diff --git a/etk/Noise.cpp b/etk/Noise.cpp
index 15750d7..a9e5620 100644
--- a/etk/Noise.cpp
+++ b/etk/Noise.cpp
@@ -13,7 +13,7 @@
 #include <time.h>
 #include <cmath>
 #include <unistd.h>
-#include <stdlib.h>
+#include <cstdlib>
 
 etk::BaseNoise::BaseNoise(const ivec2& _size, float _min, float _max) :
   m_data(_size.x()*_size.y()),
diff --git a/etk/math/Matrix4.hpp b/etk/math/Matrix4.hpp
index 47f3726..c6e2819 100644
--- a/etk/math/Matrix4.hpp
+++ b/etk/math/Matrix4.hpp
@@ -10,7 +10,7 @@
 
 #include <etk/math/Vector3D.hpp>
 
-#include <math.h>
+#include <cmath>
 
 namespace etk {
 	/**
diff --git a/etk/math/Vector4D.hpp b/etk/math/Vector4D.hpp
index a004328..3f07d6e 100644
--- a/etk/math/Vector4D.hpp
+++ b/etk/math/Vector4D.hpp
@@ -7,7 +7,7 @@
 
 #include <etk/types.hpp>
 
-#include <math.h>
+#include <cmath>
 #ifdef ETK_BUILD_LINEARMATH
 	#include <LinearMath/btScalar.h>
 	#include <LinearMath/btMinMax.h>
diff --git a/etk/os/FSNode.cpp b/etk/os/FSNode.cpp
index 1acc4b7..72dcc95 100644
--- a/etk/os/FSNode.cpp
+++ b/etk/os/FSNode.cpp
@@ -8,7 +8,7 @@
 #include <etk/types.hpp>
 #include <etk/os/FSNode.hpp>
 #include <unistd.h>
-#include <stdlib.h>
+#include <cstdlib>
 #include <etk/tool.hpp>
 #include <etk/debug.hpp>
 #include <map>
@@ -22,7 +22,7 @@ extern "C" {
 	// file browsing ...
 	#include <dirent.h>
 	#include <sys/stat.h>
-	#include <errno.h>
+	#include <cerrno>
 }
 
 #ifdef HAVE_ZIP_DATA
diff --git a/etk/tool.cpp b/etk/tool.cpp
index bcd36eb..4120332 100644
--- a/etk/tool.cpp
+++ b/etk/tool.cpp
@@ -10,7 +10,7 @@
 #include <time.h>
 #include <cmath>
 #include <unistd.h>
-#include <stdlib.h>
+#include <cstdlib>
 
 double etk::tool::frand(double _a, double _b) {
 	return (float)(( (double)rand()/(double)RAND_MAX ) * ((double)_b-(double)_a) + (double)_a);
diff --git a/etk/types.hpp b/etk/types.hpp
index 1a15f4f..7822d46 100644
--- a/etk/types.hpp
+++ b/etk/types.hpp
@@ -7,15 +7,15 @@
 
 #include <iostream>
 
-#include <stdlib.h>
-#include <stdio.h>
+#include <cstdlib>
+#include <cstdio>
 #include <stdarg.h>
-#include <string.h>
-#include <assert.h>
+#include <cstring>
+#include <cassert>
 // defien type : uintXX_t and intXX_t
 #define __STDC_LIMIT_MACROS
 // note in android include the macro of min max are overwitten
-#include <stdint.h>
+#include <cstdint>
 // in case of android error ...
 #ifdef __TARGET_OS__Android
 	#if __ANDROID_BOARD_ID__ <= 20
@@ -39,7 +39,7 @@
 	#endif
 #endif
 
-#include <etk/stdTools.h>
+#include <etk/stdTools.hpp>
 
 #ifndef ETK_BUILD_LINEARMATH
 	//! @brief If not using linear math from bullet lib, we need to define the basic element of a btScalar (float)
@@ -47,7 +47,7 @@
 #endif
 
 #ifndef _WIN32
-	#include <math.h>
+	#include <cmath>
 	#ifndef _MATH_H_MATHDEF
 		//! @brief Generate a basic type for floating point unit selection (not finished)
 		using float_t = float;