diff --git a/CMakeLists.txt b/CMakeLists.txt
index 7b8e74b41..f8e450921 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -107,15 +107,26 @@ if((WIN32 AND NOT MINGW) OR APPLE)
     add_subdirectory("${PROJECT_SOURCE_DIR}/lib/libpng")
     include_directories("${PROJECT_SOURCE_DIR}/lib/libpng")
 
-    #build jpeg library
-    add_subdirectory("${PROJECT_SOURCE_DIR}/lib/jpeglib")
-    include_directories("${PROJECT_SOURCE_DIR}/lib/jpeglib")
-
     set(PNG_PNG_INCLUDE_DIR "${PROJECT_SOURCE_DIR}/lib/libpng/")
     set(PNG_LIBRARY png15_static)
+endif()
+
+# Add jpeg-turbo library
+if (APPLE)
+    add_subdirectory("${PROJECT_SOURCE_DIR}/lib/jpeglib")
+    include_directories("${PROJECT_SOURCE_DIR}/lib/jpeglib")
     set(JPEG_INCLUDE_DIR "${PROJECT_SOURCE_DIR}/lib/jpeglib/")
     set(JPEG_LIBRARY jpeglib)
+else()
+    find_package(JPEG REQUIRED)
+    include_directories(${JPEG_INCLUDE_DIR})
 endif()
+
+if(NOT SERVER_ONLY AND NOT USE_GLES2)
+    add_subdirectory("${PROJECT_SOURCE_DIR}/lib/graphics_utils")
+    include_directories("${PROJECT_SOURCE_DIR}/lib/graphics_utils")
+endif()
+
 # Build the irrlicht library
 add_subdirectory("${PROJECT_SOURCE_DIR}/lib/irrlicht")
 include_directories("${PROJECT_SOURCE_DIR}/lib/irrlicht/include")
@@ -368,11 +379,12 @@ target_link_libraries(supertuxkart
     ${OGGVORBIS_LIBRARIES}
     ${OPENAL_LIBRARY}
     ${FREETYPE_LIBRARIES}
+    ${JPEG_LIBRARIES}
     )
 
 if(NOT SERVER_ONLY)
     if(NOT USE_GLES2)
-        target_link_libraries(supertuxkart ${OPENGL_LIBRARIES} glew)
+        target_link_libraries(supertuxkart ${OPENGL_LIBRARIES} glew graphics_utils)
     else()
         target_link_libraries(supertuxkart EGL GLESv2)
     endif()
diff --git a/android/generate_assets.sh b/android/generate_assets.sh
index 2fdec9214..30bf65df2 100755
--- a/android/generate_assets.sh
+++ b/android/generate_assets.sh
@@ -29,13 +29,16 @@ export ASSETS_PATHS="../data                    \
 export ASSETS_DIRS="library models music sfx textures"
 
 export TEXTURE_SIZE=256
-export SOUND_QUALITY=64
+export SOUND_QUALITY=42
+export SOUND_MONO=1
+export SOUND_SAMPLE=32000
 
 export RUN_OPTIMIZE_SCRIPT=0
 export DECREASE_QUALITY=1
 
 ################################################################################
 
+export LANG=C
 
 cd "`dirname "$0"`"
 
@@ -191,7 +194,22 @@ convert_sound()
    oggdec "$FILE" -o tmp.wav
 
    if [ -s tmp.wav ]; then
-      oggenc --downmix -b $SOUND_QUALITY tmp.wav -o tmp.ogg
+      OGGENC_CMD=""
+      
+      if [ "$SOUND_MONO" -gt 0 ]; then
+         OGGENC_CMD="$OGGENC_CMD --downmix"
+      fi
+      
+      OGG_RATE=`ogginfo "$FILE" | grep "Rate: " | cut -f 2 -d " " \
+                                                            | grep -o '[0-9]*'`
+      
+      if [ ! -z "$OGG_RATE" ] && [ "$OGG_RATE" -gt "32000" ]; then
+         OGGENC_CMD="$OGGENC_CMD --resample 32000"
+      fi
+      
+      OGGENC_CMD="$OGGENC_CMD -b $SOUND_QUALITY"
+      
+      oggenc $OGGENC_CMD tmp.wav -o tmp.ogg
    fi
 
    if [ -s tmp.ogg ]; then
diff --git a/data/gui/custom_video_settings.stkgui b/data/gui/custom_video_settings.stkgui
index 3a0a1b9a8..be2b29448 100644
--- a/data/gui/custom_video_settings.stkgui
+++ b/data/gui/custom_video_settings.stkgui
@@ -147,16 +147,8 @@
                 <spacer width="10" height="10"/>
                 <label text="Texture compression" I18N="Video settings"/>
             </div>
-            
-            <spacer height="4" width="10" />
-            
-            <div layout="horizontal-row" proportion="1" height="fit">
-                <checkbox id="hd-textures"/>
-                <spacer width="10" height="10"/>
-                <label text="Use high definition textures" I18N="Video settings"/>
-            </div>
         </div>
-        
+
         <spacer height="20" width="10" />
 
         <div layout="horizontal-row" width="100%" proportion="1">
@@ -168,9 +160,9 @@
         <spacer height="4" width="10" />
 
         <div layout="horizontal-row" width="100%" proportion="1">
-            <label text="Texture filtering" I18N="Video settings" width="40%"/>
+            <label text="Rendered image quality" I18N="Video settings" width="40%"/>
             <spacer width="10" height="10"/>
-            <gauge id="filtering" min_value="0" max_value="5" width="50%" />
+            <gauge id="image_quality" min_value="0" max_value="3" width="50%" />
         </div>
 
         <spacer height="4" width="10" />
diff --git a/data/shaders/object_pass2.frag b/data/shaders/object_pass2.frag
index 838570ead..93f0a0fb0 100644
--- a/data/shaders/object_pass2.frag
+++ b/data/shaders/object_pass2.frag
@@ -40,7 +40,7 @@ void main(void)
         col = vec4(new_color.r, new_color.g, new_color.b, col.a);
     }
 
-#ifdef GL_ES
+#if defined(GL_ES) && !defined(Advanced_Lighting_Enabled)
     col.xyz *= color.xyz;
 #else
     col.xyz *= pow(color.xyz, vec3(2.2));
diff --git a/data/shaders/object_unlit.frag b/data/shaders/object_unlit.frag
index 15c775fbd..d13305612 100644
--- a/data/shaders/object_unlit.frag
+++ b/data/shaders/object_unlit.frag
@@ -16,7 +16,13 @@ void main(void)
     col.xyz = pow(col.xyz, vec3(2.2));
 #endif
 #endif
-    col.xyz *= pow(color.xyz, vec3(2.2));
     if (col.a < 0.5) discard;
+    
+#if defined(GL_ES) && !defined(Advanced_Lighting_Enabled)
+    col.xyz *= color.xyz;
+#else
+    col.xyz *= pow(color.xyz, vec3(2.2));
+#endif
+
     FragColor = vec4(col.xyz, 1.);
 }
diff --git a/data/shaders/objectref_pass2.frag b/data/shaders/objectref_pass2.frag
index a6374bd0d..3dacf6bdb 100644
--- a/data/shaders/objectref_pass2.frag
+++ b/data/shaders/objectref_pass2.frag
@@ -25,8 +25,13 @@ void main(void)
     col.xyz = pow(col.xyz, vec3(2.2));
 #endif
 #endif
-    col.xyz *= pow(color.xyz, vec3(2.2));
     if (col.a * color.a < 0.5) discard;
+    
+#if defined(GL_ES) && !defined(Advanced_Lighting_Enabled)
+    col.xyz *= color.xyz;
+#else
+    col.xyz *= pow(color.xyz, vec3(2.2));
+#endif
 
     float mask = texture(colorization_mask, uv).a;
     if (color_change.x > 0.0)
diff --git a/data/shaders/tonemap.frag b/data/shaders/tonemap.frag
index 13c6b32ce..251f18ac1 100644
--- a/data/shaders/tonemap.frag
+++ b/data/shaders/tonemap.frag
@@ -13,7 +13,9 @@ void main()
 
     // Uncharted2 tonemap with Auria's custom coefficients
     vec4 perChannel = (col * (6.9 * col + .5)) / (col * (5.2 * col + 1.7) + 0.06);
+#if !(defined(GL_ES) && defined(Advanced_Lighting_Enabled))
     perChannel = pow(perChannel, vec4(2.2));
+#endif
 
