Test main

.gitignore

@@ -8,4 +8,9 @@
 .nfs*
 *.png
 raytracer
-profile.*
+profile.*
+array_bvh
+shared
+global
+realtime
+cpu

Makefile

@@ -1,7 +1,35 @@
 NVCC = /usr/local/cuda/bin/nvcc
 
-raytracer:
-	$(NVCC) raytracer.cu -o raytracer -O3 -arch=sm_75 -std=c++17 -I/usr/local/cuda/include
+LDLIBS = -lglut -lGL -lGLU -lm -lGLEW
+
+all: global optimized realtime array_bvh cpu optimized_vertices-in-shared optimized_non-coalesced optimized_bvh-tree optimized_bvh-texture
+
+realtime:
+	$(NVCC) realtime_render.cu -o realtime -O3 -arch=sm_75 -std=c++17 -I/usr/local/cuda/include $(LDLIBS)
+
+global:
+	$(NVCC) global_launcher.cu -o global -O3 -arch=sm_75 -std=c++17 -I/usr/local/cuda/include
+
+array_bvh:
+	$(NVCC) array_bvh.cu -o array_bvh -O3 -arch=sm_75 -std=c++17 -I/usr/local/cuda/include
+
+optimized:
+	$(NVCC) optimized.cu -o optimized -O3 -arch=sm_75 -std=c++17 -I/usr/local/cuda/include
+
+optimized_vertices-in-shared:
+	$(NVCC) optimized_vertices-in-shared.cu -o optimized_vertices-in-shared -O3 -arch=sm_75 -std=c++17 -I/usr/local/cuda/include
+
+optimized_non-coalesced:
+	$(NVCC) optimized_non-coalesced.cu -o optimized_non-coalesced -O3 -arch=sm_75 -std=c++17 -I/usr/local/cuda/include
+
+optimized_bvh-tree:
+	$(NVCC) optimized_bvh-tree.cu -o optimized_bvh-tree -O3 -arch=sm_75 -std=c++17 -I/usr/local/cuda/include
+
+optimized_bvh-texture:
+	$(NVCC) optimized_bvh-texture.cu -o optimized_bvh-texture -O3 -arch=sm_75 -std=c++17 -I/usr/local/cuda/include
+
+cpu:
+	g++ cpu_launcher.cpp -o cpu -O3 -fopenmp -std=c++17
 
 clean:
-	rm -f raytracer
+	rm -f *.out realtime global optimized array_bvh optimized_vertices-in-shared optimized_non-coalesced optimized_bvh-tree optimized_bvh-texture

README.md

@@ -3,3 +3,6 @@
 Run `make` in the repo directory which will generate an executable `./raytracer`. Run that executable to generate an image.
 
 The code has only been tested on a Polytechnique machine (Ubuntu) with CUDA pre-installed.
+
+![Spheres with circulating light source](gif/spheres.gif)
+![Playing with cat](gif/cat.gif)