__];
 
-
-void print_timers()
+void print_timers_()
 {
     for (uint32_t i = 0;
          i < sizeof(debug_record_array) / sizeof(debug_record_array[0]);
@@ -211,3 +215,5 @@ void print_timers()
                 (double)record->cycles / CLOCKS_PER_SEC);
     }
 }
+
+#endif

random.h

@@ -0,0 +1,15 @@
+// *Really* minimal PCG32 code / (c) 2014 M.E. O'Neill / pcg-random.org
+// Licensed under Apache License 2.0 (NO WARRANTY, etc. see website)
+
+typedef struct { uint64_t state;  uint64_t inc; } pcg32_random_t;
+
+uint32_t pcg32_random_r(pcg32_random_t* rng)
+{
+    uint64_t oldstate = rng->state;
+    // Advance internal state
+    rng->state = oldstate * 6364136223846793005ULL + (rng->inc|1);
+    // Calculate output function (XSH RR), uses old state for max ILP
+    uint32_t xorshifted = ((oldstate >> 18u) ^ oldstate) >> 27u;
+    uint32_t rot = oldstate >> 59u;
+    return (xorshifted >> rot) | (xorshifted << ((-rot) & 31));
+}

rtweekend.hpp

@@ -4,9 +4,16 @@
 #include <math.h>
 #include <memory>
 #include "timer.hpp"
+#include "random.h"
+
+#define FAST_RANDOM 1
+#ifdef FAST_RANDOM
+pcg32_random_t pcg = { 0x853c49e6748fea9bULL, 0xda3e39cb94b95bdbULL };
+#define rand() pcg32_random_r(&pcg)
+#endif
 
 /* Utility functions */
-float degrees_to_radians(float d)
+inline float degrees_to_radians(float d)
 {
     return d * M_PI / 180;
 }

timer.hpp

@@ -1,11 +1,15 @@
 #include <x86intrin.h>
 #include "rtweekend.hpp"
 
-
+#ifdef DEBUG
 #define TIMED_BLOCK__(number, ...) timed_block timed_block_##Number(__COUNTER__, __FILE__, __LINE__, __FUNCTION__, ##__VA_ARGS__)
 #define TIMED_BLOCK_(number, ...) TIMED_BLOCK__(number, ##__VA_ARGS__)
 #define TIMED_BLOCK(...) TIMED_BLOCK_(__LINE__, ##__VA_ARGS__)
+#else
+#define TIMED_BLOCK(...)
+#endif
 
+#ifdef DEBUG
 struct debug_record
 {
     uint64_t cycles;
@@ -17,6 +21,7 @@ struct debug_record
     
 };
 
+
 extern debug_record debug_record_array[];
 
 struct timed_block {
@@ -37,3 +42,4 @@ struct timed_block {
         record->cycles += __rdtsc();
     }
 };
+#endif