     vec2 inside = uv - 0.5;
     float vignette = 1. - dot(inside, inside) * vignette_weight;
diff --git a/data/shaders/utils/getLightFactor.frag b/data/shaders/utils/getLightFactor.frag
index 4fb754069..af0feed66 100644
--- a/data/shaders/utils/getLightFactor.frag
+++ b/data/shaders/utils/getLightFactor.frag
@@ -14,10 +14,6 @@ vec3 getLightFactor(vec3 diffuseMatColor, vec3 specularMatColor, float specMapVa
     vec3 DiffuseComponent = texture(DiffuseMap, tc).xyz;
     vec3 SpecularComponent = texture(SpecularMap, tc).xyz;
     float ao = texture(SSAO, tc).x;
-#ifdef GL_ES
-    DiffuseComponent = pow(DiffuseComponent, vec3(1. / 2.2));
-    SpecularComponent = pow(SpecularComponent, vec3(1. / 2.2));
-#endif
     vec3 tmp = diffuseMatColor * DiffuseComponent * (1. - specMapValue) + specularMatColor * SpecularComponent * specMapValue;
     vec3 emitCol = diffuseMatColor.xyz * diffuseMatColor.xyz * diffuseMatColor.xyz * 15.;
     return tmp * ao + (emitMapValue * emitCol);
diff --git a/lib/graphics_utils/CMakeLists.txt b/lib/graphics_utils/CMakeLists.txt
new file mode 100644
index 000000000..0c238f0b6
--- /dev/null
+++ b/lib/graphics_utils/CMakeLists.txt
@@ -0,0 +1,8 @@
+cmake_minimum_required(VERSION 2.6)
+if (UNIX OR MINGW)
+    add_definitions(-O3)
+endif()
+add_library(graphics_utils STATIC
+    mipmap/cpusimd.c
+    mipmap/imgresize.c
+)
diff --git a/lib/graphics_utils/mipmap/cpusimd.c b/lib/graphics_utils/mipmap/cpusimd.c
new file mode 100644
index 000000000..6e29dd3bb
--- /dev/null
+++ b/lib/graphics_utils/mipmap/cpusimd.c
@@ -0,0 +1,571 @@
+/* -----------------------------------------------------------------------------
+ *
+ * Copyright (c) 2008-2016 Alexis Naveros.
+ *
+ *
+ * The SIMD trigonometry functions are Copyright (C) 2007  Julien Pommier
+ * See copyright notice for simd4f_sin_ps(), simd4f_cos_ps(), simd4f_sincos_ps()
+ *
+ *
+ * Some functions are Copyright (C) 2008  José Fonseca
+ * See copyright notice for simd4f_exp2_ps(), simd4f_log2_ps(), simd4f_pow_ps()
+ *
+ *
+ * Portions developed under contract to the SURVICE Engineering Company.
+ *
+ * This software is provided 'as-is', without any express or implied
+ * warranty. In no event will the authors be held liable for any damages
+ * arising from the use of this software.
+ *
+ * Permission is granted to anyone to use this software for any purpose,
+ * including commercial applications, and to alter it and redistribute it
+ * freely, subject to the following restrictions:
+ *
+ * 1. The origin of this software must not be misrepresented; you must not
+ * claim that you wrote the original software. If you use this software
+ * in a product, an acknowledgment in the product documentation would be
+ * appreciated but is not required.
+ * 2. Altered source versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ * 3. This notice may not be removed or altered from any source distribution.
+ *
+ * -----------------------------------------------------------------------------
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <stddef.h>
+#include <stdint.h>
+#include <string.h>
+#include <math.h>
+#include <float.h>
+#include <limits.h>
+
+
+#include "cpusimd.h"
+
+
+#ifndef M_PI
+ #define M_PI 3.14159265358979323846
+#endif
+
+
+////
+
+
+#if CPU_SSE_SUPPORT
+
+const uint32_t CPU_ALIGN16 simd4fSignMask[4] = { 0x80000000, 0x80000000, 0x80000000, 0x80000000 };
+const uint32_t CPU_ALIGN16 simd4fSignMaskInv[4] = { 0x7fffffff, 0x7fffffff, 0x7fffffff, 0x7fffffff };
+const float CPU_ALIGN16 simd4fHalf[4] = { 0.5, 0.5, 0.5, 0.5 };
+const float CPU_ALIGN16 simd4fOne[4] = { 1.0, 1.0, 1.0, 1.0 };
+const float CPU_ALIGN16 simd4fTwo[4] = { 2.0, 2.0, 2.0, 2.0 };
+const float CPU_ALIGN16 simd4fThree[4] = { 3.0, 3.0, 3.0, 3.0 };
+const uint32_t CPU_ALIGN16 simd4uOne[4] = { 1, 1, 1, 1 };
+const uint32_t CPU_ALIGN16 simd4uOneInv[4] = { ~1, ~1, ~1, ~1 };
+const uint32_t CPU_ALIGN16 simd4uTwo[4] = { 2, 2, 2, 2 };
+const uint32_t CPU_ALIGN16 simd4uFour[4] = { 4, 4, 4, 4 };
+const float CPU_ALIGN16 simd4fQuarter[4] = { 0.25, 0.25, 0.25, 0.25 };
+const float CPU_ALIGN16 simd4fPi[4] = { M_PI, M_PI, M_PI, M_PI };
+const float CPU_ALIGN16 simd4fZeroOneTwoThree[4] = { 0.0, 1.0, 2.0, 3.0 };
+const uint32_t CPU_ALIGN16 simd4fAlphaMask[4] = { 0x00000000, 0x00000000, 0x00000000, 0xffffffff };
+const float CPU_ALIGN16 simd4f255[4] = { 255.0f, 255.0f, 255.0f, 255.0f };
+const float CPU_ALIGN16 simd4f255Inv[4] = { 1.0f/255.0f, 1.0f/255.0f, 1.0f/255.0f, 1.0f/255.0f };
+
+#endif
+
+
+////
+
+
+#if CPU_SSE2_SUPPORT
+
+
+/* Copyright (C) 2007  Julien Pommier
+
+  This software is provided 'as-is', without any express or implied
+  warranty.  In no event will the authors be held liable for any damages
+  arising from the use of this software.
+
+  Permission is granted to anyone to use this software for any purpose,
+  including commercial applications, and to alter it and redistribute it
+  freely, subject to the following restrictions:
+
+  1. The origin of this software must not be misrepresented; you must not
+     claim that you wrote the original software. If you use this software
+     in a product, an acknowledgment in the product documentation would be
+     appreciated but is not required.
+  2. Altered source versions must be plainly marked as such, and must not be
+     misrepresented as being the original software.
+  3. This notice may not be removed or altered from any source distribution.
+
+  (this is the zlib license)
+*/
+
+static const float CPU_ALIGN16 simd4f_cephes_FOPI[4] = { 1.27323954473516, 1.27323954473516, 1.27323954473516, 1.27323954473516 };
+static const float CPU_ALIGN16 simd4f_minus_cephes_DP1[4] = { -0.78515625, -0.78515625, -0.78515625, -0.78515625 };
+static const float CPU_ALIGN16 simd4f_minus_cephes_DP2[4] = { -2.4187564849853515625e-4, -2.4187564849853515625e-4, -2.4187564849853515625e-4, -2.4187564849853515625e-4 };
+static const float CPU_ALIGN16 simd4f_minus_cephes_DP3[4] = { -3.77489497744594108e-8, -3.77489497744594108e-8, -3.77489497744594108e-8, -3.77489497744594108e-8 };
+static const float CPU_ALIGN16 simd4f_sincof_p0[4] = { -1.9515295891E-4, -1.9515295891E-4, -1.9515295891E-4, -1.9515295891E-4 };
+static const float CPU_ALIGN16 simd4f_sincof_p1[4] = { 8.3321608736E-3, 8.3321608736E-3, 8.3321608736E-3, 8.3321608736E-3 };
+static const float CPU_ALIGN16 simd4f_sincof_p2[4] = { -1.6666654611E-1, -1.6666654611E-1, -1.6666654611E-1, -1.6666654611E-1 };
+static const float CPU_ALIGN16 simd4f_coscof_p0[4] = { 2.443315711809948E-005, 2.443315711809948E-005, 2.443315711809948E-005, 2.443315711809948E-005 };
+static const float CPU_ALIGN16 simd4f_coscof_p1[4] = { -1.388731625493765E-003, -1.388731625493765E-003, -1.388731625493765E-003, -1.388731625493765E-003 };
+static const float CPU_ALIGN16 simd4f_coscof_p2[4] = { 4.166664568298827E-002, 4.166664568298827E-002, 4.166664568298827E-002, 4.166664568298827E-002 };
+
+__m128 simd4f_sin_ps( __m128 x )
+{
+  __m128 xmm1, xmm2, xmm3, sign_bit, y;
+  __m128i emm0, emm2;
+
+  xmm2 = _mm_setzero_ps();
+
+  sign_bit = x;
+  /* take the absolute value */
+  x = _mm_and_ps( x, *(__m128 *)simd4fSignMaskInv );
+  /* extract the sign bit (upper one) */
+  sign_bit = _mm_and_ps(sign_bit, *(__m128 *)simd4fSignMask);
+
+  /* scale by 4/Pi */
+  y = _mm_mul_ps(x, *(__m128 *)simd4f_cephes_FOPI);
+
+  /* store the integer part of y in mm0 */
+  emm2 = _mm_cvttps_epi32(y);
+  /* j=(j+1) & (~1) (see the cephes sources) */
+  emm2 = _mm_add_epi32(emm2, *(__m128i*)simd4uOne);
+  emm2 = _mm_and_si128(emm2, *(__m128i*)simd4uOneInv);
+  y = _mm_cvtepi32_ps(emm2);
+
+  /* get the swap sign flag */
+  emm0 = _mm_and_si128(emm2, *(__m128i*)simd4uFour);
+  emm0 = _mm_slli_epi32(emm0, 29);
+  /* get the polynom selection mask
+     there is one polynom for 0 <= x <= Pi/4
+     and another one for Pi/4<x<=Pi/2
+     Both branches will be computed.
+  */
+  emm2 = _mm_and_si128(emm2, *(__m128i*)simd4uTwo);
+  emm2 = _mm_cmpeq_epi32(emm2, _mm_setzero_si128());
+
+  __m128 swap_sign_bit = _mm_castsi128_ps(emm0);
+  __m128 poly_mask = _mm_castsi128_ps(emm2);
+  sign_bit = _mm_xor_ps(sign_bit, swap_sign_bit);
+
+  /* The magic pass: "Extended precision modular arithmetic"
+     x = ((x - y * DP1) - y * DP2) - y * DP3; */
+  xmm1 = *(__m128 *)simd4f_minus_cephes_DP1;
+  xmm2 = *(__m128 *)simd4f_minus_cephes_DP2;
+  xmm3 = *(__m128 *)simd4f_minus_cephes_DP3;
+  xmm1 = _mm_mul_ps(y, xmm1);
+  xmm2 = _mm_mul_ps(y, xmm2);
+  xmm3 = _mm_mul_ps(y, xmm3);
+  x = _mm_add_ps(x, xmm1);
+  x = _mm_add_ps(x, xmm2);
+  x = _mm_add_ps(x, xmm3);
+
+  /* Evaluate the first polynom  (0 <= x <= Pi/4) */
+  y = *(__m128 *)simd4f_coscof_p0;
+  __m128 z = _mm_mul_ps(x,x);
+
+  y = _mm_mul_ps(y, z);
+  y = _mm_add_ps(y, *(__m128 *)simd4f_coscof_p1);
+  y = _mm_mul_ps(y, z);
+  y = _mm_add_ps(y, *(__m128 *)simd4f_coscof_p2);
+  y = _mm_mul_ps(y, z);
+  y = _mm_mul_ps(y, z);
+  __m128 tmp = _mm_mul_ps(z, *(__m128 *)simd4fHalf);
+  y = _mm_sub_ps(y, tmp);
+  y = _mm_add_ps(y, *(__m128 *)simd4fOne);
+
+  /* Evaluate the second polynom  (Pi/4 <= x <= 0) */
+
+  __m128 y2 = *(__m128 *)simd4f_sincof_p0;
+  y2 = _mm_mul_ps(y2, z);
+  y2 = _mm_add_ps(y2, *(__m128 *)simd4f_sincof_p1);
+  y2 = _mm_mul_ps(y2, z);
+  y2 = _mm_add_ps(y2, *(__m128 *)simd4f_sincof_p2);
+  y2 = _mm_mul_ps(y2, z);
+  y2 = _mm_mul_ps(y2, x);
+  y2 = _mm_add_ps(y2, x);
+
+  /* select the correct result from the two polynoms */
+  xmm3 = poly_mask;
+  y2 = _mm_and_ps(xmm3, y2);
+  y = _mm_andnot_ps(xmm3, y);
+  y = _mm_add_ps(y,y2);
+  /* update the sign */
+  y = _mm_xor_ps(y, sign_bit);
+
+  return y;
+}
+
+
+
+/* almost the same as sin_ps */
+__m128 simd4f_cos_ps( __m128 x )
+{
+  __m128 xmm1, xmm2, xmm3, y;
+  __m128i emm0, emm2;
+
+  xmm2 = _mm_setzero_ps();
+
+  /* take the absolute value */
+  x = _mm_and_ps(x, *(__m128*)simd4fSignMaskInv);
+  
+  /* scale by 4/Pi */
+  y = _mm_mul_ps(x, *(__m128*)simd4f_cephes_FOPI);
+  
+  /* store the integer part of y in mm0 */
+  emm2 = _mm_cvttps_epi32(y);
+  /* j=(j+1) & (~1) (see the cephes sources) */
+  emm2 = _mm_add_epi32(emm2, *(__m128i*)simd4uOne);
+  emm2 = _mm_and_si128(emm2, *(__m128i*)simd4uOneInv);
+  y = _mm_cvtepi32_ps(emm2);
+
+  emm2 = _mm_sub_epi32(emm2, *(__m128i*)simd4uTwo);
+  
+  /* get the swap sign flag */
+  emm0 = _mm_andnot_si128(emm2, *(__m128i*)simd4uFour);
+  emm0 = _mm_slli_epi32(emm0, 29);
+  /* get the polynom selection mask */
+  emm2 = _mm_and_si128(emm2, *(__m128i*)simd4uTwo);
+  emm2 = _mm_cmpeq_epi32(emm2, _mm_setzero_si128());
+  
+  __m128 sign_bit = _mm_castsi128_ps(emm0);
+  __m128 poly_mask = _mm_castsi128_ps(emm2);
+  /* The magic pass: "Extended precision modular arithmetic" 
+     x = ((x - y * DP1) - y * DP2) - y * DP3; */
+  xmm1 = *(__m128*)simd4f_minus_cephes_DP1;
+  xmm2 = *(__m128*)simd4f_minus_cephes_DP2;
+  xmm3 = *(__m128*)simd4f_minus_cephes_DP3;
+  xmm1 = _mm_mul_ps(y, xmm1);
+  xmm2 = _mm_mul_ps(y, xmm2);
+  xmm3 = _mm_mul_ps(y, xmm3);
+  x = _mm_add_ps(x, xmm1);
+  x = _mm_add_ps(x, xmm2);
+  x = _mm_add_ps(x, xmm3);
+  
+  /* Evaluate the first polynom  (0 <= x <= Pi/4) */
+  y = *(__m128*)simd4f_coscof_p0;
+  __m128 z = _mm_mul_ps(x,x);
+
+  y = _mm_mul_ps(y, z);
+  y = _mm_add_ps(y, *(__m128*)simd4f_coscof_p1);
+  y = _mm_mul_ps(y, z);
+  y = _mm_add_ps(y, *(__m128*)simd4f_coscof_p2);
+  y = _mm_mul_ps(y, z);
+  y = _mm_mul_ps(y, z);
+  __m128 tmp = _mm_mul_ps(z, *(__m128*)simd4fHalf);
+  y = _mm_sub_ps(y, tmp);
+  y = _mm_add_ps(y, *(__m128*)simd4fOne);
+  
+  /* Evaluate the second polynom  (Pi/4 <= x <= 0) */
+
+  __m128 y2 = *(__m128*)simd4f_sincof_p0;
+  y2 = _mm_mul_ps(y2, z);
+  y2 = _mm_add_ps(y2, *(__m128*)simd4f_sincof_p1);
+  y2 = _mm_mul_ps(y2, z);
+  y2 = _mm_add_ps(y2, *(__m128*)simd4f_sincof_p2);
+  y2 = _mm_mul_ps(y2, z);
+  y2 = _mm_mul_ps(y2, x);
+  y2 = _mm_add_ps(y2, x);
+
+  /* select the correct result from the two polynoms */  
+  xmm3 = poly_mask;
+  y2 = _mm_and_ps(xmm3, y2); //, xmm3);
+  y = _mm_andnot_ps(xmm3, y);
+  y = _mm_add_ps(y,y2);
+  /* update the sign */
+  y = _mm_xor_ps(y, sign_bit);
+
+  return y;
+}
+
+/* since sin_ps and cos_ps are almost identical, sincos_ps could replace both of them..
+   it is almost as fast, and gives you a free cosine with your sine */
+void simd4f_sincos_ps( __m128 x, __m128 *s, __m128 *c )
+{
+  __m128 xmm1, xmm2, xmm3, sign_bit_sin, y;
+  __m128i emm0, emm2, emm4;
+
+  xmm3 = _mm_setzero_ps();
+
+  sign_bit_sin = x;
+  /* take the absolute value */
+  x = _mm_and_ps(x, *(__m128*)simd4fSignMaskInv);
+  /* extract the sign bit (upper one) */
+  sign_bit_sin = _mm_and_ps(sign_bit_sin, *(__m128*)simd4fSignMask);
+  
+  /* scale by 4/Pi */
+  y = _mm_mul_ps(x, *(__m128*)simd4f_cephes_FOPI);
+    
+  /* store the integer part of y in emm2 */
+  emm2 = _mm_cvttps_epi32(y);
+
+  /* j=(j+1) & (~1) (see the cephes sources) */
+  emm2 = _mm_add_epi32(emm2, *(__m128i*)simd4uOne);
+  emm2 = _mm_and_si128(emm2, *(__m128i*)simd4uOneInv);
+  y = _mm_cvtepi32_ps(emm2);
+
+  emm4 = emm2;
+
+  /* get the swap sign flag for the sine */
+  emm0 = _mm_and_si128(emm2, *(__m128i*)simd4uFour);
+  emm0 = _mm_slli_epi32(emm0, 29);
+  __m128 swap_sign_bit_sin = _mm_castsi128_ps(emm0);
+
+  /* get the polynom selection mask for the sine*/
+  emm2 = _mm_and_si128(emm2, *(__m128i*)simd4uTwo);
+  emm2 = _mm_cmpeq_epi32(emm2, _mm_setzero_si128());
+  __m128 poly_mask = _mm_castsi128_ps(emm2);
+
+  /* The magic pass: "Extended precision modular arithmetic" 
+     x = ((x - y * DP1) - y * DP2) - y * DP3; */
+  xmm1 = *(__m128*)simd4f_minus_cephes_DP1;
+  xmm2 = *(__m128*)simd4f_minus_cephes_DP2;
+  xmm3 = *(__m128*)simd4f_minus_cephes_DP3;
+  xmm1 = _mm_mul_ps(y, xmm1);
+  xmm2 = _mm_mul_ps(y, xmm2);
+  xmm3 = _mm_mul_ps(y, xmm3);
+  x = _mm_add_ps(x, xmm1);
+  x = _mm_add_ps(x, xmm2);
+  x = _mm_add_ps(x, xmm3);
+
+  emm4 = _mm_sub_epi32(emm4, *(__m128i*)simd4uTwo);
+  emm4 = _mm_andnot_si128(emm4, *(__m128i*)simd4uFour);
+  emm4 = _mm_slli_epi32(emm4, 29);
+  __m128 sign_bit_cos = _mm_castsi128_ps(emm4);
+
+  sign_bit_sin = _mm_xor_ps(sign_bit_sin, swap_sign_bit_sin);
+  
+  /* Evaluate the first polynom  (0 <= x <= Pi/4) */
+  __m128 z = _mm_mul_ps(x,x);
+  y = *(__m128*)simd4f_coscof_p0;
+
+  y = _mm_mul_ps(y, z);
+  y = _mm_add_ps(y, *(__m128*)simd4f_coscof_p1);
+  y = _mm_mul_ps(y, z);
+  y = _mm_add_ps(y, *(__m128*)simd4f_coscof_p2);
+  y = _mm_mul_ps(y, z);
+  y = _mm_mul_ps(y, z);
+  __m128 tmp = _mm_mul_ps(z, *(__m128*)simd4fHalf);
+  y = _mm_sub_ps(y, tmp);
+  y = _mm_add_ps(y, *(__m128*)simd4fOne);
+  
+  /* Evaluate the second polynom  (Pi/4 <= x <= 0) */
+
+  __m128 y2 = *(__m128*)simd4f_sincof_p0;
+  y2 = _mm_mul_ps(y2, z);
+  y2 = _mm_add_ps(y2, *(__m128*)simd4f_sincof_p1);
+  y2 = _mm_mul_ps(y2, z);
+  y2 = _mm_add_ps(y2, *(__m128*)simd4f_sincof_p2);
+  y2 = _mm_mul_ps(y2, z);
+  y2 = _mm_mul_ps(y2, x);
+  y2 = _mm_add_ps(y2, x);
+
+  /* select the correct result from the two polynoms */  
+  xmm3 = poly_mask;
+  __m128 ysin2 = _mm_and_ps(xmm3, y2);
+  __m128 ysin1 = _mm_andnot_ps(xmm3, y);
+  y2 = _mm_sub_ps(y2,ysin2);
+  y = _mm_sub_ps(y, ysin1);
+
+  xmm1 = _mm_add_ps(ysin1,ysin2);
+  xmm2 = _mm_add_ps(y,y2);
+ 
+  /* update the sign */
+  *s = _mm_xor_ps(xmm1, sign_bit_sin);
+  *c = _mm_xor_ps(xmm2, sign_bit_cos);
+}
+
+
+#endif
+
+
+////
+
+
+#if CPU_SSE2_SUPPORT
+
+
+/* Copyright (C) 2008  José Fonseca
+   http://jrfonseca.blogspot.ca/2008/09/fast-sse2-pow-tables-or-polynomials.html
+   MIT license
+
+   Permission is hereby granted, free of charge, to any person obtaining a copy
+   of this software and associated documentation files (the "Software"), to deal
+   in the Software without restriction, including without limitation the rights
+   to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+   copies of the Software, and to permit persons to whom the Software is
+   furnished to do so, subject to the following conditions:
+
+   The above copyright notice and this permission notice shall be included in all
+   copies or substantial portions of the Software.
+
+   THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+   IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+   FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+   AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+   LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+   OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+   SOFTWARE.
+*/
+
+#define POLY0(x,c0) _mm_set1_ps(c0)
+#define POLY1(x,c0,c1) _mm_add_ps(_mm_mul_ps(POLY0(x, c1), x), _mm_set1_ps(c0))
+#define POLY2(x,c0,c1,c2) _mm_add_ps(_mm_mul_ps(POLY1(x, c1, c2), x), _mm_set1_ps(c0))
+#define POLY3(x,c0,c1,c2,c3) _mm_add_ps(_mm_mul_ps(POLY2(x, c1, c2, c3), x), _mm_set1_ps(c0))
+#define POLY4(x,c0,c1,c2,c3,c4) _mm_add_ps(_mm_mul_ps(POLY3(x, c1, c2, c3, c4), x), _mm_set1_ps(c0))
+#define POLY5(x,c0,c1,c2,c3,c4,c5) _mm_add_ps(_mm_mul_ps(POLY4(x, c1, c2, c3, c4, c5), x), _mm_set1_ps(c0))
+
+#define EXP_POLY_DEGREE 3
+#define LOG_POLY_DEGREE 5
+
+__m128 simd4f_exp2_ps( __m128 x )
+{
+  __m128i ipart;
+  __m128 fpart, expipart, expfpart;
+
+  x = _mm_min_ps( x, _mm_set1_ps(  129.00000f ) );
+  x = _mm_max_ps( x, _mm_set1_ps( -126.99999f ) );
+  /* ipart = int(x - 0.5) */
+  ipart = _mm_cvtps_epi32( _mm_sub_ps( x, _mm_set1_ps( 0.5f ) ) );
+  /* fpart = x - ipart */
+  fpart = _mm_sub_ps( x, _mm_cvtepi32_ps( ipart ) );
+  /* expipart = (float) (1 << ipart) */
+  expipart = _mm_castsi128_ps( _mm_slli_epi32( _mm_add_epi32( ipart, _mm_set1_epi32( 127 ) ), 23 ) );
+  /* minimax polynomial fit of 2**x, in range [-0.5, 0.5[ */
+#if EXP_POLY_DEGREE == 5
+  expfpart = POLY5( fpart, 9.9999994e-1f, 6.9315308e-1f, 2.4015361e-1f, 5.5826318e-2f, 8.9893397e-3f, 1.8775767e-3f );
+#elif EXP_POLY_DEGREE == 4
+  expfpart = POLY4( fpart, 1.0000026f, 6.9300383e-1f, 2.4144275e-1f, 5.2011464e-2f, 1.3534167e-2f );
+#elif EXP_POLY_DEGREE == 3
+  expfpart = POLY3( fpart, 9.9992520e-1f, 6.9583356e-1f, 2.2606716e-1f, 7.8024521e-2f );
+#elif EXP_POLY_DEGREE == 2
+  expfpart = POLY2( fpart, 1.0017247f, 6.5763628e-1f, 3.3718944e-1f );
+#else
+#error
+#endif
+  return _mm_mul_ps(expipart, expfpart);
+}
+
+__m128 simd4f_log2_ps( __m128 x )
+{
+  __m128i expmask, mantmask, i;
+  __m128 one, vexp, mant, logmant;
+
+  expmask = _mm_set1_epi32( 0x7f800000 );
+  mantmask = _mm_set1_epi32( 0x007fffff );
+  one = _mm_set1_ps( 1.0f );
+  i = _mm_castps_si128( x );
+  /* exp = (float) exponent(x) */
+  vexp = _mm_cvtepi32_ps( _mm_sub_epi32( _mm_srli_epi32( _mm_and_si128( i, expmask ), 23 ), _mm_set1_epi32( 127 ) ) );
+  /* mant = (float) mantissa(x) */
+  mant = _mm_or_ps( _mm_castsi128_ps( _mm_and_si128( i, mantmask ) ), one );
+  /* Minimax polynomial fit of log2(x)/(x - 1), for x in range [1, 2[ 
+   * These coefficients can be generate with 
+   * http://www.boost.org/doc/libs/1_36_0/libs/math/doc/sf_and_dist/html/math_toolkit/toolkit/internals2/minimax.html
+   */
+#if LOG_POLY_DEGREE == 6
+  logmant = POLY5( mant, 3.11578814719469302614f, -3.32419399085241980044f, 2.59883907202499966007f, -1.23152682416275988241f, 0.318212422185251071475f, -0.0344359067839062357313f );
+#elif LOG_POLY_DEGREE == 5
+  logmant = POLY4( mant, 2.8882704548164776201f, -2.52074962577807006663f, 1.48116647521213171641f, -0.465725644288844778798f, 0.0596515482674574969533f );
+#elif LOG_POLY_DEGREE == 4
+  logmant = POLY3( mant, 2.61761038894603480148f, -1.75647175389045657003f, 0.688243882994381274313f, -0.107254423828329604454f );
+#elif LOG_POLY_DEGREE == 3
+  logmant = POLY2( mant, 2.28330284476918490682f, -1.04913055217340124191f, 0.204446009836232697516f );
+#else
+#error
+#endif
+  /* This effectively increases the polynomial degree by one, but ensures that log2(1) == 0*/
+  logmant = _mm_mul_ps( logmant, _mm_sub_ps(mant, one ) );
+  return _mm_add_ps( logmant, vexp );
+}
+
+
+__m128 simd4f_pow_ps( __m128 x, __m128 y )
+{
+  return simd4f_exp2_ps( _mm_mul_ps( simd4f_log2_ps( x ), y ) );
+}
+
+
+#endif
+
+
+////
+
+
+#if CPU_SSE2_SUPPORT
+
+
+/*
+By Potatoswatter
+http://stackoverflow.com/questions/6475373/optimizations-for-pow-with-const-non-integer-exponent
+*/
+
+#ifndef CC_ALWAYSINLINE
+ #if defined(__GNUC__) || defined(__INTEL_COMPILER)
+  #define CC_ALWAYSINLINE __attribute__((always_inline))
+ #else
+  #define CC_ALWAYSINLINE
+ #endif
+#endif
+
+static inline CC_ALWAYSINLINE __m128 simd4f_fastpow_ps( __m128 arg, uint32_t expnum, uint32_t expden, uint32_t coeffnum, uint32_t coeffden )
+{
+  __m128 ret = arg;
+  float corrfactor, powfactor;
+  /* Apply a constant pre-correction factor. */
+  corrfactor = exp2( 127.0 * expden / expnum - 127.0 ) * pow( 1.0 * coeffnum / coeffden, 1.0 * expden / expnum );
+  powfactor = 1.0 * expnum / expden;
+  ret = _mm_mul_ps( ret, _mm_set1_ps( corrfactor ) );
+  /* Reinterpret arg as integer to obtain logarithm. */
+  ret = _mm_cvtepi32_ps( _mm_castps_si128( ret ) );
+  /* Multiply logarithm by power. */
+  ret = _mm_mul_ps( ret, _mm_set1_ps( powfactor ) );
+  /* Convert back to "integer" to exponentiate. */
+  ret = _mm_castsi128_ps( _mm_cvtps_epi32( ret ) );
+  return ret;
+}
+
+__m128 simd4f_pow12d5_ps( __m128 arg )
+{
+  /* Lower exponents provide lower initial error, but too low causes overflow. */
+  __m128 xf = simd4f_fastpow_ps( arg, 4, 5, (int)( 1.38316186f * (float)1e9 ), (int)1e9 );
+  /* Imprecise 4-cycle sqrt is still far better than fastpow, good enough. */
+  __m128 xfm4 = _mm_rsqrt_ps( xf );
+  __m128 xf4 = _mm_mul_ps( xf, xfm4 );
+  /* Precisely calculate x^2 and x^3 */
+  __m128 x2 = _mm_mul_ps( arg, arg );
+  __m128 x3 = _mm_mul_ps( x2, arg );
+  /* Overestimate of x^2 * x^0.4 */
+  x2 = _mm_mul_ps( x2, xf4 );
+  /* Get x^-0.2 from x^0.4, and square it for x^-0.4. Combine into x^-0.6. */
+  __m128 xfm2 = _mm_rsqrt_ps( xf4 );
+  x3 = _mm_mul_ps( x3, xfm4 );
+  x3 = _mm_mul_ps( x3, xfm2 );
+  return _mm_mul_ps( _mm_add_ps( x2, x3 ), _mm_set1_ps( 1.0f/1.960131704207789f * 0.9999f ) );
+}
+
+__m128 simd4f_pow5d12_ps( __m128 arg )
+{
+  /* 5/12 is too small, so compute the 4th root of 20/12 instead. */
+  /* 20/12 = 5/3 = 1 + 2/3 = 2 - 1/3. 2/3 is a suitable argument for fastpow. */
+  /* weighting coefficient: a^-1/2 = 2 a; a = 2^-2/3 */
+  __m128 xf = simd4f_fastpow_ps( arg, 2, 3, (int)( 0.629960524947437f * (float)1e9 ), (int)1e9 );
+  __m128 xover = _mm_mul_ps( arg, xf );
+  __m128 xfm1 = _mm_rsqrt_ps( xf );
+  __m128 x2 = _mm_mul_ps( arg, arg );
+  __m128 xunder = _mm_mul_ps( x2, xfm1 );
+  /* sqrt2 * over + 2 * sqrt2 * under */
+  __m128 xavg = _mm_mul_ps( _mm_set1_ps( 1.0f/( 3.0f * 0.629960524947437f ) * 0.999852f ), _mm_add_ps( xover, xunder ) );
+  xavg = _mm_mul_ps( xavg, _mm_rsqrt_ps( xavg ) );
+  xavg = _mm_mul_ps( xavg, _mm_rsqrt_ps( xavg ) );
+  return xavg;
+}
+
+#endif
+
+
+////
+
diff --git a/lib/graphics_utils/mipmap/cpusimd.h b/lib/graphics_utils/mipmap/cpusimd.h
new file mode 100644
index 000000000..1dc2c730b
--- /dev/null
+++ b/lib/graphics_utils/mipmap/cpusimd.h
@@ -0,0 +1,410 @@
+/* -----------------------------------------------------------------------------
+ *
+ * Copyright (c) 2008-2016 Alexis Naveros.
+ *
+ * The SIMD trigonometry functions are Copyright (C) 2007  Julien Pommier
+ * See copyright notice for simd4f_sin_ps(), simd4f_cos_ps(), simd4f_sincos_ps()
+ *
+ * Portions developed under contract to the SURVICE Engineering Company.
+ *
+ * This software is provided 'as-is', without any express or implied
+ * warranty. In no event will the authors be held liable for any damages
+ * arising from the use of this software.
+ *
+ * Permission is granted to anyone to use this software for any purpose,
+ * including commercial applications, and to alter it and redistribute it
+ * freely, subject to the following restrictions:
+ *
+ * 1. The origin of this software must not be misrepresented; you must not
+ * claim that you wrote the original software. If you use this software
+ * in a product, an acknowledgment in the product documentation would be
+ * appreciated but is not required.
+ * 2. Altered source versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ * 3. This notice may not be removed or altered from any source distribution.
+ *
+ * -----------------------------------------------------------------------------
+ */
+
+
+#ifndef CPUSIMD_H
+#define CPUSIMD_H
+
+
+////
+
+
+#if __MMX__ || CPU_ENABLE_MMX
+ #include <mmintrin.h>
+ #define CPU_MMX_SUPPORT (1)
+#endif
+#if __SSE__ || _M_X64 || _M_IX86_FP >= 1  || CPU_ENABLE_SSE
+ #include <xmmintrin.h>
+ #define CPU_SSE_SUPPORT (1)
+#endif
+#if __SSE2__ || _M_X64 || _M_IX86_FP >= 2  || CPU_ENABLE_SSE2
+ #include <emmintrin.h>
+ #define CPU_SSE2_SUPPORT (1)
+#endif
+#if __SSE3__ || __AVX__ || CPU_ENABLE_SSE3
+ #include <pmmintrin.h>
+ #define CPU_SSE3_SUPPORT (1)
+#endif
+#if __SSSE3__ || __AVX__  || CPU_ENABLE_SSSE3
+ #include <tmmintrin.h>
+ #define CPU_SSSE3_SUPPORT (1)
+#endif
+#if __SSE4_1__ || __AVX__  || CPU_ENABLE_SSE4_1
+ #include <smmintrin.h>
+ #define CPU_SSE4_1_SUPPORT (1)
+#endif
+#if __SSE4_2__ || CPU_ENABLE_SSE4_2
+ #include <nmmintrin.h>
+ #define CPU_SSE4_2_SUPPORT (1)
+#endif
+#if __SSE4A__ || CPU_ENABLE_SSE4A
+ #include <ammintrin.h>
+ #define CPU_SSE4A_SUPPORT (1)
+#endif
+#if __AVX__ || CPU_ENABLE_AVX
+ #include <immintrin.h>
+ #define CPU_AVX_SUPPORT (1)
+#endif
+#if __AVX2__ || CPU_ENABLE_AVX2
+ #include <immintrin.h>
+ #define CPU_AVX2_SUPPORT (1)
+#endif
+#if __XOP__ || CPU_ENABLE_XOP
+ #include <immintrin.h>
+ #define CPU_XOP_SUPPORT (1)
+#endif
+#if __FMA3__ || CPU_ENABLE_FMA3
+ #include <immintrin.h>
+ #define CPU_FMA3_SUPPORT (1)
+#endif
+#if __FMA4__ || CPU_ENABLE_FMA4
+ #include <immintrin.h>
+ #define CPU_FMA4_SUPPORT (1)
+#endif
+#if __RDRND__ || CPU_ENABLE_RDRND
+ #include <immintrin.h>
+ #define CPU_RDRND_SUPPORT (1)
+#endif
+#if __POPCNT__ || CPU_ENABLE_POPCNT
+ #include <popcntintrin.h>
+ #define CPU_POPCNT_SUPPORT (1)
+#endif
+#if __LZCNT__ || CPU_ENABLE_LZCNT
+ #include <lzcntintrin.h>
+ #define CPU_LZCNT_SUPPORT (1)
+#endif
+#if __F16C__ || CPU_ENABLE_F16C
+ #include <f16cintrin.h>
+ #define CPU_F16C_SUPPORT (1)
+#endif
+#if __BMI__ || CPU_ENABLE_BMI
+ #include <bmiintrin.h>
+ #define CPU_BMI_SUPPORT (1)
+#endif
+#if __BMI2__ || CPU_ENABLE_BMI2
+ #include <bmi2intrin.h>
+ #define CPU_BMI2_SUPPORT (1)
+#endif
+#if __TBM__ || CPU_ENABLE_TBM
+ #include <tbmintrin.h>
+ #define CPU_TBM_SUPPORT (1)
+#endif
+
+
+
+#if defined(__GNUC__) || defined(__INTEL_COMPILER)
+ #define CPU_ALIGN16 __attribute__((aligned(16)))
+ #define CPU_ALIGN32 __attribute__((aligned(32)))
+ #define CPU_ALIGN64 __attribute__((aligned(64)))
+#elif defined(_MSC_VER)
+ #define CPU_ALIGN16 __declspec(align(16))
+ #define CPU_ALIGN64 __declspec(align(64))
+#else
+ #define CPU_ALIGN16
+ #define CPU_ALIGN32
+ #define CPU_ALIGN64
+ #warning "SSE/AVX Disabled: Unsupported Compiler."
+ #undef CPU_SSE_SUPPORT
+ #undef CPU_SSE2_SUPPORT
+ #undef CPU_SSE3_SUPPORT
+ #undef CPU_SSSE3_SUPPORT
+ #undef CPU_SSE4_1_SUPPORT
+ #undef CPU_SSE4_2_SUPPORT
+ #undef CPU_AVX_SUPPORT
+ #undef CPU_AVX2_SUPPORT
+ #undef CPU_XOP_SUPPORT
+ #undef CPU_FMA3_SUPPORT
+ #undef CPU_FMA4_SUPPORT
+#endif
+
+
+////
+
+
+#if CPU_SSE_SUPPORT
+ #define CPU_APPROX_DIV_FLOAT(z,w) _mm_cvtss_f32(_mm_mul_ss(_mm_set_ss(z),_mm_rcp_ss(_mm_set_ss(w))))
+ #define CPU_APPROX_SQRT_FLOAT(z) _mm_cvtss_f32(_mm_mul_ss(_mm_set_ss(z),_mm_rsqrt_ss(_mm_set_ss(z))))
+ #define CPU_APPROX_RSQRT_FLOAT(z) _mm_cvtss_f32(_mm_rsqrt_ss(_mm_set_ss(z)))
+ #define CPU_APPROX_DIVSQRT_FLOAT(z,w) _mm_cvtss_f32(_mm_mul_ss(_mm_set_ss(z),_mm_rsqrt_ss(_mm_set_ss(w))))
+#else
+ #define CPU_APPROX_DIV_FLOAT(z,w) ((z)/(w))
+ #define CPU_APPROX_SQRT_FLOAT(z) (sqrtf(z))
+ #define CPU_APPROX_RSQRT_FLOAT(z) (1.0/sqrtf(z))
+ #define CPU_APPROX_DIVSQRT_FLOAT(z,w) ((z)/sqrtf(w))
+#endif
+
+
+#if CPU_SSE3_SUPPORT
+ #define CPU_HADD_PS(vx,vy) _mm_hadd_ps(vx,vy)
+ #define CPU_HADD_PD(vx,vy) _mm_hadd_pd(vx,vy)
+#elif CPU_SSE_SUPPORT
+ static inline __m128 CPU_HADD_PS( __m128 vx, __m128 vy )
+ {
+   __m128 vh, vl;
+   vh = _mm_shuffle_ps( vx, vy, _MM_SHUFFLE(3,1,3,1) );
+   vl = _mm_shuffle_ps( vx, vy, _MM_SHUFFLE(2,0,2,0) );
+   return _mm_add_ps( vh, vl );
+ }
+ #define CPU_HADD_PD(vx,vy) _mm_add_sd(vx,_mm_unpackhi_pd(vy,vy))
+#endif
+
+
+#if CPU_SSE4_1_SUPPORT
+ #define CPU_CVT_U8_TO_I32(x,vzero) _mm_cvtepu8_epi32(x)
+ #define CPU_CVT_S8_TO_I32(x,vzero) _mm_cvtepi8_epi32(x)
+#elif CPU_SSE2_SUPPORT
+ #define CPU_CVT_U8_TO_I32(x,vzero) _mm_unpacklo_epi16(_mm_unpacklo_epi8((x),(vzero)),(vzero))
+static inline __m128i CPU_CVT_S8_TO_I32( __m128i vx, __m128i vzero )
+{
+  __m128i vsign;
+  vsign = _mm_cmpgt_epi8( vzero, vx );
+  return _mm_unpacklo_epi16( _mm_unpacklo_epi8( vx, vsign ), _mm_unpacklo_epi8( vsign, vsign ) );
+}
+#endif
+
+
+#if CPU_SSE4_1_SUPPORT
+ #define CPU_BLENDV_PS(x,y,mask) _mm_blendv_ps(x,y,mask)
+ #define CPU_BLENDV_PD(x,y,mask) _mm_blendv_pd(x,y,mask)
+#elif CPU_SSE2_SUPPORT
+ #define CPU_BLENDV_PS(x,y,mask) _mm_or_ps(_mm_andnot_ps(mask,x),_mm_and_ps(y,mask))
+ #define CPU_BLENDV_PD(x,y,mask) _mm_or_pd(_mm_andnot_pd(mask,x),_mm_and_pd(y,mask))
+#endif
+
+
+
+/*
+CPU_FMADD = ((f0*f1)+t0)
+CPU_FMSUB = ((f0*f1)-t0)
+*/
+#if CPU_FMA3_SUPPORT
+ #define CPU_FMADD_SS(f0,f1,t0) _mm_fmadd_ss(f0,f1,t0)
+ #define CPU_FMADD_PS(f0,f1,t0) _mm_fmadd_ps(f0,f1,t0)
+ #define CPU_FMADD_SD(f0,f1,t0) _mm_fmadd_sd(f0,f1,t0)
+ #define CPU_FMADD_PD(f0,f1,t0) _mm_fmadd_pd(f0,f1,t0)
+ #define CPU_FMSUB_SS(f0,f1,t0) _mm_fmsub_ss(f0,f1,t0)
+ #define CPU_FMSUB_PS(f0,f1,t0) _mm_fmsub_ps(f0,f1,t0)
+ #define CPU_FMSUB_SD(f0,f1,t0) _mm_fmsub_sd(f0,f1,t0)
+ #define CPU_FMSUB_PD(f0,f1,t0) _mm_fmsub_pd(f0,f1,t0)
+ #define CPU_FMADD256_SS(f0,f1,t0) _mm256_fmadd_ss(f0,f1,t0)
+ #define CPU_FMADD256_PS(f0,f1,t0) _mm256_fmadd_ps(f0,f1,t0)
+ #define CPU_FMADD256_SD(f0,f1,t0) _mm256_fmadd_sd(f0,f1,t0)
+ #define CPU_FMADD256_PD(f0,f1,t0) _mm256_fmadd_pd(f0,f1,t0)
+ #define CPU_FMSUB256_SS(f0,f1,t0) _mm256_fmsub_ss(f0,f1,t0)
+ #define CPU_FMSUB256_PS(f0,f1,t0) _mm256_fmsub_ps(f0,f1,t0)
+ #define CPU_FMSUB256_SD(f0,f1,t0) _mm256_fmsub_sd(f0,f1,t0)
+ #define CPU_FMSUB256_PD(f0,f1,t0) _mm256_fmsub_pd(f0,f1,t0)
+#elif CPU_FMA4_SUPPORT
+ #define CPU_FMADD_SS(f0,f1,t0) _mm_macc_ss(f0,f1,t0)
+ #define CPU_FMADD_PS(f0,f1,t0) _mm_macc_ps(f0,f1,t0)
+ #define CPU_FMADD_SD(f0,f1,t0) _mm_macc_sd(f0,f1,t0)
+ #define CPU_FMADD_PD(f0,f1,t0) _mm_macc_pd(f0,f1,t0)
+ #define CPU_FMSUB_SS(f0,f1,t0) _mm_msub_ss(f0,f1,t0)
+ #define CPU_FMSUB_PS(f0,f1,t0) _mm_msub_ps(f0,f1,t0)
+ #define CPU_FMSUB_SD(f0,f1,t0) _mm_msub_sd(f0,f1,t0)
+ #define CPU_FMSUB_PD(f0,f1,t0) _mm_msub_pd(f0,f1,t0)
+ #define CPU_FMADD256_SS(f0,f1,t0) _mm256_macc_ss(f0,f1,t0)
+ #define CPU_FMADD256_PS(f0,f1,t0) _mm256_macc_ps(f0,f1,t0)
+ #define CPU_FMADD256_SD(f0,f1,t0) _mm256_macc_sd(f0,f1,t0)
+ #define CPU_FMADD256_PD(f0,f1,t0) _mm256_macc_pd(f0,f1,t0)
+ #define CPU_FMSUB256_SS(f0,f1,t0) _mm256_msub_ss(f0,f1,t0)
+ #define CPU_FMSUB256_PS(f0,f1,t0) _mm256_msub_ps(f0,f1,t0)
+ #define CPU_FMSUB256_SD(f0,f1,t0) _mm256_msub_sd(f0,f1,t0)
+ #define CPU_FMSUB256_PD(f0,f1,t0) _mm256_msub_pd(f0,f1,t0)
+#else
+ #define CPU_FMADD_SS(f0,f1,t0) _mm_add_ss(_mm_mul_ss(f0,f1),t0)
+ #define CPU_FMADD_PS(f0,f1,t0) _mm_add_ps(_mm_mul_ps(f0,f1),t0)
+ #define CPU_FMADD_SD(f0,f1,t0) _mm_add_sd(_mm_mul_sd(f0,f1),t0)
+ #define CPU_FMADD_PD(f0,f1,t0) _mm_add_pd(_mm_mul_pd(f0,f1),t0)
+ #define CPU_FMSUB_SS(f0,f1,t0) _mm_sub_ss(_mm_mul_ss(f0,f1),t0)
+ #define CPU_FMSUB_PS(f0,f1,t0) _mm_sub_ps(_mm_mul_ps(f0,f1),t0)
+ #define CPU_FMSUB_SD(f0,f1,t0) _mm_sub_sd(_mm_mul_sd(f0,f1),t0)
+ #define CPU_FMSUB_PD(f0,f1,t0) _mm_sub_pd(_mm_mul_pd(f0,f1),t0)
+ #define CPU_FMADD256_SS(f0,f1,t0) _mm256_add_ss(_mm256_mul_ss(f0,f1),t0)
+ #define CPU_FMADD256_PS(f0,f1,t0) _mm256_add_ps(_mm256_mul_ps(f0,f1),t0)
+ #define CPU_FMADD256_SD(f0,f1,t0) _mm256_add_sd(_mm256_mul_sd(f0,f1),t0)
+ #define CPU_FMADD256_PD(f0,f1,t0) _mm256_add_pd(_mm256_mul_pd(f0,f1),t0)
+ #define CPU_FMSUB256_SS(f0,f1,t0) _mm256_sub_ss(_mm256_mul_ss(f0,f1),t0)
+ #define CPU_FMSUB256_PS(f0,f1,t0) _mm256_sub_ps(_mm256_mul_ps(f0,f1),t0)
+ #define CPU_FMSUB256_SD(f0,f1,t0) _mm256_sub_sd(_mm256_mul_sd(f0,f1),t0)
+ #define CPU_FMSUB256_PD(f0,f1,t0) _mm256_sub_pd(_mm256_mul_pd(f0,f1),t0)
+#endif
+
+
+////
+
+
+#if CPU_SSE_SUPPORT
+
+extern const uint32_t simd4fSignMask[4];
+extern const uint32_t simd4fSignMaskInv[4];
+extern const float simd4fHalf[4];
+extern const float simd4fOne[4];
+extern const float simd4fTwo[4];
+extern const float simd4fThree[4];
+extern const uint32_t simd4uOne[4];
+extern const uint32_t simd4uOneInv[4];
+extern const uint32_t simd4uTwo[4];
+extern const uint32_t simd4uFour[4];
+extern const float simd4fQuarter[4];
+extern const float simd4fPi[4];
+extern const float simd4fZeroOneTwoThree[4];
+extern const uint32_t simd4fAlphaMask[4];
+extern const float simd4f255[4];
+extern const float simd4f255Inv[4];
+
+#endif
+
+
+#if CPU_SSE2_SUPPORT
+
+/* Input range between -8192 and 8192 */
+__m128 simd4f_sin_ps( __m128 x );
+__m128 simd4f_cos_ps( __m128 x );
+void simd4f_sincos_ps( __m128 x, __m128 *s, __m128 *c );
+
+#endif
+
+#if CPU_SSE2_SUPPORT
+
+__m128 simd4f_exp2_ps( __m128 x );
+__m128 simd4f_log2_ps( __m128 x );
+__m128 simd4f_pow_ps( __m128 x, __m128 y );
+
+#endif
+
+#if CPU_SSE2_SUPPORT
+
+__m128 simd4f_pow12d5_ps( __m128 arg );
+__m128 simd4f_pow5d12_ps( __m128 arg );
+
+#endif
+
+
+////
+
+
+#if CPU_SSE2_SUPPORT
+
+#ifndef CC_ALWAYSINLINE
+ #if defined(__GNUC__) || defined(__INTEL_COMPILER)
+  #define CC_ALWAYSINLINE __attribute__((always_inline))
+ #else
+  #define CC_ALWAYSINLINE
+ #endif
+#endif
+
+static inline CC_ALWAYSINLINE __m128 simd4f_pow12d5_inline_ps( __m128 vx )
+{
+  __m128 vpow, vpwsqrtinv, vpwsqrt, vx2;
+  vx2 = _mm_mul_ps( vx, vx );
+  vpow = _mm_castsi128_ps( _mm_cvtps_epi32( _mm_mul_ps( _mm_cvtepi32_ps( _mm_castps_si128( _mm_mul_ps( vx, _mm_set1_ps( 5417434112.0f ) ) ) ), _mm_set1_ps( 0.8f ) ) ) );
+  vpwsqrtinv = _mm_rsqrt_ps( vpow );
+  vpwsqrt = _mm_mul_ps( vpow, vpwsqrtinv );
+  return _mm_mul_ps( _mm_add_ps( _mm_mul_ps( vx2, vpwsqrt ), _mm_mul_ps( _mm_mul_ps( _mm_mul_ps( vx2, vx ), vpwsqrtinv ), _mm_rsqrt_ps( vpwsqrt ) ) ), _mm_set1_ps( 0.51011878327f ) );
+}
+
+static inline CC_ALWAYSINLINE __m128 simd4f_pow5d12_inline_ps( __m128 vx )
+{
+  __m128 vpow;
+  vpow = _mm_castsi128_ps( _mm_cvtps_epi32( _mm_mul_ps( _mm_cvtepi32_ps( _mm_castps_si128( _mm_mul_ps( vx, _mm_set1_ps( 6521909350804488192.0f ) ) ) ), _mm_set1_ps( 0.666666666666f ) ) ) );
+  vx = _mm_mul_ps( _mm_add_ps( _mm_mul_ps( vx, vpow ), _mm_mul_ps( _mm_mul_ps( vx, vx ), _mm_rsqrt_ps( vpow ) ) ), _mm_set1_ps( 0.5290553722f ) );
+#if 0
+  vx = _mm_mul_ps( vx, _mm_rsqrt_ps( vx ) );
+  vx = _mm_mul_ps( vx, _mm_rsqrt_ps( vx ) );
+#else
+  vx = _mm_sqrt_ps( vx );
+  vx = _mm_sqrt_ps( vx );
+#endif
+  return vx;
+}
+
+#endif
+
+
+////
+
+
+#if CPU_SSE_SUPPORT
+
+static inline void simdPrintDebugSSE4f( char *str, __m128 v )
+{
+  float CPU_ALIGN16 store[4];
+  _mm_store_ps( (void *)store, v );
+  printf( "%s %f %f %f %f\n", str, (double)store[0], (double)store[1], (double)store[2], (double)store[3] );
+  return;
+}
+
+static inline void simdPrintDebugSSE2d( char *str, __m128d v )
+{
+  double CPU_ALIGN16 store[2];
+  _mm_store_pd( (void *)store, v );
+  printf( "%s %f %f\n", str, store[0], store[1] );
+  return;
+}
+
+static inline void simdPrintDebugSSE16u8( char *str, __m128i v )
+{
+  uint8_t CPU_ALIGN16 store[16];
+  _mm_store_si128( (void *)store, v );
+  printf( "%s %d %d %d %d %d %d %d %d %d %d %d %d %d %d %d %d\n", str, store[0], store[1], store[2], store[3], store[4], store[5], store[6], store[7], store[8], store[9], store[10], store[11], store[12], store[13], store[14], store[15] );
+  return;
+}
+
+static inline void simdPrintDebugSSE8u16( char *str, __m128i v )
+{
+  uint16_t CPU_ALIGN16 store[8];
+  _mm_store_si128( (void *)store, v );
+  printf( "%s %d %d %d %d %d %d %d %d\n", str, store[0], store[1], store[2], store[3], store[4], store[5], store[6], store[7] );
+  return;
+}
+
+static inline void simdPrintDebugSSE4u32( char *str, __m128i v )
+{
+  uint32_t CPU_ALIGN16 store[4];
+  _mm_store_si128( (void *)store, v );
+  printf( "%s %d %d %d %d\n", str, store[0], store[1], store[2], store[3] );
+  return;
+}
+
+static inline void simdPrintDebugSSE2u64( char *str, __m128i v )
+{
+  uint64_t CPU_ALIGN16 store[2];
+  _mm_store_si128( (void *)store, v );
+  printf( "%s %lld %lld\n", str, (long long)store[0], (long long)store[1] );
+  return;
+}
+
+#endif
+
+
+////
+
+
+#endif
+
diff --git a/lib/graphics_utils/mipmap/img.h b/lib/graphics_utils/mipmap/img.h
new file mode 100644
index 000000000..17044a517
--- /dev/null
+++ b/lib/graphics_utils/mipmap/img.h
@@ -0,0 +1,77 @@
+/* -----------------------------------------------------------------------------
+ *
+ * Copyright (c) 2007-2017 Alexis Naveros.
+ * Portions developed under contract to the SURVICE Engineering Company.
+ *
+ * This software is provided 'as-is', without any express or implied
+ * warranty. In no event will the authors be held liable for any damages
+ * arising from the use of this software.
+ *
+ * Permission is granted to anyone to use this software for any purpose,
+ * including commercial applications, and to alter it and redistribute it
+ * freely, subject to the following restrictions:
+ *
+ * 1. The origin of this software must not be misrepresented; you must not
+ * claim that you wrote the original software. If you use this software
+ * in a product, an acknowledgment in the product documentation would be
+ * appreciated but is not required.
+ * 2. Altered source versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ * 3. This notice may not be removed or altered from any source distribution.
+ *
+ * -----------------------------------------------------------------------------
+ */
+
+#ifndef IMG_H
+#define IMG_H
+
+
+typedef struct
+{
+  int width;
+  int height;
+  int type;
+  int bytesperpixel;
+  int bytesperline;
+} imgFormat;
+
+enum
+{
+  IMG_FORMAT_TYPE_ANY,
+  IMG_FORMAT_TYPE_RGB24,
+  IMG_FORMAT_TYPE_BGR24,
+  IMG_FORMAT_TYPE_RGBX32,
+  IMG_FORMAT_TYPE_BGRX32,
+  IMG_FORMAT_TYPE_RGBA32,
+  IMG_FORMAT_TYPE_BGRA32,
+  IMG_FORMAT_TYPE_GRAYSCALE,
+  IMG_FORMAT_TYPE_GRAYALPHA
+};
+
+typedef struct
+{
+  imgFormat format;
+  void *data;
+} imgImage;
+
+
+////
+
+
+void imgCopyRect( imgImage *image, int dstx, int dsty, int srcx, int srcy, int sizex, int sizey );
+
+void (*imgBlendGetFunction( imgImage *dstimage, imgImage *srcimage ))( imgImage *dstimage, int dstx, int dsty, imgImage *srcimage );
+int imgBlendImage( imgImage *dstimage, int dstx, int dsty, imgImage *srcimage );
+
+void imgAllocCopy( imgImage *dst, imgImage *src );
+void imgAllocCopyExtendBorder( imgImage *dstimage, imgImage *srcimage, int extendsize );
+void imgAllocExtractChannel( imgImage *dst, imgImage *src, int channelindex );
+void imgAllocExtractChannelExtendBorder( imgImage *dstimage, imgImage *srcimage, int channelindex, int extendsize );
+void imgAllocCopyChannelToAlpha( imgImage *dstimage, imgImage *srcimage, int channelindex, unsigned char r, unsigned char g, unsigned char b );
+void imgAllocAdjustBrightnessContrast( imgImage *dstimage, imgImage *srcimage, float brightness, float contrast );
+
+void imgFree( imgImage *image );
+
+
+#endif
+
diff --git a/lib/graphics_utils/mipmap/imgresize.c b/lib/graphics_utils/mipmap/imgresize.c
new file mode 100644
index 000000000..b376bc46d
--- /dev/null
+++ b/lib/graphics_utils/mipmap/imgresize.c
@@ -0,0 +1,4098 @@
+/* -----------------------------------------------------------------------------
+ *
+ * Copyright (c) 2014-2017 Alexis Naveros.
+ * Portions developed under contract to the SURVICE Engineering Company.
+ *
+ * This software is provided 'as-is', without any express or implied
+ * warranty. In no event will the authors be held liable for any damages
+ * arising from the use of this software.
+ *
+ * Permission is granted to anyone to use this software for any purpose,
+ * including commercial applications, and to alter it and redistribute it
+ * freely, subject to the following restrictions:
+ *
+ * 1. The origin of this software must not be misrepresented; you must not
+ * claim that you wrote the original software. If you use this software
+ * in a product, an acknowledgment in the product documentation would be
+ * appreciated but is not required.
+ * 2. Altered source versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ * 3. This notice may not be removed or altered from any source distribution.
+ *
+ * -----------------------------------------------------------------------------
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <stddef.h>
+#include <stdint.h>
+#include <string.h>
+#include <math.h>
+#include <float.h>
+#include <limits.h>
+
+
+#include "cpusimd.h"
+
+#include "img.h"
+#include "imgresize.h"
+
+
+////
+
+
+#define IM_RESIZE_DEBUG (0)
+#define IM_RESIZE_DEBUG_PROGRESS (0)
+
+
+////
+
+
+#ifndef M_PI
+ #define M_PI (3.14159265358979323846)
+#endif
+
+#ifndef ADDRESS
+ #define ADDRESS(p,o) ((void *)(((char *)p)+(o)))
+#endif
+
+#if defined(__GNUC__) || defined(__INTEL_COMPILER) || defined(__clang__)
+ #define CC_ALWAYSINLINE __attribute__((always_inline))
+ #if __STDC_VERSION__ >= 199901L
+  #define CC_RESTRICT restrict
+ #else
+  #define CC_RESTRICT
+ #endif
+#elif defined(_MSC_VER)
+ #define CC_ALWAYSINLINE __forceinline
+ #define CC_RESTRICT __restrict
+#else
+ #define CC_ALWAYSINLINE
+ #define CC_RESTRICT
+#endif
+
+static inline CC_ALWAYSINLINE uint32_t ccIsPow2Int32( uint32_t v )
+{
+  return ( ( v & ( v - 1 ) ) == 0 );
+}
+
+#define ROUND_POSITIVE_FLOAT(x) ((int)((x)+0.5f))
+
+
+////
+
+
+#if defined(__x86_64__) || defined(__x86_64) || defined(__amd64__) || defined(__amd64) || defined(__i386__) || defined(__i386) || defined(i386) || defined(_M_X64) || defined(_M_AMD64) || defined(_M_IX86) || defined(_X86_)
+
+/* Input is 0.0,255.0, output is 0.0,1.0 */
+static inline CC_ALWAYSINLINE float srgb2linear( float v )
+{
+  float v2, vpow, vpwsqrt;
+  union
+  {
+    int32_t i;
+    float f;
+  } u;
+  if( v <= (0.04045f*255.0f) )
+    v = v * ( (1.0f/12.92f)*(1.0f/255.0f) );
+  else
+  {
+    v = ( v + (0.055f*255.0f) ) * ( (1.0f/1.055f)*(1.0f/255.0f) );
+    v2 = v * v;
+    u.f = v * 5417434112.0f;
+    u.i = (int32_t)ROUND_POSITIVE_FLOAT( (float)u.i * 0.8f );
+    vpow = u.f;
+    vpwsqrt = sqrtf( vpow );
+    v = ( ( v2 * vpwsqrt ) + ( ( ( v2 * v ) / vpwsqrt ) / sqrtf( vpwsqrt ) ) ) * 0.51011878327f;
+  }
+  return v;
+}
+
+/* Input is 0.0,1.0, output is 0.0,255.0 */
+static inline CC_ALWAYSINLINE float linear2srgb( float v )
+{
+  float vpow;
+  union
+  {
+    int32_t i;
+    float f;
+  } u;
+  if( v <= 0.0031308f )
+    v = v * (12.92f*255.0f);
+  else
+  {
+    u.f = ( v * 6521909350804488192.0f );
+    u.i = (int32_t)ROUND_POSITIVE_FLOAT( (float)u.i * 0.666666666666f );
+    vpow = u.f;
+    v = ( v * vpow ) + ( ( v * v ) / sqrtf( vpow ) );
+    v = ( (1.055f*255.0f) * sqrtf( sqrtf( v * 0.5290553722f ) ) - (0.055f*255.0f) );
+  }
+  return v;
+}
+
+#else
+
+/* Input is 0.0,255.0, output is 0.0,1.0 */
+/* Only for reference, this is waayyy too slow and should never be used */
+static inline CC_ALWAYSINLINE float srgb2linear( float v )
+{
+  v *= (1.0f/255.0f);
+  if( v <= 0.04045f )
+    v = v * (1.0f/12.92);
+  else
+    v = powf( ( v + 0.055f ) * (1.0f/1.055f), 2.4f );
+  return v;
+}
+
+/* Input is 0.0,1.0, output is 0.0,255.0 */
+/* Only for reference, this is waayyy too slow and should never be used */
+static inline CC_ALWAYSINLINE float linear2srgb( float v )
+{
+  if( v <= 0.0031308f )
+    v = v * 12.92f;
+  else
+    v = 1.055f * powf( v, 1.0f/2.4f ) - 0.055f;
+  return 255.0f * v;
+}
+
+#endif
+
+
+////
+
+
+#if CPU_SSE2_SUPPORT
+
+static const float CPU_ALIGN16 srgbLinearConst00[4] = { 0.04045f*255.0f, 0.04045f*255.0f, 0.04045f*255.0f, 1024.0f };
+static const float CPU_ALIGN16 srgbLinearConst01[4] = { (1.0f/12.92f)*(1.0f/255.0f), (1.0f/12.92f)*(1.0f/255.0f), (1.0f/12.92f)*(1.0f/255.0f), 1.0f };
+static const float CPU_ALIGN16 srgbLinearConst02[4] = { 0.055f*255.0f, 0.055f*255.0f, 0.055f*255.0f, 0.055f*255.0f };
+static const float CPU_ALIGN16 srgbLinearConst03[4] = { (1.0f/1.055f)*(1.0f/255.0f), (1.0f/1.055f)*(1.0f/255.0f), (1.0f/1.055f)*(1.0f/255.0f), (1.0f/1.055f)*(1.0f/255.0f) };
+static const float CPU_ALIGN16 srgbLinearConst04[4] = { 5417434112.0f, 5417434112.0f, 5417434112.0f, 5417434112.0f };
+static const float CPU_ALIGN16 srgbLinearConst05[4] = { 0.8f, 0.8f, 0.8f, 0.8f };
+static const float CPU_ALIGN16 srgbLinearConst06[4] = { 0.51011878327f, 0.51011878327f, 0.51011878327f, 0.51011878327f };
+static const float CPU_ALIGN16 srgbLinearConst07[4] = { 0.0031308f, 0.0031308f, 0.0031308f, 1024.0f };
+static const float CPU_ALIGN16 srgbLinearConst08[4] = { 12.92f*255.0f, 12.92f*255.0f, 12.92f*255.0f, 1.0f };
+static const float CPU_ALIGN16 srgbLinearConst09[4] = { 6521909350804488192.0f, 6521909350804488192.0f, 6521909350804488192.0f, 6521909350804488192.0f };
+static const float CPU_ALIGN16 srgbLinearConst10[4] = { 0.666666666666f, 0.666666666666f, 0.666666666666f, 0.666666666666f };
+static const float CPU_ALIGN16 srgbLinearConst11[4] = { 0.5290553722f, 0.5290553722f, 0.5290553722f, 0.5290553722f };
+static const float CPU_ALIGN16 srgbLinearConst12[4] = { 1.055f*255.0f, 1.055f*255.0f, 1.055f*255.0f, 1.055f*255.0f };
+static const float CPU_ALIGN16 srgbLinearConst13[4] = { -0.055f*255.0f, -0.055f*255.0f, -0.055f*255.0f, -0.055f*255.0f };
+static const float CPU_ALIGN16 srgbLinearConst14[4] = { 0.04045f*255.0f, 0.04045f*255.0f, 0.04045f*255.0f, 0.04045f*255.0f };
+static const float CPU_ALIGN16 srgbLinearConst15[4] = { (1.0f/12.92f)*(1.0f/255.0f), (1.0f/12.92f)*(1.0f/255.0f), (1.0f/12.92f)*(1.0f/255.0f), (1.0f/12.92f)*(1.0f/255.0f) };
+
+/* Input is 0.0,255.0 ~ output is 0.0,1.0 ~ alpha channel is passed as-is */
+static inline CC_ALWAYSINLINE __m128 srgb2linear3( __m128 vx )
+{
+  __m128 vmask, vbase;
+  __m128 vpow, vpwsqrtinv, vpwsqrt, vx2;
+  vmask = _mm_cmple_ps( vx, *(__m128*)srgbLinearConst00 );
+  vbase = _mm_mul_ps( vx, *(__m128*)srgbLinearConst01 );
+  vx = _mm_mul_ps( _mm_add_ps( vx, *(__m128*)srgbLinearConst02 ), *(__m128*)srgbLinearConst03 );
+  vx2 = _mm_mul_ps( vx, vx );
+  vpow = _mm_castsi128_ps( _mm_cvtps_epi32( _mm_mul_ps( _mm_cvtepi32_ps( _mm_castps_si128( _mm_mul_ps( vx, *(__m128*)srgbLinearConst04 ) ) ), *(__m128*)srgbLinearConst05 ) ) );
+  vpwsqrtinv = _mm_rsqrt_ps( vpow );
+  vpwsqrt = _mm_mul_ps( vpow, vpwsqrtinv );
+  vx = _mm_mul_ps( _mm_add_ps( _mm_mul_ps( vx2, vpwsqrt ), _mm_mul_ps( _mm_mul_ps( _mm_mul_ps( vx2, vx ), vpwsqrtinv ), _mm_rsqrt_ps( vpwsqrt ) ) ), *(__m128*)srgbLinearConst06 );
+  return CPU_BLENDV_PS( vx, vbase, vmask );
+}
+
+/* Input is 0.0,1.0 ~ output is 0.0,255.0 ~ alpha channel is passed as-is */
+static inline CC_ALWAYSINLINE __m128 linear2srgb3( __m128 vx )
+{
+  __m128 vmask, vbase, vpow;
+  vmask = _mm_cmple_ps( vx, *(__m128*)srgbLinearConst07 );
+  vbase = _mm_mul_ps( vx, *(__m128*)srgbLinearConst08 );
+  vpow = _mm_castsi128_ps( _mm_cvtps_epi32( _mm_mul_ps( _mm_cvtepi32_ps( _mm_castps_si128( _mm_mul_ps( vx, *(__m128*)srgbLinearConst09 ) ) ), *(__m128*)srgbLinearConst10 ) ) );
+  vx = _mm_add_ps( _mm_mul_ps( _mm_sqrt_ps( _mm_sqrt_ps( _mm_mul_ps( _mm_add_ps( _mm_mul_ps( vx, vpow ), _mm_mul_ps( _mm_mul_ps( vx, vx ), _mm_rsqrt_ps( vpow ) ) ), *(__m128*)srgbLinearConst11 ) ) ), *(__m128*)srgbLinearConst12 ), *(__m128*)srgbLinearConst13 );
+  return CPU_BLENDV_PS( vx, vbase, vmask );
+}
+
+/* Input is 0.0,255.0 ~ output is 0.0,1.0 ~ alpha channel is passed as-is */
+static inline CC_ALWAYSINLINE __m128 srgb2linear4( __m128 vx )
+{
+  __m128 vmask, vbase;
+  __m128 vpow, vpwsqrtinv, vpwsqrt, vx2;
+  vmask = _mm_cmple_ps( vx, *(__m128*)srgbLinearConst14 );
+  vbase = _mm_mul_ps( vx, *(__m128*)srgbLinearConst15 );
+  vx = _mm_mul_ps( _mm_add_ps( vx, *(__m128*)srgbLinearConst02 ), *(__m128*)srgbLinearConst03 );
+  vx2 = _mm_mul_ps( vx, vx );
+  vpow = _mm_castsi128_ps( _mm_cvtps_epi32( _mm_mul_ps( _mm_cvtepi32_ps( _mm_castps_si128( _mm_mul_ps( vx, *(__m128*)srgbLinearConst04 ) ) ), *(__m128*)srgbLinearConst05 ) ) );
+  vpwsqrtinv = _mm_rsqrt_ps( vpow );
+  vpwsqrt = _mm_mul_ps( vpow, vpwsqrtinv );
+  vx = _mm_mul_ps( _mm_add_ps( _mm_mul_ps( vx2, vpwsqrt ), _mm_mul_ps( _mm_mul_ps( _mm_mul_ps( vx2, vx ), vpwsqrtinv ), _mm_rsqrt_ps( vpwsqrt ) ) ), *(__m128*)srgbLinearConst06 );
+  return CPU_BLENDV_PS( vx, vbase, vmask );
+}
+
+#endif
+
+
+////
+
+
+static inline CC_ALWAYSINLINE double bessel( double x )
+{
+  double sum, t, y;
+
+  /* Zeroth order Bessel function of the first kind. */
+  sum = 1.0;
+  y = x * x * 0.25;
+  t = y;
+  sum += t;
+  t *= y * (1.0/(2.0*2.0));
+  sum += t;
+  t *= y * (1.0/(3.0*3.0));
+  sum += t;
+  t *= y * (1.0/(4.0*4.0));
+  sum += t;
+  t *= y * (1.0/(5.0*5.0));
+  sum += t;
+  t *= y * (1.0/(6.0*6.0));
+  sum += t;
+  t *= y * (1.0/(7.0*7.0));
+  sum += t;
+  t *= y * (1.0/(8.0*8.0));
+  sum += t;
+  t *= y * (1.0/(9.0*9.0));
+  sum += t;
+  t *= y * (1.0/(10.0*10.0));
+  sum += t;
+  t *= y * (1.0/(11.0*11.0));
+  sum += t;
+  t *= y * (1.0/(12.0*12.0));
+  sum += t;
+  t *= y * (1.0/(13.0*13.0));
+  sum += t;
+  t *= y * (1.0/(14.0*14.0));
+  sum += t;
+
+  return sum;
+}
+
+static inline CC_ALWAYSINLINE double kaiser( double x, double beta )
+{
+  return bessel( beta * sqrt( fmax( 0.0, 1.0 - ( x * x ) ) ) );
+}
+
+static inline CC_ALWAYSINLINE double sinc( double x )
+{
+  if( x == 0.0 )
+    return 1.0;
+  x = sin( x * M_PI ) / ( x * M_PI );
+  return x;
+}
+
+
+////
+
+
+#if CPU_SSE2_SUPPORT
+
+static inline CC_ALWAYSINLINE __m128 simd4f_bessel( __m128 x )
+{
+  __m128 sum, t, y;
+
+  sum = *(__m128 *)simd4fOne;
+  y = _mm_mul_ps( *(__m128 *)simd4fQuarter, _mm_mul_ps( x, x ) );
+  t = y;
+  sum = _mm_add_ps( sum, t );
+  t = _mm_mul_ps( t, _mm_mul_ps( y, _mm_set1_ps( 1.0f/(2.0f*2.0f) ) ) );
+  sum = _mm_add_ps( sum, t );
+  t = _mm_mul_ps( t, _mm_mul_ps( y, _mm_set1_ps( 1.0f/(3.0f*3.0f) ) ) );
+  sum = _mm_add_ps( sum, t );
+  t = _mm_mul_ps( t, _mm_mul_ps( y, _mm_set1_ps( 1.0f/(4.0f*4.0f) ) ) );
+  sum = _mm_add_ps( sum, t );
+  t = _mm_mul_ps( t, _mm_mul_ps( y, _mm_set1_ps( 1.0f/(5.0f*5.0f) ) ) );
+  sum = _mm_add_ps( sum, t );
+  t = _mm_mul_ps( t, _mm_mul_ps( y, _mm_set1_ps( 1.0f/(6.0f*6.0f) ) ) );
+  sum = _mm_add_ps( sum, t );
+  t = _mm_mul_ps( t, _mm_mul_ps( y, _mm_set1_ps( 1.0f/(7.0f*7.0f) ) ) );
+  sum = _mm_add_ps( sum, t );
+  t = _mm_mul_ps( t, _mm_mul_ps( y, _mm_set1_ps( 1.0f/(8.0f*8.0f) ) ) );
+  sum = _mm_add_ps( sum, t );
+  t = _mm_mul_ps( t, _mm_mul_ps( y, _mm_set1_ps( 1.0f/(9.0f*9.0f) ) ) );
+  sum = _mm_add_ps( sum, t );
+  t = _mm_mul_ps( t, _mm_mul_ps( y, _mm_set1_ps( 1.0f/(10.0f*10.0f) ) ) );
+  sum = _mm_add_ps( sum, t );
+  t = _mm_mul_ps( t, _mm_mul_ps( y, _mm_set1_ps( 1.0f/(11.0f*11.0f) ) ) );
+  sum = _mm_add_ps( sum, t );
+  t = _mm_mul_ps( t, _mm_mul_ps( y, _mm_set1_ps( 1.0f/(12.0f*12.0f) ) ) );
+  sum = _mm_add_ps( sum, t );
+  t = _mm_mul_ps( t, _mm_mul_ps( y, _mm_set1_ps( 1.0f/(13.0f*13.0f) ) ) );
+  sum = _mm_add_ps( sum, t );
+  t = _mm_mul_ps( t, _mm_mul_ps( y, _mm_set1_ps( 1.0f/(14.0f*14.0f) ) ) );
+  sum = _mm_add_ps( sum, t );
+
+  return sum;
+}
+
+static inline CC_ALWAYSINLINE __m128 simd4f_kaiser( __m128 x, __m128 beta )
+{
+  return simd4f_bessel( _mm_mul_ps( beta, _mm_sqrt_ps( _mm_max_ps( _mm_setzero_ps(), _mm_sub_ps( *(__m128 *)simd4fOne, _mm_mul_ps( x, x ) ) ) ) ) );
+}
+
+static inline CC_ALWAYSINLINE __m128 simd4f_sinc( __m128 x )
+{
+  __m128 zeromask;
+  zeromask = _mm_cmpeq_ps( x, _mm_setzero_ps() );
+  x = _mm_mul_ps( x, _mm_load_ps( simd4fPi ) );
+  x = _mm_div_ps( simd4f_sin_ps( x ), x );
+  x = CPU_BLENDV_PS( x, *(__m128 *)simd4fOne, zeromask );
+  return x;
+}
+
+#endif
+
+
+////
+
+
+typedef struct
+{
+  int matrixsize;
+  int matrixoffset;
+  int matrixrowwidth;
+  int matrixrowsize;
+  int rowreturn;
+  float *matrix;
+
+  int minimumalpha;
+  float dithersum;
+  float minimumalphaf;
+  float amplifynormal;
+  float normalsustainfactor;
+  void *alloc;
+
+  unsigned char *srcdata;
+  int width1;
+  int width2;
+  int width3;
+  int width4;
+  int height;
+  int bytesperline;
+} imStaticMatrixState;
+
+
+static int imBuildStaticMatrix( imStaticMatrixState * CC_RESTRICT state, int sizedivisor, float hopcount, float alpha )
+{
+  int i, j, minx, maxx;
+  double x, xshift, hopsize, offset, scalefactor, hopcountinv, beta, linsq, sum;
+  double *linear;
+  float suminv;
+  float *matrix;
+
+  if( alpha > 16.0f )
+    alpha = 16.0f;
+  beta = (double)alpha * (double)M_PI;
+  hopcountinv = 1.0 / (double)hopcount;
+
+  scalefactor = 1.0 / (double)sizedivisor;
+  hopsize = 0.5 * (double)sizedivisor;
+  offset = hopsize - 0.5;
+  minx = (int)ceil( ( (double)-hopcount * hopsize ) + offset );
+  maxx = (int)floor( ( (double)hopcount * hopsize ) + offset );
+  state->matrixoffset = minx;
+  state->matrixsize = ( maxx - minx ) + 1;
+  state->matrixrowwidth = ( state->matrixsize + 3 ) & ~3;
+  state->rowreturn = state->matrixrowwidth - state->matrixsize;
+  state->matrixrowsize = state->matrixrowwidth * sizeof(float);
+
+#if IM_RESIZE_DEBUG
+  printf( "ResizeMatrix ; scalefactor %.3f, offset %.3f, hopsize %.3f\n", scalefactor, offset, hopsize );
+#endif
+
+  linear = malloc( state->matrixrowwidth * sizeof(double) );
+  for( i = 0 ; i < state->matrixsize ; i++ )
+  {
+    x = (double)( i + state->matrixoffset );
+    xshift = 2.0 * scalefactor * ( x - offset );
+    linear[i] = sinc( xshift ) * kaiser( hopcountinv * xshift, beta );
+#if IM_RESIZE_DEBUG
+    printf( " x[%+.3f] = %+.3f ( %+.3f * %+.3f )\n", x, linear[i], sinc( xshift ), kaiser( hopcountinv * xshift, beta ) );
+#endif
+  }
+  for( ; i < state->matrixrowwidth ; i++ )
+    linear[i] = 0.0;
+
+  /* Build normalized state */
+  state->alloc = malloc( ( state->matrixsize * state->matrixrowsize ) + 16 );
+  state->matrix = (void *)( ( (uintptr_t)state->alloc + 0xf ) & ~0xf );
+  matrix = state->matrix;
+  sum = 0.0;
+  for( i = 0 ; i < state->matrixsize ; i++ )
+  {
+    for( j = 0 ; j < state->matrixsize ; j++ )
+    {
+      linsq = linear[i] * linear[j];
+      matrix[j] = (float)linsq;
+      sum += linsq;
+    }
+    for( ; j < state->matrixrowwidth ; j++ )
+      matrix[j] = 0.0f;
+    matrix += state->matrixrowwidth;
+  }
+  free( linear );
+
+#if IM_RESIZE_DEBUG
+  printf( "Matrix sum : %f\n", sum );
+#endif
+
+  suminv = (float)( 1.0 / sum );
+  j = state->matrixsize * state->matrixrowwidth;
+  for( i = 0 ; i < j ; i++ )
+    state->matrix[i] *= suminv;
+
+#if IM_RESIZE_DEBUG
+  printf( "Matrix %dx%d :\n", state->matrixsize, state->matrixsize );
+  for( i = 0 ; i < state->matrixsize ; i++ )
+  {
+    for( j = 0 ; j < state->matrixsize ; j++ )
+      printf( " %+.6f", state->matrix[ ( i * state->matrixrowwidth ) + j ] );
+    printf( "\n" );
+  }
+  printf( "Matrix Offset : %d\n", state->matrixoffset );
+  printf( "Matrix Size : %d\n", state->matrixsize );
+  printf( "Matrix Rowwidth : %d\n", state->matrixrowwidth );
+#endif
+
+  return 1;
+}
+
+
+static void imFreeStaticState( imStaticMatrixState * CC_RESTRICT state )
+{
+  free( state->alloc );
+  state->alloc = 0;
+  return;
+}
+
+
+////
+
+
+typedef struct
+{
+  int matrixsizex, matrixsizey;
+  int matrixoffsetx, matrixoffsety;
+  float *linearx;
+  float *lineary;
+  float beta;
+  float hopcountinv;
+
+  float dithersum;
+  int minimumalpha;
+  float minimumalphaf;
+  float amplifynormal;
+  float normalsustainfactor;
+  void *alloc;
+
+  unsigned char *srcdata;
+  int width1;
+  int width2;
+  int width3;
+  int width4;
+  int height;
+  int bytesperline;
+} imGenericMatrixState;
+
+
+static inline int imAllocGenericState( imGenericMatrixState *state, float scalex, float scaley, float hopcount, float alpha )
+{
+  int allocx, allocy, size;
+  void *align;
+  if( alpha > 16.0f )
+    alpha = 16.0f;
+  allocx = ( (int)ceilf( hopcount / scalex ) + 2 + 3 ) & ~0x3;
+  allocy = ( (int)ceilf( hopcount / scaley ) + 2 + 3 ) & ~0x3;
+  size = ( ( allocx + allocy ) * sizeof(float) ) + 16;
+  state->alloc = malloc( size );
+  memset( state->alloc, 0, size );
+  align = (void *)( ( (uintptr_t)state->alloc + 0xf ) & ~0xf );
+  state->linearx = align;
+  state->lineary = ADDRESS( align, allocx * sizeof(float) );
+  state->beta = alpha * (float)M_PI;
+  state->hopcountinv = 1.0f / (float)hopcount;
+  return 1;
+}
+
+static inline void imBuildGenericLinearX( imGenericMatrixState *state, float scalex, float scaleinvx, float sourcex, float hopcount, float alpha, int width )
+{
+  int i, minx, maxx;
+  float hopsizex, offsetx;
+  float *linearx;
+
+  hopsizex = 0.5f * scaleinvx;
+  offsetx = (float)sourcex;
+  minx = (int)ceil( ( -hopcount * hopsizex ) + offsetx );
+  maxx = (int)floor( ( hopcount * hopsizex ) + offsetx );
+  state->matrixsizex = ( maxx - minx ) + 1;
+  state->matrixoffsetx = ( minx + ( width << 8 ) ) % width;
+
+  linearx = state->linearx;
+  scalex *= 2.0f;
+#if CPU_SSE2_SUPPORT
+  for( i = 0 ; i < state->matrixsizex ; i += 4 )
+  {
+    __m128 vx, vxshift;
+    vx = _mm_add_ps( _mm_set1_ps( (float)( i + minx ) ), _mm_load_ps( simd4fZeroOneTwoThree ) );
+    vxshift = _mm_mul_ps( _mm_set1_ps( scalex ), _mm_sub_ps( vx, _mm_set1_ps( offsetx ) ) );
+    _mm_store_ps( &linearx[i], _mm_mul_ps( simd4f_sinc( vxshift ), simd4f_kaiser( _mm_mul_ps( _mm_set1_ps( state->hopcountinv ), vxshift ), _mm_set1_ps( state->beta ) ) ) );
+ #if IM_RESIZE_DEBUG
+    printf( " linearx[%d] = %.3f\n", i+minx+0, linearx[i+0] );
+    printf( " linearx[%d] = %.3f\n", i+minx+1, linearx[i+1] );
+    printf( " linearx[%d] = %.3f\n", i+minx+2, linearx[i+2] );
+    printf( " linearx[%d] = %.3f\n", i+minx+3, linearx[i+3] );
+ #endif
+  }
+#else
+  for( i = 0 ; i < state->matrixsizex ; i++ )
+  {
+    float x, xshift;
+    x = (float)( i + minx );
+    xshift = scalex * ( x - offsetx );
+    linearx[i] = (float)( sinc( xshift ) * kaiser( state->hopcountinv * xshift, state->beta ) );
+ #if IM_RESIZE_DEBUG
+    printf( " linearx[%+.3f] = %.3f ( %+.3f * %+.3f )\n", x, linearx[i], sinc( xshift ), kaiser( state->hopcountinv * xshift, state->beta ) );
+ #endif
+  }
+#endif
+
+  return;
+}
+
+static inline void imBuildGenericLinearY( imGenericMatrixState *state, float scaley, float scaleinvy, float sourcey, float hopcount, float alpha, int height )
+{
+  int i, miny, maxy;
+  float hopsizey, offsety;
+  float *lineary;
+
+  hopsizey = 0.5f * scaleinvy;
+  offsety = (float)sourcey;
+  miny = (int)ceil( ( -hopcount * hopsizey ) + offsety );
+  maxy = (int)floor( ( hopcount * hopsizey ) + offsety );
+  state->matrixsizey = ( maxy - miny ) + 1;
+  state->matrixoffsety = ( miny + ( height << 8 ) ) % height;
+
+  lineary = state->lineary;
+  scaley *= 2.0f;
+#if CPU_SSE2_SUPPORT
+  for( i = 0 ; i < state->matrixsizey ; i += 4 )
+  {
+    __m128 vy, vyshift;
+    vy = _mm_add_ps( _mm_set1_ps( (float)( i + miny ) ), _mm_load_ps( simd4fZeroOneTwoThree ) );
+    vyshift = _mm_mul_ps( _mm_set1_ps( scaley ), _mm_sub_ps( vy, _mm_set1_ps( offsety ) ) );
+    _mm_store_ps( &lineary[i], _mm_mul_ps( simd4f_sinc( vyshift ), simd4f_kaiser( _mm_mul_ps( _mm_set1_ps( state->hopcountinv ), vyshift ), _mm_set1_ps( state->beta ) ) ) );
+ #if IM_RESIZE_DEBUG
+    printf( " lineary[%d] = %.3f\n", i+miny+0, lineary[i+0] );
+    printf( " lineary[%d] = %.3f\n", i+miny+1, lineary[i+1] );
+    printf( " lineary[%d] = %.3f\n", i+miny+2, lineary[i+2] );
+    printf( " lineary[%d] = %.3f\n", i+miny+3, lineary[i+3] );
+ #endif
+  }
+#else
+  for( i = 0 ; i < state->matrixsizey ; i++ )
+  {
+    float y, yshift;
+    y = (float)( i + miny );
+    yshift = scaley * ( y - offsety );
+    lineary[i] = (float)( sinc( yshift ) * kaiser( state->hopcountinv * yshift, state->beta ) );
+ #if IM_RESIZE_DEBUG
+    printf( " lineary[%+.3f] = %.3f ( %+.3f * %+.3f )\n", y, lineary[i], sinc( yshift ), kaiser( state->hopcountinv * yshift, state->beta ) );
+ #endif
+  }
+#endif
+
+  return;
+}
+
+
+static inline void imFreeGenericState( imGenericMatrixState *state )
+{
+  free( state->alloc );
+  state->alloc = 0;
+  return;
+}
+
+
+
+////////////////////////////////////////////////////////////////////////////////
+
+
+
+static void imStaticKernel1Linear( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state )
+{
+  int x, y, mapx, mapy;
+  float f, sum0;
+  float *matrix;
+  unsigned char *src;
+
+  sum0 = 0.0f;
+  matrix = state->matrix;
+  mapy = pointy;
+  for( y = 0 ; y < state->matrixsize ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = pointx;
+    for( x = 0 ; x < state->matrixsize ; x++ )
+    {
+      f = matrix[x];
+      sum0 += f * (float)src[ mapx + 0 ];
+      mapx++;
+      if( mapx >= state->width1 )
+        mapx = 0;
+    }
+    matrix = ADDRESS( matrix, state->matrixrowsize );
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum0 + 0.5f ) ) );
+
+  return;
+}
+
+static void imStaticKernel2Linear( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state )
+{
+  int x, y, mapx, mapy;
+  float f, sum0, sum1;
+  float *matrix;
+  unsigned char *src;
+
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  matrix = state->matrix;
+  mapy = pointy;
+  for( y = 0 ; y < state->matrixsize ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = pointx << 1;
+    for( x = 0 ; x < state->matrixsize ; x++ )
+    {
+      f = matrix[x];
+      sum0 += f * (float)src[ mapx + 0 ];
+      sum1 += f * (float)src[ mapx + 1 ];
+      mapx += 2;
+      if( mapx >= state->width2 )
+        mapx = 0;
+    }
+    matrix = ADDRESS( matrix, state->matrixrowsize );
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum0 + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum1 + 0.5f ) ) );
+
+  return;
+}
+
+static void imStaticKernel3Linear( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state )
+{
+  int x, y, mapx, mapy;
+  float f, sum0, sum1, sum2;
+  float *matrix;
+  unsigned char *src;
+
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  matrix = state->matrix;
+  mapy = pointy;
+  for( y = 0 ; y < state->matrixsize ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = pointx + ( pointx << 1 );
+    for( x = 0 ; x < state->matrixsize ; x++ )
+    {
+      f = matrix[x];
+      sum0 += f * (float)src[ mapx + 0 ];
+      sum1 += f * (float)src[ mapx + 1 ];
+      sum2 += f * (float)src[ mapx + 2 ];
+      mapx += 3;
+      if( mapx >= state->width3 )
+        mapx = 0;
+    }
+    matrix = ADDRESS( matrix, state->matrixrowsize );
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum0 + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum1 + 0.5f ) ) );
+  dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum2 + 0.5f ) ) );
+
+  return;
+}
+
+static void imStaticKernel4Linear( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state )
+{
+  int x, y, mapx, mapy;
+#if CPU_SSE2_SUPPORT
+  __m128 vsum, vsrc;
+  __m128i vzero;
+#else
+  float f, sum0, sum1, sum2, sum3;
+#endif
+  float *matrix;
+  unsigned char *src;
+
+#if CPU_SSE2_SUPPORT
+  vsum = _mm_setzero_ps();
+  vzero = _mm_setzero_si128();
+#else
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  sum3 = 0.0f;
+#endif
+  matrix = state->matrix;
+  mapy = pointy;
+  for( y = 0 ; y < state->matrixsize ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = pointx << 2;
+    for( x = 0 ; x < state->matrixsize ; x++ )
+    {
+#if CPU_SSE2_SUPPORT
+      vsrc = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_load_ss( (void *)&src[ mapx ] ) ), vzero ) );
+      vsum = _mm_add_ps( vsum, _mm_mul_ps( _mm_set1_ps( matrix[x] ), vsrc ) );
+#else
+      f = matrix[x];
+      sum0 += f * (float)src[ mapx + 0 ];
+      sum1 += f * (float)src[ mapx + 1 ];
+      sum2 += f * (float)src[ mapx + 2 ];
+      sum3 += f * (float)src[ mapx + 3 ];
+#endif
+      mapx += 4;
+      if( mapx >= state->width4 )
+        mapx = 0;
+    }
+    matrix = ADDRESS( matrix, state->matrixrowsize );
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+#if CPU_SSE2_SUPPORT
+  _mm_store_ss( (void *)dst, _mm_castsi128_ps( _mm_packus_epi16( _mm_packs_epi32( _mm_cvtps_epi32( vsum ), vzero ), vzero ) ) );
+#else
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum0 + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum1 + 0.5f ) ) );
+  dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum2 + 0.5f ) ) );
+  dst[3] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum3 + 0.5f ) ) );
+#endif
+
+  return;
+}
+
+
+#if CPU_SSE2_SUPPORT
+
+static void imStaticKernel4Linear_Core( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state )
+{
+  int x, y, mapx, mapy;
+  __m128 vsum, vf, v0, v1, v2, v3;
+  __m128i vzero;
+  float *matrix;
+  unsigned char *src;
+
+  vsum = _mm_setzero_ps();
+  vzero = _mm_setzero_si128();
+  matrix = state->matrix;
+  mapy = pointy;
+  for( y = 0 ; y < state->matrixsize ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = pointx << 2;
+    for( x = 0 ; x < state->matrixsize ; x += 4 )
+    {
+      vf = _mm_load_ps( &matrix[x] );
+      v0 = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_load_ss( (void *)&src[ mapx + 0 ] ) ), vzero ) );
+      v1 = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_load_ss( (void *)&src[ mapx + 4 ] ) ), vzero ) );
+      v2 = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_load_ss( (void *)&src[ mapx + 8 ] ) ), vzero ) );
+      v3 = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_load_ss( (void *)&src[ mapx + 12 ] ) ), vzero ) );
+      vsum = _mm_add_ps( vsum, _mm_mul_ps( _mm_shuffle_ps( vf, vf, 0x00 ), v0 ) );
+      vsum = _mm_add_ps( vsum, _mm_mul_ps( _mm_shuffle_ps( vf, vf, 0x55 ), v1 ) );
+      vsum = _mm_add_ps( vsum, _mm_mul_ps( _mm_shuffle_ps( vf, vf, 0xaa ), v2 ) );
+      vsum = _mm_add_ps( vsum, _mm_mul_ps( _mm_shuffle_ps( vf, vf, 0xff ), v3 ) );
+      mapx += 16;
+    }
+    matrix = ADDRESS( matrix, state->matrixrowsize );
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+  _mm_store_ss( (void *)dst, _mm_castsi128_ps( _mm_packus_epi16( _mm_packs_epi32( _mm_cvtps_epi32( vsum ), vzero ), vzero ) ) );
+
+  return;
+}
+
+#endif
+
+
+////
+
+
+static void imStaticKernel4LinearAlphaNorm( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state )
+{
+  int x, y, mapx, mapy;
+  float f, sum0, sum1, sum2, sum3;
+  float *matrix;
+  unsigned char *src;
+
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  sum3 = 0.0f;
+  matrix = state->matrix;
+  mapy = pointy;
+  for( y = 0 ; y < state->matrixsize ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = pointx << 2;
+    for( x = 0 ; x < state->matrixsize ; x++ )
+    {
+      f = matrix[x] * (float)src[ mapx + 3 ];
+      sum0 += f * (float)src[ mapx + 0 ];
+      sum1 += f * (float)src[ mapx + 1 ];
+      sum2 += f * (float)src[ mapx + 2 ];
+      sum3 += f;
+      mapx += 4;
+      if( mapx >= state->width4 )
+        mapx = 0;
+    }
+    matrix = ADDRESS( matrix, state->matrixrowsize );
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+  if( sum3 >= state->minimumalphaf )
+  {
+    f = 1.0f / sum3;
+    dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, ( sum0 * f ) + 0.5f ) ) );
+    dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, ( sum1 * f ) + 0.5f ) ) );
+    dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, ( sum2 * f ) + 0.5f ) ) );
+    dst[3] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum3 + 0.5f ) ) );
+  }
+  else
+  {
+    dst[0] = 0;
+    dst[1] = 0;
+    dst[2] = 0;
+    dst[3] = 0;
+  }
+
+  return;
+}
+
+
+#if CPU_SSE2_SUPPORT
+
+static void imStaticKernel4LinearAlphaNorm_Core( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state )
+{
+  int x, y, mapx, mapy;
+  uint32_t pixel;
+  float *matrix;
+  unsigned char *src;
+  __m128 vsum0, vsum1, vsum2, vsum3;
+  __m128 vf, valpha, vr, vg, vb, va, vsrcf;
+  __m128i vsrc, vshufmask;
+  __m128i vzero;
+
+ #if CPU_SSSE3_SUPPORT
+  vshufmask = _mm_setr_epi8( 0x00,0x04,0x08,0x0c, 0x01,0x05,0x09,0x0d, 0x02,0x06,0x0a,0x0e, 0x03,0x07,0x0b,0x0f );
+ #endif
+  vsum0 = _mm_setzero_ps();
+  vsum1 = _mm_setzero_ps();
+  vsum2 = _mm_setzero_ps();
+  vsum3 = _mm_setzero_ps();
+  vzero = _mm_castps_si128( _mm_setzero_ps() );
+
+  matrix = state->matrix;
+  mapy = pointy;
+  for( y = 0 ; y < state->matrixsize ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = pointx << 2;
+    for( x = 0 ; x < state->matrixsize ; x += 4 )
+    {
+      vf = _mm_load_ps( &matrix[x] );
+      /* Load 16 bytes and unpack as RRRR,GGGG,BBBB,AAAA in one SSE register */
+      vsrc = _mm_loadu_si128( (void *)&src[ mapx ] );
+ #if CPU_SSSE3_SUPPORT
+      vsrc = _mm_shuffle_epi8( vsrc, vshufmask );
+ #else
+      vshufmask = _mm_shuffle_epi32( vsrc, 0x39 );
+      vsrc = _mm_unpacklo_epi16( _mm_unpacklo_epi8( vsrc, vshufmask ), _mm_unpackhi_epi8( vsrc, vshufmask ) );
+ #endif
+      /* Break that into 4 SSE registers as floats: vR,vG,vB,vA */
+      vsrcf = _mm_castsi128_ps( vsrc );
+      vr = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( vsrcf ), vzero ) );
+ #if CPU_SSE3_SUPPORT
+      vg = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_movehdup_ps( vsrcf ) ), vzero ) );
+ #else
+      vg = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_shuffle_ps( vsrcf, vsrcf, 0x55 ) ), vzero ) );
+ #endif
+      vb = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_movehl_ps( vsrcf, vsrcf ) ), vzero ) );
+      va = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_shuffle_ps( vsrcf, vsrcf, 0xff ) ), vzero ) );
+      valpha = _mm_mul_ps( va, vf );
+      vsum0 = _mm_add_ps( vsum0, _mm_mul_ps( vr, valpha ) );
+      vsum1 = _mm_add_ps( vsum1, _mm_mul_ps( vg, valpha ) );
+      vsum2 = _mm_add_ps( vsum2, _mm_mul_ps( vb, valpha ) );
+      vsum3 = _mm_add_ps( vsum3, valpha );
+      mapx += 16;
+    }
+    matrix = ADDRESS( matrix, state->matrixrowsize );
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+ #if CPU_SSE3_SUPPORT
+  vsum0 = _mm_hadd_ps( vsum0, vsum1 );
+  vsum2 = _mm_hadd_ps( vsum2, vsum3 );
+  vsum0 = _mm_hadd_ps( vsum0, vsum2 );
+ #else
+  vsum0 = _mm_add_ps( _mm_unpacklo_ps( vsum0, vsum2 ), _mm_unpackhi_ps( vsum0, vsum2 ) );
+  vsum1 = _mm_add_ps( _mm_unpacklo_ps( vsum1, vsum3 ), _mm_unpackhi_ps( vsum1, vsum3 ) );
+  vsum0 = _mm_add_ps( _mm_unpacklo_ps( vsum0, vsum1 ), _mm_unpackhi_ps( vsum0, vsum1 ) );
+ #endif
+
+  valpha = _mm_shuffle_ps( vsum0, vsum0, 0xff );
+  pixel = 0;
+  if( _mm_comige_ss( valpha, _mm_load_ss( &state->minimumalphaf ) ) )
+  {
+    __m128i vpixel;
+    vsum0 = _mm_mul_ps( vsum0, _mm_rcp_ps( valpha ) );
+    vsum0 = CPU_BLENDV_PS( vsum0, valpha, *(__m128 *)simd4fAlphaMask );
+    vpixel = _mm_cvtps_epi32( vsum0 );
+    vpixel = _mm_packs_epi32( vpixel, vpixel );
+    vpixel = _mm_packus_epi16( vpixel, vpixel );
+    pixel = (uint32_t)_mm_cvtsi128_si32( vpixel );
+  }
+  *(uint32_t *)dst = pixel;
+
+  return;
+}
+
+#endif
+
+
+////
+
+
+static void imStaticKernel1sRGB( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state )
+{
+  int x, y, mapx, mapy;
+#if CPU_SSE2_SUPPORT
+  __m128 vsum, vsrc;
+  __m128i vzero;
+#else
+  float f, sum0;
+#endif
+  float *matrix;
+  unsigned char *src;
+
+#if CPU_SSE2_SUPPORT
+  vsum = _mm_setzero_ps();
+  vzero = _mm_setzero_si128();
+#else
+  sum0 = 0.0f;
+#endif
+  matrix = state->matrix;
+  mapy = pointy;
+  for( y = 0 ; y < state->matrixsize ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = pointx;
+    for( x = 0 ; x < state->matrixsize ; x++ )
+    {
+#if CPU_SSE2_SUPPORT
+      vsrc = _mm_set_ss( (float)src[ mapx + 0 ] );
+      vsrc = srgb2linear3( vsrc );
+      vsum = _mm_add_ps( vsum, _mm_mul_ps( _mm_set_ss( matrix[x] ), vsrc ) );
+#else
+      f = matrix[x];
+      sum0 += f * srgb2linear( (float)src[ mapx + 0 ] );
+#endif
+      mapx++;
+      if( mapx >= state->width1 )
+        mapx = 0;
+    }
+    matrix = ADDRESS( matrix, state->matrixrowsize );
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+#if CPU_SSE2_SUPPORT
+  dst[0] = _mm_cvtsi128_si32( _mm_packus_epi16( _mm_packs_epi32( _mm_cvtps_epi32( linear2srgb3( vsum ) ), vzero ), vzero ) );
+#else
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum0 ) + 0.5f ) ) );
+#endif
+
+  return;
+}
+
+static void imStaticKernel2sRGB( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state )
+{
+  int x, y, mapx, mapy;
+#if CPU_SSE2_SUPPORT
+  __m128 vsum, vsrc;
+  __m128i vzero;
+#else
+  float f, sum0, sum1;
+#endif
+  float *matrix;
+  unsigned char *src;
+
+#if CPU_SSE2_SUPPORT
+  vsum = _mm_setzero_ps();
+  vzero = _mm_setzero_si128();
+#else
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+#endif
+  matrix = state->matrix;
+  mapy = pointy;
+  for( y = 0 ; y < state->matrixsize ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = pointx << 1;
+    for( x = 0 ; x < state->matrixsize ; x++ )
+    {
+#if CPU_SSE2_SUPPORT
+      vsrc = _mm_set_ps( 0.0f, 0.0f, (float)src[ mapx + 1 ], (float)src[ mapx + 0 ] );
+      vsrc = srgb2linear3( vsrc );
+      vsum = _mm_add_ps( vsum, _mm_mul_ps( _mm_set1_ps( matrix[x] ), vsrc ) );
+#else
+      f = matrix[x];
+      sum0 += f * srgb2linear( (float)src[ mapx + 0 ] );
+      sum1 += f * srgb2linear( (float)src[ mapx + 1 ] );
+#endif
+      mapx += 2;
+      if( mapx >= state->width2 )
+        mapx = 0;
+    }
+    matrix = ADDRESS( matrix, state->matrixrowsize );
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+#if CPU_SSE2_SUPPORT
+  union
+  {
+    char c[4];
+    uint32_t i;
+  } u;
+  vsum = linear2srgb3( vsum );
+  _mm_store_ss( (void *)&u.i, _mm_castsi128_ps( _mm_packus_epi16( _mm_packs_epi32( _mm_cvtps_epi32( vsum ), vzero ), vzero ) ) );
+  dst[0] = u.c[0];
+  dst[1] = u.c[1];
+#else
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum0 ) + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum1 ) + 0.5f ) ) );
+#endif
+
+  return;
+}
+
+static void imStaticKernel3sRGB( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state )
+{
+  int x, y, mapx, mapy;
+#if CPU_SSE2_SUPPORT
+  __m128 vsum, vsrc;
+  __m128i vzero;
+#else
+  float f, sum0, sum1, sum2;
+#endif
+  float *matrix;
+  unsigned char *src;
+
+#if CPU_SSE2_SUPPORT
+  vsum = _mm_setzero_ps();
+  vzero = _mm_setzero_si128();
+#else
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+#endif
+  matrix = state->matrix;
+  mapy = pointy;
+  for( y = 0 ; y < state->matrixsize ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = pointx + ( pointx << 1 );
+    for( x = 0 ; x < state->matrixsize ; x++ )
+    {
+#if CPU_SSE2_SUPPORT
+      vsrc = _mm_set_ps( 0.0f, (float)src[ mapx + 2 ], (float)src[ mapx + 1 ], (float)src[ mapx + 0 ] );
+      vsrc = srgb2linear3( vsrc );
+      vsum = _mm_add_ps( vsum, _mm_mul_ps( _mm_set1_ps( matrix[x] ), vsrc ) );
+#else
+      f = matrix[x];
+      sum0 += f * srgb2linear( (float)src[ mapx + 0 ] );
+      sum1 += f * srgb2linear( (float)src[ mapx + 1 ] );
+      sum2 += f * srgb2linear( (float)src[ mapx + 2 ] );
+#endif
+      mapx += 3;
+      if( mapx >= state->width3 )
+        mapx = 0;
+    }
+    matrix = ADDRESS( matrix, state->matrixrowsize );
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+#if CPU_SSE2_SUPPORT
+  union
+  {
+    char c[4];
+    uint32_t i;
+  } u;
+  vsum = linear2srgb3( vsum );
+  _mm_store_ss( (void *)&u.i, _mm_castsi128_ps( _mm_packus_epi16( _mm_packs_epi32( _mm_cvtps_epi32( vsum ), vzero ), vzero ) ) );
+  dst[0] = u.c[0];
+  dst[1] = u.c[1];
+  dst[2] = u.c[2];
+#else
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum0 ) + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum1 ) + 0.5f ) ) );
+  dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum2 ) + 0.5f ) ) );
+#endif
+
+  return;
+}
+
+static void imStaticKernel4sRGB( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state )
+{
+  int x, y, mapx, mapy;
+#if CPU_SSE2_SUPPORT
+  __m128 vsum, vsrc;
+  __m128i vzero;
+#else
+  float f, sum0, sum1, sum2, sum3;
+#endif
+  float *matrix;
+  unsigned char *src;
+
+#if CPU_SSE2_SUPPORT
+  vsum = _mm_setzero_ps();
+  vzero = _mm_setzero_si128();
+#else
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  sum3 = 0.0f;
+#endif
+  matrix = state->matrix;
+  mapy = pointy;
+  for( y = 0 ; y < state->matrixsize ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = pointx << 2;
+    for( x = 0 ; x < state->matrixsize ; x++ )
+    {
+#if CPU_SSE2_SUPPORT
+      vsrc = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_load_ss( (void *)&src[ mapx ] ) ), vzero ) );
+      vsrc = srgb2linear3( vsrc );
+      vsum = _mm_add_ps( vsum, _mm_mul_ps( _mm_set1_ps( matrix[x] ), vsrc ) );
+#else
+      f = matrix[x];
+      sum0 += f * srgb2linear( (float)src[ mapx + 0 ] );
+      sum1 += f * srgb2linear( (float)src[ mapx + 1 ] );
+      sum2 += f * srgb2linear( (float)src[ mapx + 2 ] );
+      sum3 += f * (float)src[ mapx + 3 ];
+#endif
+      mapx += 4;
+      if( mapx >= state->width4 )
+        mapx = 0;
+    }
+    matrix = ADDRESS( matrix, state->matrixrowsize );
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+#if CPU_SSE2_SUPPORT
+  vsum = linear2srgb3( vsum );
+  _mm_store_ss( (void *)dst, _mm_castsi128_ps( _mm_packus_epi16( _mm_packs_epi32( _mm_cvtps_epi32( vsum ), vzero ), vzero ) ) );
+#else
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum0 ) + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum1 ) + 0.5f ) ) );
+  dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum2 ) + 0.5f ) ) );
+  dst[3] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum3 + 0.5f ) ) );
+#endif
+
+  return;
+}
+
+
+#if CPU_SSE2_SUPPORT
+
+static void imStaticKernel3sRGB_Core( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state )
+{
+  int x, y, mapx, mapy;
+  __m128 vsum0, vsum1, vsum2, vsrc0, vsrc1, vsrc2, vf;
+  __m128i vzero;
+  float *matrix;
+  unsigned char *src;
+  union
+  {
+    char c[4];
+    uint32_t i;
+  } u;
+
+  vsum0 = _mm_setzero_ps();
+  vsum1 = _mm_setzero_ps();
+  vsum2 = _mm_setzero_ps();
+  vzero = _mm_setzero_si128();
+  matrix = state->matrix;
+  mapy = pointy;
+  for( y = 0 ; y < state->matrixsize ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = pointx + ( pointx << 1 );
+    for( x = 0 ; x < state->matrixsize ; x += 4 )
+    {
+      vsrc0 = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_load_ss( (void *)&src[ mapx+0 ] ) ), vzero ) );
+      vsrc1 = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_load_ss( (void *)&src[ mapx+4 ] ) ), vzero ) );
+      vsrc2 = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_load_ss( (void *)&src[ mapx+8 ] ) ), vzero ) );
+      vsrc0 = srgb2linear4( vsrc0 );
+      vsrc1 = srgb2linear4( vsrc1 );
+      vsrc2 = srgb2linear4( vsrc2 );
+      vf = _mm_load_ps( &matrix[x] );
+      vsum0 = _mm_add_ps( vsum0, _mm_mul_ps( _mm_shuffle_ps( vf, vf, 0x40 ), vsrc0 ) );
+      vsum1 = _mm_add_ps( vsum1, _mm_mul_ps( _mm_shuffle_ps( vf, vf, 0xA5 ), vsrc1 ) );
+      vsum2 = _mm_add_ps( vsum2, _mm_mul_ps( _mm_shuffle_ps( vf, vf, 0xFE ), vsrc2 ) );
+      mapx += 12;
+    }
+    matrix = ADDRESS( matrix, state->matrixrowsize );
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+ #if CPU_SSSE3_SUPPORT
+  vsum0 = _mm_add_ps( vsum0, _mm_castsi128_ps( _mm_alignr_epi8( _mm_castps_si128( vsum1 ), _mm_castps_si128( vsum0 ), 12 ) ) );
+  vsum0 = _mm_add_ps( vsum0, _mm_castsi128_ps( _mm_alignr_epi8( _mm_castps_si128( vsum2 ), _mm_castps_si128( vsum1 ), 8 ) ) );
+  vsum0 = _mm_add_ps( vsum0, _mm_castsi128_ps( _mm_alignr_epi8( _mm_castps_si128( vsum2 ), _mm_castps_si128( vsum2 ), 4 ) ) );
+ #else
+  vf = _mm_shuffle_ps( vsum0, vsum1, 0x4f );
+  vsum0 = _mm_add_ps( vsum0, _mm_shuffle_ps( vf, vf, 0x38 ) );
+  vsum0 = _mm_add_ps( vsum0, _mm_shuffle_ps( vsum1, vsum2, 0x0E ) );
+  vsum0 = _mm_add_ps( vsum0, _mm_shuffle_ps( vsum2, vsum2, 0x39 ) );
+ #endif
+
+  vsum0 = linear2srgb3( vsum0 );
+  _mm_store_ss( (void *)&u.i, _mm_castsi128_ps( _mm_packus_epi16( _mm_packs_epi32( _mm_cvtps_epi32( vsum0 ), vzero ), vzero ) ) );
+  dst[0] = u.c[0];
+  dst[1] = u.c[1];
+  dst[2] = u.c[2];
+
+  return;
+}
+
+static void imStaticKernel4sRGB_Core( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state )
+{
+  int x, y, mapx, mapy;
+  __m128 vsum, vsrc0, vsrc1;
+  __m128i vzero;
+  float *matrix;
+  unsigned char *src;
+
+  vsum = _mm_setzero_ps();
+  vzero = _mm_setzero_si128();
+  matrix = state->matrix;
+  mapy = pointy;
+  for( y = 0 ; y < state->matrixsize ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = pointx << 2;
+    for( x = 0 ; x < state->matrixsize ; x += 2 )
+    {
+      vsrc0 = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_load_ss( (void *)&src[ mapx+0 ] ) ), vzero ) );
+      vsrc1 = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_load_ss( (void *)&src[ mapx+4 ] ) ), vzero ) );
+      vsrc0 = srgb2linear3( vsrc0 );
+      vsrc1 = srgb2linear3( vsrc1 );
+      vsum = _mm_add_ps( vsum, _mm_mul_ps( _mm_set1_ps( matrix[x+0] ), vsrc0 ) );
+      vsum = _mm_add_ps( vsum, _mm_mul_ps( _mm_set1_ps( matrix[x+1] ), vsrc1 ) );
+      mapx += 8;
+    }
+    matrix = ADDRESS( matrix, state->matrixrowsize );
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+  vsum = linear2srgb3( vsum );
+  _mm_store_ss( (void *)dst, _mm_castsi128_ps( _mm_packus_epi16( _mm_packs_epi32( _mm_cvtps_epi32( vsum ), vzero ), vzero ) ) );
+
+  return;
+}
+
+#endif
+
+
+////
+
+
+static void imStaticKernel4sRGBAlphaNorm( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state )
+{
+  int x, y, mapx, mapy;
+#if CPU_SSE2_SUPPORT
+  __m128 vsum, vsrc, valpha;
+  __m128i vzero;
+  uint32_t pixel;
+#else
+  float f, sum0, sum1, sum2, sum3;
+#endif
+  float *matrix;
+  unsigned char *src;
+
+#if CPU_SSE2_SUPPORT
+  vsum = _mm_setzero_ps();
+  vzero = _mm_setzero_si128();
+#else
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  sum3 = 0.0f;
+#endif
+  matrix = state->matrix;
+  mapy = pointy;
+  for( y = 0 ; y < state->matrixsize ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = pointx << 2;
+    for( x = 0 ; x < state->matrixsize ; x++ )
+    {
+#if CPU_SSE2_SUPPORT
+      vsrc = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_load_ss( (void *)&src[ mapx ] ) ), vzero ) );
+      valpha = _mm_shuffle_ps( vsrc, _mm_set_ss( 1.0f ), 0x0f );
+      vsrc = srgb2linear3( vsrc );
+      vsum = _mm_add_ps( vsum, _mm_mul_ps( _mm_mul_ps( _mm_shuffle_ps( valpha, valpha, 0xC0 ), _mm_set1_ps( matrix[x] ) ), vsrc ) );
+#else
+      f = matrix[x] * (float)src[ mapx + 3 ];
+      sum0 += f * srgb2linear( (float)src[ mapx + 0 ] );
+      sum1 += f * srgb2linear( (float)src[ mapx + 1 ] );
+      sum2 += f * srgb2linear( (float)src[ mapx + 2 ] );
+      sum3 += f;
+#endif
+      mapx += 4;
+      if( mapx >= state->width4 )
+        mapx = 0;
+    }
+    matrix = ADDRESS( matrix, state->matrixrowsize );
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+#if CPU_SSE2_SUPPORT
+  valpha = _mm_shuffle_ps( vsum, vsum, 0xff );
+  pixel = 0;
+  if( _mm_comige_ss( valpha, _mm_load_ss( &state->minimumalphaf ) ) )
+  {
+    __m128i vpixel;
+    vsum = _mm_mul_ps( vsum, _mm_rcp_ps( valpha ) );
+    vsum = CPU_BLENDV_PS( vsum, valpha, *(__m128 *)simd4fAlphaMask );
+    vsum = linear2srgb3( vsum );
+    vpixel = _mm_cvtps_epi32( vsum );
+    vpixel = _mm_packs_epi32( vpixel, vpixel );
+    vpixel = _mm_packus_epi16( vpixel, vpixel );
+    pixel = (uint32_t)_mm_cvtsi128_si32( vpixel );
+  }
+  *(uint32_t *)dst = pixel;
+#else
+  if( sum3 >= state->minimumalphaf )
+  {
+    f = 1.0f / sum3;
+    dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum0 * f ) + 0.5f ) ) );
+    dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum1 * f ) + 0.5f ) ) );
+    dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum2 * f ) + 0.5f ) ) );
+    dst[3] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum3 + 0.5f ) ) );
+  }
+  else
+  {
+    dst[0] = 0;
+    dst[1] = 0;
+    dst[2] = 0;
+    dst[3] = 0;
+  }
+#endif
+
+  return;
+}
+
+
+#if CPU_SSE2_SUPPORT
+
+static void imStaticKernel4sRGBAlphaNorm_Core( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state )
+{
+  int x, y, mapx, mapy;
+  __m128 vsum, vsrc0, vsrc1, valpha0, valpha1;
+  __m128i vzero;
+  uint32_t pixel;
+  float *matrix;
+  unsigned char *src;
+
+  vsum = _mm_setzero_ps();
+  vzero = _mm_setzero_si128();
+  matrix = state->matrix;
+  mapy = pointy;
+  for( y = 0 ; y < state->matrixsize ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = pointx << 2;
+    for( x = 0 ; x < state->matrixsize ; x += 2 )
+    {
+      vsrc0 = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_load_ss( (void *)&src[ mapx+0 ] ) ), vzero ) );
+      vsrc1 = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_load_ss( (void *)&src[ mapx+4 ] ) ), vzero ) );
+      valpha0 = _mm_shuffle_ps( vsrc0, _mm_set_ss( 1.0f ), 0x0f );
+      valpha1 = _mm_shuffle_ps( vsrc1, _mm_set_ss( 1.0f ), 0x0f );
+      vsrc0 = srgb2linear3( vsrc0 );
+      vsrc1 = srgb2linear3( vsrc1 );
+      vsum = _mm_add_ps( vsum, _mm_mul_ps( _mm_mul_ps( _mm_shuffle_ps( valpha0, valpha0, 0xC0 ), _mm_set1_ps( matrix[x+0] ) ), vsrc0 ) );
+      vsum = _mm_add_ps( vsum, _mm_mul_ps( _mm_mul_ps( _mm_shuffle_ps( valpha1, valpha1, 0xC0 ), _mm_set1_ps( matrix[x+1] ) ), vsrc1 ) );
+      mapx += 8;
+    }
+    matrix = ADDRESS( matrix, state->matrixrowsize );
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+  valpha0 = _mm_shuffle_ps( vsum, vsum, 0xff );
+  pixel = 0;
+  if( _mm_comige_ss( valpha0, _mm_load_ss( &state->minimumalphaf ) ) )
+  {
+    __m128i vpixel;
+    vsum = _mm_mul_ps( vsum, _mm_rcp_ps( valpha0 ) );
+    vsum = CPU_BLENDV_PS( vsum, valpha0, *(__m128 *)simd4fAlphaMask );
+    vsum = linear2srgb3( vsum );
+    vpixel = _mm_cvtps_epi32( vsum );
+    vpixel = _mm_packs_epi32( vpixel, vpixel );
+    vpixel = _mm_packus_epi16( vpixel, vpixel );
+    pixel = (uint32_t)_mm_cvtsi128_si32( vpixel );
+  }
+  *(uint32_t *)dst = pixel;
+
+  return;
+}
+
+#endif
+
+////
+
+
+static void imStaticKernel3Normal( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state )
+{
+  int x, y, mapx, mapy;
+  float f, sum0, sum1, sum2, suminv;
+  float *matrix;
+  unsigned char *src;
+
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  matrix = state->matrix;
+  mapy = pointy;
+  for( y = 0 ; y < state->matrixsize ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = pointx + ( pointx << 1 );
+    for( x = 0 ; x < state->matrixsize ; x++ )
+    {
+      f = matrix[x];
+      sum0 += f * (float)src[ mapx + 0 ];
+      sum1 += f * (float)src[ mapx + 1 ];
+      sum2 += f * (float)src[ mapx + 2 ];
+      mapx += 3;
+      if( mapx >= state->width3 )
+        mapx = 0;
+    }
+    matrix = ADDRESS( matrix, state->matrixrowsize );
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+  sum0 -= 0.5f*255.0f;
+  sum1 -= 0.5f*255.0f;
+  sum2 -= 0.5f*255.0f;
+  sum0 *= state->amplifynormal;
+  sum1 *= state->amplifynormal;
+  suminv = (0.5f*255.0f) / sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) + ( sum2 * sum2 ) );
+  sum0 = (0.5f*255.0f) + ( sum0 * suminv );
+  sum1 = (0.5f*255.0f) + ( sum1 * suminv );
+  sum2 = (0.5f*255.0f) + ( sum2 * suminv );
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum0 + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum1 + 0.5f ) ) );
+  dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum2 + 0.5f ) ) );
+
+  return;
+}
+
+static void imStaticKernel4Normal( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state )
+{
+  int x, y, mapx, mapy;
+#if CPU_SSE2_SUPPORT
+  __m128 vsum, vsrc;
+  __m128i vzero;
+#else
+  float f;
+#endif
+  float sum0, sum1, sum2, sum3, suminv;
+  float *matrix;
+  unsigned char *src;
+
+#if CPU_SSE2_SUPPORT
+  vsum = _mm_setzero_ps();
+  vzero = _mm_setzero_si128();
+#else
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  sum3 = 0.0f;
+#endif
+  matrix = state->matrix;
+  mapy = pointy;
+  for( y = 0 ; y < state->matrixsize ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = pointx << 2;
+    for( x = 0 ; x < state->matrixsize ; x++ )
+    {
+#if CPU_SSE2_SUPPORT
+      vsrc = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_load_ss( (void *)&src[ mapx ] ) ), vzero ) );
+      vsum = _mm_add_ps( vsum, _mm_mul_ps( _mm_set1_ps( matrix[x] ), vsrc ) );
+#else
+      f = matrix[x];
+      sum0 += f * (float)src[ mapx + 0 ];
+      sum1 += f * (float)src[ mapx + 1 ];
+      sum2 += f * (float)src[ mapx + 2 ];
+      sum3 += f * (float)src[ mapx + 3 ];
+#endif
+      mapx += 4;
+      if( mapx >= state->width4 )
+        mapx = 0;
+    }
+    matrix = ADDRESS( matrix, state->matrixrowsize );
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+#if CPU_SSE2_SUPPORT
+  vsum = _mm_sub_ps( vsum, _mm_set_ps( 0.0f, 0.5f*255.0f, 0.5f*255.0f, 0.5f*255.0f ) );
+  sum0 = _mm_cvtss_f32( vsum );
+ #if CPU_SSE3_SUPPORT
+  sum1 = _mm_cvtss_f32( _mm_movehdup_ps( vsum ) );
+ #else
+  sum1 = _mm_cvtss_f32( _mm_shuffle_ps( vsum, vsum, 0x55 ) );
+ #endif
+  sum2 = _mm_cvtss_f32( _mm_movehl_ps( vsum, vsum ) );
+  sum3 = _mm_cvtss_f32( _mm_shuffle_ps( vsum, vsum, 0xff ) );
+#else
+  sum0 -= 0.5f*255.0f;
+  sum1 -= 0.5f*255.0f;
+  sum2 -= 0.5f*255.0f;
+#endif
+  sum0 *= state->amplifynormal;
+  sum1 *= state->amplifynormal;
+  suminv = (0.5f*255.0f) / sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) + ( sum2 * sum2 ) );
+  sum0 = (0.5f*255.0f) + ( sum0 * suminv );
+  sum1 = (0.5f*255.0f) + ( sum1 * suminv );
+  sum2 = (0.5f*255.0f) + ( sum2 * suminv );
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum0 + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum1 + 0.5f ) ) );
+  dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum2 + 0.5f ) ) );
+  dst[3] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum3 + 0.5f ) ) );
+
+  return;
+}
+
+
+////
+
+
+static void imStaticKernel3NormalSustain( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state )
+{
+  int x, y, mapx, mapy;
+  float f, v0, v1, v2, energy, sum0, sum1, sum2, sumenergy, suminv;
+  float *matrix;
+  unsigned char *src;
+
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  sumenergy = 0.0f;
+  matrix = state->matrix;
+  mapy = pointy;
+  for( y = 0 ; y < state->matrixsize ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = pointx + ( pointx << 1 );
+    for( x = 0 ; x < state->matrixsize ; x++ )
+    {
+      f = matrix[x];
+      v0 = f * ( (float)src[ mapx + 0 ] - 127.5f );
+      v1 = f * ( (float)src[ mapx + 1 ] - 127.5f );
+      v2 = f * ( (float)src[ mapx + 2 ] - 127.5f );
+      sum0 += v0;
+      sum1 += v1;
+      sum2 += v2;
+      energy = ( v0 * v0 ) + ( v1 * v1 );
+      if( energy )
+        sumenergy += sqrtf( energy ) / sqrtf( energy + ( v2 * v2 ) );
+      mapx += 3;
+      if( mapx >= state->width3 )
+        mapx = 0;
+    }
+    matrix = ADDRESS( matrix, state->matrixrowsize );
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+  sum0 *= state->amplifynormal;
+  sum1 *= state->amplifynormal;
+  suminv = (0.5f*255.0f) / fmaxf( 0.0625f, sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) + ( sum2 * sum2 ) ) );
+  sum0 *= suminv;
+  sum1 *= suminv;
+  sum2 *= suminv;
+  energy = sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) );
+  sumenergy *= state->normalsustainfactor;
+  if( energy < sumenergy )
+  {
+    f = fminf( sumenergy / energy, 8.0f );
+    sum0 *= f;
+    sum1 *= f;
+    suminv = (0.5f*255.0f) / fmaxf( 0.0625f, sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) + ( sum2 * sum2 ) ) );
+    sum0 *= suminv;
+    sum1 *= suminv;
+    sum2 *= suminv;
+  }
+  sum0 += (0.5f*255.0f);
+  sum1 += (0.5f*255.0f);
+  sum2 += (0.5f*255.0f);
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum0 + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum1 + 0.5f ) ) );
+  dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum2 + 0.5f ) ) );
+
+  return;
+}
+
+static void imStaticKernel4NormalSustain( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state )
+{
+  int x, y, mapx, mapy;
+  float f, v0, v1, v2, v3, energy, sum0, sum1, sum2, sum3, sumenergy, suminv;
+  float *matrix;
+  unsigned char *src;
+
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  sum3 = 0.0f;
+  sumenergy = 0.0f;
+  matrix = state->matrix;
+  mapy = pointy;
+  for( y = 0 ; y < state->matrixsize ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = pointx << 2;
+    for( x = 0 ; x < state->matrixsize ; x++ )
+    {
+      f = matrix[x];
+      v0 = f * ( (float)src[ mapx + 0 ] - 127.5f );
+      v1 = f * ( (float)src[ mapx + 1 ] - 127.5f );
+      v2 = f * ( (float)src[ mapx + 2 ] - 127.5f );
+      v3 = f * (float)src[ mapx + 3 ];
+      sum0 += v0;
+      sum1 += v1;
+      sum2 += v2;
+      sum3 += v3;
+      energy = ( v0 * v0 ) + ( v1 * v1 );
+      if( energy )
+        sumenergy += sqrtf( energy ) / sqrtf( energy + ( v2 * v2 ) );
+      mapx += 4;
+      if( mapx >= state->width4 )
+        mapx = 0;
+    }
+    matrix = ADDRESS( matrix, state->matrixrowsize );
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+  sum0 *= state->amplifynormal;
+  sum1 *= state->amplifynormal;
+  suminv = (0.5f*255.0f) / fmaxf( 0.0625f, sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) + ( sum2 * sum2 ) ) );
+  sum0 *= suminv;
+  sum1 *= suminv;
+  sum2 *= suminv;
+  energy = sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) );
+  sumenergy *= state->normalsustainfactor;
+  if( energy < sumenergy )
+  {
+    f = fminf( sumenergy / energy, 8.0f );
+    sum0 *= f;
+    sum1 *= f;
+    suminv = (0.5f*255.0f) / fmaxf( 0.0625f, sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) + ( sum2 * sum2 ) ) );
+    sum0 *= suminv;
+    sum1 *= suminv;
+    sum2 *= suminv;
+  }
+  sum0 += (0.5f*255.0f);
+  sum1 += (0.5f*255.0f);
+  sum2 += (0.5f*255.0f);
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum0 + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum1 + 0.5f ) ) );
+  dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum2 + 0.5f ) ) );
+  dst[3] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum3 + 0.5f ) ) );
+
+  return;
+}
+
+
+////
+
+
+static void imStaticKernel4NormalSustainAlphaNorm( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state )
+{
+  int x, y, mapx, mapy;
+  float f, v0, v1, v2, v3, energy, sum0, sum1, sum2, sum3, sumenergy, suminv;
+  float *matrix;
+  unsigned char *src;
+
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  sum3 = 0.0f;
+  sumenergy = 0.0f;
+  matrix = state->matrix;
+  mapy = pointy;
+  for( y = 0 ; y < state->matrixsize ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = pointx << 2;
+    for( x = 0 ; x < state->matrixsize ; x++ )
+    {
+      f = matrix[x] * (float)src[ mapx + 3 ];
+      v0 = f * ( (float)src[ mapx + 0 ] - 127.5f );
+      v1 = f * ( (float)src[ mapx + 1 ] - 127.5f );
+      v2 = f * ( (float)src[ mapx + 2 ] - 127.5f );
+      v3 = f;
+      sum0 += v0;
+      sum1 += v1;
+      sum2 += v2;
+      sum3 += v3;
+      energy = ( v0 * v0 ) + ( v1 * v1 );
+      if( energy )
+        sumenergy += sqrtf( energy ) / sqrtf( energy + ( v2 * v2 ) );
+      mapx += 4;
+      if( mapx >= state->width4 )
+        mapx = 0;
+    }
+    matrix = ADDRESS( matrix, state->matrixrowsize );
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+  if( sum3 >= state->minimumalphaf )
+  {
+    f = 1.0f / sum3;
+    sum0 *= f;
+    sum1 *= f;
+    sum2 *= f;
+    sum0 *= state->amplifynormal;
+    sum1 *= state->amplifynormal;
+    suminv = (0.5f*255.0f) / fmaxf( 0.0625f, sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) + ( sum2 * sum2 ) ) );
+    sum0 *= suminv;
+    sum1 *= suminv;
+    sum2 *= suminv;
+    energy = sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) );
+    sumenergy *= state->normalsustainfactor;
+    if( energy < sumenergy )
+    {
+      f = fminf( sumenergy / energy, 8.0f );
+      sum0 *= f;
+      sum1 *= f;
+      suminv = (0.5f*255.0f) / fmaxf( 0.0625f, sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) + ( sum2 * sum2 ) ) );
+      sum0 *= suminv;
+      sum1 *= suminv;
+      sum2 *= suminv;
+    }
+    sum0 += (0.5f*255.0f);
+    sum1 += (0.5f*255.0f);
+    sum2 += (0.5f*255.0f);
+    dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum0 + 0.5f ) ) );
+    dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum1 + 0.5f ) ) );
+    dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum2 + 0.5f ) ) );
+    dst[3] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum3 + 0.5f ) ) );
+  }
+  else
+  {
+    dst[0] = 0;
+    dst[1] = 0;
+    dst[2] = 0;
+    dst[3] = 0;
+  }
+
+  return;
+}
+
+
+////
+
+
+static void imStaticKernelPoT3Water( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state )
+{
+  int x, y, mapx, mapy, heightmask, widthmask;
+  int minx, maxx, miny, maxy;
+  float f, sum0, sum1, sum2, suminv;
+  float *matrix;
+  unsigned char *src;
+
+  minx = pointx;
+  maxx = minx + state->matrixsize;
+  miny = pointy;
+  maxy = miny + state->matrixsize;
+  heightmask = state->height - 1;
+  widthmask = state->width1 - 1;
+
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  matrix = state->matrix;
+  for( y = miny ; y < maxy ; y++ )
+  {
+    mapy = y & heightmask;
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    for( x = minx ; x < maxx ; x++, matrix++ )
+    {
+      mapx = x & widthmask;
+      mapx += mapx << 1;
+      f = *matrix;
+      sum0 += (float)src[ mapx + 0 ] * f;
+      sum1 += (float)src[ mapx + 1 ] * f;
+      sum2 += (float)src[ mapx + 2 ] * f;
+    }
+    matrix += state->rowreturn;
+  }
+
+  sum0 *= 1.0f/255.0f;
+  sum1 *= 1.0f/255.0f;
+  sum2 *= 1.0f/255.0f;
+  sum0 = 2.0f * ( sum0 - 0.5f );
+  sum1 = 2.0f * ( sum1 - 0.5f );
+  suminv = sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) );
+  if( suminv < 0.75f )
+  {
+    suminv = 0.5f / suminv;
+    sum0 = 0.5f + ( sum0 * suminv );
+    sum1 = 0.5f + ( sum1 * suminv );
+  }
+  if( sum2 > 0.1f )
+  {
+    state->dithersum += sum2;
+    if( sum2 > 0.45f )
+      sum2 = 1.0f;
+    else if( ( sum2 < 0.3f ) && ( state->dithersum < 1.0f ) )
+      sum2 = 0.0f;
+    else
+      sum2 = ( ( sum2 + state->dithersum ) < 0.45f ? 0.0f : 1.0f );
+    state->dithersum -= sum2;
+  }
+  sum0 *= 255.0f;
+  sum1 *= 255.0f;
+  sum2 *= 255.0f;
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum0 + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum1 + 0.5f ) ) );
+  dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum2 + 0.5f ) ) );
+
+  return;
+}
+
+static void imStaticKernelPoT4Water( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state )
+{
+  int x, y, mapx, mapy, heightmask, widthmask;
+  int minx, maxx, miny, maxy;
+  float f, sum0, sum1, sum2, sum3, suminv;
+  float *matrix;
+  unsigned char *src;
+
+  minx = pointx;
+  maxx = minx + state->matrixsize;
+  miny = pointy;
+  maxy = miny + state->matrixsize;
+  heightmask = state->height - 1;
+  widthmask = state->width1 - 1;
+
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  sum3 = 0.0f;
+  matrix = state->matrix;
+  for( y = miny ; y < maxy ; y++ )
+  {
+    mapy = y & heightmask;
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    for( x = minx ; x < maxx ; x++, matrix++ )
+    {
+      mapx = x & widthmask;
+      mapx <<= 2;
+      f = *matrix;
+      sum0 += (float)src[ mapx + 0 ] * f;
+      sum1 += (float)src[ mapx + 1 ] * f;
+      sum2 += (float)src[ mapx + 2 ] * f;
+      sum3 += (float)src[ mapx + 3 ] * f;
+    }
+    matrix += state->rowreturn;
+  }
+
+  sum0 *= 1.0f/255.0f;
+  sum1 *= 1.0f/255.0f;
+  sum2 *= 1.0f/255.0f;
+  sum0 = 2.0f * ( sum0 - 0.5f );
+  sum1 = 2.0f * ( sum1 - 0.5f );
+  suminv = sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) );
+  if( suminv < 0.75f )
+  {
+    suminv = 0.5f / suminv;
+    sum0 = 0.5f + ( sum0 * suminv );
+    sum1 = 0.5f + ( sum1 * suminv );
+  }
+  if( sum2 > 0.1f )
+  {
+    state->dithersum += sum2;
+    if( sum2 > 0.45f )
+      sum2 = 1.0f;
+    else if( ( sum2 < 0.3f ) && ( state->dithersum < 1.0f ) )
+      sum2 = 0.0f;
+    else
+      sum2 = ( ( sum2 + state->dithersum ) < 0.45f ? 0.0f : 1.0f );
+    state->dithersum -= sum2;
+  }
+  sum0 *= 255.0f;
+  sum1 *= 255.0f;
+  sum2 *= 255.0f;
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum0 + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum1 + 0.5f ) ) );
+  dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum2 + 0.5f ) ) );
+  dst[3] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum3 + 0.5f ) ) );
+
+  return;
+}
+
+
+////
+
+
+static void imStaticKernelPoT4Plant( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state )
+{
+  int x, y, mapx, mapy, heightmask, widthmask;
+  int minx, maxx, miny, maxy;
+  float f, sum0, sum1, sum2, sum3;
+  float *matrix;
+  unsigned char *src;
+
+  minx = pointx;
+  maxx = minx + state->matrixsize;
+  miny = pointy;
+  maxy = miny + state->matrixsize;
+  heightmask = state->height - 1;
+  widthmask = state->width1 - 1;
+
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  sum3 = 0.0f;
+  matrix = state->matrix;
+  for( y = miny ; y < maxy ; y++ )
+  {
+    mapy = y & heightmask;
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    for( x = minx ; x < maxx ; x++, matrix++ )
+    {
+      mapx = x & widthmask;
+      mapx <<= 2;
+      f = *matrix;
+      sum0 += (float)src[ mapx + 0 ] * f;
+      sum1 += (float)src[ mapx + 1 ] * f;
+      sum2 += (float)src[ mapx + 2 ] * f;
+      sum3 += (float)src[ mapx + 3 ] * f;
+    }
+    matrix += state->rowreturn;
+  }
+
+  sum3 *= 1.25f;
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum0 + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum1 + 0.5f ) ) );
+  dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum2 + 0.5f ) ) );
+  dst[3] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum3 + 0.5f ) ) );
+
+  return;
+}
+
+
+////
+
+
+int imReduceImageKaiserDataDivisor( unsigned char *dstdata, unsigned char *srcdata, int width, int height, int bytesperpixel, int bytesperline, int sizedivisor, imReduceOptions *options )
+{
+  int filter, x, y, pointx, pointy, basex, basey, pow2flag;
+  int newwidth, newheight;
+  unsigned char *dst;
+  imStaticMatrixState state;
+  void (*applykernel)( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state );
+#if CPU_SSE2_SUPPORT
+  int corebase, corerange;
+  void (*applykernelcore)( unsigned char *dst, int pointx, int pointy, imStaticMatrixState * CC_RESTRICT state );
+#endif
+
+  filter = options->filter;
+  imBuildStaticMatrix( &state, sizedivisor, options->hopcount, options->alpha );
+
+  newwidth = ( width < sizedivisor ) ? 1 : ( ( width + sizedivisor - 1 ) / sizedivisor );
+  newheight = ( height < sizedivisor ) ? 1 : ( ( height + sizedivisor - 1 ) / sizedivisor );
+
+  pow2flag = ccIsPow2Int32( width ) && ccIsPow2Int32( height );
+  applykernel = 0;
+#if CPU_SSE2_SUPPORT
+  applykernelcore = 0;
+#endif
+
+  if( filter == IM_REDUCE_FILTER_LINEAR )
+  {
+    if( bytesperpixel == 4 )
+    {
+      applykernel = imStaticKernel4Linear;
+#if CPU_SSE2_SUPPORT
+      applykernelcore = imStaticKernel4Linear_Core;
+#endif
+    }
+    else if( bytesperpixel == 3 )
+      applykernel = imStaticKernel3Linear;
+    else if( bytesperpixel == 2 )
+      applykernel = imStaticKernel2Linear;
+    else if( bytesperpixel == 1 )
+      applykernel = imStaticKernel1Linear;
+  }
+  else if( filter == IM_REDUCE_FILTER_LINEAR_ALPHANORM )
+  {
+    if( bytesperpixel == 4 )
+    {
+      applykernel = imStaticKernel4LinearAlphaNorm;
+#if CPU_SSE2_SUPPORT
+      applykernelcore = imStaticKernel4LinearAlphaNorm_Core;
+#endif
+    }
+    else if( bytesperpixel == 3 )
+      applykernel = imStaticKernel3Linear;
+    else if( bytesperpixel == 2 )
+      applykernel = imStaticKernel2Linear;
+    else if( bytesperpixel == 1 )
+      applykernel = imStaticKernel1Linear;
+  }
+  else if( filter == IM_REDUCE_FILTER_SRGB )
+  {
+    if( bytesperpixel == 4 )
+    {
+      applykernel = imStaticKernel4sRGB;
+#if CPU_SSE2_SUPPORT
+      applykernelcore = imStaticKernel4sRGB_Core;
+#endif
+    }
+    else if( bytesperpixel == 3 )
+    {
+      applykernel = imStaticKernel3sRGB;
+#if CPU_SSE2_SUPPORT
+      applykernelcore = imStaticKernel3sRGB_Core;
+#endif
+    }
+    else if( bytesperpixel == 2 )
+      applykernel = imStaticKernel2sRGB;
+    else if( bytesperpixel == 1 )
+      applykernel = imStaticKernel1sRGB;
+  }
+  else if( filter == IM_REDUCE_FILTER_SRGB_ALPHANORM )
+  {
+    if( bytesperpixel == 4 )
+    {
+      applykernel = imStaticKernel4sRGBAlphaNorm;
+#if CPU_SSE2_SUPPORT
+      applykernelcore = imStaticKernel4sRGBAlphaNorm_Core;
+#endif
+    }
+    else if( bytesperpixel == 3 )
+      applykernel = imStaticKernel3sRGB;
+    else if( bytesperpixel == 2 )
+      applykernel = imStaticKernel2sRGB;
+    else if( bytesperpixel == 1 )
+      applykernel = imStaticKernel1sRGB;
+  }
+  else if( filter == IM_REDUCE_FILTER_NORMALMAP )
+  {
+    if( bytesperpixel == 4 )
+      applykernel = imStaticKernel4Normal;
+    else if( bytesperpixel == 3 )
+      applykernel = imStaticKernel3Normal;
+  }
+  else if( filter == IM_REDUCE_FILTER_NORMALMAP_SUSTAIN )
+  {
+    if( bytesperpixel == 4 )
+      applykernel = imStaticKernel4NormalSustain;
+    else if( bytesperpixel == 3 )
+      applykernel = imStaticKernel3NormalSustain;
+  }
+  else if( filter == IM_REDUCE_FILTER_NORMALMAP_SUSTAIN_ALPHANORM )
+  {
+    if( bytesperpixel == 4 )
+      applykernel = imStaticKernel4NormalSustainAlphaNorm;
+    else if( bytesperpixel == 3 )
+      applykernel = imStaticKernel3NormalSustain;
+  }
+  else if( filter == IM_REDUCE_FILTER_WATERMAP )
+  {
+    if( ( bytesperpixel == 4 ) && ( pow2flag ) )
+      applykernel = imStaticKernelPoT4Water;
+    else if( ( bytesperpixel == 3 ) && ( pow2flag ) )
+      applykernel = imStaticKernelPoT3Water;
+  }
+  else if( filter == IM_REDUCE_FILTER_PLANTMAP )
+  {
+    if( ( bytesperpixel == 4 ) && ( pow2flag ) )
+      applykernel = imStaticKernelPoT4Plant;
+  }
+
+  if( !applykernel )
+    return 0;
+
+#if CPU_SSE2_SUPPORT
+  corebase = -state.matrixoffset;
+  corerange = ( newwidth + state.matrixoffset ) - corebase;
+#endif
+
+  state.dithersum = 0.0f;
+  if( ( newwidth | newheight ) > 2 )
+    state.dithersum = 0.5f;
+
+  state.srcdata = srcdata;
+  state.width1 = width * 1;
+  state.width2 = width * 2;
+  state.width3 = width * 3;
+  state.width4 = width * 4;
+  state.height = height;
+  state.bytesperline = bytesperline;
+
+  state.minimumalpha = 4;
+  state.minimumalphaf = (float)state.minimumalpha;
+  state.amplifynormal = fmaxf( 1.0f, options->amplifynormal );
+  state.normalsustainfactor = options->normalsustainfactor;
+
+  basex = ( state.matrixoffset + ( width << 8 ) ) % width;
+  basey = ( state.matrixoffset + ( height << 8 ) ) % height;
+  while( basex < 0 )
+    basex += width;
+  while( basey < 0 )
+    basey += height;
+
+#if CPU_SSE2_SUPPORT
+  if( applykernelcore )
+  {
+    dst = dstdata;
+    pointy = basey;
+    for( y = 0 ; y < newheight ; y++ )
+    {
+      pointx = basex;
+      for( x = 0 ; x < newwidth ; x++, dst += bytesperpixel )
+      {
+        ( (unsigned int)( x - corebase ) < corerange ? applykernelcore : applykernel )( dst, pointx, pointy, &state );
+        pointx += sizedivisor;
+        while( pointx >= width )
+          pointx -= width;
+      }
+      pointy += sizedivisor;
+      while( pointy >= height )
+        pointy -= height;
+    }
+  }
+  else
+#endif
+  {
+    dst = dstdata;
+    pointy = basey;
+    for( y = 0 ; y < newheight ; y++ )
+    {
+      pointx = basex;
+      for( x = 0 ; x < newwidth ; x++, dst += bytesperpixel )
+      {
+        applykernel( dst, pointx, pointy, &state );
+        pointx += sizedivisor;
+        while( pointx >= width )
+          pointx -= width;
+      }
+      pointy += sizedivisor;
+      while( pointy >= height )
+        pointy -= height;
+    }
+  }
+
+  imFreeStaticState( &state );
+
+  return 1;
+}
+
+
+int imReduceImageKaiserDivisor( imgImage *imgdst, imgImage *imgsrc, int sizedivisor, imReduceOptions *options )
+{
+  int width, height;
+  int newwidth, newheight, retvalue;
+
+  width = imgsrc->format.width;
+  height = imgsrc->format.height;
+  newwidth = ( width < sizedivisor ) ? 1 : ( ( width + sizedivisor - 1 ) / sizedivisor );
+  newheight = ( height < sizedivisor ) ? 1 : ( ( height + sizedivisor - 1 ) / sizedivisor );
+
+  imgdst->format.width = newwidth;
+  imgdst->format.height = newheight;
+  imgdst->format.type = imgsrc->format.type;
+  imgdst->format.bytesperpixel = imgsrc->format.bytesperpixel;
+  imgdst->format.bytesperline = imgdst->format.width * imgdst->format.bytesperpixel;
+  imgdst->data = malloc( imgdst->format.height * imgdst->format.bytesperline );
+  if( !( imgdst->data ) )
+    return 0;
+
+  retvalue = imReduceImageKaiserDataDivisor( imgdst->data, imgsrc->data, width, height, imgsrc->format.bytesperpixel, imgsrc->format.bytesperline, sizedivisor, options );
+
+  return retvalue;
+}
+
+
+
+////////////////////////////////////////////////////////////////////////////////
+
+
+
+static void imDynamicKernel1Linear( unsigned char *dst, imGenericMatrixState *state )
+{
+  int x, y, mapx, mapy;
+  float f, sum0;
+  float matrixsum;
+  unsigned char *src;
+
+  sum0 = 0.0f;
+  matrixsum = 0.0f;
+  mapy = state->matrixoffsety;
+  for( y = 0 ; y < state->matrixsizey ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = state->matrixoffsetx;
+    for( x = 0 ; x < state->matrixsizex ; x++ )
+    {
+      f = state->linearx[x] * state->lineary[y];
+      sum0 += f * (float)src[ mapx + 0 ];
+      matrixsum += f;
+      mapx++;
+      if( mapx >= state->width1 )
+        mapx = 0;
+    }
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+  sum0 /= matrixsum;
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum0 + 0.5f ) ) );
+
+  return;
+}
+
+
+static void imDynamicKernel2Linear( unsigned char *dst, imGenericMatrixState *state )
+{
+  int x, y, mapx, mapy;
+  float f, sum0, sum1;
+  float matrixsum;
+  unsigned char *src;
+
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  matrixsum = 0.0f;
+  mapy = state->matrixoffsety;
+  for( y = 0 ; y < state->matrixsizey ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = state->matrixoffsetx << 1;
+    for( x = 0 ; x < state->matrixsizex ; x++ )
+    {
+      f = state->linearx[x] * state->lineary[y];
+      sum0 += f * (float)src[ mapx + 0 ];
+      sum1 += f * (float)src[ mapx + 1 ];
+      matrixsum += f;
+      mapx += 2;
+      if( mapx >= state->width2 )
+        mapx = 0;
+    }
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+  matrixsum = 1.0f / matrixsum;
+  sum0 *= matrixsum;
+  sum1 *= matrixsum;
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum0 + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum1 + 0.5f ) ) );
+
+  return;
+}
+
+
+static void imDynamicKernel3Linear( unsigned char *dst, imGenericMatrixState *state )
+{
+  int x, y, mapx, mapy;
+  float f, sum0, sum1, sum2;
+  float matrixsum;
+  unsigned char *src;
+
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  matrixsum = 0.0f;
+  mapy = state->matrixoffsety;
+  for( y = 0 ; y < state->matrixsizey ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = state->matrixoffsetx + ( state->matrixoffsetx << 1 );
+    for( x = 0 ; x < state->matrixsizex ; x++ )
+    {
+      f = state->linearx[x] * state->lineary[y];
+      sum0 += f * (float)src[ mapx + 0 ];
+      sum1 += f * (float)src[ mapx + 1 ];
+      sum2 += f * (float)src[ mapx + 2 ];
+      matrixsum += f;
+      mapx += 3;
+      if( mapx >= state->width3 )
+        mapx = 0;
+    }
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+  matrixsum = 1.0f / matrixsum;
+  sum0 *= matrixsum;
+  sum1 *= matrixsum;
+  sum2 *= matrixsum;
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum0 + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum1 + 0.5f ) ) );
+  dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum2 + 0.5f ) ) );
+
+  return;
+}
+
+
+static void imDynamicKernel4Linear( unsigned char *dst, imGenericMatrixState *state )
+{
+  int x, y, mapx, mapy;
+  float f, sum0, sum1, sum2, sum3;
+  float matrixsum;
+  unsigned char *src;
+
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  sum3 = 0.0f;
+  matrixsum = 0.0f;
+  mapy = state->matrixoffsety;
+  for( y = 0 ; y < state->matrixsizey ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = state->matrixoffsetx << 2;
+    for( x = 0 ; x < state->matrixsizex ; x++ )
+    {
+      f = state->linearx[x] * state->lineary[y];
+      sum0 += f * (float)src[ mapx + 0 ];
+      sum1 += f * (float)src[ mapx + 1 ];
+      sum2 += f * (float)src[ mapx + 2 ];
+      sum3 += f * (float)src[ mapx + 3 ];
+      matrixsum += f;
+      mapx += 4;
+      if( mapx >= state->width4 )
+        mapx = 0;
+    }
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+  matrixsum = 1.0f / matrixsum;
+  sum0 *= matrixsum;
+  sum1 *= matrixsum;
+  sum2 *= matrixsum;
+  sum3 *= matrixsum;
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum0 + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum1 + 0.5f ) ) );
+  dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum2 + 0.5f ) ) );
+  dst[3] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum3 + 0.5f ) ) );
+
+  return;
+}
+
+
+////
+
+
+static void imDynamicKernel4LinearAlphaNorm( unsigned char *dst, imGenericMatrixState *state )
+{
+  int x, y, mapx, mapy;
+  float f, sum0, sum1, sum2, sum3, alpha;
+  float matrixsum;
+  unsigned char *src;
+
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  sum3 = 0.0f;
+  matrixsum = 0.0f;
+  mapy = state->matrixoffsety;
+  for( y = 0 ; y < state->matrixsizey ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = state->matrixoffsetx << 2;
+    for( x = 0 ; x < state->matrixsizex ; x++ )
+    {
+      f = state->linearx[x] * state->lineary[y];
+      alpha = (float)src[ mapx + 3 ] * f;
+      sum0 += alpha * (float)src[ mapx + 0 ];
+      sum1 += alpha * (float)src[ mapx + 1 ];
+      sum2 += alpha * (float)src[ mapx + 2 ];
+      sum3 += alpha;
+      matrixsum += f;
+      mapx += 4;
+      if( mapx >= state->width4 )
+        mapx = 0;
+    }
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+  matrixsum = 1.0f / matrixsum;
+  sum0 *= matrixsum;
+  sum1 *= matrixsum;
+  sum2 *= matrixsum;
+  sum3 *= matrixsum;
+  if( sum3 >= state->minimumalphaf )
+  {
+    f = 1.0f / sum3;
+    dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, ( sum0 * f ) + 0.5f ) ) );
+    dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, ( sum1 * f ) + 0.5f ) ) );
+    dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, ( sum2 * f ) + 0.5f ) ) );
+    dst[3] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum3 + 0.5f ) ) );
+  }
+  else
+  {
+    dst[0] = 0;
+    dst[1] = 0;
+    dst[2] = 0;
+    dst[3] = 0;
+  }
+
+  return;
+}
+
+
+#if CPU_SSE2_SUPPORT
+
+static void imDynamicKernel4LinearAlphaNorm_Core( unsigned char *dst, imGenericMatrixState *state )
+{
+  int x, y, mapx, mapy;
+  uint32_t pixel;
+  unsigned char *src;
+  __m128 vmatrixsum, vsum0, vsum1, vsum2, vsum3;
+  __m128 vlx, vly, vf, valpha, vr, vg, vb, va, vsrcf;
+  __m128i vsrc, vshufmask;
+  __m128i vzero;
+
+ #if CPU_SSSE3_SUPPORT
+  vshufmask = _mm_setr_epi8( 0x00,0x04,0x08,0x0c, 0x01,0x05,0x09,0x0d, 0x02,0x06,0x0a,0x0e, 0x03,0x07,0x0b,0x0f );
+ #endif
+  vsum0 = _mm_setzero_ps();
+  vsum1 = _mm_setzero_ps();
+  vsum2 = _mm_setzero_ps();
+  vsum3 = _mm_setzero_ps();
+  vmatrixsum = _mm_setzero_ps();
+  vzero = _mm_castps_si128( _mm_setzero_ps() );
+
+  mapy = state->matrixoffsety;
+  for( y = 0 ; y < state->matrixsizey ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = state->matrixoffsetx << 2;
+    vly = _mm_set1_ps( state->lineary[y] );
+    for( x = 0 ; x < state->matrixsizex ; x += 4 )
+    {
+      vlx = _mm_load_ps( &state->linearx[x] );
+      /* Load 16 bytes and unpack as RRRR,GGGG,BBBB,AAAA in one SSE register */
+      vsrc = _mm_loadu_si128( (void *)&src[ mapx ] );
+ #if CPU_SSSE3_SUPPORT
+      vsrc = _mm_shuffle_epi8( vsrc, vshufmask );
+ #else
+      vshufmask = _mm_shuffle_epi32( vsrc, 0x39 );
+      vsrc = _mm_unpacklo_epi16( _mm_unpacklo_epi8( vsrc, vshufmask ), _mm_unpackhi_epi8( vsrc, vshufmask ) );
+ #endif
+      /* Break that into 4 SSE registers as floats: vR,vG,vB,vA */
+      vsrcf = _mm_castsi128_ps( vsrc );
+      vr = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( vsrcf ), vzero ) );
+ #if CPU_SSE3_SUPPORT
+      vg = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_movehdup_ps( vsrcf ) ), vzero ) );
+ #else
+      vg = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_shuffle_ps( vsrcf, vsrcf, 0x55 ) ), vzero ) );
+ #endif
+      vb = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_movehl_ps( vsrcf, vsrcf ) ), vzero ) );
+      va = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_shuffle_ps( vsrcf, vsrcf, 0xff ) ), vzero ) );
+      vf = _mm_mul_ps( vlx, vly );
+      valpha = _mm_mul_ps( va, vf );
+      vsum0 = _mm_add_ps( vsum0, _mm_mul_ps( vr, valpha ) );
+      vsum1 = _mm_add_ps( vsum1, _mm_mul_ps( vg, valpha ) );
+      vsum2 = _mm_add_ps( vsum2, _mm_mul_ps( vb, valpha ) );
+      vsum3 = _mm_add_ps( vsum3, valpha );
+      vmatrixsum = _mm_add_ps( vmatrixsum, vf );      
+      mapx += 16;
+    }
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+ #if CPU_SSE3_SUPPORT
+  vmatrixsum = _mm_hadd_ps( vmatrixsum, vmatrixsum );
+  vmatrixsum = _mm_hadd_ps( vmatrixsum, vmatrixsum );
+ #else
+  vmatrixsum = _mm_add_ps( vmatrixsum, _mm_shuffle_ps( vmatrixsum, vmatrixsum, 0x4e ) );
+  vmatrixsum = _mm_add_ps( vmatrixsum, _mm_shuffle_ps( vmatrixsum, vmatrixsum, 0x39 ) );
+ #endif
+
+ #if CPU_SSE3_SUPPORT
+  vsum0 = _mm_hadd_ps( vsum0, vsum1 );
+  vsum2 = _mm_hadd_ps( vsum2, vsum3 );
+  vsum0 = _mm_hadd_ps( vsum0, vsum2 );
+ #else
+  vsum0 = _mm_add_ps( _mm_unpacklo_ps( vsum0, vsum2 ), _mm_unpackhi_ps( vsum0, vsum2 ) );
+  vsum1 = _mm_add_ps( _mm_unpacklo_ps( vsum1, vsum3 ), _mm_unpackhi_ps( vsum1, vsum3 ) );
+  vsum0 = _mm_add_ps( _mm_unpacklo_ps( vsum0, vsum1 ), _mm_unpackhi_ps( vsum0, vsum1 ) );
+ #endif
+  vsum0 = _mm_div_ps( vsum0, vmatrixsum );
+
+  valpha = _mm_shuffle_ps( vsum0, vsum0, 0xff );
+  pixel = 0;
+  if( _mm_comige_ss( valpha, _mm_load_ss( &state->minimumalphaf ) ) )
+  {
+    __m128i vpixel;
+    vsum0 = _mm_mul_ps( vsum0, _mm_rcp_ps( valpha ) );
+    vsum0 = CPU_BLENDV_PS( vsum0, valpha, *(__m128 *)simd4fAlphaMask );
+    vpixel = _mm_cvtps_epi32( vsum0 );
+    vpixel = _mm_packs_epi32( vpixel, vpixel );
+    vpixel = _mm_packus_epi16( vpixel, vpixel );
+    pixel = (uint32_t)_mm_cvtsi128_si32( vpixel );
+  }
+  *(uint32_t *)dst = pixel;
+
+  return;
+}
+
+#endif
+
+
+////
+
+
+static void imDynamicKernel1sRGB( unsigned char *dst, imGenericMatrixState *state )
+{
+  int x, y, mapx, mapy;
+#if CPU_SSE2_SUPPORT
+  __m128 vsum, vsrc;
+  __m128i vzero;
+#else
+  float sum0;
+#endif
+  float f, matrixsum;
+  unsigned char *src;
+
+#if CPU_SSE2_SUPPORT
+  vsum = _mm_setzero_ps();
+  vzero = _mm_setzero_si128();
+#else
+  sum0 = 0.0f;
+#endif
+  matrixsum = 0.0f;
+  mapy = state->matrixoffsety;
+  for( y = 0 ; y < state->matrixsizey ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = state->matrixoffsetx;
+    for( x = 0 ; x < state->matrixsizex ; x++ )
+    {
+#if CPU_SSE2_SUPPORT
+      f = state->linearx[x] * state->lineary[y];
+      vsrc = _mm_set_ss( (float)src[ mapx + 0 ] );
+      vsrc = srgb2linear3( vsrc );
+      vsum = _mm_add_ps( vsum, _mm_mul_ps( _mm_set1_ps( f ), vsrc ) );
+#else
+      f = state->linearx[x] * state->lineary[y];
+      sum0 += f * srgb2linear( (float)src[ mapx + 0 ] );
+#endif
+      matrixsum += f;
+      mapx++;
+      if( mapx >= state->width1 )
+        mapx = 0;
+    }
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+#if CPU_SSE2_SUPPORT
+  vsum = linear2srgb3( _mm_div_ps( vsum, _mm_set1_ps( matrixsum ) ) );
+  dst[0] = _mm_cvtsi128_si32( _mm_packus_epi16( _mm_packs_epi32( _mm_cvtps_epi32( vsum ), vzero ), vzero ) );
+#else
+  sum0 /= matrixsum;
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum0 ) + 0.5f ) ) );
+#endif
+
+  return;
+}
+
+
+static void imDynamicKernel2sRGB( unsigned char *dst, imGenericMatrixState *state )
+{
+  int x, y, mapx, mapy;
+#if CPU_SSE2_SUPPORT
+  __m128 vsum, vsrc;
+  __m128i vzero;
+#else
+  float sum0, sum1;
+#endif
+  float f, matrixsum;
+  unsigned char *src;
+
+#if CPU_SSE2_SUPPORT
+  vsum = _mm_setzero_ps();
+  vzero = _mm_setzero_si128();
+#else
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+#endif
+  matrixsum = 0.0f;
+  mapy = state->matrixoffsety;
+  for( y = 0 ; y < state->matrixsizey ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = state->matrixoffsetx << 1;
+    for( x = 0 ; x < state->matrixsizex ; x++ )
+    {
+#if CPU_SSE2_SUPPORT
+      f = state->linearx[x] * state->lineary[y];
+      vsrc = _mm_set_ps( 0.0f, 0.0f, (float)src[ mapx + 1 ], (float)src[ mapx + 0 ] );
+      vsrc = srgb2linear3( vsrc );
+      vsum = _mm_add_ps( vsum, _mm_mul_ps( _mm_set1_ps( f ), vsrc ) );
+#else
+      f = state->linearx[x] * state->lineary[y];
+      sum0 += f * srgb2linear( (float)src[ mapx + 0 ] );
+      sum1 += f * srgb2linear( (float)src[ mapx + 1 ] );
+#endif
+      matrixsum += f;
+      mapx += 2;
+      if( mapx >= state->width2 )
+        mapx = 0;
+    }
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+#if CPU_SSE2_SUPPORT
+  union
+  {
+    char c[4];
+    uint32_t i;
+  } u;
+  vsum = linear2srgb3( _mm_div_ps( vsum, _mm_set1_ps( matrixsum ) ) );
+  _mm_store_ss( (void *)&u.i, _mm_castsi128_ps( _mm_packus_epi16( _mm_packs_epi32( _mm_cvtps_epi32( vsum ), vzero ), vzero ) ) );
+  dst[0] = u.c[0];
+  dst[1] = u.c[1];
+#else
+  matrixsum = 1.0f / matrixsum;
+  sum0 *= matrixsum;
+  sum1 *= matrixsum;
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum0 ) + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum1 ) + 0.5f ) ) );
+#endif
+
+  return;
+}
+
+
+static void imDynamicKernel3sRGB( unsigned char *dst, imGenericMatrixState *state )
+{
+  int x, y, mapx, mapy;
+#if CPU_SSE2_SUPPORT
+  __m128 vsum, vsrc;
+  __m128i vzero;
+#else
+  float sum0, sum1, sum2;
+#endif
+  float f, matrixsum;
+  unsigned char *src;
+
+#if CPU_SSE2_SUPPORT
+  vsum = _mm_setzero_ps();
+  vzero = _mm_setzero_si128();
+#else
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+#endif
+  matrixsum = 0.0f;
+  mapy = state->matrixoffsety;
+  for( y = 0 ; y < state->matrixsizey ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = state->matrixoffsetx + ( state->matrixoffsetx << 1 );
+    for( x = 0 ; x < state->matrixsizex ; x++ )
+    {
+#if CPU_SSE2_SUPPORT
+      f = state->linearx[x] * state->lineary[y];
+      vsrc = _mm_set_ps( 0.0f, (float)src[ mapx + 2 ], (float)src[ mapx + 1 ], (float)src[ mapx + 0 ] );
+      vsrc = srgb2linear3( vsrc );
+      vsum = _mm_add_ps( vsum, _mm_mul_ps( _mm_set1_ps( f ), vsrc ) );
+#else
+      f = state->linearx[x] * state->lineary[y];
+      sum0 += f * srgb2linear( (float)src[ mapx + 0 ] );
+      sum1 += f * srgb2linear( (float)src[ mapx + 1 ] );
+      sum2 += f * srgb2linear( (float)src[ mapx + 2 ] );
+#endif
+      matrixsum += f;
+      mapx += 3;
+      if( mapx >= state->width3 )
+        mapx = 0;
+    }
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+#if CPU_SSE2_SUPPORT
+  union
+  {
+    char c[4];
+    uint32_t i;
+  } u;
+  vsum = linear2srgb3( _mm_div_ps( vsum, _mm_set1_ps( matrixsum ) ) );
+  _mm_store_ss( (void *)&u.i, _mm_castsi128_ps( _mm_packus_epi16( _mm_packs_epi32( _mm_cvtps_epi32( vsum ), vzero ), vzero ) ) );
+  dst[0] = u.c[0];
+  dst[1] = u.c[1];
+  dst[2] = u.c[2];
+#else
+  matrixsum = 1.0f / matrixsum;
+  sum0 *= matrixsum;
+  sum1 *= matrixsum;
+  sum2 *= matrixsum;
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum0 ) + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum1 ) + 0.5f ) ) );
+  dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum2 ) + 0.5f ) ) );
+#endif
+
+  return;
+}
+
+
+static void imDynamicKernel4sRGB( unsigned char *dst, imGenericMatrixState *state )
+{
+  int x, y, mapx, mapy;
+#if CPU_SSE2_SUPPORT
+  __m128 vsum, vsrc;
+  __m128i vzero;
+#else
+  float sum0, sum1, sum2, sum3;
+#endif
+  float f, matrixsum;
+  unsigned char *src;
+
+#if CPU_SSE2_SUPPORT
+  vsum = _mm_setzero_ps();
+  vzero = _mm_setzero_si128();
+#else
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  sum3 = 0.0f;
+#endif
+  matrixsum = 0.0f;
+  mapy = state->matrixoffsety;
+  for( y = 0 ; y < state->matrixsizey ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = state->matrixoffsetx << 2;
+    for( x = 0 ; x < state->matrixsizex ; x++ )
+    {
+#if CPU_SSE2_SUPPORT
+      f = state->linearx[x] * state->lineary[y];
+      vsrc = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_load_ss( (void *)&src[ mapx ] ) ), vzero ) );
+      vsrc = srgb2linear3( vsrc );
+      vsum = _mm_add_ps( vsum, _mm_mul_ps( _mm_set1_ps( f ), vsrc ) );
+#else
+      f = state->linearx[x] * state->lineary[y];
+      sum0 += f * srgb2linear( (float)src[ mapx + 0 ] );
+      sum1 += f * srgb2linear( (float)src[ mapx + 1 ] );
+      sum2 += f * srgb2linear( (float)src[ mapx + 2 ] );
+      sum3 += f * (float)src[ mapx + 3 ];
+#endif
+      matrixsum += f;
+      mapx += 4;
+      if( mapx >= state->width4 )
+        mapx = 0;
+    }
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+#if CPU_SSE2_SUPPORT
+  vsum = linear2srgb3( _mm_div_ps( vsum, _mm_set1_ps( matrixsum ) ) );
+  _mm_store_ss( (void *)dst, _mm_castsi128_ps( _mm_packus_epi16( _mm_packs_epi32( _mm_cvtps_epi32( vsum ), vzero ), vzero ) ) );
+#else
+  matrixsum = 1.0f / matrixsum;
+  sum0 *= matrixsum;
+  sum1 *= matrixsum;
+  sum2 *= matrixsum;
+  sum3 *= matrixsum;
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum0 ) + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum1 ) + 0.5f ) ) );
+  dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum2 ) + 0.5f ) ) );
+  dst[3] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum3 + 0.5f ) ) );
+#endif
+
+  return;
+}
+
+
+////
+
+
+static void imDynamicKernel4sRGBAlphaNorm( unsigned char *dst, imGenericMatrixState *state )
+{
+  int x, y, mapx, mapy;
+#if CPU_SSE2_SUPPORT
+  __m128 vsum, vsrc, valpha;
+  __m128i vzero;
+  uint32_t pixel;
+#else
+  float sum0, sum1, sum2, sum3, alpha;
+#endif
+  float f, matrixsum;
+  unsigned char *src;
+
+#if CPU_SSE2_SUPPORT
+  vsum = _mm_setzero_ps();
+  vzero = _mm_setzero_si128();
+#else
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  sum3 = 0.0f;
+#endif
+  matrixsum = 0.0f;
+  mapy = state->matrixoffsety;
+  for( y = 0 ; y < state->matrixsizey ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = state->matrixoffsetx << 2;
+    for( x = 0 ; x < state->matrixsizex ; x++ )
+    {
+#if CPU_SSE2_SUPPORT
+      f = state->linearx[x] * state->lineary[y];
+      vsrc = _mm_cvtepi32_ps( CPU_CVT_U8_TO_I32( _mm_castps_si128( _mm_load_ss( (void *)&src[ mapx ] ) ), vzero ) );
+      valpha = _mm_shuffle_ps( vsrc, _mm_set_ss( 1.0f ), 0x0f );
+      vsrc = srgb2linear3( vsrc );
+      vsum = _mm_add_ps( vsum, _mm_mul_ps( _mm_mul_ps( _mm_shuffle_ps( valpha, valpha, 0xC0 ), _mm_set1_ps( f ) ), vsrc ) );
+#else
+      f = state->linearx[x] * state->lineary[y];
+      alpha = (float)src[ mapx + 3 ] * f;
+      sum0 += alpha * srgb2linear( (float)src[ mapx + 0 ] );
+      sum1 += alpha * srgb2linear( (float)src[ mapx + 1 ] );
+      sum2 += alpha * srgb2linear( (float)src[ mapx + 2 ] );
+      sum3 += alpha;
+#endif
+      matrixsum += f;
+      mapx += 4;
+      if( mapx >= state->width4 )
+        mapx = 0;
+    }
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+#if CPU_SSE2_SUPPORT
+  vsum = _mm_div_ps( vsum, _mm_set1_ps( matrixsum ) );
+  valpha = _mm_shuffle_ps( vsum, vsum, 0xff );
+  pixel = 0;
+  if( _mm_comige_ss( valpha, _mm_load_ss( &state->minimumalphaf ) ) )
+  {
+    __m128i vpixel;
+    vsum = _mm_mul_ps( vsum, _mm_rcp_ps( valpha ) );
+    vsum = CPU_BLENDV_PS( vsum, valpha, *(__m128 *)simd4fAlphaMask );
+    vsum = linear2srgb3( vsum );
+    vpixel = _mm_cvtps_epi32( vsum );
+    vpixel = _mm_packs_epi32( vpixel, vpixel );
+    vpixel = _mm_packus_epi16( vpixel, vpixel );
+    pixel = (uint32_t)_mm_cvtsi128_si32( vpixel );
+  }
+  *(uint32_t *)dst = pixel;
+#else
+  matrixsum = 1.0f / matrixsum;
+  sum0 *= matrixsum;
+  sum1 *= matrixsum;
+  sum2 *= matrixsum;
+  sum3 *= matrixsum;
+  if( sum3 >= state->minimumalphaf )
+  {
+    f = 1.0f / sum3;
+    dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum0 * f ) + 0.5f ) ) );
+    dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum1 * f ) + 0.5f ) ) );
+    dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum2 * f ) + 0.5f ) ) );
+    dst[3] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum3 + 0.5f ) ) );
+  }
+  else
+  {
+    dst[0] = 0;
+    dst[1] = 0;
+    dst[2] = 0;
+    dst[3] = 0;
+  }
+#endif
+
+  return;
+}
+
+
+////
+
+
+static void imDynamicKernel3Normal( unsigned char *dst, imGenericMatrixState *state )
+{
+  int x, y, mapx, mapy;
+  float f, sum0, sum1, sum2;
+  float matrixsum, suminv;
+  unsigned char *src;
+
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  matrixsum = 0.0f;
+  mapy = state->matrixoffsety;
+  for( y = 0 ; y < state->matrixsizey ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = state->matrixoffsetx + ( state->matrixoffsetx << 1 );
+    for( x = 0 ; x < state->matrixsizex ; x++ )
+    {
+      f = state->linearx[x] * state->lineary[y];
+      sum0 += f * (float)src[ mapx + 0 ];
+      sum1 += f * (float)src[ mapx + 1 ];
+      sum2 += f * (float)src[ mapx + 2 ];
+      matrixsum += f;
+      mapx += 3;
+      if( mapx >= state->width3 )
+        mapx = 0;
+    }
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+  matrixsum = (1.0f/255.0f) / matrixsum;
+  sum0 *= matrixsum;
+  sum1 *= matrixsum;
+  sum2 *= matrixsum;
+  sum0 -= 0.5f;
+  sum1 -= 0.5f;
+  sum2 -= 0.5f;
+  sum0 *= state->amplifynormal;
+  sum1 *= state->amplifynormal;
+  suminv = (0.5f*255.0f) / sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) + ( sum2 * sum2 ) );
+  sum0 = (0.5f*255.0f) + ( sum0 * suminv );
+  sum1 = (0.5f*255.0f) + ( sum1 * suminv );
+  sum2 = (0.5f*255.0f) + ( sum2 * suminv );
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum0 + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum1 + 0.5f ) ) );
+  dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum2 + 0.5f ) ) );
+
+  return;
+}
+
+
+static void imDynamicKernel4Normal( unsigned char *dst, imGenericMatrixState *state )
+{
+  int x, y, mapx, mapy;
+  float f, sum0, sum1, sum2, sum3;
+  float matrixsum, suminv;
+  unsigned char *src;
+
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  sum3 = 0.0f;
+  matrixsum = 0.0f;
+  mapy = state->matrixoffsety;
+  for( y = 0 ; y < state->matrixsizey ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = state->matrixoffsetx << 2;
+    for( x = 0 ; x < state->matrixsizex ; x++ )
+    {
+      f = state->linearx[x] * state->lineary[y];
+      sum0 += f * (float)src[ mapx + 0 ];
+      sum1 += f * (float)src[ mapx + 1 ];
+      sum2 += f * (float)src[ mapx + 2 ];
+      sum3 += f * (float)src[ mapx + 3 ];
+      matrixsum += f;
+      mapx += 4;
+      if( mapx >= state->width4 )
+        mapx = 0;
+    }
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+  matrixsum = (1.0f/255.0f) / matrixsum;
+  sum0 *= matrixsum;
+  sum1 *= matrixsum;
+  sum2 *= matrixsum;
+  sum3 *= matrixsum;
+  sum0 -= 0.5f;
+  sum1 -= 0.5f;
+  sum2 -= 0.5f;
+  sum0 *= state->amplifynormal;
+  sum1 *= state->amplifynormal;
+  suminv = (0.5f*255.0f) / sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) + ( sum2 * sum2 ) );
+  sum0 = (0.5f*255.0f) + ( sum0 * suminv );
+  sum1 = (0.5f*255.0f) + ( sum1 * suminv );
+  sum2 = (0.5f*255.0f) + ( sum2 * suminv );
+  sum3 *= 255.0f;
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum0 + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum1 + 0.5f ) ) );
+  dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum2 + 0.5f ) ) );
+  dst[3] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum3 + 0.5f ) ) );
+
+  return;
+}
+
+
+////
+
+
+static void imDynamicKernel3NormalSustain( unsigned char *dst, imGenericMatrixState *state )
+{
+  int x, y, mapx, mapy;
+  float f, v0, v1, v2, energy, sum0, sum1, sum2, sumenergy;
+  float matrixsum, suminv;
+  unsigned char *src;
+
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  sumenergy = 0.0f;
+  matrixsum = 0.0f;
+  mapy = state->matrixoffsety;
+  for( y = 0 ; y < state->matrixsizey ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = state->matrixoffsetx + ( state->matrixoffsetx << 1 );
+    for( x = 0 ; x < state->matrixsizex ; x++ )
+    {
+      f = state->linearx[x] * state->lineary[y];
+      v0 = f * ( (float)src[ mapx + 0 ] - 127.5f );
+      v1 = f * ( (float)src[ mapx + 1 ] - 127.5f );
+      v2 = f * ( (float)src[ mapx + 2 ] - 127.5f );
+      sum0 += v0;
+      sum1 += v1;
+      sum2 += v2;
+      energy = ( v0 * v0 ) + ( v1 * v1 );
+      if( energy )
+        sumenergy += sqrtf( energy ) / sqrtf( energy + ( v2 * v2 ) );
+      matrixsum += f;
+      mapx += 3;
+      if( mapx >= state->width3 )
+        mapx = 0;
+    }
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+  matrixsum = (1.0f/255.0f) / matrixsum;
+  sum0 *= matrixsum;
+  sum1 *= matrixsum;
+  sum2 *= matrixsum;
+  sum0 *= state->amplifynormal;
+  sum1 *= state->amplifynormal;
+  suminv = (0.5f*255.0f) / fmaxf( 0.0625f, sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) + ( sum2 * sum2 ) ) );
+  sum0 *= suminv;
+  sum1 *= suminv;
+  sum2 *= suminv;
+  energy = sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) );
+  sumenergy *= state->normalsustainfactor;
+  if( energy < sumenergy )
+  {
+    f = fminf( sumenergy / energy, 8.0f );
+    sum0 *= f;
+    sum1 *= f;
+    suminv = (0.5f*255.0f) / fmaxf( 0.0625f, sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) + ( sum2 * sum2 ) ) );
+    sum0 *= suminv;
+    sum1 *= suminv;
+    sum2 *= suminv;
+  }
+  sum0 += (0.5f*255.0f);
+  sum1 += (0.5f*255.0f);
+  sum2 += (0.5f*255.0f);
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum0 + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum1 + 0.5f ) ) );
+  dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum2 + 0.5f ) ) );
+
+  return;
+}
+
+static void imDynamicKernel4NormalSustain( unsigned char *dst, imGenericMatrixState *state )
+{
+  int x, y, mapx, mapy;
+  float f, v0, v1, v2, v3, energy, sum0, sum1, sum2, sum3, sumenergy;
+  float matrixsum, suminv;
+  unsigned char *src;
+
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  sum3 = 0.0f;
+  sumenergy = 0.0f;
+  matrixsum = 0.0f;
+  mapy = state->matrixoffsety;
+  for( y = 0 ; y < state->matrixsizey ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = state->matrixoffsetx << 2;
+    for( x = 0 ; x < state->matrixsizex ; x++ )
+    {
+      f = state->linearx[x] * state->lineary[y];
+      v0 = f * ( (float)src[ mapx + 0 ] - 127.5f );
+      v1 = f * ( (float)src[ mapx + 1 ] - 127.5f );
+      v2 = f * ( (float)src[ mapx + 2 ] - 127.5f );
+      v3 = f * (float)src[ mapx + 3 ];
+      sum0 += v0;
+      sum1 += v1;
+      sum2 += v2;
+      sum3 += v3;
+      energy = ( v0 * v0 ) + ( v1 * v1 );
+      if( energy )
+        sumenergy += sqrtf( energy ) / sqrtf( energy + ( v2 * v2 ) );
+      matrixsum += f;
+      mapx += 4;
+      if( mapx >= state->width4 )
+        mapx = 0;
+    }
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+  matrixsum = (1.0f/255.0f) / matrixsum;
+  sum0 *= matrixsum;
+  sum1 *= matrixsum;
+  sum2 *= matrixsum;
+  sum3 *= matrixsum;
+  sum0 *= state->amplifynormal;
+  sum1 *= state->amplifynormal;
+  suminv = (0.5f*255.0f) / fmaxf( 0.0625f, sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) + ( sum2 * sum2 ) ) );
+  sum0 *= suminv;
+  sum1 *= suminv;
+  sum2 *= suminv;
+  energy = sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) );
+  sumenergy *= state->normalsustainfactor;
+  if( energy < sumenergy )
+  {
+    f = fminf( sumenergy / energy, 8.0f );
+    sum0 *= f;
+    sum1 *= f;
+    suminv = (0.5f*255.0f) / fmaxf( 0.0625f, sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) + ( sum2 * sum2 ) ) );
+    sum0 *= suminv;
+    sum1 *= suminv;
+    sum2 *= suminv;
+  }
+  sum0 += (0.5f*255.0f);
+  sum1 += (0.5f*255.0f);
+  sum2 += (0.5f*255.0f);
+  sum3 *= 255.0f;
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum0 + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum1 + 0.5f ) ) );
+  dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum2 + 0.5f ) ) );
+  dst[3] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum3 + 0.5f ) ) );
+
+
+  return;
+}
+
+
+////
+
+
+static void imDynamicKernel4NormalSustainAlphaNorm( unsigned char *dst, imGenericMatrixState *state )
+{
+  int x, y, mapx, mapy;
+  float f, alpha, v0, v1, v2, v3, energy, sum0, sum1, sum2, sum3, sumenergy;
+  float matrixsum, suminv;
+  unsigned char *src;
+
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  sum3 = 0.0f;
+  sumenergy = 0.0f;
+  matrixsum = 0.0f;
+  mapy = state->matrixoffsety;
+  for( y = 0 ; y < state->matrixsizey ; y++ )
+  {
+    src = ADDRESS( state->srcdata, ( mapy * state->bytesperline ) );
+    mapx = state->matrixoffsetx << 2;
+    for( x = 0 ; x < state->matrixsizex ; x++ )
+    {
+      f = state->linearx[x] * state->lineary[y];
+      alpha = (float)src[ mapx + 3 ] * f;
+      v0 = alpha * ( (float)src[ mapx + 0 ] - 127.5f );
+      v1 = alpha * ( (float)src[ mapx + 1 ] - 127.5f );
+      v2 = alpha * ( (float)src[ mapx + 2 ] - 127.5f );
+      v3 = alpha;
+      sum0 += v0;
+      sum1 += v1;
+      sum2 += v2;
+      sum3 += v3;
+      energy = ( v0 * v0 ) + ( v1 * v1 );
+      if( energy )
+        sumenergy += sqrtf( energy ) / sqrtf( energy + ( v2 * v2 ) );
+      matrixsum += f;
+      mapx += 4;
+      if( mapx >= state->width4 )
+        mapx = 0;
+    }
+    mapy++;
+    if( mapy >= state->height )
+      mapy = 0;
+  }
+
+  matrixsum = 1.0f / matrixsum;
+  sum0 *= matrixsum;
+  sum1 *= matrixsum;
+  sum2 *= matrixsum;
+  sum3 *= matrixsum;
+  if( sum3 >= state->minimumalphaf )
+  {
+    f = 1.0f / sum3;
+    sum0 *= f;
+    sum1 *= f;
+    sum2 *= f;
+    sum0 *= state->amplifynormal;
+    sum1 *= state->amplifynormal;
+    suminv = (0.5f*255.0f) / fmaxf( 0.0625f, sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) + ( sum2 * sum2 ) ) );
+    sum0 *= suminv;
+    sum1 *= suminv;
+    sum2 *= suminv;
+    energy = sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) );
+    sumenergy *= state->normalsustainfactor;
+    if( energy < sumenergy )
+    {
+      f = fminf( sumenergy / energy, 8.0f );
+      sum0 *= f;
+      sum1 *= f;
+      suminv = (0.5f*255.0f) / fmaxf( 0.0625f, sqrtf( ( sum0 * sum0 ) + ( sum1 * sum1 ) + ( sum2 * sum2 ) ) );
+      sum0 *= suminv;
+      sum1 *= suminv;
+      sum2 *= suminv;
+    }
+    sum0 += (0.5f*255.0f);
+    sum1 += (0.5f*255.0f);
+    sum2 += (0.5f*255.0f);
+    dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum0 + 0.5f ) ) );
+    dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum1 + 0.5f ) ) );
+    dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum2 + 0.5f ) ) );
+    dst[3] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, sum3 + 0.5f ) ) );
+  }
+  else
+  {
+    dst[0] = 0;
+    dst[1] = 0;
+    dst[2] = 0;
+    dst[3] = 0;
+  }
+
+  return;
+}
+
+
+////
+
+
+int imReduceImageKaiserData( unsigned char *dstdata, unsigned char *srcdata, int width, int height, int bytesperpixel, int bytesperline, int newwidth, int newheight, imReduceOptions *options )
+{
+  int filter, x, y;
+  float scalex, scaley, scaleinvx, scaleinvy;
+  float sourcex, sourcey;
+  unsigned char *dst;
+  imGenericMatrixState state;
+  void (*applykernel)( unsigned char *dst, imGenericMatrixState *state );
+#if CPU_SSE2_SUPPORT
+  void (*applykernelcore)( unsigned char *dst, imGenericMatrixState *state );
+#endif
+
+  filter = options->filter;
+  if( ( newwidth > width ) || ( newheight > height ) )
+    return 0;
+
+  applykernel = 0;
+#if CPU_SSE2_SUPPORT
+  applykernelcore = 0;
+#endif
+
+  if( filter == IM_REDUCE_FILTER_LINEAR )
+  {
+    if( bytesperpixel == 4 )
+      applykernel = imDynamicKernel4Linear;
+    else if( bytesperpixel == 3 )
+      applykernel = imDynamicKernel3Linear;
+    else if( bytesperpixel == 2 )
+      applykernel = imDynamicKernel2Linear;
+    else if( bytesperpixel == 1 )
+      applykernel = imDynamicKernel1Linear;
+  }
+  else if( filter == IM_REDUCE_FILTER_LINEAR_ALPHANORM )
+  {
+    if( bytesperpixel == 4 )
+    {
+      applykernel = imDynamicKernel4LinearAlphaNorm;
+#if CPU_SSE2_SUPPORT
+      applykernelcore = imDynamicKernel4LinearAlphaNorm_Core;
+#endif
+    }
+    else if( bytesperpixel == 3 )
+      applykernel = imDynamicKernel3Linear;
+    else if( bytesperpixel == 2 )
+      applykernel = imDynamicKernel2Linear;
+    else if( bytesperpixel == 1 )
+      applykernel = imDynamicKernel1Linear;
+  }
+  else if( filter == IM_REDUCE_FILTER_SRGB )
+  {
+    if( bytesperpixel == 4 )
+      applykernel = imDynamicKernel4sRGB;
+    else if( bytesperpixel == 3 )
+      applykernel = imDynamicKernel3sRGB;
+    else if( bytesperpixel == 2 )
+      applykernel = imDynamicKernel2sRGB;
+    else if( bytesperpixel == 1 )
+      applykernel = imDynamicKernel1sRGB;
+  }
+  else if( filter == IM_REDUCE_FILTER_SRGB_ALPHANORM )
+  {
+    if( bytesperpixel == 4 )
+      applykernel = imDynamicKernel4sRGBAlphaNorm;
+    else if( bytesperpixel == 3 )
+      applykernel = imDynamicKernel3sRGB;
+    else if( bytesperpixel == 2 )
+      applykernel = imDynamicKernel2sRGB;
+    else if( bytesperpixel == 1 )
+      applykernel = imDynamicKernel1sRGB;
+  }
+  else if( filter == IM_REDUCE_FILTER_NORMALMAP )
+  {
+    if( bytesperpixel == 4 )
+      applykernel = imDynamicKernel4Normal;
+    else if( bytesperpixel == 3 )
+      applykernel = imDynamicKernel3Normal;
+  }
+  else if( filter == IM_REDUCE_FILTER_NORMALMAP_SUSTAIN )
+  {
+    if( bytesperpixel == 4 )
+      applykernel = imDynamicKernel4NormalSustain;
+    else if( bytesperpixel == 3 )
+      applykernel = imDynamicKernel3NormalSustain;
+  }
+  else if( filter == IM_REDUCE_FILTER_NORMALMAP_SUSTAIN_ALPHANORM )
+  {
+    if( bytesperpixel == 4 )
+      applykernel = imDynamicKernel4NormalSustainAlphaNorm;
+    else if( bytesperpixel == 3 )
+      applykernel = imDynamicKernel3NormalSustain;
+  }
+
+  if( !applykernel )
+    return 0;
+
+  state.minimumalpha = 4;
+  state.minimumalphaf = (float)state.minimumalpha;
+  state.amplifynormal = fmaxf( 1.0f, options->amplifynormal );
+  state.normalsustainfactor = options->normalsustainfactor;
+
+  state.dithersum = 0.0f;
+  if( ( newwidth | newheight ) > 2 )
+    state.dithersum = 0.5f;
+
+  state.srcdata = srcdata;
+  state.width1 = width * 1;
+  state.width2 = width * 2;
+  state.width3 = width * 3;
+  state.width4 = width * 4;
+  state.height = height;
+  state.bytesperline = bytesperline;
+
+  scalex = (float)newwidth / (float)width;
+  scaley = (float)newheight / (float)height;
+  scaleinvx = (float)width / (float)newwidth;
+  scaleinvy = (float)height / (float)newheight;
+
+  imAllocGenericState( &state, scalex, scaley, options->hopcount, options->alpha );
+
+#if CPU_SSE2_SUPPORT
+  if( applykernelcore )
+  {
+    dst = dstdata;
+    for( y = 0 ; y < newheight ; y++ )
+    {
+      sourcey = ( ( (float)y + 0.5f ) * scaleinvy ) - 0.5f;
+      imBuildGenericLinearY( &state, scaley, scaleinvy, sourcey, options->hopcount, options->alpha, height );
+      for( x = 0 ; x < newwidth ; x++, dst += bytesperpixel )
+      {
+        sourcex = ( ( (float)x + 0.5f ) * scaleinvx ) - 0.5f;
+        imBuildGenericLinearX( &state, scalex, scaleinvx, sourcex, options->hopcount, options->alpha, width );
+        if( ( state.matrixoffsetx + ( ( state.matrixsizex + 3 ) & ~3 ) ) < width )
+          applykernelcore( dst, &state );
+        else
+          applykernel( dst, &state );
+      }
+    }
+  }
+  else
+#endif
+  {
+    dst = dstdata;
+    for( y = 0 ; y < newheight ; y++ )
+    {
+      sourcey = ( ( (float)y + 0.5f ) * scaleinvy ) - 0.5f;
+      imBuildGenericLinearY( &state, scaley, scaleinvy, sourcey, options->hopcount, options->alpha, height );
+      for( x = 0 ; x < newwidth ; x++, dst += bytesperpixel )
+      {
+        sourcex = ( ( (float)x + 0.5f ) * scaleinvx ) - 0.5f;
+        imBuildGenericLinearX( &state, scalex, scaleinvx, sourcex, options->hopcount, options->alpha, width );
+        applykernel( dst, &state );
+      }
+    }
+  }
+
+  imFreeGenericState( &state );
+
+  return 1;
+}
+
+
+int imReduceImageKaiser( imgImage *imgdst, imgImage *imgsrc, int newwidth, int newheight, imReduceOptions *options )
+{
+  int retvalue;
+
+  imgdst->format.width = newwidth;
+  imgdst->format.height = newheight;
+  imgdst->format.type = imgsrc->format.type;
+  imgdst->format.bytesperpixel = imgsrc->format.bytesperpixel;
+  imgdst->format.bytesperline = imgdst->format.width * imgdst->format.bytesperpixel;
+  imgdst->data = malloc( imgdst->format.height * imgdst->format.bytesperline );
+  if( !( imgdst->data ) )
+    return 0;
+
+  retvalue = imReduceImageKaiserData( imgdst->data, imgsrc->data, imgsrc->format.width, imgsrc->format.height, imgsrc->format.bytesperpixel, imgsrc->format.bytesperline, newwidth, newheight, options );
+
+  return retvalue;
+}
+
+
+
+////////////////////////////////////////////////////////////////////////////////
+
+
+
+static inline CC_ALWAYSINLINE void imReduceHalfBox1Linear( unsigned char *dst, unsigned char *src, int bytesperpixel, int bytesperline, float *dithersum )
+{
+  dst[0] = (unsigned char)( ( (int)src[0] + (int)src[bytesperpixel+0] + (int)src[bytesperline+0] + (int)src[bytesperpixel+bytesperline+0] + 2 ) >> 2 );
+  return;
+}
+
+static inline CC_ALWAYSINLINE void imReduceHalfBox2Linear( unsigned char *dst, unsigned char *src, int bytesperpixel, int bytesperline, float *dithersum )
+{
+  dst[0] = (unsigned char)( ( (int)src[0] + (int)src[bytesperpixel+0] + (int)src[bytesperline+0] + (int)src[bytesperpixel+bytesperline+0] + 2 ) >> 2 );
+  dst[1] = (unsigned char)( ( (int)src[1] + (int)src[bytesperpixel+1] + (int)src[bytesperline+1] + (int)src[bytesperpixel+bytesperline+1] + 2 ) >> 2 );
+  return;
+}
+
+static inline CC_ALWAYSINLINE void imReduceHalfBox3Linear( unsigned char *dst, unsigned char *src, int bytesperpixel, int bytesperline, float *dithersum )
+{
+  dst[0] = (unsigned char)( ( (int)src[0] + (int)src[bytesperpixel+0] + (int)src[bytesperline+0] + (int)src[bytesperpixel+bytesperline+0] + 2 ) >> 2 );
+  dst[1] = (unsigned char)( ( (int)src[1] + (int)src[bytesperpixel+1] + (int)src[bytesperline+1] + (int)src[bytesperpixel+bytesperline+1] + 2 ) >> 2 );
+  dst[2] = (unsigned char)( ( (int)src[2] + (int)src[bytesperpixel+2] + (int)src[bytesperline+2] + (int)src[bytesperpixel+bytesperline+2] + 2 ) >> 2 );
+  return;
+}
+
+static inline CC_ALWAYSINLINE void imReduceHalfBox4Linear( unsigned char *dst, unsigned char *src, int bytesperpixel, int bytesperline, float *dithersum )
+{
+  dst[0] = (unsigned char)( ( (int)src[0] + (int)src[bytesperpixel+0] + (int)src[bytesperline+0] + (int)src[bytesperpixel+bytesperline+0] + 2 ) >> 2 );
+  dst[1] = (unsigned char)( ( (int)src[1] + (int)src[bytesperpixel+1] + (int)src[bytesperline+1] + (int)src[bytesperpixel+bytesperline+1] + 2 ) >> 2 );
+  dst[2] = (unsigned char)( ( (int)src[2] + (int)src[bytesperpixel+2] + (int)src[bytesperline+2] + (int)src[bytesperpixel+bytesperline+2] + 2 ) >> 2 );
+  dst[3] = (unsigned char)( ( (int)src[3] + (int)src[bytesperpixel+3] + (int)src[bytesperline+3] + (int)src[bytesperpixel+bytesperline+3] + 2 ) >> 2 );
+  return;
+}
+
+static inline CC_ALWAYSINLINE void imReduceHalfBox1sRGB( unsigned char *dst, unsigned char *src, int bytesperpixel, int bytesperline, float *dithersum )
+{
+  int i, offset[4];
+  float sum0;
+  offset[0] = 0;
+  offset[1] = bytesperpixel;
+  offset[2] = bytesperline;
+  offset[3] = bytesperline + bytesperpixel;
+  sum0 = 0.0f;
+  for( i = 0 ; i < 4 ; i++ )
+    sum0 += srgb2linear( (float)src[offset[i]+0] );
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum0 * 0.25f ) + 0.5f ) ) );
+  return;
+}
+
+static inline CC_ALWAYSINLINE void imReduceHalfBox2sRGB( unsigned char *dst, unsigned char *src, int bytesperpixel, int bytesperline, float *dithersum )
+{
+  int i, offset[4];
+  float sum0, sum1;
+  offset[0] = 0;
+  offset[1] = bytesperpixel;
+  offset[2] = bytesperline;
+  offset[3] = bytesperline + bytesperpixel;
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  for( i = 0 ; i < 4 ; i++ )
+  {
+    sum0 += srgb2linear( (float)src[offset[i]+0] );
+    sum1 += srgb2linear( (float)src[offset[i]+1] );
+  }
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum0 * 0.25f ) + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum1 * 0.25f ) + 0.5f ) ) );
+  return;
+}
+
+static inline CC_ALWAYSINLINE void imReduceHalfBox3sRGB( unsigned char *dst, unsigned char *src, int bytesperpixel, int bytesperline, float *dithersum )
+{
+  int i, offset[4];
+  float sum0, sum1, sum2;
+  offset[0] = 0;
+  offset[1] = bytesperpixel;
+  offset[2] = bytesperline;
+  offset[3] = bytesperline + bytesperpixel;
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  for( i = 0 ; i < 4 ; i++ )
+  {
+    sum0 += srgb2linear( (float)src[offset[i]+0] );
+    sum1 += srgb2linear( (float)src[offset[i]+1] );
+    sum2 += srgb2linear( (float)src[offset[i]+2] );
+  }
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum0 * 0.25f ) + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum1 * 0.25f ) + 0.5f ) ) );
+  dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum2 * 0.25f ) + 0.5f ) ) );
+  return;
+}
+
+static inline CC_ALWAYSINLINE void imReduceHalfBox4sRGB( unsigned char *dst, unsigned char *src, int bytesperpixel, int bytesperline, float *dithersum )
+{
+  int i, offset[4], sum3;
+  float sum0, sum1, sum2;
+  offset[0] = 0;
+  offset[1] = bytesperpixel;
+  offset[2] = bytesperline;
+  offset[3] = bytesperline + bytesperpixel;
+  sum0 = 0.0f;
+  sum1 = 0.0f;
+  sum2 = 0.0f;
+  sum3 = 2;
+  for( i = 0 ; i < 4 ; i++ )
+  {
+    sum0 += srgb2linear( (float)src[offset[i]+0] );
+    sum1 += srgb2linear( (float)src[offset[i]+1] );
+    sum2 += srgb2linear( (float)src[offset[i]+2] );
+    sum3 += (int)src[offset[i]+2];
+  }
+  dst[0] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum0 * 0.25f ) + 0.5f ) ) );
+  dst[1] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum1 * 0.25f ) + 0.5f ) ) );
+  dst[2] = (unsigned char)( fmaxf( 0.0f, fminf( 255.0f, linear2srgb( sum2 * 0.25f ) + 0.5f ) ) );
+  dst[3] = (unsigned char)( sum3 >> 2 );
+  return;
+}
+
+static inline CC_ALWAYSINLINE void imReduceHalfBox3Normal( unsigned char *dst, unsigned char *src, int bytesperpixel, int bytesperline, float *dithersum )
+{
+  float v0, v1, v2, suminv;
+
+  v0 = (1.0f/1020.0f) * (float)( (int)src[0] + (int)src[bytesperpixel+0] + (int)src[bytesperline+0] + (int)src[bytesperpixel+bytesperline+0] );
+  v1 = (1.0f/1020.0f) * (float)( (int)src[1] + (int)src[bytesperpixel+1] + (int)src[bytesperline+1] + (int)src[bytesperpixel+bytesperline+1] );
+  v2 = (1.0f/1020.0f) * (float)( (int)src[2] + (int)src[bytesperpixel+2] + (int)src[bytesperline+2] + (int)src[bytesperpixel+bytesperline+2] );
+  v0 = 2.0f * ( v0 - 0.5f );
+  v1 = 2.0f * ( v1 - 0.5f );
+  v2 = 2.0f * ( v2 - 0.5f );
+  suminv = 0.5f / sqrtf( ( v0 * v0 ) + ( v1 * v1 ) + ( v2 * v2 ) );
+  v0 = 0.5f + ( v0 * suminv );
+  v1 = 0.5f + ( v1 * suminv );
+  v2 = 0.5f + ( v2 * suminv );
+  dst[0] = (unsigned char)ROUND_POSITIVE_FLOAT( 255.0f * v0 );
+  dst[1] = (unsigned char)ROUND_POSITIVE_FLOAT( 255.0f * v1 );
+  dst[2] = (unsigned char)ROUND_POSITIVE_FLOAT( 255.0f * v2 );
+
+  return;
+}
+
+static inline CC_ALWAYSINLINE void imReduceHalfBox4Normal( unsigned char *dst, unsigned char *src, int bytesperpixel, int bytesperline, float *dithersum )
+{
+  float v0, v1, v2, suminv;
+
+  v0 = (1.0f/1020.0f) * (float)( (int)src[0] + (int)src[bytesperpixel+0] + (int)src[bytesperline+0] + (int)src[bytesperpixel+bytesperline+0] );
+  v1 = (1.0f/1020.0f) * (float)( (int)src[1] + (int)src[bytesperpixel+1] + (int)src[bytesperline+1] + (int)src[bytesperpixel+bytesperline+1] );
+  v2 = (1.0f/1020.0f) * (float)( (int)src[2] + (int)src[bytesperpixel+2] + (int)src[bytesperline+2] + (int)src[bytesperpixel+bytesperline+2] );
+  v0 = 2.0f * ( v0 - 0.5f );
+  v1 = 2.0f * ( v1 - 0.5f );
+  v2 = 2.0f * ( v2 - 0.5f );
+  suminv = 0.5f / sqrtf( ( v0 * v0 ) + ( v1 * v1 ) + ( v2 * v2 ) );
+  v0 = 0.5f + ( v0 * suminv );
+  v1 = 0.5f + ( v1 * suminv );
+  v2 = 0.5f + ( v2 * suminv );
+  dst[0] = (unsigned char)ROUND_POSITIVE_FLOAT( 255.0f * v0 );
+  dst[1] = (unsigned char)ROUND_POSITIVE_FLOAT( 255.0f * v1 );
+  dst[2] = (unsigned char)ROUND_POSITIVE_FLOAT( 255.0f * v2 );
+  dst[3] = (unsigned char)( ( (int)src[3] + (int)src[bytesperpixel+3] + (int)src[bytesperline+3] + (int)src[bytesperpixel+bytesperline+3] + 2 ) >> 2 );
+
+  return;
+}
+
+static inline CC_ALWAYSINLINE void imReduceHalfBox3Water( unsigned char *dst, unsigned char *src, int bytesperpixel, int bytesperline, float *dithersum )
+{
+  float v0, v1, v2, suminv;
+
+  v0 = (1.0f/1020.0f) * (float)( (int)src[0] + (int)src[bytesperpixel+0] + (int)src[bytesperline+0] + (int)src[bytesperpixel+bytesperline+0] );
+  v1 = (1.0f/1020.0f) * (float)( (int)src[1] + (int)src[bytesperpixel+1] + (int)src[bytesperline+1] + (int)src[bytesperpixel+bytesperline+1] );
+  v2 = (1.0f/1020.0f) * (float)( (int)src[2] + (int)src[bytesperpixel+2] + (int)src[bytesperline+2] + (int)src[bytesperpixel+bytesperline+2] );
+
+  v0 = 2.0f * ( v0 - 0.5f );
+  v1 = 2.0f * ( v1 - 0.5f );
+  suminv = sqrtf( ( v0 * v0 ) + ( v1 * v1 ) );
+  if( suminv < 0.75f )
+  {
+    suminv = 0.5f / suminv;
+    v0 = 0.5f + ( v0 * suminv );
+    v1 = 0.5f + ( v1 * suminv );
+  }
+  if( v2 > 0.1f )
+  {
+    *dithersum += v2;
+    if( v2 > 0.45f )
+      v2 = 1.0f;
+    else if( ( v2 < 0.3f ) && ( *dithersum < 1.0f ) )
+      v2 = 0.0f;
+    else
+      v2 = ( ( v2 + *dithersum ) < 0.45f ? 0.0f : 1.0f );
+    *dithersum -= v2;
+  }
+  v0 *= 255.0f;
+  v1 *= 255.0f;
+  v2 *= 255.0f;
+
+  dst[0] = (int)( fmaxf( 0.0f, fminf( 255.0f, v0 + 0.5f ) ) );
+  dst[1] = (int)( fmaxf( 0.0f, fminf( 255.0f, v1 + 0.5f ) ) );
+  dst[2] = (int)( fmaxf( 0.0f, fminf( 255.0f, v2 + 0.5f ) ) );
+
+  return;
+}
+
+static inline CC_ALWAYSINLINE void imReduceHalfBox4Water( unsigned char *dst, unsigned char *src, int bytesperpixel, int bytesperline, float *dithersum )
+{
+  float v0, v1, v2, suminv;
+
+  v0 = (1.0f/1020.0f) * (float)( (int)src[0] + (int)src[bytesperpixel+0] + (int)src[bytesperline+0] + (int)src[bytesperpixel+bytesperline+0] );
+  v1 = (1.0f/1020.0f) * (float)( (int)src[1] + (int)src[bytesperpixel+1] + (int)src[bytesperline+1] + (int)src[bytesperpixel+bytesperline+1] );
+  v2 = (1.0f/1020.0f) * (float)( (int)src[2] + (int)src[bytesperpixel+2] + (int)src[bytesperline+2] + (int)src[bytesperpixel+bytesperline+2] );
+
+  v0 = 2.0f * ( v0 - 0.5f );
+  v1 = 2.0f * ( v1 - 0.5f );
+  suminv = sqrtf( ( v0 * v0 ) + ( v1 * v1 ) );
+  if( suminv < 0.75f )
+  {
+    suminv = 0.5f / suminv;
+    v0 = 0.5f + ( v0 * suminv );
+    v1 = 0.5f + ( v1 * suminv );
+  }
+  if( v2 > 0.1f )
+  {
+    *dithersum += v2;
+    if( v2 > 0.45f )
+      v2 = 1.0f;
+    else if( ( v2 < 0.3f ) && ( *dithersum < 1.0f ) )
+      v2 = 0.0f;
+    else
+      v2 = ( ( v2 + *dithersum ) < 0.45f ? 0.0f : 1.0f );
+    *dithersum -= v2;
+  }
+  v0 *= 255.0f;
+  v1 *= 255.0f;
+  v2 *= 255.0f;
+
+  dst[0] = (int)( fmaxf( 0.0f, fminf( 255.0f, v0 + 0.5f ) ) );
+  dst[1] = (int)( fmaxf( 0.0f, fminf( 255.0f, v1 + 0.5f ) ) );
+  dst[2] = (int)( fmaxf( 0.0f, fminf( 255.0f, v2 + 0.5f ) ) );
+  dst[3] = (unsigned char)( ( (int)src[3] + (int)src[bytesperpixel+3] + (int)src[bytesperline+3] + (int)src[bytesperpixel+bytesperline+3] + 2 ) >> 2 );
+
+  return;
+}
+
+static inline CC_ALWAYSINLINE void imReduceHalfBox4Plant( unsigned char *dst, unsigned char *src, int bytesperpixel, int bytesperline, float *dithersum )
+{
+  int alpha;
+
+  dst[0] = (unsigned char)( ( (int)src[0] + (int)src[bytesperpixel+0] + (int)src[bytesperline+0] + (int)src[bytesperpixel+bytesperline+0] + 2 ) >> 2 );
+  dst[1] = (unsigned char)( ( (int)src[1] + (int)src[bytesperpixel+1] + (int)src[bytesperline+1] + (int)src[bytesperpixel+bytesperline+1] + 2 ) >> 2 );
+  dst[2] = (unsigned char)( ( (int)src[2] + (int)src[bytesperpixel+2] + (int)src[bytesperline+2] + (int)src[bytesperpixel+bytesperline+2] + 2 ) >> 2 );
+
+  alpha = ( (int)src[3] + (int)src[bytesperpixel+3] + (int)src[bytesperline+3] + (int)src[bytesperpixel+bytesperline+3] );
+  alpha += alpha >> 2;
+  alpha = ( alpha + 2 ) >> 2;
+  if( alpha > 255 )
+    alpha = 255;
+  dst[3] = (unsigned char)alpha;
+
+  return;
+}
+
+
+static inline CC_ALWAYSINLINE void imReduceImageHalfBoxWork( unsigned char *dst, unsigned char *src, int width, int height, int bytesperpixel, int bytesperline, void (*work)( unsigned char *dst, unsigned char *src, int bytesperpixel, int bytesperline, float *dithersum ) )
+{
+  int x, y, newwidth, newheight, rowoffset;
+  float dithersum;
+  newwidth = ( width < 2 ) ? 1 : ( ( width + 1 ) / 2 );
+  newheight = ( height < 2 ) ? 1 : ( ( height + 1 ) / 2 );
+  rowoffset = bytesperline + ( bytesperpixel * ( width - ( newwidth << 1 ) ) );
+  dithersum = 0.0f;
+  if( ( newwidth | newheight ) > 2 )
+    dithersum = 0.5f;
+  for( y = 0 ; y < newheight ; y++ )
+  {
+    for( x = 0 ; x < newwidth ; x++, src += bytesperpixel, dst += bytesperpixel )
+    {
+      work( dst, src, bytesperpixel, bytesperline, &dithersum );
+      src += bytesperpixel;
+    }
+    src += rowoffset;
+  }
+  return;
+}
+
+
+int imReduceImageHalfBoxData( unsigned char *dstdata, unsigned char *srcdata, int width, int height, int bytesperpixel, int bytesperline, imReduceOptions *options )
+{
+  int filter, retval;
+
+  filter = options->filter;
+  retval = 1;
+  if( ( filter == IM_REDUCE_FILTER_LINEAR ) || ( filter == IM_REDUCE_FILTER_LINEAR_ALPHANORM ) )
+  {
+    if( bytesperpixel == 4 )
+      imReduceImageHalfBoxWork( dstdata, srcdata, width, height, bytesperpixel, bytesperline, imReduceHalfBox4Linear );
+    else if( bytesperpixel == 3 )
+      imReduceImageHalfBoxWork( dstdata, srcdata, width, height, bytesperpixel, bytesperline, imReduceHalfBox3Linear );
+    else if( bytesperpixel == 2 )
+      imReduceImageHalfBoxWork( dstdata, srcdata, width, height, bytesperpixel, bytesperline, imReduceHalfBox2Linear );
+    else if( bytesperpixel == 1 )
+      imReduceImageHalfBoxWork( dstdata, srcdata, width, height, bytesperpixel, bytesperline, imReduceHalfBox1Linear );
+    else
+      retval = 0;
+  }
+  else if( ( filter == IM_REDUCE_FILTER_SRGB ) || ( filter == IM_REDUCE_FILTER_SRGB_ALPHANORM ) )
+  {
+    if( bytesperpixel == 4 )
+      imReduceImageHalfBoxWork( dstdata, srcdata, width, height, bytesperpixel, bytesperline, imReduceHalfBox4sRGB );
+    else if( bytesperpixel == 3 )
+      imReduceImageHalfBoxWork( dstdata, srcdata, width, height, bytesperpixel, bytesperline, imReduceHalfBox3sRGB );
+    else if( bytesperpixel == 2 )
+      imReduceImageHalfBoxWork( dstdata, srcdata, width, height, bytesperpixel, bytesperline, imReduceHalfBox2sRGB );
+    else if( bytesperpixel == 1 )
+      imReduceImageHalfBoxWork( dstdata, srcdata, width, height, bytesperpixel, bytesperline, imReduceHalfBox1sRGB );
+    else
+      retval = 0;
+  }
+  else if( ( filter == IM_REDUCE_FILTER_NORMALMAP ) || ( filter == IM_REDUCE_FILTER_NORMALMAP_ALPHANORM ) || ( filter == IM_REDUCE_FILTER_NORMALMAP_SUSTAIN ) || ( filter == IM_REDUCE_FILTER_NORMALMAP_SUSTAIN_ALPHANORM ) )
+  {
+    if( bytesperpixel == 4 )
+      imReduceImageHalfBoxWork( dstdata, srcdata, width, height, bytesperpixel, bytesperline, imReduceHalfBox4Normal );
+    else if( bytesperpixel == 3 )
+      imReduceImageHalfBoxWork( dstdata, srcdata, width, height, bytesperpixel, bytesperline, imReduceHalfBox3Normal );
+    else
+      retval = 0;
+  }
+  else if( filter == IM_REDUCE_FILTER_WATERMAP )
+  {
+    if( bytesperpixel == 4 )
+      imReduceImageHalfBoxWork( dstdata, srcdata, width, height, bytesperpixel, bytesperline, imReduceHalfBox4Water );
+    else if( bytesperpixel == 3 )
+      imReduceImageHalfBoxWork( dstdata, srcdata, width, height, bytesperpixel, bytesperline, imReduceHalfBox3Water );
+    else
+      retval = 0;
+  }
+  else if( filter == IM_REDUCE_FILTER_PLANTMAP )
+  {
+    if( bytesperpixel == 4 )
+      imReduceImageHalfBoxWork( dstdata, srcdata, width, height, bytesperpixel, bytesperline, imReduceHalfBox4Plant );
+    else
+      retval = 0;
+  }
+  else
+    retval = 0;
+
+  return retval;
+}
+
+
+int imReduceImageHalfBox( imgImage *imgdst, imgImage *imgsrc, imReduceOptions *options )
+{
+  int newwidth, newheight, retvalue;
+
+  newwidth = ( ( imgsrc->format.width < 2 ) ? 1 : ( ( imgsrc->format.width + 1 ) / 2 ) );
+  newheight = ( ( imgsrc->format.height < 2 ) ? 1 : ( ( imgsrc->format.height + 1 ) / 2 ) );
+
+  imgdst->format.width = newwidth;
+  imgdst->format.height = newheight;
+  imgdst->format.type = imgsrc->format.type;
+  imgdst->format.bytesperpixel = imgsrc->format.bytesperpixel;
+  imgdst->format.bytesperline = imgdst->format.width * imgdst->format.bytesperpixel;
+  imgdst->data = malloc( imgdst->format.height * imgdst->format.bytesperline );
+
+  retvalue = imReduceImageHalfBoxData( imgdst->data, imgsrc->data, imgsrc->format.width, imgsrc->format.height, imgsrc->format.bytesperpixel, imgsrc->format.bytesperline, options );
+
+  return retvalue;
+}
+
+
+
+////////////////////////////////////////////////////////////////////////////////
+
+
+
+int imBuildMipmapCascade( imMipmapCascade *cascade, void *imagedata, int width, int height, int layercount, int bytesperpixel, int bytesperline, imReduceOptions *options, int cascadeflags )
+{
+  int layerindex, level, srclevel, srcwidth, srcheight, method, divisor;
+  int levelwidth, levelheight;
+  void *src, *dst;
+
+  cascade->width = width;
+  cascade->height = height;
+  cascade->layercount = layercount;
+  cascade->bytesperpixel = bytesperpixel;
+  cascade->bytesperline = bytesperline;
+  cascade->options = options;
+
+  /* No need for mipmaps */
+  if( ( cascade->width == 1 ) && ( cascade->height == 1 ) )
+    return 1;
+  if( bytesperpixel != 4 )
+    cascadeflags &= ~( IM_CASCADE_FLAGS_COLOR_BORDER_BASE | IM_CASCADE_FLAGS_COLOR_BORDER_MIPMAPS );
+
+  /* Allocate all the mipmap levels */
+  if( !( layercount ) )
+    layercount = 1;
+  cascade->mipmap[0] = imagedata;
+  levelwidth = cascade->width;
+  levelheight = cascade->height;
+  for( level = 1 ; ; level++ )
+  {
+    levelwidth = ( levelwidth < 2 ) ? 1 : ( levelwidth >> 1 );
+    levelheight = ( levelheight < 2 ) ? 1 : ( levelheight >> 1 );
+    if( !( cascade->mipmap[level] = malloc( levelwidth * levelheight * layercount * bytesperpixel ) ) )
+      return 0;
+    if( ( levelwidth == 1 ) && ( levelheight == 1 ) )
+      break;
+  }
+  cascade->mipmap[level+1] = 0;
+
+  if( cascadeflags & IM_CASCADE_FLAGS_COLOR_BORDER_BASE )
+    imPropagateAlphaBorder( imagedata, width, height * layercount, bytesperpixel, bytesperline );
+
+  /* For every layer, compute all its mipmap */
+  for( layerindex = 0 ; layerindex < layercount ; layerindex++ )
+  {
+    levelwidth = cascade->width;
+    levelheight = cascade->height;
+    for( level = 1 ; cascade->mipmap[level] ; level++ )
+    {
+      levelwidth = ( levelwidth < 2 ) ? 1 : ( levelwidth >> 1 );
+      levelheight = ( levelheight < 2 ) ? 1 : ( levelheight >> 1 );
+      dst = ADDRESS( cascade->mipmap[level], layerindex * levelwidth * levelheight * bytesperpixel );
+
+      /* Decide what method and source level to pick */
+      if( ( levelwidth | levelheight ) >= 16 )
+      {
+        srclevel = level - 2;
+        if( srclevel < 0 )
+          srclevel = 0;
+        method = 1;
+      }
+      else
+      {
+        srclevel = level - 1;
+        method = 0;
+      }
+#if DEBUG_VERBOSE
+      printf( "Tex level %d, srclevel %d, layer %d, filter %d, method %d : %d x %d\n", level, srclevel, layerindex, options->filter, method, levelwidth, levelheight );
+#endif
+      srcwidth = width >> srclevel;
+      if( !( srcwidth ) )
+        srcwidth = 1;
+      srcheight = height >> srclevel;
+      if( !( srcheight ) )
+        srcheight = 1;
+      if( srclevel )
+        src = ADDRESS( cascade->mipmap[srclevel], layerindex * srcheight * srcwidth * bytesperpixel );
+      else
+        src = ADDRESS( cascade->mipmap[srclevel], layerindex * srcheight * cascade->bytesperline );
+
+      divisor = 1 << ( level - srclevel );
+      if( ( ( levelwidth * divisor ) != srcwidth ) || ( ( levelheight * divisor ) != srcheight ) )
+        method = 2;
+
+      if( method == 2 )
+      {
+        if( !( imReduceImageKaiserData( dst, src, srcwidth, srcheight, bytesperpixel, srcwidth * bytesperpixel, levelwidth, levelheight, options ) ) )
+        {
+          printf( "ERROR AT %s:%d\n", __FILE__, __LINE__ );
+          return 0;
+        }
+      }
+      else if( method == 1 )
+      {
+        if( !( imReduceImageKaiserDataDivisor( dst, src, srcwidth, srcheight, bytesperpixel, srcwidth * bytesperpixel, divisor, options ) ) )
+        {
+          printf( "ERROR AT %s:%d\n", __FILE__, __LINE__ );
+          return 0;
+        }
+      }
+      else
+      {
+        if( !( imReduceImageHalfBoxData( dst, src, srcwidth, srcheight, bytesperpixel, srcwidth * bytesperpixel, options ) ) )
+        {
+          printf( "ERROR AT %s:%d\n", __FILE__, __LINE__ );
+          return 0;
+        }
+      }
+
+      if( cascadeflags & IM_CASCADE_FLAGS_COLOR_BORDER_MIPMAPS )
+        imPropagateAlphaBorder( dst, levelwidth, levelheight, bytesperpixel, levelwidth * bytesperpixel );
+    }
+  }
+
+  return 1;
+}
+
+
+void imFreeMipmapCascade( imMipmapCascade *cascade )
+{
+  int level;
+  for( level = 1 ; ; level++ )
+  {
+    if( !( cascade->mipmap[level] ) )
+      break;
+    free( cascade->mipmap[level] );
+    cascade->mipmap[level] = 0;
+  }
+  return;
+}
+
+
+////
+
+
+#define IM_PIXEL_ALPHA_MASK (0xff000000)
+#define IM_PIXEL_RGB_MASK (0x00ffffff)
+
+void imPropagateAlphaBorder( unsigned char *imagedata, int width, int height, int bytesperpixel, int bytesperline )
+{
+  int x, y, backtrackflag;
+  uint32_t pixel, refcolor, prevrowpixel;
+  uint32_t *row, *prevrow;
+
+  if( bytesperpixel != 4 )
+    return;
+  row = (uint32_t *)imagedata;
+  prevrow = row;
+  for( y = 0 ; y < height ; y++ )
+  {
+    refcolor = 0;
+    backtrackflag = 0;
+    for( x = 0 ; x < width ; x++ )
+    {
+      pixel = row[x];
+      prevrowpixel = prevrow[x];
+      if( pixel & IM_PIXEL_ALPHA_MASK )
+      {
+        /* Pixel has some color, spread to neighbor if applicable */
+        refcolor = pixel & IM_PIXEL_RGB_MASK;
+        if( backtrackflag )
+        {
+          row[x-1] = refcolor;
+          backtrackflag = 0;
+        }
+        if( !( prevrowpixel & IM_PIXEL_ALPHA_MASK ) )
+          prevrow[x] = refcolor;
+      }
+      else
+      {
+        /* Pixel is fully transparent, spread from neighbor if applicable */
+        if( refcolor )
+        {
+          row[x] = refcolor;
+          backtrackflag = 0;
+          refcolor = 0;
+        }
+        else if( prevrowpixel & IM_PIXEL_ALPHA_MASK )
+        {
+          row[x] = prevrowpixel & IM_PIXEL_RGB_MASK;
+          backtrackflag = 0;
+        }
+        else
+          backtrackflag = 1;
+      }
+    }
+    prevrow = row;
+    row = ADDRESS( row, bytesperline );
+  }
+
+  return;
+}
+
+
diff --git a/lib/graphics_utils/mipmap/imgresize.h b/lib/graphics_utils/mipmap/imgresize.h
new file mode 100644
index 000000000..b47d07a6b
--- /dev/null
+++ b/lib/graphics_utils/mipmap/imgresize.h
@@ -0,0 +1,150 @@
+/* -----------------------------------------------------------------------------
+ *
+ * Copyright (c) 2014-2017 Alexis Naveros.
+ * Portions developed under contract to the SURVICE Engineering Company.
+ *
+ * This software is provided 'as-is', without any express or implied
+ * warranty. In no event will the authors be held liable for any damages
+ * arising from the use of this software.
+ *
+ * Permission is granted to anyone to use this software for any purpose,
+ * including commercial applications, and to alter it and redistribute it
+ * freely, subject to the following restrictions:
+ *
+ * 1. The origin of this software must not be misrepresented; you must not
+ * claim that you wrote the original software. If you use this software
+ * in a product, an acknowledgment in the product documentation would be
+ * appreciated but is not required.
+ * 2. Altered source versions must be plainly marked as such, and must not be
+ * misrepresented as being the original software.
+ * 3. This notice may not be removed or altered from any source distribution.
+ *
+ * -----------------------------------------------------------------------------
+ */
+
+
+#ifndef IMGRESIZE_H
+#define IMGRESIZE_H
+
+
+typedef struct
+{
+  /* Specify filter type, from the IM_REDUCE_FILTER_* list */
+  int filter;
+  /* High quality, a little slow: hopcount=3; */
+  /* Good quality, much faster: hopcount=2; */
+  int hopcount;
+  /* Strong preservation/amplification of details: alpha=2.0f; */
+  /* Mild preservation/amplification of details: alpha=6.0f; */
+  float alpha;
+  /* NORMALMAP filters: factor to amyplify normals on X and Y before normalization */
+  float amplifynormal;
+  /* NORMALMAP_SUSTAIN filters: Preserve a factor of deviation "energy" as calculated by sqrtf(x*x+y*y) */
+  float normalsustainfactor;
+} imReduceOptions;
+
+static inline void imReduceSetOptions( imReduceOptions *options, int filter, int hopcount, float alpha, float amplifynormal, float normalsustainfactor )
+{
+  options->filter = filter;
+  options->hopcount = hopcount;
+  options->alpha = alpha;
+  options->amplifynormal = amplifynormal;
+  options->normalsustainfactor = normalsustainfactor;
+  return;
+}
+
+
+/* Reduce the image's dimensions by an integer divisor ~ this is fairly fast */
+int imReduceImageKaiserDataDivisor( unsigned char *dstdata, unsigned char *srcdata, int width, int height, int bytesperpixel, int bytesperline, int sizedivisor, imReduceOptions *options );
+/* Same as imReduceImageKaiserDataDivisor(), but imgdst is allocated */
+int imReduceImageKaiserDivisor( imgImage *imgdst, imgImage *imgsrc, int sizedivisor, imReduceOptions *options );
+
+
+/* Reduce the image's dimensions to match the newwidth and newheight ~ this is a little slower */
+int imReduceImageKaiserData( unsigned char *dstdata, unsigned char *srcdata, int width, int height, int bytesperpixel, int bytesperline, int newwidth, int newheight, imReduceOptions *options );
+/* Same as imReduceImageKaiserData(), but imgdst is allocated */
+int imReduceImageKaiser( imgImage *imgdst, imgImage *imgsrc, int newwidth, int newheight, imReduceOptions *options );
+
+
+/* Resize by half with a dumb box filter ~ don't use that except for the smallest mipmaps */
+/* Filters with ALPHANORM and/or SUSTAIN keywords are processed as the regular base filter only */
+int imReduceImageHalfBoxData( unsigned char *dstdata, unsigned char *srcdata, int width, int height, int bytesperpixel, int bytesperline, imReduceOptions *options );
+int imReduceImageHalfBox( imgImage *imgdst, imgImage *imgsrc, imReduceOptions *options );
+
+
+/*
+Keywords for image reduction filters
+
+LINEAR: Data is linear, note that this is *not* the format of typical diffuse textures
+SRGB: Color is in sRGB space, any alpha is presumed linear
+NORMALMAP: RGB represents a XYZ vector as (2.0*RGB)-1.0f, any alpha is presumed linear
+
+ALPHANORM: Alpha normalization, the weight of pixels is proportional to their alpha values
+           (do you have "black" fully transparent pixels? please use an ALPHANORM filter)
+SUSTAIN: The "energy" of the normal map is sustained, amplified to preserve the level of details
+         Note that this filter is rather slow (set options->normalsustainfactor to 0.75 or so)
+*/
+
+enum
+{
+  /* Linear space */
+  IM_REDUCE_FILTER_LINEAR,
+  IM_REDUCE_FILTER_LINEAR_ALPHANORM,
+
+  /* sRGB space (probably what you want for diffuse textures) */
+  IM_REDUCE_FILTER_SRGB,
+  IM_REDUCE_FILTER_SRGB_ALPHANORM,
+
+  /* RGB represents a XYZ vector as (2.0*RGB)-1.0f, any alpha is presumed linear */
+  IM_REDUCE_FILTER_NORMALMAP,
+  IM_REDUCE_FILTER_NORMALMAP_ALPHANORM,
+  IM_REDUCE_FILTER_NORMALMAP_SUSTAIN,
+  IM_REDUCE_FILTER_NORMALMAP_SUSTAIN_ALPHANORM,
+
+  /* Custom specialized filters */
+  IM_REDUCE_FILTER_WATERMAP,
+  IM_REDUCE_FILTER_PLANTMAP,
+  IM_REDUCE_FILTER_FOLLIAGE,
+  IM_REDUCE_FILTER_SKY,
+  IM_REDUCE_FILTER_FOG
+};
+
+
+////
+
+
+#define IM_MIPMAP_CASCADE_MAX (16)
+
+typedef struct
+{
+  int width;
+  int height;
+  int layercount;
+  int bytesperpixel;
+  int bytesperline;
+  imReduceOptions *options;
+  void *mipmap[IM_MIPMAP_CASCADE_MAX];
+} imMipmapCascade;
+
+
+int imBuildMipmapCascade( imMipmapCascade *cascade, void *imagedata, int width, int height, int layercount, int bytesperpixel, int bytesperline, imReduceOptions *options, int cascadeflags );
+
+void imFreeMipmapCascade( imMipmapCascade *cascade );
+
+/* For base texture, propagate RGB channels to neighbors if they are fully transparent (ignored if bytesperpixel != 4 ) */
+#define IM_CASCADE_FLAGS_COLOR_BORDER_BASE (0x1)
+/* For generated mipmaps, propagate RGB channels to neighbors if they are fully transparent (ignored if bytesperpixel != 4 ) */
+#define IM_CASCADE_FLAGS_COLOR_BORDER_MIPMAPS (0x2)
+
+
+////
+
+
+void imPropagateAlphaBorder( unsigned char *imagedata, int width, int height, int bytesperpixel, int bytesperline );
+
+
+////
+
+
+#endif
+
diff --git a/lib/irrlicht/include/IImage.h b/lib/irrlicht/include/IImage.h
index a2a93b252..7918759b5 100644
--- a/lib/irrlicht/include/IImage.h
+++ b/lib/irrlicht/include/IImage.h
@@ -98,6 +98,8 @@ public:
 	//! fills the surface with given color
 	virtual void fill(const SColor &color) =0;
 
+	virtual void setDeleteMemory(bool val) = 0;
+
 	//! get the amount of Bits per Pixel of the given color format
 	static u32 getBitsPerPixelFromFormat(const ECOLOR_FORMAT format)
 	{
diff --git a/lib/irrlicht/include/IImageLoader.h b/lib/irrlicht/include/IImageLoader.h
index 5ff091d9e..ba0da6886 100644
--- a/lib/irrlicht/include/IImageLoader.h
+++ b/lib/irrlicht/include/IImageLoader.h
@@ -42,7 +42,9 @@ public:
 	//! Creates a surface from the file
 	/** \param file File handle to check.
 	\return Pointer to newly created image, or 0 upon error. */
-	virtual IImage* loadImage(io::IReadFile* file) const = 0;
+	virtual IImage* loadImage(io::IReadFile* file, bool skip_checking = false) const = 0;
+	virtual core::dimension2du getImageSize(io::IReadFile* file) const { return core::dimension2du(0, 0); }
+	virtual bool supportThreadedLoading() const { return false; }
 };
 
 
diff --git a/lib/irrlicht/include/ITexture.h b/lib/irrlicht/include/ITexture.h
index 0586ec673..950544ce5 100644
--- a/lib/irrlicht/include/ITexture.h
+++ b/lib/irrlicht/include/ITexture.h
@@ -198,6 +198,16 @@ public:
 	virtual u64 getHandle() = 0;
 
 	virtual void unloadHandle() {}
+
+	virtual u32 getTextureSize() const { return 0; }
+
+	virtual void threadedReload(void* ptr, void* param) const {}
+
+	virtual void threadedSubImage(void* ptr) const {}
+
+	virtual void cleanThreadedLoader() {}
+
+	virtual int getThreadedLoadTextureCounter() const { return 0; }
 protected:
 
 	//! Helper function, helps to get the desired texture creation format from the flags.
diff --git a/lib/irrlicht/include/IVideoDriver.h b/lib/irrlicht/include/IVideoDriver.h
index 3f3851425..77e854cf0 100644
--- a/lib/irrlicht/include/IVideoDriver.h
+++ b/lib/irrlicht/include/IVideoDriver.h
@@ -1186,7 +1186,7 @@ namespace video
 		\return The created image.
 		If you no longer need the image, you should call IImage::drop().
 		See IReferenceCounted::drop() for more information. */
-		virtual IImage* createImageFromFile(io::IReadFile* file) =0;
+		virtual IImage* createImageFromFile(io::IReadFile* file, video::IImageLoader** loader = NULL) =0;
 
 		//! Writes the provided image to a file.
 		/** Requires that there is a suitable image writer registered
diff --git a/lib/irrlicht/include/utfwrapping.h b/lib/irrlicht/include/utfwrapping.h
index 8d67c6ea3..604b3c7f6 100644
--- a/lib/irrlicht/include/utfwrapping.h
+++ b/lib/irrlicht/include/utfwrapping.h
@@ -104,9 +104,10 @@ bool breakable (wchar_t c)
 	if ((c > 12287 && c < 40960) || //Common CJK words
 		(c > 44031 && c < 55204)  || //Hangul
 		(c > 63743 && c < 64256)  || //More Chinese
-		c == 173 || c == L' ' || c == 0) //Soft hyphen and white space
-			return true;
-			return false;
+		c == 173 || c == L' ' || //Soft hyphen and white space
+		c == 47 || c == 92) //Slash and blackslash
+		return true;
+	return false;
 }
 } // end namespace gui
 } // end namespace irr
diff --git a/lib/irrlicht/source/Irrlicht/CImage.cpp b/lib/irrlicht/source/Irrlicht/CImage.cpp
index d5677320e..60ff3702a 100644
--- a/lib/irrlicht/source/Irrlicht/CImage.cpp
+++ b/lib/irrlicht/source/Irrlicht/CImage.cpp
@@ -457,6 +457,10 @@ inline SColor CImage::getPixelBox( s32 x, s32 y, s32 fx, s32 fy, s32 bias ) cons
 	return c;
 }
 
+void CImage::setDeleteMemory(bool val)
+{
+	DeleteMemory = val;
+}
 
 } // end namespace video
 } // end namespace irr
diff --git a/lib/irrlicht/source/Irrlicht/CImage.h b/lib/irrlicht/source/Irrlicht/CImage.h
index 82b34c776..8eb8459bc 100644
--- a/lib/irrlicht/source/Irrlicht/CImage.h
+++ b/lib/irrlicht/source/Irrlicht/CImage.h
@@ -103,6 +103,8 @@ public:
 	//! fills the surface with given color
 	virtual void fill(const SColor &color);
 
+	virtual void setDeleteMemory(bool val);
+
 private:
 
 	//! assumes format and size has been set and creates the rest
diff --git a/lib/irrlicht/source/Irrlicht/CImageLoaderBMP.cpp b/lib/irrlicht/source/Irrlicht/CImageLoaderBMP.cpp
index 2ccb64cca..7bed41196 100644
--- a/lib/irrlicht/source/Irrlicht/CImageLoaderBMP.cpp
+++ b/lib/irrlicht/source/Irrlicht/CImageLoaderBMP.cpp
@@ -216,7 +216,7 @@ void CImageLoaderBMP::decompress4BitRLE(u8*& bmpData, s32 size, s32 width, s32 h
 
 
 //! creates a surface from the file
-IImage* CImageLoaderBMP::loadImage(io::IReadFile* file) const
+IImage* CImageLoaderBMP::loadImage(io::IReadFile* file, bool skip_checking) const
 {
 	SBMPHeader header;
 
diff --git a/lib/irrlicht/source/Irrlicht/CImageLoaderBMP.h b/lib/irrlicht/source/Irrlicht/CImageLoaderBMP.h
index 9321ecc23..c5708dcc5 100644
--- a/lib/irrlicht/source/Irrlicht/CImageLoaderBMP.h
+++ b/lib/irrlicht/source/Irrlicht/CImageLoaderBMP.h
@@ -81,7 +81,7 @@ public:
 	virtual bool isALoadableFileFormat(io::IReadFile* file) const;
 
 	//! creates a surface from the file
-	virtual IImage* loadImage(io::IReadFile* file) const;
+	virtual IImage* loadImage(io::IReadFile* file, bool skip_checking = false) const;
 
 private:
 
diff --git a/lib/irrlicht/source/Irrlicht/CImageLoaderJPG.cpp b/lib/irrlicht/source/Irrlicht/CImageLoaderJPG.cpp
index 1811f31dd..f9e77dff0 100644
--- a/lib/irrlicht/source/Irrlicht/CImageLoaderJPG.cpp
+++ b/lib/irrlicht/source/Irrlicht/CImageLoaderJPG.cpp
@@ -135,7 +135,7 @@ bool CImageLoaderJPG::isALoadableFileFormat(io::IReadFile* file) const
 }
 
 //! creates a surface from the file
-IImage* CImageLoaderJPG::loadImage(io::IReadFile* file) const
+IImage* CImageLoaderJPG::loadImage(io::IReadFile* file, bool skip_checking) const
 {
 	#ifndef _IRR_COMPILE_WITH_LIBJPEG_
 	os::Printer::log("Can't load as not compiled with _IRR_COMPILE_WITH_LIBJPEG_:", file->getFileName(), ELL_DEBUG);
diff --git a/lib/irrlicht/source/Irrlicht/CImageLoaderJPG.h b/lib/irrlicht/source/Irrlicht/CImageLoaderJPG.h
index d3d0633f3..d15c06dbe 100644
--- a/lib/irrlicht/source/Irrlicht/CImageLoaderJPG.h
+++ b/lib/irrlicht/source/Irrlicht/CImageLoaderJPG.h
@@ -49,7 +49,7 @@ public:
 	virtual bool isALoadableFileFormat(io::IReadFile* file) const;
 
 	//! creates a surface from the file
-	virtual IImage* loadImage(io::IReadFile* file) const;
+	virtual IImage* loadImage(io::IReadFile* file, bool skip_checking = false) const;
 
 private:
 
diff --git a/lib/irrlicht/source/Irrlicht/CImageLoaderPNG.cpp b/lib/irrlicht/source/Irrlicht/CImageLoaderPNG.cpp
index e612330ce..b7478c541 100644
--- a/lib/irrlicht/source/Irrlicht/CImageLoaderPNG.cpp
+++ b/lib/irrlicht/source/Irrlicht/CImageLoaderPNG.cpp
@@ -27,14 +27,14 @@ namespace video
 // PNG function for error handling
 static void png_cpexcept_error(png_structp png_ptr, png_const_charp msg)
 {
-	os::Printer::log("PNG fatal error", msg, ELL_ERROR);
+	printf("PNG fatal error: %s\n", msg);
 	longjmp(png_jmpbuf(png_ptr), 1);
 }
 
 // PNG function for warning handling
 static void png_cpexcept_warn(png_structp png_ptr, png_const_charp msg)
 {
-	os::Printer::log("PNG warning", msg, ELL_WARNING);
+	//os::Printer::log("PNG warning", msg, ELL_WARNING);
 }
 
 // PNG function for file reading
@@ -86,7 +86,7 @@ bool CImageLoaderPng::isALoadableFileFormat(io::IReadFile* file) const
 
 
 // load in the image data
-IImage* CImageLoaderPng::loadImage(io::IReadFile* file) const
+IImage* CImageLoaderPng::loadImage(io::IReadFile* file, bool skip_checking) const
 {
 #ifdef _IRR_COMPILE_WITH_LIBPNG_
 	if (!file)
@@ -96,27 +96,17 @@ IImage* CImageLoaderPng::loadImage(io::IReadFile* file) const
 	//Used to point to image rows
 	u8** RowPointers = 0;
 
-	png_byte buffer[8];
-	// Read the first few bytes of the PNG file
-	if( file->read(buffer, 8) != 8 )
-	{
-		os::Printer::log("LOAD PNG: can't read file\n", file->getFileName(), ELL_ERROR);
+	if (skip_checking)
+		file->seek(8);
+	else if (!isALoadableFileFormat(file))
 		return 0;
-	}
-
-	// Check if it really is a PNG file
-	if( png_sig_cmp(buffer, 0, 8) )
-	{
-		os::Printer::log("LOAD PNG: not really a png\n", file->getFileName(), ELL_ERROR);
-		return 0;
-	}
 
 	// Allocate the png read struct
 	png_structp png_ptr = png_create_read_struct(PNG_LIBPNG_VER_STRING,
 		NULL, (png_error_ptr)png_cpexcept_error, (png_error_ptr)png_cpexcept_warn);
 	if (!png_ptr)
 	{
-		os::Printer::log("LOAD PNG: Internal PNG create read struct failure\n", file->getFileName(), ELL_ERROR);
+		//os::Printer::log("LOAD PNG: Internal PNG create read struct failure\n", file->getFileName(), ELL_ERROR);
 		return 0;
 	}
 
@@ -124,7 +114,7 @@ IImage* CImageLoaderPng::loadImage(io::IReadFile* file) const
 	png_infop info_ptr = png_create_info_struct(png_ptr);
 	if (!info_ptr)
 	{
-		os::Printer::log("LOAD PNG: Internal PNG create info struct failure\n", file->getFileName(), ELL_ERROR);
+		//os::Printer::log("LOAD PNG: Internal PNG create info struct failure\n", file->getFileName(), ELL_ERROR);
 		png_destroy_read_struct(&png_ptr, NULL, NULL);
 		return 0;
 	}
@@ -184,20 +174,6 @@ IImage* CImageLoaderPng::loadImage(io::IReadFile* file) const
 	if (ColorType==PNG_COLOR_TYPE_GRAY || ColorType==PNG_COLOR_TYPE_GRAY_ALPHA)
 		png_set_gray_to_rgb(png_ptr);
 
-	int intent;
-	const double screen_gamma = 2.2;
-
-	if (png_get_sRGB(png_ptr, info_ptr, &intent))
-		png_set_gamma(png_ptr, screen_gamma, 0.45455);
-	else
-	{
-		double image_gamma;
-		if (png_get_gAMA(png_ptr, info_ptr, &image_gamma))
-			png_set_gamma(png_ptr, screen_gamma, image_gamma);
-		else
-			png_set_gamma(png_ptr, screen_gamma, 0.45455);
-	}
-
 	// Update the changes in between, as we need to get the new color type
 	// for proper processing of the RGBA type
 	png_read_update_info(png_ptr, info_ptr);
@@ -229,7 +205,7 @@ IImage* CImageLoaderPng::loadImage(io::IReadFile* file) const
 		image = new CImage(ECF_R8G8B8, core::dimension2d<u32>(Width, Height));
 	if (!image)
 	{
-		os::Printer::log("LOAD PNG: Internal PNG create image struct failure\n", file->getFileName(), ELL_ERROR);
+		//os::Printer::log("LOAD PNG: Internal PNG create image struct failure\n", file->getFileName(), ELL_ERROR);
 		png_destroy_read_struct(&png_ptr, NULL, NULL);
 		return 0;
 	}
@@ -238,7 +214,7 @@ IImage* CImageLoaderPng::loadImage(io::IReadFile* file) const
 	RowPointers = new png_bytep[Height];
 	if (!RowPointers)
 	{
-		os::Printer::log("LOAD PNG: Internal PNG create row pointers failure\n", file->getFileName(), ELL_ERROR);
+		//os::Printer::log("LOAD PNG: Internal PNG create row pointers failure\n", file->getFileName(), ELL_ERROR);
 		png_destroy_read_struct(&png_ptr, NULL, NULL);
 		delete image;
 		return 0;
@@ -276,6 +252,26 @@ IImage* CImageLoaderPng::loadImage(io::IReadFile* file) const
 #endif // _IRR_COMPILE_WITH_LIBPNG_
 }
 
+core::dimension2du CImageLoaderPng::getImageSize(io::IReadFile* file) const
+{
+#ifdef _IRR_COMPILE_WITH_LIBPNG_
+	if (!file || !isALoadableFileFormat(file))
+		return core::dimension2du(0, 0);
+	core::dimension2d<u32> dim;
+	file->seek(16);
+	file->read(&dim.Width, 4);
+	file->seek(20);
+	file->read(&dim.Height, 4);
+	file->seek(0);
+#ifndef __BIG_ENDIAN__
+	dim.Width = os::Byteswap::byteswap(dim.Width);
+	dim.Height = os::Byteswap::byteswap(dim.Height);
+#endif
+	return dim;
+#else
+	return core::dimension2du(0, 0);
+#endif // _IRR_COMPILE_WITH_LIBPNG_
+}
 
 IImageLoader* createImageLoaderPNG()
 {
@@ -287,4 +283,3 @@ IImageLoader* createImageLoaderPNG()
 }//end namespace video
 
 #endif
-
diff --git a/lib/irrlicht/source/Irrlicht/CImageLoaderPNG.h b/lib/irrlicht/source/Irrlicht/CImageLoaderPNG.h
index a3faac2ab..789cfdd18 100644
--- a/lib/irrlicht/source/Irrlicht/CImageLoaderPNG.h
+++ b/lib/irrlicht/source/Irrlicht/CImageLoaderPNG.h
@@ -25,15 +25,17 @@ class CImageLoaderPng : public IImageLoader
 {
 public:
 
-   //! returns true if the file maybe is able to be loaded by this class
-   //! based on the file extension (e.g. ".png")
-   virtual bool isALoadableFileExtension(const io::path& filename) const;
+	//! returns true if the file maybe is able to be loaded by this class
+	//! based on the file extension (e.g. ".png")
+	virtual bool isALoadableFileExtension(const io::path& filename) const;
 
-   //! returns true if the file maybe is able to be loaded by this class
-   virtual bool isALoadableFileFormat(io::IReadFile* file) const;
+	//! returns true if the file maybe is able to be loaded by this class
+	virtual bool isALoadableFileFormat(io::IReadFile* file) const;
 
-   //! creates a surface from the file
-   virtual IImage* loadImage(io::IReadFile* file) const;
+	//! creates a surface from the file
+	virtual IImage* loadImage(io::IReadFile* file, bool skip_checking = false) const;
+	virtual core::dimension2du getImageSize(io::IReadFile* file) const;
+	virtual bool supportThreadedLoading() const { return true; }
 };
 
 
diff --git a/lib/irrlicht/source/Irrlicht/CNullDriver.cpp b/lib/irrlicht/source/Irrlicht/CNullDriver.cpp
index 4ef9c44de..404e410c3 100644
--- a/lib/irrlicht/source/Irrlicht/CNullDriver.cpp
+++ b/lib/irrlicht/source/Irrlicht/CNullDriver.cpp
@@ -103,6 +103,7 @@ CNullDriver::CNullDriver(io::IFileSystem* io, const core::dimension2d<u32>& scre
 //	DriverAttributes->addInt("MaxGeometryVerticesOut", 0);
 //	DriverAttributes->addFloat("MaxTextureLODBias", 0.f);
 	DriverAttributes->addInt("Version", 1);
+	DriverAttributes->setAttribute("MAX_TEXTURE_SIZE", core::dimension2du(2048, 2048));
 //	DriverAttributes->addInt("ShaderLanguageVersion", 0);
 //	DriverAttributes->addInt("AntiAlias", 0);
 
@@ -1278,7 +1279,7 @@ IImage* CNullDriver::createImageFromFile(const io::path& filename)
 
 
 //! Creates a software image from a file.
-IImage* CNullDriver::createImageFromFile(io::IReadFile* file)
+IImage* CNullDriver::createImageFromFile(io::IReadFile* file, video::IImageLoader** loader)
 {
 	if (!file)
 		return 0;
@@ -1292,6 +1293,11 @@ IImage* CNullDriver::createImageFromFile(io::IReadFile* file)
 	{
 		if (SurfaceLoader[i]->isALoadableFileExtension(file->getFileName()))
 		{
+			if (loader)
+			{
+				*loader = SurfaceLoader[i];
+				return 0;
+			}
 			// reset file position which might have changed due to previous loadImage calls
 			file->seek(0);
 			image = SurfaceLoader[i]->loadImage(file);
@@ -1307,6 +1313,11 @@ IImage* CNullDriver::createImageFromFile(io::IReadFile* file)
 		file->seek(0);
 		if (SurfaceLoader[i]->isALoadableFileFormat(file))
 		{
+			if (loader)
+			{
+				*loader = SurfaceLoader[i];
+				return 0;
+			}
 			file->seek(0);
 			image = SurfaceLoader[i]->loadImage(file);
 			if (image)
diff --git a/lib/irrlicht/source/Irrlicht/CNullDriver.h b/lib/irrlicht/source/Irrlicht/CNullDriver.h
index 7a10cccad..3d19cfc86 100644
--- a/lib/irrlicht/source/Irrlicht/CNullDriver.h
+++ b/lib/irrlicht/source/Irrlicht/CNullDriver.h
@@ -348,7 +348,7 @@ namespace video
 		virtual IImage* createImageFromFile(const io::path& filename);
 
 		//! Creates a software image from a file.
-		virtual IImage* createImageFromFile(io::IReadFile* file);
+		virtual IImage* createImageFromFile(io::IReadFile* file, video::IImageLoader** loader = NULL);
 
 		//! Creates a software image from a byte array.
 		/** \param useForeignMemory: If true, the image will use the data pointer
diff --git a/sources.cmake b/sources.cmake
index ddc029d4f..ba4868d71 100644
--- a/sources.cmake
+++ b/sources.cmake
@@ -1,5 +1,5 @@
-# Modify this file to change the last-modified date when you add/remove a file. 
-# This will then trigger a new cmake run automatically.
+# Modify this file to change the last-modified date when you add/remove a file.
+# This will then trigger a new cmake run automatically. 
 file(GLOB_RECURSE STK_HEADERS RELATIVE ${CMAKE_CURRENT_SOURCE_DIR} "src/*.hpp")
 file(GLOB_RECURSE STK_SOURCES RELATIVE ${CMAKE_CURRENT_SOURCE_DIR} "src/*.cpp")
 file(GLOB_RECURSE STK_SHADERS RELATIVE ${CMAKE_CURRENT_SOURCE_DIR} "data/shaders/*")
diff --git a/src/config/hardware_stats.cpp b/src/config/hardware_stats.cpp
index f657c6409..d3f532f86 100644
--- a/src/config/hardware_stats.cpp
+++ b/src/config/hardware_stats.cpp
@@ -102,7 +102,7 @@ int getRAM()
  */
 int getNumProcessors()
 {
-#ifdef __linux__
+#if defined(__linux__) || defined(__CYGWIN__)
     return sysconf(_SC_NPROCESSORS_CONF);
 #endif
 #ifdef WIN32
diff --git a/src/config/hardware_stats.hpp b/src/config/hardware_stats.hpp
index d2dfd0044..7558c040c 100644
--- a/src/config/hardware_stats.hpp
+++ b/src/config/hardware_stats.hpp
@@ -82,6 +82,7 @@ namespace HardwareStats
     // ========================================================================
     void reportHardwareStats();
     const std::string& getOSVersion();
+    int getNumProcessors();
 };   // HardwareStats
 
 #endif
diff --git a/src/config/user_config.hpp b/src/config/user_config.hpp
index 6bdfa1ae8..0ecc320e3 100644
--- a/src/config/user_config.hpp
+++ b/src/config/user_config.hpp
@@ -553,6 +553,33 @@ namespace UserConfigParams
         &m_video_group, "Max texture size when high definition textures are "
                         "disabled"));
 
+    PARAM_PREFIX BoolUserConfigParam        m_hq_mipmap
+        PARAM_DEFAULT(BoolUserConfigParam(false, "hq_mipmap",
+        &m_video_group, "Generate mipmap for textures using "
+                        "high quality method with SSE"));
+                        
+    // ---- Recording
+    PARAM_PREFIX GroupUserConfigParam        m_recording_group
+        PARAM_DEFAULT( GroupUserConfigParam("Recording", "Recording Settings") );
+
+    PARAM_PREFIX BoolUserConfigParam        m_record_bmp
+        PARAM_DEFAULT(BoolUserConfigParam(false, "record_bmp",
+        &m_recording_group, "Record video using uncompressed bitmap, notice: this "
+                        "will require a lot of space and fast disk access."));
+
+    PARAM_PREFIX BoolUserConfigParam        m_limit_game_fps
+        PARAM_DEFAULT(BoolUserConfigParam(true, "limit_game_fps",
+        &m_recording_group, "Limit game framerate not beyond the fps of recording "
+                        "video."));
+
+    PARAM_PREFIX IntUserConfigParam         m_record_compression
+        PARAM_DEFAULT(IntUserConfigParam(90, "record_compression",
+        &m_recording_group, "Specify the compression level of recording video"));
+
+    PARAM_PREFIX IntUserConfigParam         m_record_fps
+        PARAM_DEFAULT(IntUserConfigParam(30, "record_fps",
+        &m_recording_group, "Specify the fps of recording video"));
+
     // ---- Debug - not saved to config file
     /** If gamepad debugging is enabled. */
     PARAM_PREFIX bool m_unit_testing PARAM_DEFAULT(false);
@@ -617,9 +644,6 @@ namespace UserConfigParams
     /** True if graphical profiler should be displayed */
     PARAM_PREFIX bool m_profiler_enabled  PARAM_DEFAULT( false );
 
-    /** True if hardware skinning should be enabled */
-    PARAM_PREFIX bool m_hw_skinning_enabled  PARAM_DEFAULT( false );
-
     // not saved to file
 
     // ---- Networking
diff --git a/src/graphics/camera_normal.cpp b/src/graphics/camera_normal.cpp
index 0c0e1aa65..2e4c12bc2 100644
--- a/src/graphics/camera_normal.cpp
+++ b/src/graphics/camera_normal.cpp
@@ -127,8 +127,13 @@ void CameraNormal::smoothMoveCamera(float dt)
         delta2 = 1;
 
     btTransform btt = m_kart->getTrans();
-    m_kart_position = btt.getOrigin();// m_kart_position + (btt.getOrigin() - m_kart_position) * delta2;
-    m_kart_rotation = m_kart_rotation.normalized().slerp(btt.getRotation().normalized(), delta2);
+    m_kart_position = btt.getOrigin();
+    btQuaternion q1, q2;
+    q1 = m_kart_rotation.normalized();
+    q2 = btt.getRotation().normalized();
+    if (dot(q1, q2) < 0.0f)
+        q2 = -q2;
+    m_kart_rotation = q1.slerp(q2, delta2);
 
     btt.setOrigin(m_kart_position);
     btt.setRotation(m_kart_rotation);
diff --git a/src/graphics/central_settings.cpp b/src/graphics/central_settings.cpp
index d3ce43759..18b07ca2d 100644
--- a/src/graphics/central_settings.cpp
+++ b/src/graphics/central_settings.cpp
@@ -52,6 +52,7 @@ void CentralVideoSettings::init()
     hasGS = false;
     hasTextureFilterAnisotropic = false;
     hasTextureSwizzle = false;
+    hasPixelBufferObject = false;
 
 #if defined(USE_GLES2)
     hasBGRA = false;
@@ -196,6 +197,11 @@ void CentralVideoSettings::init()
             hasTextureSwizzle = true;
             Log::info("GLDriver", "ARB Texture Swizzle Present");
         }
+        if (hasGLExtension("GL_ARB_pixel_buffer_object"))
+        {
+            hasPixelBufferObject = true;
+            Log::info("GLDriver", "ARB Pixel Buffer Object Present");
+        }
         // Only unset the high def textures if they are set as default. If the
         // user has enabled them (bit 1 set), then leave them enabled.
         if (GraphicsRestrictions::isDisabled(GraphicsRestrictions::GR_HIGHDEFINITION_TEXTURES) &&
@@ -238,7 +244,7 @@ void CentralVideoSettings::init()
             hasTextureStorage = true;
             hasTextureSwizzle = true;
         }
-        
+
         if (!GraphicsRestrictions::isDisabled(GraphicsRestrictions::GR_EXPLICIT_ATTRIB_LOCATION) &&
             m_glsl == true)
         {
@@ -476,4 +482,14 @@ bool CentralVideoSettings::isARBTextureSwizzleUsable() const
     return m_glsl && hasTextureSwizzle;
 }
 
+bool CentralVideoSettings::isARBPixelBufferObjectUsable() const
+{
+    return hasPixelBufferObject;
+}
+
+bool CentralVideoSettings::supportsThreadedTextureLoading() const
+{
+    return isARBPixelBufferObjectUsable() && isARBBufferStorageUsable() && isARBTextureStorageUsable();
+}
+
 #endif   // !SERVER_ONLY
diff --git a/src/graphics/central_settings.hpp b/src/graphics/central_settings.hpp
index 146160e0b..c2f0c4937 100644
--- a/src/graphics/central_settings.hpp
+++ b/src/graphics/central_settings.hpp
@@ -44,6 +44,7 @@ private:
     bool hasMultiDrawIndirect;
     bool hasTextureFilterAnisotropic;
     bool hasTextureSwizzle;
+    bool hasPixelBufferObject;
 
 #if defined(USE_GLES2)
     bool hasBGRA;
@@ -84,6 +85,7 @@ public:
     bool isARBExplicitAttribLocationUsable() const;
     bool isEXTTextureFilterAnisotropicUsable() const;
     bool isARBTextureSwizzleUsable() const;
+    bool isARBPixelBufferObjectUsable() const;
 
 #if defined(USE_GLES2)
     bool isEXTTextureFormatBGRA8888Usable() const;
@@ -98,6 +100,7 @@ public:
     bool supportsComputeShadersFiltering() const;
     bool supportsAsyncInstanceUpload() const;
     bool supportsHardwareSkinning() const;
+    bool supportsThreadedTextureLoading() const;
 
     // "Macro" around feature support and user config
     bool isShadowEnabled() const;
diff --git a/src/graphics/hq_mipmap_generator.cpp b/src/graphics/hq_mipmap_generator.cpp
new file mode 100644
index 000000000..d0531d9b4
--- /dev/null
+++ b/src/graphics/hq_mipmap_generator.cpp
@@ -0,0 +1,120 @@
+//  SuperTuxKart - a fun racing game with go-kart
+//  Copyright (C) 2017 SuperTuxKart-Team
+//
+//  This program is free software; you can redistribute it and/or
+//  modify it under the terms of the GNU General Public License
+//  as published by the Free Software Foundation; either version 3
+//  of the License, or (at your option) any later version.
+//
+//  This program 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.
+//
+//  You should have received a copy of the GNU General Public License
+//  along with this program; if not, write to the Free Software
+//  Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
+
+#if !(defined(SERVER_ONLY) || defined(USE_GLES2))
+
+#include "graphics/hq_mipmap_generator.hpp"
+#include "graphics/stk_tex_manager.hpp"
+#undef DUMP_MIPMAP
+#ifdef DUMP_MIPMAP
+#include "graphics/irr_driver.hpp"
+#include "utils/string_utils.hpp"
+#endif
+#include <cassert>
+
+extern "C"
+{
+    #include <mipmap/img.h>
+    #include <mipmap/imgresize.h>
+}
+
+// ----------------------------------------------------------------------------
+HQMipmapGenerator::HQMipmapGenerator(const io::path& name, uint8_t* data,
+                                     const core::dimension2d<u32>& size,
+                                     GLuint texture_name, TexConfig* tc)
+                 : video::ITexture(name), m_orig_data(data), m_size(size),
+                   m_texture_name(texture_name), m_texture_size(0),
+                   m_mipmap_data(NULL), m_tex_config(tc)
+{
+    assert(m_tex_config != NULL);
+    unsigned width = m_size.Width;
+    unsigned height = m_size.Height;
+    while (true)
+    {
+        width = width < 2 ? 1 : width >> 1;
+        height = height < 2 ? 1 : height >> 1;
+        m_mipmap_sizes.emplace_back(core::dimension2du(width, height),
+            m_texture_size);
+        m_texture_size += width * height * 4;
+        if (width == 1 && height == 1)
+            break;
+    }
+    m_texture_size = unsigned(m_mipmap_sizes.back().second) + 4;
+    m_mipmap_data = malloc(sizeof(imMipmapCascade));
+}   // HQMipmapGenerator
+
+// ----------------------------------------------------------------------------
+void HQMipmapGenerator::threadedReload(void* ptr, void* param) const
+{
+    imReduceOptions options;
+    imReduceSetOptions(&options,
+        m_tex_config->m_normal_map ?
+        IM_REDUCE_FILTER_NORMALMAP: m_tex_config->m_srgb ?
+        IM_REDUCE_FILTER_SRGB : IM_REDUCE_FILTER_LINEAR/*filter*/,
+        2/*hopcount*/, 2.0f/*alpha*/, 1.0f/*amplifynormal*/,
+        0.0f/*normalsustainfactor*/);
+    imMipmapCascade* mm_cascade = (imMipmapCascade*)m_mipmap_data;
+#ifdef DEBUG
+    int ret = imBuildMipmapCascade(mm_cascade, m_orig_data, m_size.Width,
+        m_size.Height, 1/*layercount*/, 4, m_size.Width * 4, &options, 0);
+    assert(ret == 1);
+#else
+    imBuildMipmapCascade(mm_cascade, m_orig_data, m_size.Width,
+        m_size.Height, 1/*layercount*/, 4, m_size.Width * 4, &options, 0);
+#endif
+    for (unsigned int i = 0; i < m_mipmap_sizes.size(); i++)
+    {
+        const unsigned size = m_mipmap_sizes[i].first.getArea() *  4;
+        memcpy((uint8_t*)ptr + m_mipmap_sizes[i].second,
+            mm_cascade->mipmap[i + 1], size);
+#ifdef DUMP_MIPMAP
+        video::IImage* image = irr_driver->getVideoDriver()
+            ->createImageFromData(video::ECF_A8R8G8B8, m_mipmap_sizes[i].first,
+            mm_cascade->mipmap[i + 1], false/*ownForeignMemory*/);
+        irr_driver->getVideoDriver()->writeImageToFile(image, std::string
+            (StringUtils::toString(i) + "_" +
+            StringUtils::getBasename(NamedPath.getPtr())).c_str());
+        image->drop();
+#endif
+    }
+}   // threadedReload
+
+// ----------------------------------------------------------------------------
+void HQMipmapGenerator::threadedSubImage(void* ptr) const
+{
+#if !(defined(SERVER_ONLY) || defined(USE_GLES2))
+    glBindTexture(GL_TEXTURE_2D, m_texture_name);
+    for (unsigned int i = 0; i < m_mipmap_sizes.size(); i++)
+    {
+        glTexSubImage2D(GL_TEXTURE_2D, i + 1, 0, 0,
+            m_mipmap_sizes[i].first.Width, m_mipmap_sizes[i].first.Height,
+            GL_BGRA, GL_UNSIGNED_BYTE,
+            (uint8_t*)ptr + m_mipmap_sizes[i].second);
+    }
+    delete this;
+#endif
+}   // threadedSubImage
+
+// ----------------------------------------------------------------------------
+void HQMipmapGenerator::cleanThreadedLoader()
+{
+    delete[] m_orig_data;
+    imFreeMipmapCascade((imMipmapCascade*)m_mipmap_data);
+    free(m_mipmap_data);
+}   // cleanThreadedLoader
+
+#endif
diff --git a/src/graphics/hq_mipmap_generator.hpp b/src/graphics/hq_mipmap_generator.hpp
new file mode 100644
index 000000000..189827637
--- /dev/null
+++ b/src/graphics/hq_mipmap_generator.hpp
@@ -0,0 +1,103 @@
+//  SuperTuxKart - a fun racing game with go-kart
+//  Copyright (C) 2017 SuperTuxKart-Team
+//
+//  This program is free software; you can redistribute it and/or
+//  modify it under the terms of the GNU General Public License
+//  as published by the Free Software Foundation; either version 3
+//  of the License, or (at your option) any later version.
+//
+//  This program 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.
+//
+//  You should have received a copy of the GNU General Public License
+//  along with this program; if not, write to the Free Software
+//  Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
+
+#ifndef HEADER_HQ_MIPMAP_GENERATOR_HPP
+#define HEADER_HQ_MIPMAP_GENERATOR_HPP
+
+#if !(defined(SERVER_ONLY) || defined(USE_GLES2))
+
+#include "graphics/gl_headers.hpp"
+#include "utils/no_copy.hpp"
+#include "utils/types.hpp"
+
+#include <vector>
+#include <ITexture.h>
+
+using namespace irr;
+struct TexConfig;
+
+class HQMipmapGenerator : public video::ITexture, NoCopy
+{
+private:
+    uint8_t* m_orig_data;
+
+    core::dimension2d<u32> m_size;
+
+    GLuint m_texture_name;
+
+    unsigned int m_texture_size;
+
+    void* m_mipmap_data;
+
+    TexConfig* m_tex_config;
+
+    std::vector<std::pair<core::dimension2d<u32>, size_t> > m_mipmap_sizes;
+
+public:
+    // ------------------------------------------------------------------------
+    HQMipmapGenerator(const io::path& name, uint8_t* data,
+                      const core::dimension2d<u32>& size, GLuint texture_name,
+                      TexConfig* tc);
+    // ------------------------------------------------------------------------
+    virtual ~HQMipmapGenerator() {}
+    // ------------------------------------------------------------------------
+    virtual void* lock(video::E_TEXTURE_LOCK_MODE mode =
+                       video::ETLM_READ_WRITE, u32 mipmap_level = 0)
+                                                               { return NULL; }
+    // ------------------------------------------------------------------------
+    virtual void unlock()                                                    {}
+    // ------------------------------------------------------------------------
+    virtual const core::dimension2d<u32>& getOriginalSize() const
+                                                             { return m_size; }
+    // ------------------------------------------------------------------------
+    virtual const core::dimension2d<u32>& getSize() const    { return m_size; }
+    // ------------------------------------------------------------------------
+    virtual video::E_DRIVER_TYPE getDriverType() const
+    {
+#if defined(USE_GLES2)
+        return video::EDT_OGLES2;
+#else
+        return video::EDT_OPENGL;
+#endif
+    }
+    // ------------------------------------------------------------------------
+    virtual video::ECOLOR_FORMAT getColorFormat() const
+                                                { return video::ECF_A8R8G8B8; }
+    // ------------------------------------------------------------------------
+    virtual u32 getPitch() const                                  { return 0; }
+    // ------------------------------------------------------------------------
+    virtual bool hasMipMaps() const                           { return false; }
+    // ------------------------------------------------------------------------
+    virtual void regenerateMipMapLevels(void* mipmap_data = NULL)            {}
+    // ------------------------------------------------------------------------
+    virtual u32 getOpenGLTextureName() const         { return m_texture_name; }
+    // ------------------------------------------------------------------------
+    virtual u64 getHandle()                                       { return 0; }
+    // ------------------------------------------------------------------------
+    virtual unsigned int getTextureSize() const      { return m_texture_size; }
+    // ------------------------------------------------------------------------
+    virtual void threadedReload(void* ptr, void* param) const;
+    // ------------------------------------------------------------------------
+    virtual void threadedSubImage(void* ptr) const;
+    // ------------------------------------------------------------------------
+    virtual void cleanThreadedLoader();
+
+};   // HQMipmapGenerator
+
+#endif
+
+#endif
diff --git a/src/graphics/irr_driver.cpp b/src/graphics/irr_driver.cpp
index 3c0e78783..22c1ab322 100644
--- a/src/graphics/irr_driver.cpp
+++ b/src/graphics/irr_driver.cpp
@@ -61,6 +61,7 @@
 #include "states_screens/dialogs/confirm_resolution_dialog.hpp"
 #include "states_screens/state_manager.hpp"
 #include "tracks/track_manager.hpp"
+#include "utils/avi_writer.hpp"
 #include "utils/constants.hpp"
 #include "utils/log.hpp"
 #include "utils/profiler.hpp"
@@ -146,6 +147,7 @@ IrrDriver::IrrDriver()
     m_last_light_bucket_distance = 0;
     m_clear_color                = video::SColor(255, 100, 101, 140);
     m_skinning_joint             = 0;
+    m_recording = false;
 
 }   // IrrDriver
 
@@ -167,6 +169,9 @@ IrrDriver::~IrrDriver()
 #endif
     delete m_wind;
     delete m_renderer;
+#if !(defined(SERVER_ONLY) || defined(USE_GLES2))
+    AVIWriter::kill();
+#endif
 }   // ~IrrDriver
 
 // ----------------------------------------------------------------------------
@@ -716,13 +721,11 @@ void IrrDriver::initDevice()
 // ----------------------------------------------------------------------------
 void IrrDriver::setMaxTextureSize()
 {
-    if( (UserConfigParams::m_high_definition_textures & 0x01) == 0)
-    {
-        io::IAttributes &att = m_video_driver->getNonConstDriverAttributes();
-        att.setAttribute("MAX_TEXTURE_SIZE", core::dimension2du(
-                                        UserConfigParams::m_max_texture_size,
-                                        UserConfigParams::m_max_texture_size));
-    }
+    const unsigned max =
+        (UserConfigParams::m_high_definition_textures & 0x01) == 0 ?
+        UserConfigParams::m_max_texture_size : 2048;
+    io::IAttributes &att = m_video_driver->getNonConstDriverAttributes();
+    att.setAttribute("MAX_TEXTURE_SIZE", core::dimension2du(max, max));
 }   // setMaxTextureSize
 
 // ----------------------------------------------------------------------------
@@ -923,6 +926,9 @@ void IrrDriver::applyResolutionSettings()
     // (we're sure to update main.cpp at some point and forget this one...)
     VAOManager::getInstance()->kill();
     STKTexManager::getInstance()->kill();
+#if !(defined(SERVER_ONLY) || defined(USE_GLES2))
+    AVIWriter::kill();
+#endif
     // initDevice will drop the current device.
     if (CVS->isGLSL())
     {
@@ -1842,6 +1848,9 @@ void IrrDriver::update(float dt)
 
     PropertyAnimator::get()->update(dt);
 
+    STKTexManager::getInstance()
+        ->checkThreadedLoadTextures(true/*util_queue_empty*/);
+
     World *world = World::getWorld();
 
     if (world)
@@ -1884,8 +1893,41 @@ void IrrDriver::update(float dt)
     // menu.
     //if(World::getWorld() && World::getWorld()->isRacePhase())
     //    printRenderStats();
+#if !(defined(SERVER_ONLY) || defined(USE_GLES2))
+    if (m_recording)
+        AVIWriter::getInstance()->captureFrameBufferImage(dt);
+#endif
 }   // update
 
+// ----------------------------------------------------------------------------
+void IrrDriver::setRecording(bool val)
+{
+#if !(defined(SERVER_ONLY) || defined(USE_GLES2))
+    if (!CVS->isARBPixelBufferObjectUsable())
+    {
+        Log::warn("irr_driver", "PBO extension missing, can't record video.");
+        return;
+    }
+    if (m_recording == val)
+        return;
+    m_recording = val;
+    if (m_recording == true)
+    {
+        std::string track_name = World::getWorld() != NULL ?
+            race_manager->getTrackName() : "menu";
+        AVIWriter::setRecordingTarget(file_manager->getScreenshotDir() +
+            track_name);
+        AVIWriter::getInstance()->resetFrameBufferImage();
+        MessageQueue::add(MessageQueue::MT_GENERIC,
+            _("Video recording started."));
+    }
+    else
+    {
+        AVIWriter::getInstance()->stopRecording();
+    }
+#endif
+}   // setRecording
+
 // ----------------------------------------------------------------------------
 
 void IrrDriver::requestScreenshot()
@@ -2095,4 +2137,3 @@ GLuint IrrDriver::getDepthStencilTexture()
     return m_renderer->getDepthStencilTexture();
 }   // getDepthStencilTexture
 
-
diff --git a/src/graphics/irr_driver.hpp b/src/graphics/irr_driver.hpp
index f3f6ba15d..21d6a611c 100644
--- a/src/graphics/irr_driver.hpp
+++ b/src/graphics/irr_driver.hpp
@@ -163,6 +163,7 @@ private:
     bool                 m_lightviz;
     bool                 m_distortviz;
     bool                 m_boundingboxesviz;
+    bool                 m_recording;
 
     /** Background colour to reset a buffer. Can be changed by each track. */
     irr::video::SColor m_clear_color;
@@ -414,6 +415,10 @@ public:
     // ------------------------------------------------------------------------
     bool getBoundingBoxesViz()    { return m_boundingboxesviz;      }
     // ------------------------------------------------------------------------
+    bool isRecording() const { return m_recording; }
+    // ------------------------------------------------------------------------
+    void setRecording(bool val);
+    // ------------------------------------------------------------------------
     u32 getRenderPass() { return m_renderpass; }
     // ------------------------------------------------------------------------
     std::vector<LightNode *> getLights() { return m_lights; }
diff --git a/src/graphics/material.cpp b/src/graphics/material.cpp
index ce40bad83..fb82e1aa8 100644
--- a/src/graphics/material.cpp
+++ b/src/graphics/material.cpp
@@ -526,9 +526,9 @@ void Material::install(bool srgb, bool premul_alpha)
     }
     else
     {
-        m_texture = STKTexManager::getInstance()->getTexture
-            (m_original_full_path, srgb, premul_alpha, false/*set_material*/,
-            srgb/*mesh_tex*/);
+        TexConfig tc(srgb, premul_alpha, srgb/*mesh_tex*/);
+        m_texture = STKTexManager::getInstance()
+            ->getTexture(m_original_full_path, &tc);
     }
 
     if (m_texture == NULL) return;
@@ -771,9 +771,8 @@ void  Material::setMaterialProperties(video::SMaterial *m, scene::IMeshBuffer* m
         STKTexManager* stm = STKTexManager::getInstance();
         if (m_gloss_map.size() > 0 && CVS->isDefferedEnabled())
         {
-            glossytex = stm->getTexture(m_gloss_map, false/*srgb*/,
-                false/*premul_alpha*/, false/*set_material*/,
-                true/*mesh_tex*/);
+            TexConfig gtc(false/*srgb*/, false/*premul_alpha*/);
+            glossytex = stm->getTexture(m_gloss_map, &gtc);
         }
         else
         {
@@ -787,9 +786,11 @@ void  Material::setMaterialProperties(video::SMaterial *m, scene::IMeshBuffer* m
                 stm->STKTexManager::getInstance()->getUnicolorTexture(SColor(0, 0, 0, 0));
             if (m_colorization_mask.size() > 0)
             {
+                TexConfig cmtc(false/*srgb*/, false/*premul_alpha*/,
+                    true/*mesh_tex*/, false/*set_material*/,
+                    true/*color_mask*/);
                 colorization_mask_tex = stm->getTexture(m_colorization_mask,
-                    false/*srgb*/, false/*premul_alpha*/, false/*set_material*/,
-                    true/*mesh_tex*/, false/*no_upload*/, true/*single_channel*/);
+                    &cmtc);
             }
             m->setTexture(2, colorization_mask_tex);
         }
@@ -845,32 +846,29 @@ void  Material::setMaterialProperties(video::SMaterial *m, scene::IMeshBuffer* m
             m->setTexture(1, glossytex);
             return;
         case SHADERTYPE_SPLATTING:
-            tex = stm->getTexture(m_splatting_texture_1,
-                true/*srgb*/, false/*premul_alpha*/, false/*set_material*/,
-                true/*mesh_tex*/);
+        {
+            TexConfig stc(true/*srgb*/, false/*premul_alpha*/,
+                true/*mesh_tex*/, false/*set_material*/);
+            tex = stm->getTexture(m_splatting_texture_1, &stc);
             m->setTexture(3, tex);
 
             if (m_splatting_texture_2.size() > 0)
             {
-                tex = stm->getTexture(m_splatting_texture_2,
-                    true/*srgb*/, false/*premul_alpha*/, false/*set_material*/,
-                    true/*mesh_tex*/);
+                tex = stm->getTexture(m_splatting_texture_2, &stc);
             }
             m->setTexture(4, tex);
 
             if (m_splatting_texture_3.size() > 0)
             {
-                tex = stm->getTexture(m_splatting_texture_3,
-                    true/*srgb*/, false/*premul_alpha*/, false/*set_material*/,
-                    true/*mesh_tex*/);
+                tex = stm->getTexture(m_splatting_texture_3, &stc);
             }
             m->setTexture(5, tex);
 
             if (m_splatting_texture_4.size() > 0)
             {
-                tex = stm->getTexture(m_splatting_texture_4,
-                    false/*srgb*/, false/*premul_alpha*/, false/*set_material*/,
-                    true/*mesh_tex*/);
+                TexConfig s4tc(false/*srgb*/, false/*premul_alpha*/,
+                    true/*mesh_tex*/, false/*set_material*/);
+                tex = stm->getTexture(m_splatting_texture_4, &s4tc);
             }
             m->setTexture(6, tex);
             m->setTexture(7, glossytex);
@@ -878,6 +876,7 @@ void  Material::setMaterialProperties(video::SMaterial *m, scene::IMeshBuffer* m
             // Material and shaders
             m->MaterialType = Shaders::getShader(ES_SPLATTING);
             return;
+        }
         case SHADERTYPE_WATER:
             m->setTexture(1, irr_driver->getTexture(FileManager::TEXTURE,
                 "waternormals.jpg"));
@@ -912,9 +911,10 @@ void  Material::setMaterialProperties(video::SMaterial *m, scene::IMeshBuffer* m
         {
             if (CVS->isDefferedEnabled())
             {
-                tex = stm->getTexture(m_normal_map_tex, false/*srgb*/,
-                    false/*premul_alpha*/, false/*set_material*/,
-                    true/*mesh_tex*/);
+                TexConfig nmtc(false/*srgb*/, false/*premul_alpha*/,
+                    true/*mesh_tex*/, false/*set_material*/,
+                    false/*color_mask*/, true/*normal_map*/);
+                tex = stm->getTexture(m_normal_map_tex, &nmtc);
             }
             else
                 tex = stm->STKTexManager::getInstance()->getUnicolorTexture(SColor(0, 0, 0, 0));
diff --git a/src/graphics/rtts.cpp b/src/graphics/rtts.cpp
index 76cfc85e4..ea2b35d12 100644
--- a/src/graphics/rtts.cpp
+++ b/src/graphics/rtts.cpp
@@ -49,10 +49,10 @@ static GLuint generateRTT(const core::dimension2du &res, GLint internalFormat, G
     return result;
 }
 
-RTT::RTT(size_t width, size_t height)
+RTT::RTT(size_t width, size_t height, float rtt_scale)
 {
-    m_width = width;
-    m_height = height;
+    m_width = width * rtt_scale;
+    m_height = height  * rtt_scale;
     m_shadow_FBO = NULL;
     m_RH_FBO = NULL;
     m_RSM = NULL;
@@ -60,14 +60,13 @@ RTT::RTT(size_t width, size_t height)
     using namespace video;
     using namespace core;
     
-    dimension2du res(int(width * UserConfigParams::m_scale_rtts_factor),
-                     int(height * UserConfigParams::m_scale_rtts_factor) );
+    dimension2du res(m_width, m_height);
     
     const dimension2du half = res/2;
     const dimension2du quarter = res/4;
     const dimension2du eighth = res/8;
 
-    const u16 shadowside = u16(1024 * UserConfigParams::m_scale_rtts_factor);
+    const u16 shadowside = u16(1024 * rtt_scale);
     const dimension2du shadowsize0(shadowside, shadowside);
     const dimension2du shadowsize1(shadowside / 2, shadowside / 2);
     const dimension2du shadowsize2(shadowside / 4, shadowside / 4);
@@ -294,7 +293,12 @@ RTT::RTT(size_t width, size_t height)
     glClear(GL_COLOR_BUFFER_BIT);
 
     getFBO(FBO_COMBINED_DIFFUSE_SPECULAR).bind();
-    glClearColor(.5, .5, .5, .5);
+    float color = 0.5;
+#if defined(USE_GLES2)
+    if (!CVS->isDefferedEnabled())
+        color = pow(color, 1. / 2.2);
+#endif
+    glClearColor(color, color, color, color);
     glClear(GL_COLOR_BUFFER_BIT);
     glBindFramebuffer(GL_FRAMEBUFFER, 0);
 #if !defined(USE_GLES2)
diff --git a/src/graphics/rtts.hpp b/src/graphics/rtts.hpp
index f5102ff00..4e38322c8 100644
--- a/src/graphics/rtts.hpp
+++ b/src/graphics/rtts.hpp
@@ -144,7 +144,7 @@ enum TypeRTT : unsigned int
 class RTT
 {
 public:
-    RTT(size_t width, size_t height);
+    RTT(size_t width, size_t height, float rtt_scale = 1.0f);
     ~RTT();
     
     size_t getWidth () const { return m_width ; }
diff --git a/src/graphics/shader_based_renderer.cpp b/src/graphics/shader_based_renderer.cpp
index c1bcba02a..1d253cb58 100644
--- a/src/graphics/shader_based_renderer.cpp
+++ b/src/graphics/shader_based_renderer.cpp
@@ -199,10 +199,8 @@ void ShaderBasedRenderer::uploadLightingData() const
 void ShaderBasedRenderer::computeMatrixesAndCameras(scene::ICameraSceneNode *const camnode,
                                                     size_t width, size_t height)
 {
-    float w = width * UserConfigParams::m_scale_rtts_factor;
-    float h = height * UserConfigParams::m_scale_rtts_factor;
-    m_current_screen_size = core::vector2df(w, h);
-    m_shadow_matrices.computeMatrixesAndCameras(camnode, int(w), int(h),
+    m_current_screen_size = core::vector2df((float)width, (float)height);
+    m_shadow_matrices.computeMatrixesAndCameras(camnode, width, height,
         m_rtts->getDepthStencilTexture());
 }   // computeMatrixesAndCameras
 
@@ -669,7 +667,8 @@ void ShaderBasedRenderer::onLoadWorld()
     const core::recti &viewport = Camera::getCamera(0)->getViewport();
     size_t width = viewport.LowerRightCorner.X - viewport.UpperLeftCorner.X;
     size_t height = viewport.LowerRightCorner.Y - viewport.UpperLeftCorner.Y;
-    RTT* rtts = new RTT(width, height);
+    RTT* rtts = new RTT(width, height, CVS->isDefferedEnabled() ?
+                        UserConfigParams::m_scale_rtts_factor : 1.0f);
     setRTT(rtts);
 }
 
@@ -787,9 +786,7 @@ void ShaderBasedRenderer::render(float dt)
     RaceGUIBase *rg = world->getRaceGUI();
     if (rg) rg->update(dt);
     
-    bool force_rtt = UserConfigParams::m_scale_rtts_factor != 1.0f;
-    
-    if (!CVS->isDefferedEnabled() && !force_rtt)
+    if (!CVS->isDefferedEnabled())
     {
         prepareForwardRenderer();
     }
@@ -803,12 +800,10 @@ void ShaderBasedRenderer::render(float dt)
         oss << "drawAll() for kart " << cam;
         PROFILER_PUSH_CPU_MARKER(oss.str().c_str(), (cam+1)*60,
                                  0x00, 0x00);
-        camera->activate(!CVS->isDefferedEnabled() && !force_rtt);
+        camera->activate(!CVS->isDefferedEnabled());
         rg->preRenderCallback(camera);   // adjusts start referee
         irr_driver->getSceneManager()->setActiveCamera(camnode);
 
-        const core::recti &viewport = camera->getViewport();
-
         if (!CVS->isDefferedEnabled())
             glEnable(GL_FRAMEBUFFER_SRGB);
         
@@ -816,12 +811,12 @@ void ShaderBasedRenderer::render(float dt)
         m_lighting_passes.updateLightsInfo(camnode, dt);
         PROFILER_POP_CPU_MARKER();
         PROFILER_PUSH_CPU_MARKER("UBO upload", 0x0, 0xFF, 0x0);
-        computeMatrixesAndCameras(camnode, viewport.LowerRightCorner.X - viewport.UpperLeftCorner.X, viewport.LowerRightCorner.Y - viewport.UpperLeftCorner.Y);
+        computeMatrixesAndCameras(camnode, m_rtts->getWidth(), m_rtts->getHeight());
         m_shadow_matrices.updateSunOrthoMatrices();
         if(CVS->isARBUniformBufferObjectUsable())
             uploadLightingData();
         PROFILER_POP_CPU_MARKER();
-        renderScene(camnode, dt, track->hasShadows(), force_rtt); 
+        renderScene(camnode, dt, track->hasShadows(), false); 
         
         if (irr_driver->getBoundingBoxesViz())
         {        
@@ -830,7 +825,7 @@ void ShaderBasedRenderer::render(float dt)
         
         debugPhysics();
         
-        if (CVS->isDefferedEnabled() || force_rtt)
+        if (CVS->isDefferedEnabled())
         {
             renderPostProcessing(camera);
         }
diff --git a/src/graphics/shader_files_manager.cpp b/src/graphics/shader_files_manager.cpp
index e7b37de05..6682af1b6 100644
--- a/src/graphics/shader_files_manager.cpp
+++ b/src/graphics/shader_files_manager.cpp
@@ -51,6 +51,62 @@ const std::string& ShaderFilesManager::getHeader()
     return shader_header;
 }   // getHeader
 
+// ----------------------------------------------------------------------------
+void ShaderFilesManager::readFile(const std::string& file, 
+                                  std::ostringstream& code)
+{
+    std::ifstream stream(file_manager->getShader(file), std::ios::in);
+    
+    if (!stream.is_open())
+    {
+        Log::error("ShaderFilesManager", "Can not open '%s'.", file.c_str());
+        return;
+    }
+    
+    const std::string stk_include = "#stk_include";
+    std::string line;
+
+    while (std::getline(stream, line))
+    {
+        const std::size_t pos = line.find(stk_include);
+
+        // load the custom file pointed by the #stk_include directive
+        if (pos != std::string::npos)
+        {
+            // find the start "
+            std::size_t pos = line.find("\"");
+            if (pos == std::string::npos)
+            {
+                Log::error("ShaderFilesManager", "Invalid #stk_include"
+                    " line: '%s'.", line.c_str());
+                continue;
+            }
+
+            std::string filename = line.substr(pos + 1);
+
+            // find the end "
+            pos = filename.find("\"");
+            if (pos == std::string::npos)
+            {
+                Log::error("ShaderFilesManager", "Invalid #stk_include"
+                    " line: '%s'.", line.c_str());
+                continue;
+            }
+
+            filename = filename.substr(0, pos);
+            
+            // read the whole include file
+            readFile(filename, code);
+        }
+        else
+        {
+            code << "\n" << line;
+        }
+    }
+
+    stream.close();
+}
+
 // ----------------------------------------------------------------------------
 /** Loads a single shader. This is NOT cached, use addShaderFile for that.
  *  \param file Filename of the shader to load.
@@ -90,7 +146,9 @@ GLuint ShaderFilesManager::loadShader(const std::string &file, unsigned type)
 
     if (CVS->isARBExplicitAttribLocationUsable())
     {
+#if !defined(USE_GLES2)
         code << "#extension GL_ARB_explicit_attrib_location : enable\n";
+#endif
         code << "#define Explicit_Attrib_Location_Usable\n";
     }
 
@@ -106,12 +164,14 @@ GLuint ShaderFilesManager::loadShader(const std::string &file, unsigned type)
         code << "#define VSLayer\n";
     if (CVS->needsRGBBindlessWorkaround())
         code << "#define SRGBBindlessFix\n";
+    if (CVS->isDefferedEnabled())
+        code << "#define Advanced_Lighting_Enabled\n";
 
 #if !defined(USE_GLES2)
     // shader compilation fails with some drivers if there is no precision
     // qualifier
     if (type == GL_FRAGMENT_SHADER)
-        code << "precision mediump float;\n";
+        code << "precision highp float;\n";
 #else
     int range[2], precision;
     glGetShaderPrecisionFormat(GL_FRAGMENT_SHADER, GL_HIGH_FLOAT, range,
@@ -126,69 +186,7 @@ GLuint ShaderFilesManager::loadShader(const std::string &file, unsigned type)
 
     code << getHeader();
 
-    std::ifstream stream(file_manager->getShader(file), std::ios::in);
-    if (stream.is_open())
-    {
-        const std::string stk_include = "#stk_include";
-        std::string line;
-
-        while (std::getline(stream, line))
-        {
-            const std::size_t pos = line.find(stk_include);
-
-            // load the custom file pointed by the #stk_include directive
-            if (pos != std::string::npos)
-            {
-                // find the start "
-                std::size_t pos = line.find("\"");
-                if (pos == std::string::npos)
-                {
-                    Log::error("ShaderFilesManager", "Invalid #stk_include"
-                        " line: '%s'.", line.c_str());
-                    continue;
-                }
-
-                std::string filename = line.substr(pos + 1);
-
-                // find the end "
-                pos = filename.find("\"");
-                if (pos == std::string::npos)
-                {
-                    Log::error("ShaderFilesManager", "Invalid #stk_include"
-                        " line: '%s'.", line.c_str());
-                    continue;
-                }
-
-                filename = filename.substr(0, pos);
-
-                // read the whole include file
-                std::ifstream include_stream(file_manager->getShader(filename), std::ios::in);
-                if (!include_stream.is_open())
-                {
-                    Log::error("ShaderFilesManager", "Couldn't open included"
-                        " shader: '%s'.", filename.c_str());
-                    continue;
-                }
-
-                std::string include_line = "";
-                while (std::getline(include_stream, include_line))
-                {
-                    code << "\n" << include_line;
-                }
-                include_stream.close();
-            }
-            else
-            {
-                code << "\n" << line;
-            }
-        }
-
-        stream.close();
-    }
-    else
-    {
-        Log::error("ShaderFilesManager", "Can not open '%s'.", file.c_str());
-    }
+    readFile(file, code);
 
     Log::info("ShaderFilesManager", "Compiling shader : %s", file.c_str());
     const std::string &source  = code.str();
diff --git a/src/graphics/shader_files_manager.hpp b/src/graphics/shader_files_manager.hpp
index aa11d90a0..f0ad1fbd9 100644
--- a/src/graphics/shader_files_manager.hpp
+++ b/src/graphics/shader_files_manager.hpp
@@ -38,6 +38,7 @@ private:
 
     // ------------------------------------------------------------------------
     const std::string& getHeader();
+    void readFile(const std::string& file, std::ostringstream& code);
 
 public:
     // ------------------------------------------------------------------------
diff --git a/src/graphics/skybox.cpp b/src/graphics/skybox.cpp
index 9c7a27a48..f8d503cd6 100644
--- a/src/graphics/skybox.cpp
+++ b/src/graphics/skybox.cpp
@@ -163,19 +163,6 @@ void Skybox::generateCubeMapFromTextures()
         assert(img != NULL);
         img->copyToScaling(rgba[i], size, size);
 
-#if defined(USE_GLES2)
-        if (CVS->isEXTTextureFormatBGRA8888Usable())
-        {
-            // BGRA image returned by getTextureImage causes black sky in gles
-            for (unsigned int j = 0; j < size * size; j++)
-            {
-                char tmp_val = rgba[i][j * 4];
-                rgba[i][j * 4] = rgba[i][j * 4 + 2];
-                rgba[i][j * 4 + 2] = tmp_val;
-            }
-        }
-#endif
-
         if (i == 2 || i == 3)
         {
             char *tmp = new char[size * size * 4];
@@ -196,7 +183,8 @@ void Skybox::generateCubeMapFromTextures()
                                     GL_COMPRESSED_SRGB_ALPHA : GL_SRGB_ALPHA;
         GLint format = GL_BGRA;
 #else
-        GLint internal_format = GL_RGBA8;
+        GLint internal_format = CVS->isDefferedEnabled() ? GL_SRGB8_ALPHA8
+                                                         : GL_RGBA8;
         GLint format = GL_RGBA;
 #endif
 
diff --git a/src/graphics/spherical_harmonics.cpp b/src/graphics/spherical_harmonics.cpp
index 3a4a4c9f7..c94149bbf 100644
--- a/src/graphics/spherical_harmonics.cpp
+++ b/src/graphics/spherical_harmonics.cpp
@@ -565,15 +565,12 @@ void SphericalHarmonics::setTextures(const std::vector<video::ITexture *> &spher
         assert(img != NULL);
         img->copyToScaling(sh_rgba[i], sh_w, sh_h);
 #if defined(USE_GLES2)
-        if (!CVS->isEXTTextureFormatBGRA8888Usable())
+        // Code here assume color format is BGRA
+        for (unsigned int j = 0; j < sh_w * sh_h; j++)
         {
-            // Code here assume color format is BGRA
-            for (unsigned int j = 0; j < sh_w * sh_h; j++)
-            {
-                char tmp_val = sh_rgba[i][j * 4];
-                sh_rgba[i][j * 4] = sh_rgba[i][j * 4 + 2];
-                sh_rgba[i][j * 4 + 2] = tmp_val;
-            }
+            char tmp_val = sh_rgba[i][j * 4];
+            sh_rgba[i][j * 4] = sh_rgba[i][j * 4 + 2];
+            sh_rgba[i][j * 4 + 2] = tmp_val;
         }
 #endif
     } //for (unsigned i = 0; i < 6; i++)
diff --git a/src/graphics/stars.cpp b/src/graphics/stars.cpp
index d1d54a1d2..13db505ae 100644
--- a/src/graphics/stars.cpp
+++ b/src/graphics/stars.cpp
@@ -40,9 +40,9 @@ Stars::Stars(AbstractKart *kart)
     m_parent_kart_node = kart->getNode();
     m_enabled = false;
 
+    TexConfig stc(true/*srgb*/, true/*premul_alpha*/);
     video::ITexture* texture = STKTexManager::getInstance()->getTexture
-        ("starparticle.png", true/*srgb*/, true/*premul_alpha*/,
-        false/*set_material*/, true/*mesh_tex*/);
+        ("starparticle.png", &stc);
 
     Material* star_material =
         material_manager->getMaterial("starparticle.png");
diff --git a/src/graphics/stk_mesh_loader.cpp b/src/graphics/stk_mesh_loader.cpp
index b2918b61a..a11dca346 100644
--- a/src/graphics/stk_mesh_loader.cpp
+++ b/src/graphics/stk_mesh_loader.cpp
@@ -1044,10 +1044,11 @@ void STKMeshLoader::loadTextures(SB3dMaterial& material) const
             else
                 full_path = fs->getFileBasename(B3dTexture->TextureName);
 
+            TexConfig mtc(i <= 1 ? true : false/*srgb*/, false/*premul_alpha*/,
+                true/*mesh_tex*/, true/*set_material*/);
             video::ITexture* tex =
                 STKTexManager::getInstance()->getTexture(full_path.c_str(),
-                i <= 1 ? true : false/*is_srgb*/, false/*premul_alpha*/,
-                true/*set_material*/);
+                &mtc);
 
             material.Material.setTexture(i, tex);
             if (material.Textures[i]->Flags & 0x10) // Clamp U
diff --git a/src/graphics/stk_tex_manager.cpp b/src/graphics/stk_tex_manager.cpp
index aa39e60e8..4ea5ccd94 100644
--- a/src/graphics/stk_tex_manager.cpp
+++ b/src/graphics/stk_tex_manager.cpp
@@ -16,19 +16,107 @@
 //  Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
 
 #include "graphics/stk_tex_manager.hpp"
+#include "config/hardware_stats.hpp"
+#include "config/user_config.hpp"
 #include "graphics/central_settings.hpp"
 #include "graphics/materials.hpp"
+#include "graphics/threaded_tex_loader.hpp"
 #include "graphics/stk_texture.hpp"
 #include "io/file_manager.hpp"
 #include "utils/string_utils.hpp"
 #include "utils/log.hpp"
 
+#include <algorithm>
+
+// ----------------------------------------------------------------------------
+STKTexManager::STKTexManager() : m_pbo(0), m_thread_size(0),
+                                 m_threaded_load_textures_counter(0)
+{
+    createThreadedTexLoaders();
+}   // STKTexManager
+
 // ----------------------------------------------------------------------------
 STKTexManager::~STKTexManager()
 {
     removeTexture(NULL/*texture*/, true/*remove_all*/);
+    destroyThreadedTexLoaders();
 }   // ~STKTexManager
 
+// ----------------------------------------------------------------------------
+void STKTexManager::createThreadedTexLoaders()
+{
+#if !(defined(SERVER_ONLY) || defined(USE_GLES2))
+    if (CVS->supportsThreadedTextureLoading())
+    {
+        pthread_mutex_init(&m_threaded_load_textures_mutex, NULL);
+        pthread_cond_init(&m_cond_request, NULL);
+        m_thread_size = HardwareStats::getNumProcessors();
+        if (m_thread_size == 0)
+            m_thread_size = 1;
+        m_thread_size = core::clamp(m_thread_size, 1,
+            UserConfigParams::m_hq_mipmap ? m_thread_size : 3);
+        const unsigned max_tex_size =
+            (UserConfigParams::m_high_definition_textures & 0x01) == 0 ?
+            UserConfigParams::m_max_texture_size : 2048;
+        const unsigned each_capacity = max_tex_size * max_tex_size * 4;
+        const unsigned pbo_size = each_capacity * m_thread_size;
+        Log::info("STKTexManager", "%d thread(s) for texture loading,"
+            " each capacity %d MB.", m_thread_size,
+            each_capacity / 1024 / 1024);
+        if (UserConfigParams::m_hq_mipmap)
+            Log::info("STKTexManager", "High quality mipmap enabled.");
+        glGenBuffers(1, &m_pbo);
+        glBindBuffer(GL_PIXEL_UNPACK_BUFFER, m_pbo);
+        glBufferStorage(GL_PIXEL_UNPACK_BUFFER, pbo_size, NULL,
+            GL_MAP_WRITE_BIT | GL_MAP_PERSISTENT_BIT |
+            GL_MAP_COHERENT_BIT);
+        uint8_t* pbo_ptr = (uint8_t*)glMapBufferRange(GL_PIXEL_UNPACK_BUFFER,
+            0, pbo_size, GL_MAP_WRITE_BIT | GL_MAP_PERSISTENT_BIT |
+            GL_MAP_COHERENT_BIT);
+        size_t offset = 0;
+        for (int i = 0; i < m_thread_size; i++)
+        {
+            m_all_tex_loaders.push_back(new ThreadedTexLoader(each_capacity,
+                offset, pbo_ptr + offset, &m_threaded_load_textures_mutex,
+                &m_cond_request, this));
+            offset += each_capacity;
+        }
+        glBindBuffer(GL_PIXEL_UNPACK_BUFFER, 0);
+    }
+#endif
+}   // createThreadedTexLoaders
+
+// ----------------------------------------------------------------------------
+void STKTexManager::destroyThreadedTexLoaders()
+{
+#if !(defined(SERVER_ONLY) || defined(USE_GLES2))
+    if (CVS->supportsThreadedTextureLoading())
+    {
+        STKTexture* delete_ttl = new STKTexture((uint8_t*)NULL, "delete_ttl",
+            0, false, true);
+        for (int i = 0; i < m_thread_size; i++)
+            addThreadedLoadTexture(delete_ttl);
+        for (int i = 0; i < m_thread_size; i++)
+        {
+            if (!m_all_tex_loaders[i]->waitForReadyToDeleted(2.0f))
+            {
+                Log::info("STKTexManager", "ThreadedTexLoader %d not stopping,"
+                    "exiting anyway.", i);
+            }
+            delete m_all_tex_loaders[i];
+        }
+        delete delete_ttl;
+        glDeleteBuffers(1, &m_pbo);
+        pthread_mutex_destroy(&m_threaded_load_textures_mutex);
+        pthread_cond_destroy(&m_cond_request);
+        m_pbo = 0;
+        m_thread_size = 0;
+        m_threaded_load_textures_counter = 0;
+        m_all_tex_loaders.clear();
+    }
+#endif
+}   // destroyThreadedTexLoaders
+
 // ----------------------------------------------------------------------------
 STKTexture* STKTexManager::findTextureInFileSystem(const std::string& filename,
                                                    std::string* full_path)
@@ -55,10 +143,8 @@ STKTexture* STKTexManager::findTextureInFileSystem(const std::string& filename,
 }   // findTextureInFileSystem
 
 // ----------------------------------------------------------------------------
-video::ITexture* STKTexManager::getTexture(const std::string& path, bool srgb,
-                                           bool premul_alpha,
-                                           bool set_material, bool mesh_tex,
-                                           bool no_upload, bool single_channel,
+video::ITexture* STKTexManager::getTexture(const std::string& path,
+                                           TexConfig* tc, bool no_upload,
                                            bool create_if_unfound)
 {
     auto ret = m_all_textures.find(path);
@@ -79,8 +165,7 @@ video::ITexture* STKTexManager::getTexture(const std::string& path, bool srgb,
     if (create_if_unfound)
     {
         new_texture = new STKTexture(full_path.empty() ? path : full_path,
-            srgb, premul_alpha, set_material, mesh_tex, no_upload,
-            single_channel);
+            tc, no_upload);
         if (new_texture->getOpenGLTextureName() == 0 && !no_upload)
         {
             const char* name = new_texture->getName().getPtr();
@@ -95,6 +180,11 @@ video::ITexture* STKTexManager::getTexture(const std::string& path, bool srgb,
             delete new_texture;
             return NULL;
         }
+        if (new_texture->useThreadedLoading())
+        {
+            addThreadedLoadTexture(new_texture);
+            checkThreadedLoadTextures(false/*util_queue_empty*/);
+        }
     }
 
     if (create_if_unfound && !no_upload)
@@ -204,6 +294,10 @@ core::stringw STKTexManager::reloadTexture(const irr::core::stringw& name)
             if (p.second == NULL || !p.second->isMeshTexture())
                 continue;
             p.second->reload();
+            if (p.second->useThreadedLoading())
+            {
+                addThreadedLoadTexture(p.second);
+            }
             Log::info("STKTexManager", "%s reloaded",
                 p.second->getName().getPtr());
         }
@@ -226,6 +320,10 @@ core::stringw STKTexManager::reloadTexture(const irr::core::stringw& name)
             if (fname == tex_name || fname == tex_path)
             {
                 p.second->reload();
+                if (p.second->useThreadedLoading())
+                {
+                    addThreadedLoadTexture(p.second);
+                }
                 result += tex_name.c_str();
                 result += L" ";
                 break;
@@ -271,3 +369,78 @@ void STKTexManager::setTextureErrorMessage(const std::string &error,
     else
         m_texture_error_message = StringUtils::insertValues(error, detail);
 }   // setTextureErrorMessage
+
+// ----------------------------------------------------------------------------
+void STKTexManager::checkThreadedLoadTextures(bool util_queue_empty)
+{
+#if !(defined(SERVER_ONLY) || defined(USE_GLES2))
+    if (!CVS->supportsThreadedTextureLoading()) return;
+    bool uploaded = false;
+    bool empty_queue = false;
+    if (util_queue_empty)
+    {
+        while (true)
+        {
+            pthread_mutex_lock(&m_threaded_load_textures_mutex);
+            empty_queue = m_threaded_load_textures_counter == 0;
+            pthread_mutex_unlock(&m_threaded_load_textures_mutex);
+            if (empty_queue)
+            {
+                for (ThreadedTexLoader* ttl : m_all_tex_loaders)
+                {
+                    if (ttl->lastQueueReady())
+                    {
+                        ttl->lock();
+                        ttl->setFinishLoading();
+                        uploaded = true;
+                        ttl->unlock(false/*finish_it*/);
+                    }
+                }
+                break;
+            }
+            else
+            {
+                checkThreadedLoadTextures(false/*util_queue_empty*/);
+            }
+        }
+    }
+    if (empty_queue && !uploaded)
+        return;
+    uploaded = false;
+    for (ThreadedTexLoader* ttl : m_all_tex_loaders)
+    {
+        ttl->lock();
+        if (ttl->finishedLoading())
+        {
+            if (!uploaded)
+            {
+                glBindBuffer(GL_PIXEL_UNPACK_BUFFER, m_pbo);
+                uploaded = true;
+            }
+            ttl->handleCompletedTextures();
+        }
+        else
+        {
+            ttl->unlock(false/*finish_it*/);
+        }
+    }
+    if (uploaded)
+    {
+        GLsync sync = glFenceSync(GL_SYNC_GPU_COMMANDS_COMPLETE, 0);
+        GLenum reason = glClientWaitSync(sync, GL_SYNC_FLUSH_COMMANDS_BIT, 0);
+        if (reason != GL_ALREADY_SIGNALED)
+        {
+            do
+            {
+                reason = glClientWaitSync(sync, GL_SYNC_FLUSH_COMMANDS_BIT,
+                    1000000);
+            }
+            while (reason == GL_TIMEOUT_EXPIRED);
+        }
+        glDeleteSync(sync);
+        glBindBuffer(GL_PIXEL_UNPACK_BUFFER, 0);
+        for (ThreadedTexLoader* ttl : m_all_tex_loaders)
+            ttl->unlock(true/*finish_it*/);
+    }
+#endif
+}   // checkThreadedLoadTextures
diff --git a/src/graphics/stk_tex_manager.hpp b/src/graphics/stk_tex_manager.hpp
index 47b151154..80163f491 100644
--- a/src/graphics/stk_tex_manager.hpp
+++ b/src/graphics/stk_tex_manager.hpp
@@ -23,17 +23,43 @@
 #include "utils/singleton.hpp"
 
 #include "irrString.h"
+#include "ITexture.h"
+#include <pthread.h>
 
-#include <algorithm>
+#include <cassert>
 #include <string>
+#include <queue>
 #include <unordered_map>
+#include <vector>
 
 class STKTexture;
+class ThreadedTexLoader;
 namespace irr
 {
-    namespace video { class ITexture; class SColor; }
+    namespace video { class SColor; }
 }
 
+struct TexConfig
+{
+     bool m_srgb;
+     bool m_premul_alpha;
+     bool m_mesh_tex;
+     bool m_set_material;
+     bool m_colorization_mask;
+     bool m_normal_map;
+     TexConfig(bool srgb = false, bool premul_alpha = false,
+               bool mesh_tex = true, bool set_material = false,
+               bool color_mask = false, bool normal_map = false)
+     {
+         m_srgb = srgb;
+         m_premul_alpha = premul_alpha;
+         m_mesh_tex = mesh_tex;
+         m_set_material = set_material;
+         m_colorization_mask = color_mask;
+         m_normal_map = normal_map;
+     }
+};
+
 class STKTexManager : public Singleton<STKTexManager>, NoCopy
 {
 private:
@@ -43,22 +69,43 @@ private:
      *  This is used to specify details like: "while loading kart '...'" */
     std::string m_texture_error_message;
 
+    std::vector<ThreadedTexLoader*> m_all_tex_loaders;
+
+    GLuint m_pbo;
+
+    int m_thread_size;
+
+    class SmallestTexture
+    {
+    public:
+        inline bool operator()(const irr::video::ITexture* a,
+            const irr::video::ITexture* b) const
+        {
+            return a->getTextureSize() > b->getTextureSize();
+        }
+    };
+    std::priority_queue<irr::video::ITexture*,
+        std::vector<irr::video::ITexture*>, SmallestTexture>
+        m_threaded_load_textures;
+
+    int m_threaded_load_textures_counter;
+
+    pthread_mutex_t m_threaded_load_textures_mutex;
+
+    pthread_cond_t m_cond_request;
+
     // ------------------------------------------------------------------------
     STKTexture* findTextureInFileSystem(const std::string& filename,
                                         std::string* full_path);
 public:
     // ------------------------------------------------------------------------
-    STKTexManager() {}
+    STKTexManager();
     // ------------------------------------------------------------------------
     ~STKTexManager();
     // ------------------------------------------------------------------------
     irr::video::ITexture* getTexture(const std::string& path,
-                                     bool srgb = false,
-                                     bool premul_alpha = false,
-                                     bool set_material = false,
-                                     bool mesh_tex = false,
+                                     TexConfig* tc = NULL,
                                      bool no_upload = false,
-                                     bool single_channel = false,
                                      bool create_if_unfound = true);
     // ------------------------------------------------------------------------
     irr::video::ITexture* getUnicolorTexture(const irr::video::SColor &c);
@@ -127,6 +174,35 @@ public:
         return getTexture(filename, std::string(error_message),
                           std::string(detail));
     }   // getTexture
+    // ------------------------------------------------------------------------
+    void checkThreadedLoadTextures(bool util_queue_empty);
+    // ------------------------------------------------------------------------
+    irr::video::ITexture* getThreadedLoadTexture()
+                                     { return m_threaded_load_textures.top(); }
+    // ------------------------------------------------------------------------
+    void setThreadedLoadTextureCounter(int val)
+    {
+        m_threaded_load_textures_counter += val;
+        assert(m_threaded_load_textures_counter >= 0);
+    }
+    // ------------------------------------------------------------------------
+    void addThreadedLoadTexture(irr::video::ITexture* t)
+    {
+        pthread_mutex_lock(&m_threaded_load_textures_mutex);
+        m_threaded_load_textures.push(t);
+        setThreadedLoadTextureCounter(t->getThreadedLoadTextureCounter());
+        pthread_cond_signal(&m_cond_request);
+        pthread_mutex_unlock(&m_threaded_load_textures_mutex);
+    }
+    // ------------------------------------------------------------------------
+    void removeThreadedLoadTexture()        { m_threaded_load_textures.pop(); }
+    // ------------------------------------------------------------------------
+    bool isThreadedLoadTexturesEmpty()
+                                   { return m_threaded_load_textures.empty(); }
+    // ------------------------------------------------------------------------
+    void createThreadedTexLoaders();
+    // ------------------------------------------------------------------------
+    void destroyThreadedTexLoaders();
 
 };   // STKTexManager
 
diff --git a/src/graphics/stk_texture.cpp b/src/graphics/stk_texture.cpp
index 6139ca850..593f3f000 100644
--- a/src/graphics/stk_texture.cpp
+++ b/src/graphics/stk_texture.cpp
@@ -18,10 +18,12 @@
 #include "graphics/stk_texture.hpp"
 #include "config/user_config.hpp"
 #include "graphics/central_settings.hpp"
+#include "graphics/hq_mipmap_generator.hpp"
 #include "graphics/irr_driver.hpp"
 #include "graphics/material.hpp"
 #include "graphics/material_manager.hpp"
 #include "graphics/materials.hpp"
+#include "graphics/stk_tex_manager.hpp"
 #include "modes/profile_world.hpp"
 #include "utils/log.hpp"
 #include "utils/string_utils.hpp"
@@ -33,23 +35,27 @@
 static const uint8_t CACHE_VERSION = 1;
 #endif
 // ----------------------------------------------------------------------------
-STKTexture::STKTexture(const std::string& path, bool srgb, bool premul_alpha,
-                       bool set_material, bool mesh_tex, bool no_upload,
-                       bool single_channel)
-          : video::ITexture(path.c_str()), m_texture_handle(0), m_srgb(srgb),
-            m_premul_alpha(premul_alpha), m_mesh_texture(mesh_tex),
-            m_single_channel(single_channel), m_material(NULL),
-            m_texture_name(0), m_texture_size(0), m_texture_image(NULL)
+STKTexture::STKTexture(const std::string& path, TexConfig* tc, bool no_upload)
+          : video::ITexture(path.c_str()), m_texture_handle(0),
+            m_single_channel(false), m_tex_config(NULL), m_material(NULL),
+            m_texture_name(0), m_texture_size(0), m_texture_image(NULL),
+            m_file(NULL), m_img_loader(NULL)
 {
-    if (set_material)
+    if (tc != NULL)
     {
-        m_material = material_manager->getMaterialFor(this);
-        m_mesh_texture = true;
+        m_tex_config = (TexConfig*)malloc(sizeof(TexConfig));
+        memcpy(m_tex_config, tc, sizeof(TexConfig));
+        m_single_channel = m_tex_config->m_colorization_mask;
+        if (m_tex_config->m_set_material)
+            m_material = material_manager->getMaterialFor(this);
     }
-
 #ifndef SERVER_ONLY
-    if (!CVS->isGLSL())
-        m_srgb = false;
+    if (m_tex_config && !CVS->isGLSL())
+        m_tex_config->m_srgb = false;
+#ifdef USE_GLES2
+    if (m_tex_config && !CVS->isDefferedEnabled())
+        m_tex_config->m_srgb = false;
+#endif
     if (!CVS->isARBTextureSwizzleUsable())
         m_single_channel = false;
 #endif
@@ -58,24 +64,25 @@ STKTexture::STKTexture(const std::string& path, bool srgb, bool premul_alpha,
 
 // ----------------------------------------------------------------------------
 STKTexture::STKTexture(uint8_t* data, const std::string& name, size_t size,
-                       bool single_channel)
-          : video::ITexture(name.c_str()), m_texture_handle(0), m_srgb(false),
-            m_premul_alpha(false), m_mesh_texture(false),
-            m_single_channel(single_channel), m_material(NULL),
-            m_texture_name(0), m_texture_size(0), m_texture_image(NULL)
+                       bool single_channel, bool delete_ttl)
+          : video::ITexture(name.c_str()), m_texture_handle(0),
+            m_single_channel(single_channel), m_tex_config(NULL),
+            m_material(NULL), m_texture_name(0), m_texture_size(0),
+            m_texture_image(NULL), m_file(NULL), m_img_loader(NULL)
 {
     m_size.Width = size;
     m_size.Height = size;
     m_orig_size = m_size;
-    reload(false/*no_upload*/, data);
+    if (!delete_ttl)
+        reload(false/*no_upload*/, data);
 }   // STKTexture
 
 // ----------------------------------------------------------------------------
 STKTexture::STKTexture(video::IImage* img, const std::string& name)
-          : video::ITexture(name.c_str()), m_texture_handle(0), m_srgb(false),
-            m_premul_alpha(false), m_mesh_texture(false),
-            m_single_channel(false), m_material(NULL), m_texture_name(0),
-            m_texture_size(0), m_texture_image(NULL)
+          : video::ITexture(name.c_str()), m_texture_handle(0),
+            m_single_channel(false), m_tex_config(NULL), m_material(NULL),
+            m_texture_name(0), m_texture_size(0), m_texture_image(NULL),
+            m_file(NULL), m_img_loader(NULL)
 {
     reload(false/*no_upload*/, NULL/*preload_data*/, img);
 }   // STKTexture
@@ -92,6 +99,7 @@ STKTexture::~STKTexture()
 #endif   // !SERVER_ONLY
     if (m_texture_image != NULL)
         m_texture_image->drop();
+    free(m_tex_config);
 }   // ~STKTexture
 
 // ----------------------------------------------------------------------------
@@ -108,11 +116,10 @@ void STKTexture::reload(bool no_upload, uint8_t* preload_data,
         return;
     }
 #ifndef SERVER_ONLY
-    irr_driver->getDevice()->getLogger()->setLogLevel(ELL_NONE);
 
     std::string compressed_texture;
 #if !defined(USE_GLES2)
-    if (!no_upload && m_mesh_texture && CVS->isTextureCompressionEnabled())
+    if (!no_upload && isMeshTexture() && CVS->isTextureCompressionEnabled())
     {
         std::string orig_file = NamedPath.getPtr();
 
@@ -157,77 +164,75 @@ void STKTexture::reload(bool no_upload, uint8_t* preload_data,
     uint8_t* data = preload_data;
     if (data == NULL)
     {
-        orig_img = preload_img ? preload_img :
-            irr_driver->getVideoDriver()->createImageFromFile(NamedPath);
-        if (orig_img == NULL)
+        if (preload_img)
+            orig_img = preload_img;
+        else
         {
-            return;
-        }
-
-        if (orig_img->getDimension().Width  == 0 ||
-            orig_img->getDimension().Height == 0)
-        {
-            orig_img->drop();
-            return;
+            m_file = irr_driver->getDevice()->getFileSystem()
+                ->createAndOpenFile(NamedPath);
+            if (m_file == NULL)
+                return;
+            irr_driver->getVideoDriver()->createImageFromFile(m_file,
+                &m_img_loader);
+            if (m_img_loader == NULL)
+                return;
+            m_file->seek(0);
+            m_orig_size = m_img_loader->getImageSize(m_file);
+            if ((!m_material || m_material->getAlphaMask().empty()) &&
+                useThreadedLoading() && !no_upload)
+            {
+                if (m_orig_size.Width == 0 || m_orig_size.Height == 0)
+                {
+                    m_file->drop();
+                    m_file = NULL;
+                    m_img_loader = NULL;
+                    return;
+                }
+            }
+            else
+            {
+                orig_img = m_img_loader->loadImage(m_file);
+                m_file->drop();
+                m_file = NULL;
+                if (orig_img == NULL || orig_img->getDimension().Width == 0 ||
+                    orig_img->getDimension().Height == 0)
+                {
+                    if (orig_img)
+                        orig_img->drop();
+                    return;
+                }
+                m_img_loader = NULL;
+            }
         }
         orig_img = resizeImage(orig_img, &m_orig_size, &m_size);
         applyMask(orig_img);
-        data = (uint8_t*)orig_img->lock();
-        if (m_single_channel)
+        data = orig_img ? (uint8_t*)orig_img->lock() : NULL;
+        if (m_single_channel && !useThreadedLoading())
         {
-            uint8_t* sc = new uint8_t[m_size.Width * m_size.Height];
-            for (unsigned int i = 0; i < m_size.Width * m_size.Height; i++)
-                sc[i] = data[4 * i + 3];
+            data = singleChannelConversion(data);
             orig_img->unlock();
             orig_img->drop();
             orig_img = NULL;
-            data = sc;
         }
     }
 
     const unsigned int w = m_size.Width;
     const unsigned int h = m_size.Height;
     unsigned int format = m_single_channel ? GL_RED : GL_BGRA;
-    unsigned int internal_format = m_single_channel ? GL_R8 : GL_RGBA;
+    unsigned int internal_format = m_single_channel ? GL_R8 : isSrgb() ? 
+                                                    GL_SRGB8_ALPHA8 : GL_RGBA8;
 
 #if !defined(USE_GLES2)
-    if (m_mesh_texture && CVS->isTextureCompressionEnabled())
+    if (isMeshTexture() && CVS->isTextureCompressionEnabled())
     {
         internal_format = m_single_channel ? GL_COMPRESSED_RED_RGTC1 :
-            m_srgb ? GL_COMPRESSED_SRGB_ALPHA_S3TC_DXT5_EXT :
+            isSrgb() ? GL_COMPRESSED_SRGB_ALPHA_S3TC_DXT5_EXT :
             GL_COMPRESSED_RGBA_S3TC_DXT5_EXT;
     }
-    else
-    {
-        internal_format =
-            m_single_channel ? GL_R8 : m_srgb ? GL_SRGB_ALPHA : GL_RGBA;
-    }
 #endif
 
-#if defined(USE_GLES2)
-    if (!CVS->isEXTTextureFormatBGRA8888Usable() && !m_single_channel)
-    {
-        format = GL_RGBA;
-        for (unsigned int i = 0; i < w * h; i++)
-        {
-            uint8_t tmp_val = data[i * 4];
-            data[i * 4] = data[i * 4 + 2];
-            data[i * 4 + 2] = tmp_val;
-        }
-    }
-#endif
-    if (m_premul_alpha && !m_single_channel)
-    {
-        for (unsigned int i = 0; i < w * h; i++)
-        {
-            float alpha = data[4 * i + 3];
-            if (alpha > 0.0f)
-                alpha = pow(alpha / 255.f, 1.f / 2.2f);
-            data[i * 4] = (uint8_t)(data[i * 4] * alpha);
-            data[i * 4 + 1] = (uint8_t)(data[i * 4 + 1] * alpha);
-            data[i * 4 + 2] = (uint8_t)(data[i * 4 + 2] * alpha);
-        }
-    }
+    if (!useThreadedLoading())
+        formatConversion(data, &format, w, h);
 
     if (!no_upload)
     {
@@ -246,17 +251,35 @@ void STKTexture::reload(bool no_upload, uint8_t* preload_data,
                 glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_SWIZZLE_B, GL_ONE);
                 glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_SWIZZLE_A, GL_RED);
             }
-            glTexImage2D(GL_TEXTURE_2D, 0, internal_format, w, h, 0, format,
-                GL_UNSIGNED_BYTE, data);
+            if (useThreadedLoading())
+            {
+                int levels = 1;
+                int width = w;
+                int height = h;
+                while (true)
+                {
+                    width = width < 2 ? 1 : width >> 1;
+                    height = height < 2 ? 1 : height >> 1;
+                    levels++;
+                    if (width == 1 && height == 1)
+                        break;
+                }
+                glTexStorage2D(GL_TEXTURE_2D, levels, internal_format, w, h);
+            }
+            else
+            {
+                glTexImage2D(GL_TEXTURE_2D, 0, internal_format, w, h, 0, format,
+                    GL_UNSIGNED_BYTE, data);
+            }
         }
-        else
+        else if (!useThreadedLoading())
         {
             glTexSubImage2D(GL_TEXTURE_2D, 0, 0, 0, w, h, format,
                 GL_UNSIGNED_BYTE, data);
         }
         if (orig_img)
             orig_img->unlock();
-        if (hasMipMaps())
+        if (!useThreadedLoading() && hasMipMaps())
             glGenerateMipmap(GL_TEXTURE_2D);
     }
 
@@ -273,54 +296,74 @@ void STKTexture::reload(bool no_upload, uint8_t* preload_data,
     if (!no_upload)
         glBindTexture(GL_TEXTURE_2D, 0);
 
-    irr_driver->getDevice()->getLogger()->setLogLevel(ELL_WARNING);
 #endif   // !SERVER_ONLY
 }   // reload
 
+//-----------------------------------------------------------------------------
+void STKTexture::formatConversion(uint8_t* data, unsigned int* format,
+                                  unsigned int w, unsigned int h) const
+{
+#if defined(USE_GLES2)
+    if (!m_single_channel)
+    {
+        if (format)
+            *format = GL_RGBA;
+        for (unsigned int i = 0; i < w * h; i++)
+        {
+            uint8_t tmp_val = data[i * 4];
+            data[i * 4] = data[i * 4 + 2];
+            data[i * 4 + 2] = tmp_val;
+        }
+    }
+#endif
+    if (isPremulAlpha() && !m_single_channel)
+    {
+        for (unsigned int i = 0; i < w * h; i++)
+        {
+            float alpha = data[4 * i + 3];
+            if (alpha > 0.0f)
+            {
+                alpha /= 255.0f;
+#if defined(USE_GLES2)
+                if (CVS->isDefferedEnabled())
+#endif
+                    alpha = pow(alpha, 1.0f / 2.2f);
+            }
+            data[i * 4] = (uint8_t)(data[i * 4] * alpha);
+            data[i * 4 + 1] = (uint8_t)(data[i * 4 + 1] * alpha);
+            data[i * 4 + 2] = (uint8_t)(data[i * 4 + 2] * alpha);
+        }
+    }
+}   // formatConversion
+
 // ----------------------------------------------------------------------------
 video::IImage* STKTexture::resizeImage(video::IImage* orig_img,
-                                       core::dimension2du* new_img_size,
-                                       core::dimension2du* new_tex_size)
+                                       core::dimension2du* orig_size,
+                                       core::dimension2du* final_size) const
 {
     video::IImage* image = orig_img;
 #ifndef SERVER_ONLY
-    const core::dimension2du& old_size = image->getDimension();
-    core::dimension2du img_size = old_size;
-
-    const float ratio = float(img_size.Width) / float(img_size.Height);
-    const unsigned int drv_max_size =
-        irr_driver->getVideoDriver()->getMaxTextureSize().Width;
-
-    if ((img_size.Width > drv_max_size) && (ratio >= 1.0f))
-    {
-        img_size.Width = drv_max_size;
-        img_size.Height = (unsigned)(drv_max_size / ratio);
-    }
-    else if (img_size.Height > drv_max_size)
-    {
-        img_size.Height = drv_max_size;
-        img_size.Width = (unsigned)(drv_max_size * ratio);
-    }
-
-    if (img_size != old_size)
-    {
-        video::IImage* new_img = irr_driver->getVideoDriver()
-            ->createImage(video::ECF_A8R8G8B8, img_size);
-        image->copyToScaling(new_img);
-        image->drop();
-        image = new_img;
-    }
-
+    if (image == NULL)
+        assert(orig_size && orig_size->Width > 0 && orig_size->Height > 0);
+    core::dimension2du img_size = image ? image->getDimension() : *orig_size;
     core::dimension2du tex_size = img_size.getOptimalSize
         (!irr_driver->getVideoDriver()->queryFeature(video::EVDF_TEXTURE_NPOT));
     const core::dimension2du& max_size = irr_driver->getVideoDriver()
         ->getDriverAttributes().getAttributeAsDimension2d("MAX_TEXTURE_SIZE");
 
-    if (max_size.Width > 0 && tex_size.Width > max_size.Width)
+    if (tex_size.Width > max_size.Width)
         tex_size.Width = max_size.Width;
-    if (max_size.Height > 0 && tex_size.Height > max_size.Height)
+    if (tex_size.Height > max_size.Height)
         tex_size.Height = max_size.Height;
 
+    if (orig_size && final_size)
+    {
+        *orig_size = img_size;
+        *final_size = tex_size;
+    }
+    if (image == NULL)
+        return NULL;
+
     if (image->getColorFormat() != video::ECF_A8R8G8B8 ||
         tex_size != img_size)
     {
@@ -334,11 +377,6 @@ video::IImage* STKTexture::resizeImage(video::IImage* orig_img,
         image = new_texture;
     }
 
-    if (new_img_size && new_tex_size)
-    {
-        *new_img_size = img_size;
-        *new_tex_size = tex_size;
-    }
 #endif   // !SERVER_ONLY
     return image;
 }   // resizeImage
@@ -548,3 +586,101 @@ void STKTexture::unloadHandle()
     }
 #endif
 }   // unloadHandle
+
+//-----------------------------------------------------------------------------
+bool STKTexture::useThreadedLoading() const
+{
+#ifdef SERVER_ONLY
+    return false;
+#else
+    return CVS->supportsThreadedTextureLoading() &&
+        !CVS->isTextureCompressionEnabled() && isMeshTexture() &&
+        m_img_loader && m_img_loader->supportThreadedLoading();
+#endif
+}   // useThreadedLoading
+
+//-----------------------------------------------------------------------------
+void STKTexture::threadedReload(void* ptr, void* param) const
+{
+#if !(defined(SERVER_ONLY) || defined(USE_GLES2))
+    video::IImage* orig_img =
+        m_img_loader->loadImage(m_file, true/*skip_checking*/);
+    orig_img = resizeImage(orig_img);
+    uint8_t* data = (uint8_t*)orig_img->lock();
+    if (m_single_channel)
+    {
+        data = singleChannelConversion(data);
+        orig_img->unlock();
+        orig_img->drop();
+        orig_img = NULL;
+    }
+    formatConversion(data, NULL, m_size.Width, m_size.Height);
+    memcpy(ptr, data, m_texture_size);
+
+    if (orig_img)
+    {
+        orig_img->unlock();
+        orig_img->setDeleteMemory(false);
+        orig_img->drop();
+    }
+    if (useHQMipmap())
+    {
+        HQMipmapGenerator* hqmg = new HQMipmapGenerator(NamedPath, data,
+            m_size, m_texture_name, m_tex_config);
+        ((STKTexManager*)(param))->addThreadedLoadTexture(hqmg);
+    }
+    else
+        delete[] data;
+#endif
+}   // threadedReload
+
+//-----------------------------------------------------------------------------
+void STKTexture::threadedSubImage(void* ptr) const
+{
+#if !(defined(SERVER_ONLY) || defined(USE_GLES2))
+    glBindTexture(GL_TEXTURE_2D, m_texture_name);
+    glTexSubImage2D(GL_TEXTURE_2D, 0, 0, 0, m_size.Width, m_size.Height,
+        m_single_channel ? GL_RED : GL_BGRA, GL_UNSIGNED_BYTE, ptr);
+    if (useHQMipmap())
+        return;
+    if (hasMipMaps())
+        glGenerateMipmap(GL_TEXTURE_2D);
+
+#endif
+}   // threadedSubImage
+
+//-----------------------------------------------------------------------------
+void STKTexture::cleanThreadedLoader()
+{
+#if !(defined(SERVER_ONLY) || defined(USE_GLES2))
+    assert(m_file);
+    m_file->drop();
+    m_file = NULL;
+    m_img_loader = NULL;
+#endif
+}   // cleanThreadedLoader
+
+//-----------------------------------------------------------------------------
+bool STKTexture::useHQMipmap() const
+{
+    return !m_single_channel && UserConfigParams::m_hq_mipmap &&
+        m_size.Width > 1 && m_size.Height > 1;
+}   // useHQMipmap
+
+//-----------------------------------------------------------------------------
+bool STKTexture::isSrgb() const
+{
+    return m_tex_config && m_tex_config->m_srgb;
+}   // isSrgb
+
+//-----------------------------------------------------------------------------
+bool STKTexture::isPremulAlpha() const
+{
+    return m_tex_config && m_tex_config->m_premul_alpha;
+}   // isPremulAlpha
+
+//-----------------------------------------------------------------------------
+bool STKTexture::isMeshTexture() const
+{
+    return m_tex_config && m_tex_config->m_mesh_tex;
+}   // isMeshTexture
diff --git a/src/graphics/stk_texture.hpp b/src/graphics/stk_texture.hpp
index 34be33cbc..3eb09c0b2 100644
--- a/src/graphics/stk_texture.hpp
+++ b/src/graphics/stk_texture.hpp
@@ -24,8 +24,15 @@
 #include <string>
 #include <ITexture.h>
 
+namespace irr
+{
+    namespace io { class IReadFile; }
+    namespace video { class IImageLoader; }
+}
+
 using namespace irr;
 
+struct TexConfig;
 class Material;
 
 class STKTexture : public video::ITexture, NoCopy
@@ -35,7 +42,9 @@ private:
 
     uint64_t m_texture_handle;
 
-    bool m_srgb, m_premul_alpha, m_mesh_texture, m_single_channel;
+    bool m_single_channel;
+
+    TexConfig* m_tex_config;
 
     Material* m_material;
 
@@ -45,26 +54,44 @@ private:
 
     video::IImage* m_texture_image;
 
+    io::IReadFile* m_file;
+
+    video::IImageLoader* m_img_loader;
+
     // ------------------------------------------------------------------------
     video::IImage* resizeImage(video::IImage* orig_img,
-                               core::dimension2du* new_img_size = NULL,
-                               core::dimension2du* new_tex_size = NULL);
+                               core::dimension2du* orig_size = NULL,
+                               core::dimension2du* final_size = NULL) const;
     // ------------------------------------------------------------------------
     void applyMask(video::IImage* orig_img);
     // ------------------------------------------------------------------------
     bool loadCompressedTexture(const std::string& file_name);
     // ------------------------------------------------------------------------
     void saveCompressedTexture(const std::string& file_name);
+    // ------------------------------------------------------------------------
+    void formatConversion(uint8_t* data, unsigned int* format, unsigned int w,
+                          unsigned int h) const;
+    // ------------------------------------------------------------------------
+    uint8_t* singleChannelConversion(uint8_t* data) const
+    {
+        uint8_t* sc = new uint8_t[m_size.Width * m_size.Height];
+        for (unsigned int i = 0; i < m_size.Width * m_size.Height; i++)
+            sc[i] = data[4 * i + 3];
+        return sc;
+    }
+    // ------------------------------------------------------------------------
+    bool useHQMipmap() const;
+    // ------------------------------------------------------------------------
+    bool isSrgb() const;
+    // ------------------------------------------------------------------------
+    bool isPremulAlpha() const;
 
 public:
     // ------------------------------------------------------------------------
-    STKTexture(const std::string& path, bool srgb = false,
-               bool premul_alpha = false, bool set_material = false,
-               bool mesh_tex = false, bool no_upload = false,
-               bool single_channel = false);
+    STKTexture(const std::string& path, TexConfig* tc, bool no_upload = false);
     // ------------------------------------------------------------------------
     STKTexture(uint8_t* data, const std::string& name, size_t size,
-               bool single_channel = false);
+               bool single_channel = false, bool delete_ttl = false);
     // ------------------------------------------------------------------------
     STKTexture(video::IImage* img, const std::string& name);
     // ------------------------------------------------------------------------
@@ -108,20 +135,27 @@ public:
     // ------------------------------------------------------------------------
     virtual void unloadHandle();
     // ------------------------------------------------------------------------
-    bool isSrgb() const                                      { return m_srgb; }
-    // ------------------------------------------------------------------------
-    bool isPremulAlpha() const                       { return m_premul_alpha; }
-    // ------------------------------------------------------------------------
-    bool isMeshTexture() const                       { return m_mesh_texture; }
-    // ------------------------------------------------------------------------
-    void setMeshTexture(bool val)                     { m_mesh_texture = val; }
-    // ------------------------------------------------------------------------
-    unsigned int getTextureSize() const              { return m_texture_size; }
+    virtual unsigned int getTextureSize() const      { return m_texture_size; }
     // ------------------------------------------------------------------------
     void reload(bool no_upload = false, uint8_t* preload_data = NULL,
                 video::IImage* preload_img = NULL);
     // ------------------------------------------------------------------------
     video::IImage* getTextureImage()                { return m_texture_image; }
+    // ------------------------------------------------------------------------
+    bool useThreadedLoading() const;
+    // ------------------------------------------------------------------------
+    virtual void threadedReload(void* ptr, void* param) const;
+    // ------------------------------------------------------------------------
+    virtual void threadedSubImage(void* ptr) const;
+    // ------------------------------------------------------------------------
+    virtual void cleanThreadedLoader();
+    // ------------------------------------------------------------------------
+    virtual int getThreadedLoadTextureCounter() const
+    {
+        return useHQMipmap() ? 2 : 1;
+    }
+    // ------------------------------------------------------------------------
+    bool isMeshTexture() const;
 
 };   // STKTexture
 
diff --git a/src/graphics/threaded_tex_loader.cpp b/src/graphics/threaded_tex_loader.cpp
new file mode 100644
index 000000000..564fd75e8
--- /dev/null
+++ b/src/graphics/threaded_tex_loader.cpp
@@ -0,0 +1,100 @@
+//  SuperTuxKart - a fun racing game with go-kart
+//  Copyright (C) 2017 SuperTuxKart-Team
+//
+//  This program is free software; you can redistribute it and/or
+//  modify it under the terms of the GNU General Public License
+//  as published by the Free Software Foundation; either version 3
+//  of the License, or (at your option) any later version.
+//
+//  This program 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.
+//
+//  You should have received a copy of the GNU General Public License
+//  along with this program; if not, write to the Free Software
+//  Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
+
+#if !(defined(SERVER_ONLY) || defined(USE_GLES2))
+
+#include "graphics/threaded_tex_loader.hpp"
+#include "graphics/stk_tex_manager.hpp"
+#include "utils/string_utils.hpp"
+#include "utils/vs.hpp"
+
+#include <cassert>
+#include <ITexture.h>
+
+// ----------------------------------------------------------------------------
+void* ThreadedTexLoader::startRoutine(void *obj)
+{
+    ThreadedTexLoader* ttl = (ThreadedTexLoader*)obj;
+    VS::setThreadName((std::string("ThrTexLoader") +
+        StringUtils::toString(ttl->m_pbo_offset / 1024 / 1024)).c_str());
+    while (true)
+    {
+        pthread_mutex_lock(&ttl->m_mutex);
+        bool finished = ttl->finishedLoading();
+        pthread_mutex_unlock(&ttl->m_mutex);
+        if (finished)
+        {
+            continue;
+        }
+        pthread_mutex_lock(ttl->m_texture_queue_mutex);
+        bool waiting = ttl->m_stktm->isThreadedLoadTexturesEmpty();
+        ttl->m_last_queue_ready.setAtomic(!ttl->m_completed_textures.empty() &&
+            waiting);
+        while (waiting)
+        {
+            pthread_cond_wait(ttl->m_cond_request, ttl->m_texture_queue_mutex);
+            waiting = ttl->m_stktm->isThreadedLoadTexturesEmpty();
+        }
+        irr::video::ITexture* target_tex =
+            ttl->m_stktm->getThreadedLoadTexture();
+        if (strcmp(target_tex->getName().getPtr(), "delete_ttl") == 0)
+        {
+            ttl->m_stktm->removeThreadedLoadTexture();
+            ttl->m_stktm->setThreadedLoadTextureCounter(-1);
+            pthread_mutex_unlock(ttl->m_texture_queue_mutex);
+            ttl->setCanBeDeleted();
+            return NULL;
+        }
+        assert(target_tex->getTextureSize() <= ttl->m_tex_capacity);
+        if (target_tex->getTextureSize() + ttl->m_tex_size_loaded >
+            ttl->m_tex_capacity)
+        {
+            pthread_mutex_lock(&ttl->m_mutex);
+            ttl->setFinishLoading();
+            pthread_mutex_unlock(&ttl->m_mutex);
+            pthread_mutex_unlock(ttl->m_texture_queue_mutex);
+            continue;
+        }
+        ttl->m_stktm->removeThreadedLoadTexture();
+        pthread_mutex_unlock(ttl->m_texture_queue_mutex);
+        target_tex->threadedReload(ttl->m_pbo_ptr + ttl->m_tex_size_loaded,
+            ttl->m_stktm);
+        target_tex->cleanThreadedLoader();
+        ttl->m_tex_size_loaded += target_tex->getTextureSize();
+        ttl->m_completed_textures.push_back(target_tex);
+        pthread_mutex_lock(ttl->m_texture_queue_mutex);
+        ttl->m_stktm->setThreadedLoadTextureCounter(-1);
+        pthread_mutex_unlock(ttl->m_texture_queue_mutex);
+    }
+    return NULL;
+}   // startRoutine
+
+// ----------------------------------------------------------------------------
+void ThreadedTexLoader::handleCompletedTextures()
+{
+    assert(m_locked);
+    size_t offset = m_pbo_offset;
+    for (irr::video::ITexture* tex : m_completed_textures)
+    {
+        size_t cur_offset = tex->getTextureSize();
+        tex->threadedSubImage((void*)offset);
+        offset += cur_offset;
+    }
+    m_completed_textures.clear();
+}   // handleCompletedTextures
+
+#endif
diff --git a/src/graphics/threaded_tex_loader.hpp b/src/graphics/threaded_tex_loader.hpp
new file mode 100644
index 000000000..22bd15d08
--- /dev/null
+++ b/src/graphics/threaded_tex_loader.hpp
@@ -0,0 +1,118 @@
+//  SuperTuxKart - a fun racing game with go-kart
+//  Copyright (C) 2017 SuperTuxKart-Team
+//
+//  This program is free software; you can redistribute it and/or
+//  modify it under the terms of the GNU General Public License
+//  as published by the Free Software Foundation; either version 3
+//  of the License, or (at your option) any later version.
+//
+//  This program 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.
+//
+//  You should have received a copy of the GNU General Public License
+//  along with this program; if not, write to the Free Software
+//  Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
+
+#ifndef HEADER_THREADED_TEX_LOADER_HPP
+#define HEADER_THREADED_TEX_LOADER_HPP
+
+#if !(defined(SERVER_ONLY) || defined(USE_GLES2))
+
+#include "utils/can_be_deleted.hpp"
+#include "utils/no_copy.hpp"
+#include "utils/synchronised.hpp"
+#include "utils/types.hpp"
+
+#include <vector>
+
+namespace irr
+{
+    namespace video { class ITexture; }
+}
+class STKTexManager;
+
+class ThreadedTexLoader : public NoCopy, public CanBeDeleted
+{
+private:
+    const unsigned m_tex_capacity;
+
+    const size_t m_pbo_offset;
+
+    uint8_t* m_pbo_ptr;
+
+    pthread_mutex_t m_mutex;
+
+    pthread_mutex_t* m_texture_queue_mutex;
+
+    pthread_cond_t* m_cond_request;
+
+    STKTexManager* m_stktm;
+
+    unsigned m_tex_size_loaded;
+
+    pthread_t m_thread;
+
+    bool m_finished_loading, m_locked;
+
+    Synchronised<bool> m_last_queue_ready;
+
+    std::vector<irr::video::ITexture*> m_completed_textures;
+
+public:
+    // ------------------------------------------------------------------------
+    static void* startRoutine(void *obj);
+    // ------------------------------------------------------------------------
+    ThreadedTexLoader(unsigned capacity, size_t offset, uint8_t* pbo_ptr,
+                      pthread_mutex_t* mutex, pthread_cond_t* cond,
+                      STKTexManager* stktm)
+                    : m_tex_capacity(capacity), m_pbo_offset(offset),
+                      m_pbo_ptr(pbo_ptr), m_texture_queue_mutex(mutex),
+                      m_cond_request(cond), m_stktm(stktm),
+                      m_tex_size_loaded(0), m_finished_loading(false),
+                      m_locked(false), m_last_queue_ready(false)
+    {
+        pthread_mutex_init(&m_mutex, NULL);
+        pthread_create(&m_thread, NULL, &startRoutine, this);
+    }
+    // ------------------------------------------------------------------------
+    ~ThreadedTexLoader()
+    {
+        pthread_mutex_destroy(&m_mutex);
+        pthread_join(m_thread, NULL);
+    }
+    // ------------------------------------------------------------------------
+    bool finishedLoading() const                 { return m_finished_loading; }
+    // ------------------------------------------------------------------------
+    void setFinishLoading()
+    {
+        m_last_queue_ready.setAtomic(false);
+        m_finished_loading = true;
+        m_tex_size_loaded = 0;
+    }
+    // ------------------------------------------------------------------------
+    bool lastQueueReady() const      { return m_last_queue_ready.getAtomic(); }
+    // ------------------------------------------------------------------------
+    void handleCompletedTextures();
+    // ------------------------------------------------------------------------
+    void lock()
+    {
+        pthread_mutex_lock(&m_mutex);
+        m_locked = true;
+    }
+    // ------------------------------------------------------------------------
+    void unlock(bool finish_it)
+    {
+        if (!m_locked) return;
+        m_locked = false;
+        if (finish_it)
+            m_finished_loading = false;
+        pthread_mutex_unlock(&m_mutex);
+    }
+
+};   // ThreadedTexLoader
+
+#endif
+
+#endif
diff --git a/src/guiengine/message_queue.cpp b/src/guiengine/message_queue.cpp
index 548c5a86c..2063d1e2a 100644
--- a/src/guiengine/message_queue.cpp
+++ b/src/guiengine/message_queue.cpp
@@ -20,18 +20,25 @@
 
 #include "guiengine/message_queue.hpp"
 
-#include "config/user_config.hpp"
+#include "graphics/irr_driver.hpp"
 #include "guiengine/engine.hpp"
-#include "guiengine/scalable_font.hpp"
 #include "guiengine/skin.hpp"
+#include "utils/synchronised.hpp"
+#include "utils/translation.hpp"
 
 #include "IGUIElement.h"
+#include "IGUIEnvironment.h"
+#include "IGUIStaticText.h"
 
 using namespace GUIEngine;
 
 namespace MessageQueue
 {
+// ============================================================================
+/** The area which the message is drawn. */
+core::recti g_area;
 
+// ============================================================================
 /** A small helper class to store and sort messages to be displayed. */
 class Message
 {
@@ -45,11 +52,15 @@ private:
      *  or friend-message::neutral. */
     std::string m_render_type;
 
+    /** The text label, can do linebreak if needed. */
+    gui::IGUIStaticText* m_text;
+
 public:
     Message(MessageQueue::MessageType mt, const core::stringw &message)
     {
         m_message_type = mt;
         m_message      = message;
+        m_text         = NULL;
         if(mt==MessageQueue::MT_ACHIEVEMENT)
             m_render_type = "achievement-message::neutral";
         else if (mt==MessageQueue::MT_ERROR)
@@ -60,6 +71,11 @@ public:
             m_render_type = "friend-message::neutral";
     }   // Message
     // ------------------------------------------------------------------------
+    ~Message()
+    {
+        assert(m_text != NULL);
+        m_text->drop();
+    }
     /** Returns the message. */
     const core::stringw & getMessage() const { return m_message; }
     // ------------------------------------------------------------------------
@@ -72,6 +88,39 @@ public:
     {
         return m_render_type;
     }
+    // ------------------------------------------------------------------------
+    /** Init the message text, do linebreak as required. */
+    void init()
+    {
+        const GUIEngine::BoxRenderParams &brp =
+            GUIEngine::getSkin()->getBoxRenderParams(m_render_type);
+        const unsigned width = irr_driver->getActualScreenSize().Width;
+        const unsigned height = irr_driver->getActualScreenSize().Height;
+        const unsigned max_width = width - (brp.m_left_border +
+            brp.m_right_border);
+        m_text =
+            GUIEngine::getGUIEnv()->addStaticText(m_message.c_str(),
+            core::recti(0, 0, max_width, height));
+        m_text->setRightToLeft(translations->isRTLText(m_message));
+        core::dimension2du dim(m_text->getTextWidth(),
+            m_text->getTextHeight());
+        dim.Width += brp.m_left_border + brp.m_right_border;
+        int x = (width - dim.Width) / 2;
+        int y = height - int(1.5f * dim.Height);
+        g_area = irr::core::recti(x, y, x + dim.Width, y + dim.Height);
+        m_text->setRelativePosition(g_area);
+        m_text->setTextAlignment(gui::EGUIA_CENTER, gui::EGUIA_CENTER);
+        m_text->grab();
+        m_text->remove();
+    }
+    // ------------------------------------------------------------------------
+    /** Draw the message. */
+    void draw()
+    {
+        assert(m_text != NULL);
+        m_text->draw();
+    }
+
 };   // class Message
 
 // ============================================================================
@@ -88,11 +137,10 @@ public:
     }   // operator ()
 };   // operator()
 
-
 // ============================================================================
 /** List of all messages. */
-std::priority_queue<Message*, std::vector<Message*>,
-                   CompareMessages> g_all_messages;
+Synchronised<std::priority_queue<Message*, std::vector<Message*>,
+                   CompareMessages> > g_all_messages;
 
 /** How long the current message has been displayed. The special value
  *  -1 indicates that a new message was added when the queue was empty. */
@@ -103,26 +151,18 @@ float        g_max_display_time     = 5.0f;
 
 /** The label widget used to show the current message. */
 SkinWidgetContainer *g_container    = NULL;
-core::recti g_area;
 
 // ============================================================================
 
-void createLabel(const Message *message)
+void createLabel(Message *message)
 {
     if(!g_container)
         g_container = new SkinWidgetContainer();
 
-    gui::ScalableFont *font = GUIEngine::getFont();
-    core::dimension2du dim = font->getDimension(message->getMessage().c_str());
     g_current_display_time = 0.0f;
     // Maybe make this time dependent on message length as well?
     g_max_display_time     = 5.0f;
-    const GUIEngine::BoxRenderParams &brp =
-        GUIEngine::getSkin()->getBoxRenderParams(message->getRenderType());
-    dim.Width +=brp.m_left_border + brp.m_right_border;
-    int x = (UserConfigParams::m_width - dim.Width) / 2;
-    int y = UserConfigParams::m_height - int(1.5f*dim.Height);
-    g_area = irr::core::recti(x, y, x+dim.Width, y+dim.Height);
+    message->init();
 }   // createLabel
 
 // ----------------------------------------------------------------------------
@@ -130,9 +170,16 @@ void createLabel(const Message *message)
  *  position of the message. */
 void updatePosition()
 {
-    if (g_all_messages.empty()) return;
-    Message *last = g_all_messages.top();
+    g_all_messages.lock();
+    bool empty = g_all_messages.getData().empty();
+    if (empty)
+    {
+        g_all_messages.unlock();
+        return;
+    }
+    Message *last = g_all_messages.getData().top();
     createLabel(last);
+    g_all_messages.unlock();
 }   // updatePosition
 
 // ----------------------------------------------------------------------------
@@ -143,13 +190,15 @@ void updatePosition()
 void add(MessageType mt, const irr::core::stringw &message)
 {
     Message *m = new Message(mt, message);
-    if(g_all_messages.empty())
+    g_all_messages.lock();
+    if (g_all_messages.getData().empty())
     {
         // Indicate that there is a new message, which should
         // which needs a new label etc. to be computed.
         g_current_display_time =-1.0f;
     }
-    g_all_messages.push(m);
+    g_all_messages.getData().push(m);
+    g_all_messages.unlock();
 }   // add
 
 // ----------------------------------------------------------------------------
@@ -161,32 +210,38 @@ void add(MessageType mt, const irr::core::stringw &message)
  */
 void update(float dt)
 {
-    if(g_all_messages.empty()) return;
+    g_all_messages.lock();
+    bool empty = g_all_messages.getData().empty();
+    g_all_messages.unlock();
+    if (empty) return;
 
+    g_all_messages.lock();
     g_current_display_time += dt;
-    if(g_current_display_time > g_max_display_time)
+    if (g_current_display_time > g_max_display_time)
     {
-        Message *last = g_all_messages.top();
-        g_all_messages.pop();
+        Message *last = g_all_messages.getData().top();
+        g_all_messages.getData().pop();
         delete last;
-        if(g_all_messages.empty()) return;
+        if (g_all_messages.getData().empty())
+        {
+            g_all_messages.unlock();
+            return;
+        }
         g_current_display_time = -1.0f;
     }
 
+    Message *current = g_all_messages.getData().top();
     // Create new data for the display.
-    if(g_current_display_time < 0)
+    if (g_current_display_time < 0)
     {
-        createLabel(g_all_messages.top());
+        createLabel(current);
     }
+    g_all_messages.unlock();
 
-    Message *current = g_all_messages.top();
     GUIEngine::getSkin()->drawMessage(g_container, g_area,
                                       current->getRenderType());
-    gui::ScalableFont *font = GUIEngine::getFont();
-    
-    video::SColor color(255, 0, 0, 0);
-    font->draw(current->getMessage(), g_area, color, true, true);
-    
+    current->draw();
+
 }   // update
 
 }   // namespace GUIEngine
diff --git a/src/guiengine/widgets/icon_button_widget.cpp b/src/guiengine/widgets/icon_button_widget.cpp
index 730710f4d..379af20fe 100644
--- a/src/guiengine/widgets/icon_button_widget.cpp
+++ b/src/guiengine/widgets/icon_button_widget.cpp
@@ -342,9 +342,7 @@ video::ITexture* IconButtonWidget::getDeactivatedTexture(video::ITexture* textur
     name += "_disabled";
     STKTexManager* stkm = STKTexManager::getInstance();
     STKTexture* disabled_stk_tex = static_cast<STKTexture*>(stkm->getTexture
-        (name, false/*srgb*/, false/*premul_alpha*/, false/*set_material*/,
-        false/*mesh_tex*/, false /*no_upload*/, false/*single_channel*/,
-        false/*create_if_unfound*/));
+        (name, NULL/*tc*/, false /*no_upload*/, false/*create_if_unfound*/));
     if (disabled_stk_tex == NULL)
     {
         SColor c;
diff --git a/src/input/input_manager.cpp b/src/input/input_manager.cpp
index df3308a32..9594ca1ef 100644
--- a/src/input/input_manager.cpp
+++ b/src/input/input_manager.cpp
@@ -273,8 +273,18 @@ void InputManager::handleStaticAction(int key, int value)
         case KEY_PRINT:
             // on windows we don't get a press event, only release.  So
             // save on release only (to avoid saving twice on other platforms)
-            if (value ==0 )
-                irr_driver->requestScreenshot();
+            if (value == 0)
+            {
+                if (control_is_pressed)
+                {
+                    const bool is_recording = irr_driver->isRecording();
+                    irr_driver->setRecording(!is_recording);
+                }
+                else
+                {
+                    irr_driver->requestScreenshot();
+                }
+            }
             break;
         case KEY_F11:
             if(value && shift_is_pressed && world && RewindManager::isEnabled())
diff --git a/src/main_loop.cpp b/src/main_loop.cpp
index 80b1aec28..6f97697f2 100644
--- a/src/main_loop.cpp
+++ b/src/main_loop.cpp
@@ -117,7 +117,9 @@ float MainLoop::getLimitedDt()
         // Throttle fps if more than maximum, which can reduce
         // the noise the fan on a graphics card makes.
         // When in menus, reduce FPS much, it's not necessary to push to the maximum for plain menus
-        const int max_fps = (StateManager::get()->throttleFPS() ? 30 : UserConfigParams::m_max_fps);
+        const int max_fps = (irr_driver->isRecording() &&
+            UserConfigParams::m_limit_game_fps ? UserConfigParams::m_record_fps :
+            StateManager::get()->throttleFPS() ? 60 : UserConfigParams::m_max_fps);
         if (dt > 0)
         {
             const int current_fps = (int)(1000.0f / dt);
diff --git a/src/modes/soccer_world.cpp b/src/modes/soccer_world.cpp
index 60bcaa37c..1f660ada1 100644
--- a/src/modes/soccer_world.cpp
+++ b/src/modes/soccer_world.cpp
@@ -331,13 +331,12 @@ void SoccerWorld::initKartList()
     std::string blue_path =
             file_manager->getAsset(FileManager::GUI, "soccer_player_blue.png");
 
+    TexConfig btc(true/*srgb*/, true/*premul_alpha*/);
     video::ITexture* red = STKTexManager::getInstance()->getTexture
-        (red_path, true/*srgb*/, true/*premul_alpha*/, false/*set_material*/,
-        true/*mesh_tex*/);
+        (red_path, &btc);
 
     video::ITexture* blue = STKTexManager::getInstance()->getTexture
-        (blue_path, true/*srgb*/, true/*premul_alpha*/, false/*set_material*/,
-        true/*mesh_tex*/);
+        (blue_path, &btc);
 
     //Assigning indicators
     for(unsigned int i = 0; i < kart_amount; i++)
diff --git a/src/modes/three_strikes_battle.cpp b/src/modes/three_strikes_battle.cpp
index 87052c086..86f789243 100644
--- a/src/modes/three_strikes_battle.cpp
+++ b/src/modes/three_strikes_battle.cpp
@@ -179,9 +179,9 @@ void ThreeStrikesBattle::kartAdded(AbstractKart* kart, scene::ISceneNode* node)
         // Add heart billboard above it
         std::string heart_path =
             file_manager->getAsset(FileManager::GUI, "heart.png");
+        TexConfig btc(true/*srgb*/, true/*premul_alpha*/);
         video::ITexture* heart = STKTexManager::getInstance()->getTexture
-            (heart_path, true/*srgb*/, true/*premul_alpha*/,
-            false/*set_material*/, true/*mesh_tex*/);
+            (heart_path, &btc);
 
         float height = kart->getKartHeight() + 0.5f;
 
diff --git a/src/states_screens/dialogs/custom_video_settings.cpp b/src/states_screens/dialogs/custom_video_settings.cpp
index b0435834d..b7b5a6dd5 100644
--- a/src/states_screens/dialogs/custom_video_settings.cpp
+++ b/src/states_screens/dialogs/custom_video_settings.cpp
@@ -61,8 +61,6 @@ void CustomVideoSettingsDialog::beforeAddingWidgets()
     getWidget<CheckBoxWidget>("anim_gfx")->setState(UserConfigParams::m_graphical_effects);
     getWidget<CheckBoxWidget>("weather_gfx")->setState(UserConfigParams::m_weather_effects);
     getWidget<CheckBoxWidget>("dof")->setState(UserConfigParams::m_dof);
-    getWidget<CheckBoxWidget>("hd-textures")
-        ->setState((UserConfigParams::m_high_definition_textures & 0x01)==0x01);
 
     SpinnerWidget* kart_anim = getWidget<SpinnerWidget>("steering_animations");
     kart_anim->addLabel(_("Disabled")); // 0
@@ -79,33 +77,24 @@ void CustomVideoSettingsDialog::beforeAddingWidgets()
     //I18N: Geometry level disabled : lowest level, no details
     geometry_level->addLabel(_("Disabled"));
     //I18N: Geometry level low : few details are displayed
-    geometry_level->addLabel(_("low"));
+    geometry_level->addLabel(_("Low"));
     //I18N: Geometry level high : everything is displayed
-    geometry_level->addLabel(_("high"));
+    geometry_level->addLabel(_("High"));
     geometry_level->setValue(
         UserConfigParams::m_geometry_level == 2 ? 0 :
         UserConfigParams::m_geometry_level == 0 ? 2 : 1);
 
-    SpinnerWidget* filtering = getWidget<SpinnerWidget>("filtering");
-    int value = 0;
-    if (UserConfigParams::m_anisotropic == 2)  value = 2;
-    else if (UserConfigParams::m_anisotropic == 4)  value = 3;
-    else if (UserConfigParams::m_anisotropic == 8)  value = 4;
-    else if (UserConfigParams::m_anisotropic == 16) value = 5;
-    else if (UserConfigParams::m_trilinear)         value = 1;
-    filtering->addLabel(_("Bilinear"));        // 0
-    filtering->addLabel(_("Trilinear"));       // 1
-    filtering->addLabel(_("Anisotropic x2"));  // 2
-    filtering->addLabel(_("Anisotropic x4"));  // 3
-    filtering->addLabel(_("Anisotropic x8"));  // 4
-    filtering->addLabel(_("Anisotropic x16")); // 5
-
-    filtering->setValue(value);
+    SpinnerWidget* filtering = getWidget<SpinnerWidget>("image_quality");
+    filtering->addLabel(_("Very Low"));
+    filtering->addLabel(_("Low"));
+    filtering->addLabel(_("High"));
+    filtering->addLabel(_("Very High"));
+    filtering->setValue(OptionsScreenVideo::getImageQuality());
 
     SpinnerWidget* shadows = getWidget<SpinnerWidget>("shadows");
     shadows->addLabel(_("Disabled"));   // 0
-    shadows->addLabel(_("low"));        // 1
-    shadows->addLabel(_("high"));       // 2
+    shadows->addLabel(_("Low"));        // 1
+    shadows->addLabel(_("High"));       // 2
     if (CVS->supportsShadows())
         shadows->setValue(UserConfigParams::m_shadows_resolution / 512);
     else
@@ -195,11 +184,6 @@ GUIEngine::EventPropagation CustomVideoSettingsDialog::processEvent(const std::s
         UserConfigParams::m_weather_effects =
             getWidget<CheckBoxWidget>("weather_gfx")->getState();
 
-        // Set bit 0 for enabled/disabled, and set bit 1 to indicate that this
-        // is now a user's choice and should not be overwritten by any default
-        UserConfigParams::m_high_definition_textures =
-            getWidget<CheckBoxWidget>("hd-textures")->getState() ? 0x03 : 0x02;
-
         UserConfigParams::m_show_steering_animations =
             getWidget<SpinnerWidget>("steering_animations")->getValue();
 
@@ -207,33 +191,8 @@ GUIEngine::EventPropagation CustomVideoSettingsDialog::processEvent(const std::s
             getWidget<SpinnerWidget>("geometry_detail")->getValue();
         UserConfigParams::m_geometry_level = val == 2 ? 0 : val == 0 ? 2 : 1;
 
-        switch (getWidget<SpinnerWidget>("filtering")->getValue())
-        {
-            case 0:
-                UserConfigParams::m_anisotropic = 0;
-                UserConfigParams::m_trilinear   = false;
-                break;
-            case 1:
-                UserConfigParams::m_anisotropic = 0;
-                UserConfigParams::m_trilinear   = true;
-                break;
-            case 2:
-                UserConfigParams::m_anisotropic = 2;
-                UserConfigParams::m_trilinear   = true;
-                break;
-            case 3:
-                UserConfigParams::m_anisotropic = 4;
-                UserConfigParams::m_trilinear   = true;
-                break;
-            case 4:
-                UserConfigParams::m_anisotropic = 8;
-                UserConfigParams::m_trilinear   = true;
-                break;
-            case 5:
-                UserConfigParams::m_anisotropic = 16;
-                UserConfigParams::m_trilinear   = true;
-                break;
-        }
+        OptionsScreenVideo::setImageQuality(getWidget<SpinnerWidget>
+            ("image_quality")->getValue());
 
         user_config->saveConfig();
 
diff --git a/src/states_screens/dialogs/debug_slider.cpp b/src/states_screens/dialogs/debug_slider.cpp
index 464977cb3..23dd0da74 100644
--- a/src/states_screens/dialogs/debug_slider.cpp
+++ b/src/states_screens/dialogs/debug_slider.cpp
@@ -39,7 +39,6 @@ DebugSliderDialog::DebugSliderDialog() : ModalDialog(0.85f, 0.25f, MODAL_DIALOG_
     loadFromFile("debug_slider.stkgui");
 }
 
-#if !defined(__APPLE__)
 void DebugSliderDialog::setSliderHook(std::string id, unsigned min, unsigned max, std::function<int()> G, std::function<void(int)> S)
 {
     getWidget<SpinnerWidget>(id.c_str())->setValue(G());
@@ -47,7 +46,6 @@ void DebugSliderDialog::setSliderHook(std::string id, unsigned min, unsigned max
     getWidget<SpinnerWidget>(id.c_str())->setMax(max);
     Setters[id] = S;
 }
-#endif
 
 // ------------------------------------------------------------------------------------------------------
 
@@ -66,7 +64,6 @@ void DebugSliderDialog::onEnterPressedInternal()
 
 GUIEngine::EventPropagation DebugSliderDialog::processEvent(const std::string& eventSource)
 {
-#if !defined(__APPLE__)
     if (Setters.find(eventSource) == Setters.end())
         return GUIEngine::EVENT_LET;
 
@@ -74,9 +71,6 @@ GUIEngine::EventPropagation DebugSliderDialog::processEvent(const std::string& e
     Log::info("DebugSlider", "Value for <%s> : %i", eventSource.c_str(), value);
     Setters[eventSource](value);
     return GUIEngine::EVENT_BLOCK;
-#else
-    return GUIEngine::EVENT_LET;
-#endif
 }
 
 // ------------------------------------------------------------------------------------------------------
diff --git a/src/states_screens/dialogs/debug_slider.hpp b/src/states_screens/dialogs/debug_slider.hpp
index 8e0768db6..36b5dba84 100644
--- a/src/states_screens/dialogs/debug_slider.hpp
+++ b/src/states_screens/dialogs/debug_slider.hpp
@@ -33,17 +33,13 @@ class DebugSliderDialog : public GUIEngine::ModalDialog
 private:
 
     std::string m_id;
-#if !defined(__APPLE__)
     std::map<std::string, std::function<void(int)> >Setters;
-#endif
 
 public:
     DebugSliderDialog();
 
     ~DebugSliderDialog() {};
-#if !defined(__APPLE__)
     void setSliderHook(std::string id, unsigned min, unsigned max, std::function<int()> G, std::function<void(int)> S);
-#endif
     void changeLabel(std::string id, std::string new_label);
 
     virtual void onEnterPressedInternal() OVERRIDE;
diff --git a/src/states_screens/options_screen_video.cpp b/src/states_screens/options_screen_video.cpp
index b85e5b158..765b35268 100644
--- a/src/states_screens/options_screen_video.cpp
+++ b/src/states_screens/options_screen_video.cpp
@@ -23,6 +23,7 @@
 #include "graphics/central_settings.hpp"
 #include "graphics/irr_driver.hpp"
 #include "graphics/shared_gpu_objects.hpp"
+#include "graphics/stk_tex_manager.hpp"
 #include "guiengine/screen.hpp"
 #include "guiengine/widgets/button_widget.hpp"
 #include "guiengine/widgets/check_box_widget.hpp"
@@ -53,32 +54,32 @@ void OptionsScreenVideo::initPresets()
     ({
         false /* light */, 0 /* shadow */, false /* bloom */, false /* motionblur */,
         false /* lightshaft */, false /* glow */, false /* mlaa */, false /* ssao */, false /* weather */,
-        false /* animatedScenery */, 0 /* animatedCharacters */, 0 /* anisotropy */,
-        false /* depth of field */, false /* global illumination */, true /* degraded IBL */, 0 /* hd_textures */
+        false /* animatedScenery */, 0 /* animatedCharacters */, 0 /* image_quality */,
+        false /* depth of field */, false /* global illumination */, true /* degraded IBL */
     });
 
     m_presets.push_back
     ({
         false /* light */, 0 /* shadow */, false /* bloom */, false /* motionblur */,
         false /* lightshaft */, false /* glow */, false /* mlaa */, false /* ssao */, false /* weather */,
-        true /* animatedScenery */, 1 /* animatedCharacters */, 4 /* anisotropy */,
-        false /* depth of field */, false /* global illumination */, true /* degraded IBL */, 0 /* hd_textures */
+        true /* animatedScenery */, 1 /* animatedCharacters */, 1 /* image_quality */,
+        false /* depth of field */, false /* global illumination */, true /* degraded IBL */
     });
 
     m_presets.push_back
     ({
         true /* light */, 0 /* shadow */, false /* bloom */, false /* motionblur */,
         false /* lightshaft */, false /* glow */, false /* mlaa */, false /* ssao */, true /* weather */,
-        true /* animatedScenery */, 1 /* animatedCharacters */, 4 /* anisotropy */,
-        false /* depth of field */, false /* global illumination */, true /* degraded IBL */, 1 /* hd_textures */
+        true /* animatedScenery */, 1 /* animatedCharacters */, 2 /* image_quality */,
+        false /* depth of field */, false /* global illumination */, true /* degraded IBL */
     });
 
     m_presets.push_back
     ({
         true /* light */, 0 /* shadow */, false /* bloom */, true /* motionblur */,
         true /* lightshaft */, true /* glow */, true /* mlaa */, false /* ssao */, true /* weather */,
-        true /* animatedScenery */, 1 /* animatedCharacters */, 8 /* anisotropy */,
-        false /* depth of field */, false /* global illumination */, false /* degraded IBL */, 1 /* hd_textures */
+        true /* animatedScenery */, 1 /* animatedCharacters */, 2 /* image_quality */,
+        false /* depth of field */, false /* global illumination */, false /* degraded IBL */
     });
 
     m_presets.push_back
@@ -91,8 +92,8 @@ void OptionsScreenVideo::initPresets()
 #else
         2 /* animatedCharacters */,
 #endif
-        16 /* anisotropy */,
-        true /* depth of field */, false /* global illumination */, false /* degraded IBL */, 1 /* hd_textures */
+        3 /* image_quality */,
+        true /* depth of field */, false /* global illumination */, false /* degraded IBL */
     });
 
     m_presets.push_back
@@ -105,8 +106,8 @@ void OptionsScreenVideo::initPresets()
 #else
         2 /* animatedCharacters */,
 #endif
-        16 /* anisotropy */,
-        true /* depth of field */, true /* global illumination */, false /* degraded IBL */, 1 /* hd_textures */
+        3 /* image_quality */,
+        true /* depth of field */, true /* global illumination */, false /* degraded IBL */
     });
 
 }   // initPresets
@@ -139,10 +140,79 @@ struct Resolution
     }
 };
 
+// ----------------------------------------------------------------------------
+int OptionsScreenVideo::getImageQuality()
+{
+    if (UserConfigParams::m_scale_rtts_factor == 0.8f &&
+        UserConfigParams::m_trilinear == false &&
+        UserConfigParams::m_anisotropic == 0 &&
+        (UserConfigParams::m_high_definition_textures & 0x01) == 0x00 &&
+        UserConfigParams::m_hq_mipmap == false)
+        return 0;
+    if (UserConfigParams::m_scale_rtts_factor == 1.0f &&
+        UserConfigParams::m_trilinear == true &&
+        UserConfigParams::m_anisotropic == 2 &&
+        (UserConfigParams::m_high_definition_textures & 0x01) == 0x00 &&
+        UserConfigParams::m_hq_mipmap == false)
+        return 1;
+    if (UserConfigParams::m_scale_rtts_factor == 1.0f &&
+        UserConfigParams::m_trilinear == true &&
+        UserConfigParams::m_anisotropic == 4 &&
+        (UserConfigParams::m_high_definition_textures & 0x01) == 0x01 &&
+        UserConfigParams::m_hq_mipmap == false)
+        return 2;
+    if (UserConfigParams::m_scale_rtts_factor == 1.0f &&
+        UserConfigParams::m_trilinear == true &&
+        UserConfigParams::m_anisotropic == 16 &&
+        (UserConfigParams::m_high_definition_textures & 0x01) == 0x01 &&
+        UserConfigParams::m_hq_mipmap == true)
+        return 3;
+    return 2;
+}   // getImageQuality
+
+// ----------------------------------------------------------------------------
+void OptionsScreenVideo::setImageQuality(int quality)
+{
+    switch (quality)
+    {
+        case 0:
+            UserConfigParams::m_scale_rtts_factor = 0.8f;
+            UserConfigParams::m_trilinear = false;
+            UserConfigParams::m_anisotropic = 0;
+            UserConfigParams::m_high_definition_textures = 0x02;
+            UserConfigParams::m_hq_mipmap = false;
+            break;
+        case 1:
+            UserConfigParams::m_scale_rtts_factor = 1.0f;
+            UserConfigParams::m_trilinear = true;
+            UserConfigParams::m_anisotropic = 2;
+            UserConfigParams::m_high_definition_textures = 0x02;
+            UserConfigParams::m_hq_mipmap = false;
+            break;
+        case 2:
+            UserConfigParams::m_scale_rtts_factor = 1.0f;
+            UserConfigParams::m_trilinear = true;
+            UserConfigParams::m_anisotropic = 4;
+            UserConfigParams::m_high_definition_textures = 0x03;
+            UserConfigParams::m_hq_mipmap = false;
+            break;
+        case 3:
+            UserConfigParams::m_scale_rtts_factor = 1.0f;
+            UserConfigParams::m_trilinear = true;
+            UserConfigParams::m_anisotropic = 16;
+            UserConfigParams::m_high_definition_textures = 0x03;
+            UserConfigParams::m_hq_mipmap = true;
+            break;
+        default:
+            assert(false);
+    }
+}   // setImageQuality
+
 // ----------------------------------------------------------------------------
 
 OptionsScreenVideo::OptionsScreenVideo() : Screen("options_video.stkgui"),
-                                           m_prev_adv_pipline(false)
+                                           m_prev_adv_pipline(false),
+                                           m_prev_img_quality(-1)
 {
     m_inited = false;
     initPresets();
@@ -168,6 +238,7 @@ void OptionsScreenVideo::init()
 {
     Screen::init();
     m_prev_adv_pipline = UserConfigParams::m_dynamic_lights;
+    m_prev_img_quality = getImageQuality();
     RibbonWidget* ribbon = getWidget<RibbonWidget>("options_choice");
     assert(ribbon != NULL);
     ribbon->select( "tab_video", PLAYER_ID_GAME_MASTER );
@@ -341,7 +412,7 @@ void OptionsScreenVideo::updateGfxSlider()
     {
         if (m_presets[l].animatedCharacters == UserConfigParams::m_show_steering_animations &&
             m_presets[l].animatedScenery == UserConfigParams::m_graphical_effects &&
-            m_presets[l].anisotropy == UserConfigParams::m_anisotropic &&
+            m_presets[l].image_quality == getImageQuality() &&
             m_presets[l].bloom == UserConfigParams::m_bloom &&
             m_presets[l].glow == UserConfigParams::m_glow &&
             m_presets[l].lights == UserConfigParams::m_dynamic_lights &&
@@ -354,8 +425,7 @@ void OptionsScreenVideo::updateGfxSlider()
             m_presets[l].weather == UserConfigParams::m_weather_effects &&
             m_presets[l].dof == UserConfigParams::m_dof &&
             m_presets[l].global_illumination == UserConfigParams::m_gi &&
-            m_presets[l].degraded_ibl == UserConfigParams::m_degraded_IBL &&
-            m_presets[l].hd_textures == (UserConfigParams::m_high_definition_textures & 0x01))
+            m_presets[l].degraded_ibl == UserConfigParams::m_degraded_IBL)
         {
             gfx->setValue(l + 1);
             found = true;
@@ -395,6 +465,19 @@ void OptionsScreenVideo::updateTooltip()
     //I18N: if no kart animations are enabled
     const core::stringw none = _LTR("None");
 
+    //I18N: in the graphical options tooltip;
+    // indicates the rendered image quality is very low
+    const core::stringw very_low = _LTR("Very Low");
+    //I18N: in the graphical options tooltip;
+    // indicates the rendered image quality is low
+    const core::stringw low = _LTR("Low");
+    //I18N: in the graphical options tooltip;
+    // indicates the rendered image quality is high
+    const core::stringw high = _LTR("High");
+    //I18N: in the graphical options tooltip;
+    // indicates the rendered image quality is very high
+    const core::stringw very_high = _LTR("Very High");
+
     //I18N: in graphical options
 //    tooltip = tooltip + L"\n" + _("Pixel shaders: %s",
 //                                  UserConfigParams::m_pixel_shaders ? enabled : disabled);
@@ -448,9 +531,11 @@ void OptionsScreenVideo::updateTooltip()
         UserConfigParams::m_gi ? enabled : disabled);
     
     //I18N: in graphical options
-    tooltip = tooltip + L"\n" + _("Use high definition textures: %s",
-        (UserConfigParams::m_high_definition_textures & 0x1) == 0 ? disabled : enabled);
-    
+    int quality = getImageQuality();
+    tooltip = tooltip + L"\n" + _("Rendered image quality: %s",
+        quality == 0 ? very_low : quality == 1 ? low : quality == 2 ?
+        high : very_high);
+
     gfx->setTooltip(tooltip);
 }   // updateTooltip
 
@@ -518,7 +603,7 @@ void OptionsScreenVideo::eventCallback(Widget* widget, const std::string& name,
 
         UserConfigParams::m_show_steering_animations = m_presets[level].animatedCharacters;
         UserConfigParams::m_graphical_effects = m_presets[level].animatedScenery;
-        UserConfigParams::m_anisotropic = m_presets[level].anisotropy;
+        setImageQuality(m_presets[level].image_quality);
         UserConfigParams::m_bloom = m_presets[level].bloom;
         UserConfigParams::m_glow = m_presets[level].glow;
         UserConfigParams::m_dynamic_lights = m_presets[level].lights;
@@ -532,7 +617,6 @@ void OptionsScreenVideo::eventCallback(Widget* widget, const std::string& name,
         UserConfigParams::m_dof = m_presets[level].dof;
         UserConfigParams::m_gi = m_presets[level].global_illumination;
         UserConfigParams::m_degraded_IBL = m_presets[level].degraded_ibl;
-        UserConfigParams::m_high_definition_textures = 0x02 | m_presets[level].hd_textures;
 
         updateGfxSlider();
     }
@@ -563,6 +647,12 @@ void OptionsScreenVideo::tearDown()
 {
     if (m_prev_adv_pipline != UserConfigParams::m_dynamic_lights)
         irr_driver->sameRestart();
+    else if (m_prev_img_quality != getImageQuality())
+    {
+        irr_driver->setMaxTextureSize();
+        STKTexManager::getInstance()->destroyThreadedTexLoaders();
+        STKTexManager::getInstance()->createThreadedTexLoaders();
+    }
     Screen::tearDown();
     // save changes when leaving screen
     user_config->saveConfig();
diff --git a/src/states_screens/options_screen_video.hpp b/src/states_screens/options_screen_video.hpp
index 88444c3e9..95f966655 100644
--- a/src/states_screens/options_screen_video.hpp
+++ b/src/states_screens/options_screen_video.hpp
@@ -38,12 +38,11 @@ struct GFXPreset
     bool weather;
     bool animatedScenery;
     int animatedCharacters;
-    int anisotropy;
+    int image_quality;
     /** Depth of field */
     bool dof;
     bool global_illumination;
     bool degraded_ibl;
-    int hd_textures;
 };
 
 /**
@@ -54,6 +53,7 @@ class OptionsScreenVideo : public GUIEngine::Screen, public GUIEngine::ScreenSin
 {
 private:
     bool m_prev_adv_pipline;
+    int m_prev_img_quality;
     OptionsScreenVideo();
     bool m_inited;
     std::vector<GFXPreset> m_presets;
@@ -81,6 +81,8 @@ public:
     virtual void unloaded() OVERRIDE;
 
     void         updateGfxSlider();
+    static int getImageQuality();
+    static void setImageQuality(int quality);
 };
 
 #endif
diff --git a/src/states_screens/race_gui.cpp b/src/states_screens/race_gui.cpp
index 7ee8e0715..a052c7a92 100644
--- a/src/states_screens/race_gui.cpp
+++ b/src/states_screens/race_gui.cpp
@@ -128,8 +128,11 @@ RaceGUI::RaceGUI()
 
     m_is_tutorial = (race_manager->getTrackName() == "tutorial");
 
+    // Load speedmeter texture before rendering the first frame
     m_speed_meter_icon = material_manager->getMaterial("speedback.png");
+    m_speed_meter_icon->getTexture();
     m_speed_bar_icon   = material_manager->getMaterial("speedfore.png");
+    m_speed_bar_icon->getTexture();
     //createMarkerTexture();
 }   // RaceGUI
 
@@ -401,8 +404,7 @@ void RaceGUI::drawGlobalMiniMap()
         const Vec3& xyz = kart->getXYZ();
         Vec3 draw_at;
         track->mapPoint2MiniMap(xyz, &draw_at);
-        draw_at *= UserConfigParams::m_scale_rtts_factor;
-
+        
         video::ITexture* icon = sta ?
             irr_driver->getTexture(FileManager::GUI, "heart.png") :
             kart->getKartProperties()->getMinimapIcon();
@@ -424,7 +426,7 @@ void RaceGUI::drawGlobalMiniMap()
     {
         Vec3 draw_at;
         track->mapPoint2MiniMap(sw->getBallPosition(), &draw_at);
-        draw_at *= UserConfigParams::m_scale_rtts_factor;
+        
         video::ITexture* icon =
             irr_driver->getTexture(FileManager::GUI, "soccer_ball_normal.png");
 
diff --git a/src/states_screens/race_gui_overworld.cpp b/src/states_screens/race_gui_overworld.cpp
index bff708a30..d72e5825c 100644
--- a/src/states_screens/race_gui_overworld.cpp
+++ b/src/states_screens/race_gui_overworld.cpp
@@ -395,8 +395,7 @@ void RaceGUIOverworld::drawGlobalMiniMap()
             kart_xyz= kart->getXYZ();
             Vec3 draw_at;
             track->mapPoint2MiniMap(kart_xyz, &draw_at);
-            draw_at *= UserConfigParams::m_scale_rtts_factor;
-
+            
             video::ITexture* icon = kart->getKartProperties()->getMinimapIcon();
             core::rect<s32> source(core::position2di(0, 0), icon->getSize());
             int marker_half_size = (kart->getController()->isLocalPlayerController()
@@ -433,8 +432,7 @@ void RaceGUIOverworld::drawGlobalMiniMap()
 
         Vec3 draw_at;
         track->mapPoint2MiniMap(challenges[n].m_position, &draw_at);
-        draw_at *= UserConfigParams::m_scale_rtts_factor;
-
+        
         const ChallengeData* challenge = unlock_manager->getChallengeData(challenges[n].m_challenge_id);
         const unsigned int val = challenge->getNumTrophies();
         bool unlocked = (PlayerManager::getCurrentPlayer()->getPoints() >= val);
diff --git a/src/tracks/track.cpp b/src/tracks/track.cpp
index 99c454b7e..2640de231 100644
--- a/src/tracks/track.cpp
+++ b/src/tracks/track.cpp
@@ -2176,9 +2176,7 @@ void Track::handleSky(const XMLNode &xml_node, const std::string &filename)
             if (CVS->isGLSL())
             {
                 t = STKTexManager::getInstance()->getTexture(v[i],
-                    false/*srgb*/, false/*premul_alpha*/,
-                    false/*set_material*/, false/*mesh_tex*/,
-                    true/*no_upload*/);
+                    (TexConfig*)NULL/*tex_config*/, true/*no_upload*/);
             }
             else
 #endif   // !SERVER_ONLY
@@ -2222,9 +2220,7 @@ void Track::handleSky(const XMLNode &xml_node, const std::string &filename)
             if (CVS->isGLSL())
             {
                 t = STKTexManager::getInstance()->getTexture(v[i],
-                    false/*srgb*/, false/*premul_alpha*/,
-                    false/*set_material*/, false/*mesh_tex*/,
-                    true/*no_upload*/);
+                    (TexConfig*)NULL/*tex_config*/, true/*no_upload*/);
             }
             else
 #endif   // !SERVER_ONLY
diff --git a/src/tracks/track_object_presentation.cpp b/src/tracks/track_object_presentation.cpp
index d47bf8119..7a2914171 100644
--- a/src/tracks/track_object_presentation.cpp
+++ b/src/tracks/track_object_presentation.cpp
@@ -862,9 +862,9 @@ TrackObjectPresentationBillboard::TrackObjectPresentationBillboard(
         xml_node.get("start",  &m_fade_out_start);
         xml_node.get("end",    &m_fade_out_end  );
     }
+    TexConfig tc(true/*srgb*/, true/*premul_alpha*/);
     video::ITexture* texture = STKTexManager::getInstance()->getTexture
-        (file_manager->searchTexture(texture_name), true/*srgb*/,
-        true/*premul_alpha*/, false/*set_material*/, true/*mesh_tex*/);
+        (file_manager->searchTexture(texture_name), &tc);
 
     if (texture == NULL)
     {
diff --git a/src/utils/avi_writer.cpp b/src/utils/avi_writer.cpp
new file mode 100644
index 000000000..b6348e413
--- /dev/null
+++ b/src/utils/avi_writer.cpp
@@ -0,0 +1,625 @@
+//
+//  SuperTuxKart - a fun racing game with go-kart
+//  Copyright (C) 2015 Dawid Gan
+//
+//  This program is free software; you can redistribute it and/or
+//  modify it under the terms of the GNU General Public License
+//  as published by the Free Software Foundation; either version 3
+//  of the License, or (at your option) any later version.
+//
+//  This program 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.
+//
+//  You should have received a copy of the GNU General Public License
+//  along with this program; if not, write to the Free Software
+//  Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
+
+#if !(defined(SERVER_ONLY) || defined(USE_GLES2))
+
+#include "utils/avi_writer.hpp"
+#include "config/user_config.hpp"
+#include "graphics/irr_driver.hpp"
+#include "guiengine/message_queue.hpp"
+#include "utils/translation.hpp"
+#include "utils/vs.hpp"
+
+#include <jpeglib.h>
+#include <cstring>
+
+Synchronised<std::string> AVIWriter::m_recording_target("");
+// ----------------------------------------------------------------------------
+AVIWriter::AVIWriter() : m_idle(true)
+{
+    resetFrameBufferImage();
+    resetCaptureFormat();
+    m_file = NULL;
+    m_last_junk_chunk = 0;
+    m_end_of_header = 0;
+    m_movi_start = 0;
+    m_stream_bytes = 0;
+    m_total_frames = 0;
+    m_chunk_fcc = 0;
+    m_width = irr_driver->getActualScreenSize().Width;
+    m_height = irr_driver->getActualScreenSize().Height;
+    glGenBuffers(3, m_pbo);
+    for (int i = 0; i < 3; i++)
+    {
+        glBindBuffer(GL_PIXEL_PACK_BUFFER, m_pbo[i]);
+        glBufferData(GL_PIXEL_PACK_BUFFER, m_width * m_height * 4, NULL,
+            GL_STREAM_READ);
+    }
+    glBindBuffer(GL_PIXEL_PACK_BUFFER, 0);
+    pthread_cond_init(&m_cond_request, NULL);
+    pthread_create(&m_thread, NULL, &startRoutine, this);
+}   // AVIWriter
+
+// ----------------------------------------------------------------------------
+AVIWriter::~AVIWriter()
+{
+    glDeleteBuffers(3, m_pbo);
+    addFrameBufferImage(NULL, 0);
+    if (!waitForReadyToDeleted(2.0f))
+        Log::info("AVIWriter", "AVIWriter not stopping, exiting anyway.");
+    pthread_join(m_thread, NULL);
+    pthread_cond_destroy(&m_cond_request);
+}   // ~AVIWriter
+
+// ----------------------------------------------------------------------------
+void AVIWriter::resetFrameBufferImage()
+{
+    m_pbo_use = 0;
+    m_accumulated_time = 0.0f;
+    m_remaining_time = 0.0f;
+}   // resetFrameBufferImage
+
+// ----------------------------------------------------------------------------
+void AVIWriter::resetCaptureFormat()
+{
+    m_img_quality = UserConfigParams::m_record_compression;
+    m_msec_per_frame = unsigned(1000 / UserConfigParams::m_record_fps);
+    m_avi_format =
+        UserConfigParams::m_record_bmp ? AVI_FORMAT_BMP : AVI_FORMAT_JPG;
+}   // resetCaptureFormat
+
+// ----------------------------------------------------------------------------
+void* AVIWriter::startRoutine(void *obj)
+{
+    VS::setThreadName("AVIWriter");
+    AVIWriter* avi_writer = (AVIWriter*)obj;
+    while (true)
+    {
+        avi_writer->m_fbi_queue.lock();
+        bool waiting = avi_writer->m_fbi_queue.getData().empty();
+        while (waiting)
+        {
+            pthread_cond_wait(&avi_writer->m_cond_request,
+                avi_writer->m_fbi_queue.getMutex());
+            waiting = avi_writer->m_fbi_queue.getData().empty();
+        }
+        auto& p =  avi_writer->m_fbi_queue.getData().front();
+        uint8_t* fbi = p.first;
+        int frame_count = p.second;
+        if (frame_count == -1)
+        {
+            avi_writer->closeFile();
+            avi_writer->m_idle.setAtomic(true);
+            avi_writer->m_fbi_queue.getData().pop_front();
+            avi_writer->m_fbi_queue.unlock();
+            continue;
+        }
+        else if (fbi == NULL)
+        {
+            avi_writer->closeFile(false/*delete_file*/, true/*exiting*/);
+            avi_writer->setCanBeDeleted();
+            avi_writer->m_fbi_queue.getData().pop_front();
+            avi_writer->m_fbi_queue.unlock();
+            return NULL;
+        }
+        const bool too_slow = avi_writer->m_fbi_queue.getData().size() > 50;
+        avi_writer->m_fbi_queue.getData().pop_front();
+        avi_writer->m_fbi_queue.unlock();
+        if (too_slow)
+        {
+            MessageQueue::add(MessageQueue::MT_ERROR,
+                _("Encoding is too slow, dropping frames."));
+            delete [] fbi;
+            avi_writer->cleanAllFrameBufferImages();
+            continue;
+        }
+        if (avi_writer->m_file == NULL)
+        {
+            bool ret = avi_writer->createFile();
+            if (!ret)
+            {
+                delete [] fbi;
+                avi_writer->cleanAllFrameBufferImages();
+                continue;
+            }
+        }
+        uint8_t* orig_fbi = fbi;
+        const unsigned width = avi_writer->m_width;
+        const unsigned height = avi_writer->m_height;
+        const unsigned area = width * height;
+        int size = area * 4;
+        int dest = size - 3;
+        int src = size - 4;
+        int copied = 0;
+        while (true)
+        {
+            if (copied++ > 1)
+                memcpy(fbi + dest, fbi + src, 3);
+            else
+                memmove(fbi + dest, fbi + src, 3);
+            if (src == 0)
+                break;
+            dest -= 3;
+            src -= 4;
+        }
+        fbi = fbi + area;
+        const int pitch = width * 3;
+        uint8_t* p2 = fbi + (height - 1) * pitch;
+        uint8_t* tmp_buf = new uint8_t[pitch];
+        for (unsigned i = 0; i < height; i += 2)
+        {
+            memcpy(tmp_buf, fbi, pitch);
+            memcpy(fbi, p2, pitch);
+            memcpy(p2, tmp_buf, pitch);
+            fbi += pitch;
+            p2 -= pitch;
+        }
+        delete [] tmp_buf;
+        size = area * 3;
+        if (avi_writer->m_avi_format == AVI_FORMAT_JPG)
+        {
+            uint8_t* jpg = new uint8_t[size];
+            size = avi_writer->bmpToJpg(orig_fbi + area, jpg, size);
+            delete [] orig_fbi;
+            orig_fbi = jpg;
+        }
+        while (frame_count != 0)
+        {
+            AVIErrCode code = avi_writer->addImage
+                (avi_writer->m_avi_format == AVI_FORMAT_JPG ? orig_fbi :
+                orig_fbi + area, size);
+            if (code == AVI_SIZE_LIMIT_ERR)
+            {
+                avi_writer->createFile();
+                continue;
+            }
+            else if (code == AVI_IO_ERR)
+                break;
+            frame_count--;
+        }
+        delete [] orig_fbi;
+    }
+    return NULL;
+}   // startRoutine
+
+// ----------------------------------------------------------------------------
+int AVIWriter::getFrameCount(float dt)
+{
+    const float frame_rate = 0.001f * m_msec_per_frame;
+    m_accumulated_time += dt;
+    if (m_accumulated_time < frame_rate && m_remaining_time < frame_rate)
+    {
+        return 0;
+    }
+    int frame_count = 1;
+    m_remaining_time += m_accumulated_time - frame_rate;
+    m_accumulated_time = 0.0f;
+    while (m_remaining_time > frame_rate)
+    {
+        frame_count++;
+        m_remaining_time -= frame_rate;
+    }
+    return frame_count;
+}   // getFrameCount
+
+// ----------------------------------------------------------------------------
+void AVIWriter::captureFrameBufferImage(float dt)
+{
+    glReadBuffer(GL_BACK);
+    int pbo_read = -1;
+    if (m_pbo_use > 3 && m_pbo_use % 3 == 0)
+        m_pbo_use = 3;
+    if (m_pbo_use >= 3)
+    {
+        int frame_count = getFrameCount(dt);
+        if (frame_count != 0)
+        {
+            pbo_read = m_pbo_use % 3;
+            glBindBuffer(GL_PIXEL_PACK_BUFFER, m_pbo[pbo_read]);
+            void* ptr = glMapBuffer(GL_PIXEL_PACK_BUFFER, GL_READ_ONLY);
+            const unsigned size = m_width * m_height * 4;
+            uint8_t* fbi = new uint8_t[size];
+            memcpy(fbi, ptr, size);
+            glUnmapBuffer(GL_PIXEL_PACK_BUFFER);
+            addFrameBufferImage(fbi, frame_count);
+        }
+    }
+    int pbo_use = m_pbo_use++ % 3;
+    assert(pbo_read == -1 || pbo_use == pbo_read);
+    glBindBuffer(GL_PIXEL_PACK_BUFFER, m_pbo[pbo_use]);
+    glReadPixels(0, 0, m_width, m_height,
+        m_avi_format == AVI_FORMAT_JPG ? GL_RGBA: GL_BGRA,
+        GL_UNSIGNED_BYTE, NULL);
+    glBindBuffer(GL_PIXEL_PACK_BUFFER, 0);
+}   // captureFrameBufferImage
+
+// ----------------------------------------------------------------------------
+bool AVIWriter::addJUNKChunk(std::string str, unsigned int min_size)
+{
+    int size = str.size() < min_size ? min_size : str.size() + 1;
+    size = (size + 1) & 0xfffffffe;
+
+    CHUNK chunk;
+    chunk.fcc = FOURCC('J', 'U', 'N', 'K');
+    chunk.cb = size;
+
+    char* buffer = (char*)calloc(size, 1);
+    strcpy(buffer, str.c_str());
+
+    int num = fwrite(&chunk, 1, sizeof(chunk), m_file);
+    if (num != sizeof(chunk))
+        goto error;
+
+    num = fwrite(buffer, 1, size * sizeof(char), m_file);
+    free(buffer);
+    if (num != size)
+        goto error;
+
+    m_last_junk_chunk = ftell(m_file);
+    if (m_last_junk_chunk < 0)
+        goto error;
+
+    return true;
+
+error:
+    closeFile(true/*delete_file*/);
+    return false;
+}   // addJUNKChunk
+
+// ----------------------------------------------------------------------------
+AVIErrCode AVIWriter::addImage(unsigned char* buffer, int buf_size)
+{
+    if (m_file == NULL)
+        goto error;
+
+    int num; num = ftell(m_file);
+    if (num < 0)
+        goto error;
+
+    if (m_total_frames >= MAX_FRAMES)
+        goto size_limit;
+
+    CHUNK chunk;
+    chunk.fcc = m_chunk_fcc;
+    chunk.cb = buf_size;
+
+    m_index_table[m_total_frames].Offset = num;
+    m_index_table[m_total_frames].Length = chunk.cb;
+    m_index_table[m_total_frames].fcc = chunk.fcc;
+
+    num = fwrite(&chunk, 1, sizeof(chunk), m_file);
+    if (num != sizeof(chunk))
+        goto error;
+
+    num = fwrite(buffer, 1, buf_size, m_file);
+    if (num != buf_size)
+        goto error;
+
+    int fill_size; fill_size = (sizeof(chunk) + buf_size) & 0x00000001;
+    if (fill_size > 0)
+    {
+        uint32_t filler = 0;
+        num = fwrite(&filler, 1, fill_size, m_file);
+        if (num != fill_size)
+            goto error;
+    }
+
+    m_stream_bytes += sizeof(chunk) + buf_size + fill_size;
+    m_total_frames++;
+
+    num = ftell(m_file);
+    if (num < 0)
+        goto error;
+
+    if (((num - m_last_junk_chunk) > 20000) && (!addJUNKChunk("", 1)))
+        goto error;
+
+    // check if we reached the file size limit
+    if (num >= MAX_FILE_SIZE)
+        goto size_limit;
+
+    return AVI_SUCCESS;
+
+error:
+    closeFile(true/*delete_file*/);
+    return AVI_IO_ERR;
+
+size_limit:
+    MessageQueue::add(MessageQueue::MT_GENERIC,
+       _("Video exceeded size limit, starting a new one."));
+    closeFile();
+    return AVI_SIZE_LIMIT_ERR;
+}   // addImage
+
+// ----------------------------------------------------------------------------
+bool AVIWriter::closeFile(bool delete_file, bool exiting)
+{
+    if (m_file == NULL)
+        return false;
+
+    if (delete_file)
+        goto error;
+
+    // add the index
+    int idx_start; idx_start = ftell(m_file);
+    if (idx_start < 0)
+        goto error;
+
+    CHUNK chunk;
+    chunk.fcc = FOURCC('i', 'd', 'x', '1');
+    chunk.cb = sizeof(AVIINDEXENTRY) * m_total_frames;
+
+    int num; num = fwrite(&chunk, 1, sizeof(chunk), m_file);
+    if (num != sizeof(chunk))
+        goto error;
+
+    for (unsigned int i = 0; i < m_total_frames; i++)
+    {
+        AVIINDEXENTRY Index;
+        Index.ckid = m_index_table[i].fcc;
+        Index.dwFlags = AVIIF_KEYFRAME;
+        Index.dwChunkOffset = m_index_table[i].Offset;
+        Index.dwChunkLength = m_index_table[i].Length;
+
+        num = fwrite(&Index, 1, sizeof(Index), m_file);
+        if (num != sizeof(Index))
+            goto error;
+    }
+
+    // update the header
+    if (m_total_frames > 0 && m_msec_per_frame > 0)
+    {
+        num = fseek(m_file, 0, SEEK_END);
+        if (num < 0)
+            goto error;
+
+        int size; size = ftell(m_file);
+        if (size < 0)
+            goto error;
+
+        num = fseek(m_file, 0, SEEK_SET);
+        if (num < 0)
+            goto error;
+
+        m_avi_hdr.riff.cb = size - sizeof(m_avi_hdr.riff);
+        m_avi_hdr.avih.dwMaxBytesPerSec = (uint32_t)
+            (((m_stream_bytes / m_total_frames) * m_format_hdr.strh.dwRate) /
+            m_msec_per_frame + 0.5f);
+        m_avi_hdr.avih.dwTotalFrames = m_total_frames;
+
+        num = fwrite(&m_avi_hdr, 1, sizeof(m_avi_hdr), m_file);
+        if (num != sizeof(m_avi_hdr))
+            goto error;
+
+        m_format_hdr.strh.dwLength = m_total_frames;
+
+        num = fwrite(&m_format_hdr, 1, sizeof(m_format_hdr), m_file);
+        if (num != sizeof(m_format_hdr))
+            goto error;
+    }
+
+    // update the movi section
+    m_movi_chunk.cb = idx_start - m_movi_start;
+
+    num = fseek(m_file, m_movi_start - sizeof(m_movi_chunk), SEEK_SET);
+    if (num < 0)
+        goto error;
+
+    num = fwrite(&m_movi_chunk, 1, sizeof(m_movi_chunk), m_file);
+    if (num != sizeof(m_movi_chunk))
+        goto error;
+
+    fclose(m_file);
+    m_file = NULL;
+
+    if (!exiting)
+    {
+        MessageQueue::add(MessageQueue::MT_GENERIC,
+            _("Video saved in \"%s\".", m_filename.c_str()));
+    }
+    return true;
+
+error:
+    if (!exiting)
+    {
+        MessageQueue::add(MessageQueue::MT_ERROR,
+            _("Error when saving video."));
+    }
+    fclose(m_file);
+    remove(m_filename.c_str());
+    m_file = NULL;
+    return false;
+}   // closeFile
+
+// ----------------------------------------------------------------------------
+bool AVIWriter::createFile()
+{
+    m_idle.setAtomic(false);
+    time_t rawtime;
+    time(&rawtime);
+    tm* timeInfo = localtime(&rawtime);
+    char time_buffer[256];
+    sprintf(time_buffer, "%i.%02i.%02i_%02i.%02i.%02i",
+        timeInfo->tm_year + 1900, timeInfo->tm_mon + 1,
+        timeInfo->tm_mday, timeInfo->tm_hour,
+        timeInfo->tm_min, timeInfo->tm_sec);
+
+    m_filename = m_recording_target.getAtomic() + "-" + time_buffer + ".avi";
+    m_stream_bytes = 0;
+    m_total_frames = 0;
+    m_movi_start = 0;
+    m_last_junk_chunk = 0;
+
+    BitmapInfoHeader bitmap_hdr;
+    bitmap_hdr.biSize = sizeof(BitmapInfoHeader);
+    bitmap_hdr.biWidth = m_width;
+    bitmap_hdr.biHeight = m_height;
+    bitmap_hdr.biPlanes = 1;
+    bitmap_hdr.biBitCount = 24;
+    bitmap_hdr.biCompression = 0;
+    bitmap_hdr.biSizeImage = (m_width * m_height * 3 * bitmap_hdr.biPlanes);
+    bitmap_hdr.biXPelsPerMeter = 0;
+    bitmap_hdr.biYPelsPerMeter = 0;
+    bitmap_hdr.biClrUsed = 0;
+    bitmap_hdr.biClrImportant = 0;
+
+    memset(&m_avi_hdr, '\0', sizeof(m_avi_hdr));
+    m_avi_hdr.riff.fcc = FOURCC('R', 'I', 'F', 'F');
+    m_avi_hdr.riff.cb = 0; // update when finished (size of the file - 8)
+    m_avi_hdr.avi = FOURCC('A', 'V', 'I', ' ');
+    m_avi_hdr.list1.fcc = FOURCC('L', 'I', 'S', 'T');
+    m_avi_hdr.list1.cb = 0;
+    m_avi_hdr.hdrl = FOURCC('h', 'd', 'r', 'l');
+    m_avi_hdr.avihhdr.fcc = FOURCC('a', 'v', 'i', 'h');
+    m_avi_hdr.avihhdr.cb = sizeof(m_avi_hdr.avih);
+    m_avi_hdr.avih.dwMicroSecPerFrame = m_msec_per_frame * 1000;
+    m_avi_hdr.avih.dwMaxBytesPerSec = 0; // update when finished
+    m_avi_hdr.avih.dwPaddingGranularity = 0;
+    m_avi_hdr.avih.dwFlags = AVIF_WASCAPTUREFILE | AVIF_HASINDEX;
+    m_avi_hdr.avih.dwTotalFrames = 0; // update when finished
+    m_avi_hdr.avih.dwInitialFrames = 0;
+    m_avi_hdr.avih.dwStreams = 1; // 1 = video, 2 = video and audio
+    m_avi_hdr.avih.dwSuggestedBufferSize = 0; // can be just 0
+    m_avi_hdr.avih.dwWidth = m_width;
+    m_avi_hdr.avih.dwHeight = m_height;
+
+    m_format_hdr.list.fcc = FOURCC('L', 'I', 'S', 'T');
+    m_format_hdr.list.cb = (sizeof(m_format_hdr) - 8) +
+        sizeof(BitmapInfoHeader);
+    m_format_hdr.strl = FOURCC('s', 't', 'r', 'l');
+    m_format_hdr.strhhdr.fcc = FOURCC('s', 't', 'r', 'h');
+    m_format_hdr.strhhdr.cb = sizeof(m_format_hdr.strh);
+    m_format_hdr.strh.fccType = FOURCC('v', 'i', 'd', 's');
+    m_format_hdr.strh.fccHandler = CC_DIB;
+    m_format_hdr.strh.dwFlags = 0;
+    m_format_hdr.strh.wPriority = 0;
+    m_format_hdr.strh.wLanguage = 0;
+    m_format_hdr.strh.dwInitialFrames = 0;
+    m_format_hdr.strh.dwScale = m_msec_per_frame;
+    m_format_hdr.strh.dwRate = 1000;
+    m_format_hdr.strh.dwStart = 0;
+    m_format_hdr.strh.dwLength = 0; // update when finished
+    m_format_hdr.strh.dwSuggestedBufferSize = 0; // can be just 0
+    m_format_hdr.strh.dwQuality = m_img_quality * 100;
+    m_format_hdr.strh.dwSampleSize = 0;
+    m_format_hdr.strh.Left = 0;
+    m_format_hdr.strh.Top = 0;
+    m_format_hdr.strh.Right = m_avi_hdr.avih.dwWidth;
+    m_format_hdr.strh.Bottom = m_avi_hdr.avih.dwHeight;
+    m_format_hdr.strfhdr.fcc = FOURCC('s', 't', 'r', 'f');
+    m_format_hdr.strfhdr.cb = sizeof(BitmapInfoHeader);
+
+    // Format specific changes
+    if (m_avi_format == AVI_FORMAT_JPG)
+    {
+        m_format_hdr.strh.fccHandler = CC_MJPG;
+        bitmap_hdr.biCompression = FOURCC('M', 'J', 'P', 'G');
+        m_chunk_fcc = FOURCC('0', '0', 'd', 'c');
+    }
+    else if (m_avi_format == AVI_FORMAT_BMP)
+    {
+        bitmap_hdr.biHeight = -m_height;
+        bitmap_hdr.biCompression = 0;
+        m_chunk_fcc = FOURCC('0', '0', 'd', 'b');
+    }
+
+    const uint32_t fcc_movi = FOURCC('m', 'o', 'v', 'i');
+
+    m_file = fopen(m_filename.c_str(), "wb");
+    if (m_file == NULL)
+        return false;
+
+    int num = fwrite(&m_avi_hdr, 1, sizeof(m_avi_hdr), m_file);
+    if (num != sizeof(m_avi_hdr))
+        goto error;
+
+    num = fwrite(&m_format_hdr, 1, sizeof(m_format_hdr), m_file);
+    if (num != sizeof(m_format_hdr))
+        goto error;
+
+    num = fwrite(&bitmap_hdr, 1, sizeof(BitmapInfoHeader), m_file);
+    if (num != sizeof(BitmapInfoHeader))
+        goto error;
+
+    m_end_of_header = ftell(m_file);
+    if (m_end_of_header < 0)
+        goto error;
+
+    if (!addJUNKChunk("", 2840))
+        goto error;
+
+    m_avi_hdr.list1.cb = m_end_of_header - sizeof(m_avi_hdr.riff) -
+                        sizeof(m_avi_hdr.avi) - sizeof(m_avi_hdr.list1);
+    m_movi_chunk.fcc = FOURCC('L', 'I', 'S', 'T');
+    m_movi_chunk.cb = 0; // update when finished
+
+    num = fwrite(&m_movi_chunk, 1, sizeof(m_movi_chunk), m_file);
+    if (num != sizeof(m_movi_chunk))
+        goto error;
+
+    m_movi_start = ftell(m_file);
+    if (m_movi_start < 0)
+        goto error;
+
+    num = fwrite(&fcc_movi, 1, sizeof(fcc_movi), m_file);
+    if (num != sizeof(fcc_movi))
+        goto error;
+
+    return true;
+
+error:
+    closeFile(true/*delete_file*/);
+    return false;
+}   // createFile
+
+// ----------------------------------------------------------------------------
+int AVIWriter::bmpToJpg(unsigned char* image_data, unsigned char* image_output,
+                        unsigned long buf_length)
+{
+    struct jpeg_compress_struct cinfo;
+    struct jpeg_error_mgr jerr;
+    cinfo.err = jpeg_std_error(&jerr);
+
+    jpeg_create_compress(&cinfo);
+
+    cinfo.image_width = m_width;
+    cinfo.image_height = m_height;
+    cinfo.input_components = 3;
+    cinfo.in_color_space = JCS_RGB;
+
+    jpeg_set_defaults(&cinfo);
+    jpeg_set_quality(&cinfo, m_img_quality, true);
+
+    jpeg_mem_dest(&cinfo, &image_output, &buf_length);
+
+    jpeg_start_compress(&cinfo, true);
+
+    JSAMPROW jrow[1];
+    while (cinfo.next_scanline < cinfo.image_height)
+    {
+        jrow[0] = &image_data[cinfo.next_scanline * m_width * 3];
+        jpeg_write_scanlines(&cinfo, jrow, 1);
+    }
+
+    jpeg_finish_compress(&cinfo);
+    jpeg_destroy_compress(&cinfo);
+
+    return buf_length;
+}   // bmpToJpg
+
+#endif
diff --git a/src/utils/avi_writer.hpp b/src/utils/avi_writer.hpp
new file mode 100644
index 000000000..66e455f89
--- /dev/null
+++ b/src/utils/avi_writer.hpp
@@ -0,0 +1,259 @@
+//
+//  SuperTuxKart - a fun racing game with go-kart
+//  Copyright (C) 2015 Dawid Gan
+//
+//  This program is free software; you can redistribute it and/or
+//  modify it under the terms of the GNU General Public License
+//  as published by the Free Software Foundation; either version 3
+//  of the License, or (at your option) any later version.
+//
+//  This program 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.
+//
+//  You should have received a copy of the GNU General Public License
+//  along with this program; if not, write to the Free Software
+//  Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
+
+#if !(defined(SERVER_ONLY) || defined(USE_GLES2))
+
+#include "graphics/gl_headers.hpp"
+#include "utils/can_be_deleted.hpp"
+#include "utils/no_copy.hpp"
+#include "utils/singleton.hpp"
+#include "utils/synchronised.hpp"
+
+#include <string>
+#include <list>
+
+#include <pthread.h>
+
+#define FOURCC(a,b,c,d) ((uint32_t) (((d)<<24) | ((c)<<16) | ((b)<<8) | (a)))
+
+const uint32_t CC_MJPG = FOURCC('m', 'j', 'p', 'g');
+const uint32_t CC_DIB = FOURCC('\0', '\0', '\0', '\0');
+const uint32_t CC_VIDS = FOURCC('v', 'i', 'd', 's');
+
+const uint32_t AVIF_HASINDEX = 0x00000010;
+const uint32_t AVIF_MUSTUSEINDEX = 0x00000020;
+const uint32_t AVIF_ISINTERLEAVED = 0x00000100;
+const uint32_t AVIF_TRUSTCKTYPE = 0x00000800;
+const uint32_t AVIF_WASCAPTUREFILE = 0x00010000;
+const uint32_t AVIF_COPYRIGHTED = 0x00020000;
+
+const uint32_t AVISF_DISABLED = 0x00000001;
+const uint32_t AVISF_VIDEO_PALCHANGES = 0x00010000;
+
+const uint32_t AVIIF_LIST = 0x00000001;
+const uint32_t AVIIF_KEYFRAME = 0x00000010;
+const uint32_t AVIIF_FIRSTPART = 0x00000020;
+const uint32_t AVIIF_LASTPART = 0x00000040;
+const uint32_t AVIIF_MIDPART = 0x00000060;
+const uint32_t AVIIF_NOTIME = 0x00000100;
+const uint32_t AVIIF_COMPUSE = 0x0FFF0000;
+
+enum AVIFormat
+{
+    AVI_FORMAT_BMP,
+    AVI_FORMAT_JPG
+};
+
+enum AVIErrCode
+{
+    AVI_SUCCESS,
+    AVI_SIZE_LIMIT_ERR,
+    AVI_IO_ERR
+};
+
+const int MAX_FRAMES = 1000000;
+const int MAX_FILE_SIZE = 2000000000;
+
+struct MainAVIHeader
+{
+    uint32_t dwMicroSecPerFrame;
+    uint32_t dwMaxBytesPerSec;
+    uint32_t dwPaddingGranularity;
+    uint32_t dwFlags;
+    uint32_t dwTotalFrames;
+    uint32_t dwInitialFrames;
+    uint32_t dwStreams;
+    uint32_t dwSuggestedBufferSize;
+    uint32_t dwWidth;
+    uint32_t dwHeight;
+    uint32_t dwReserved[4];
+};
+
+struct AVIStreamHeader
+{
+    uint32_t fccType;
+    uint32_t fccHandler;
+    uint32_t dwFlags;
+    uint16_t wPriority;
+    uint16_t wLanguage;
+    uint32_t dwInitialFrames;
+    uint32_t dwScale;
+    uint32_t dwRate;
+    uint32_t dwStart;
+    uint32_t dwLength;
+    uint32_t dwSuggestedBufferSize;
+    uint32_t dwQuality;
+    uint32_t dwSampleSize;
+    uint16_t Left;
+    uint16_t Top;
+    uint16_t Right;
+    uint16_t Bottom;
+};
+
+struct BitmapInfoHeader
+{
+    uint32_t biSize;
+    uint32_t biWidth;
+    uint32_t biHeight;
+    uint16_t biPlanes;
+    uint16_t biBitCount;
+    uint32_t biCompression;
+    uint32_t biSizeImage;
+    uint32_t biXPelsPerMeter;
+    uint32_t biYPelsPerMeter;
+    uint32_t biClrUsed;
+    uint32_t biClrImportant;
+};
+
+struct AVIINDEXENTRY
+{
+    uint32_t ckid;
+    uint32_t dwFlags;
+    uint32_t dwChunkOffset;
+    uint32_t dwChunkLength;
+};
+
+struct CHUNK
+{
+    uint32_t fcc;
+    uint32_t cb;
+};
+
+struct AVIHeader
+{
+    CHUNK riff;
+    uint32_t avi;
+    CHUNK list1;
+    uint32_t hdrl;
+    CHUNK avihhdr;
+    MainAVIHeader avih;
+};
+
+struct FormatHeader
+{
+    CHUNK list;
+    uint32_t strl;
+    CHUNK strhhdr;
+    AVIStreamHeader strh;
+    CHUNK strfhdr;
+};
+
+struct IndexTable
+{
+    uint32_t Offset;
+    uint32_t Length;
+    uint32_t fcc;
+};
+
+
+class AVIWriter : public CanBeDeleted, public NoCopy,
+                  public Singleton<AVIWriter>
+{
+private:
+    FILE* m_file;
+
+    static Synchronised<std::string> m_recording_target;
+
+    std::string m_filename;
+
+    int m_last_junk_chunk, m_end_of_header, m_movi_start, m_img_quality,
+        m_width, m_height;
+
+    unsigned int m_msec_per_frame, m_stream_bytes, m_total_frames, m_pbo_use;
+
+    float m_accumulated_time, m_remaining_time;
+
+    AVIFormat m_avi_format;
+
+    AVIHeader m_avi_hdr;
+
+    CHUNK m_movi_chunk;
+
+    FormatHeader m_format_hdr;
+
+    IndexTable m_index_table[MAX_FRAMES];
+
+    uint32_t m_chunk_fcc;
+
+    Synchronised<std::list<std::pair<uint8_t*, int> > > m_fbi_queue;
+
+    Synchronised<bool> m_idle;
+
+    pthread_t m_thread;
+
+    pthread_cond_t m_cond_request;
+
+    GLuint m_pbo[3];
+
+    // ------------------------------------------------------------------------
+    int bmpToJpg(unsigned char* image_data, unsigned char* image_output,
+                 unsigned long buf_length);
+    // ------------------------------------------------------------------------
+    AVIErrCode addImage(unsigned char* buffer, int size);
+    // ------------------------------------------------------------------------
+    bool closeFile(bool delete_file = false, bool exiting = false);
+    // ------------------------------------------------------------------------
+    bool createFile();
+    // ------------------------------------------------------------------------
+    bool addJUNKChunk(std::string str, unsigned int min_size);
+    // ------------------------------------------------------------------------
+    void addFrameBufferImage(uint8_t* fbi, int frame_count)
+    {
+        m_fbi_queue.lock();
+        m_fbi_queue.getData().emplace_back(fbi, frame_count);
+        pthread_cond_signal(&m_cond_request);
+        m_fbi_queue.unlock();
+    }
+    // ------------------------------------------------------------------------
+    int getFrameCount(float dt);
+    // ------------------------------------------------------------------------
+    void cleanAllFrameBufferImages()
+    {
+        m_fbi_queue.lock();
+        for (auto& p : m_fbi_queue.getData())
+            delete [] p.first;
+        m_fbi_queue.getData().clear();
+        m_fbi_queue.unlock();
+    }
+    // ------------------------------------------------------------------------
+    bool isIdle() const                          { return m_idle.getAtomic(); }
+
+public:
+    // ------------------------------------------------------------------------
+    AVIWriter();
+    // ------------------------------------------------------------------------
+    ~AVIWriter();
+    // ------------------------------------------------------------------------
+    static void* startRoutine(void *obj);
+    // ------------------------------------------------------------------------
+    static void setRecordingTarget(const std::string& name)
+    {
+        m_recording_target.setAtomic(name);
+    }
+    // ------------------------------------------------------------------------
+    void captureFrameBufferImage(float dt);
+    // ------------------------------------------------------------------------
+    void resetFrameBufferImage();
+    // ------------------------------------------------------------------------
+    void resetCaptureFormat();
+    // ------------------------------------------------------------------------
+    void stopRecording()                     { addFrameBufferImage(NULL, -1); }
+
+};
+
+#endif
diff --git a/src/utils/debug.cpp b/src/utils/debug.cpp
index 2f39ccf5d..2c0826139 100644
--- a/src/utils/debug.cpp
+++ b/src/utils/debug.cpp
@@ -134,6 +134,8 @@ enum DebugMenuCommand
     DEBUG_SCRIPT_CONSOLE,
     DEBUG_RUN_CUTSCENE,
     DEBUG_TEXTURE_CONSOLE,
+    DEBUG_START_RECORDING,
+    DEBUG_STOP_RECORDING
 };   // DebugMenuCommand
 
 // -----------------------------------------------------------------------------
@@ -541,7 +543,6 @@ bool handleContextMenuAction(s32 cmd_id)
         break;
     case DEBUG_VISUAL_VALUES:
     {
-#if !defined(__APPLE__)
         DebugSliderDialog *dsd = new DebugSliderDialog();
         dsd->setSliderHook("red_slider", 0, 255,
             [](){ return int(irr_driver->getAmbientLight().r * 255.f); },
@@ -576,12 +577,10 @@ bool handleContextMenuAction(s32 cmd_id)
             [](){ return int(irr_driver->getSSAOSigma() * 10.f); },
             [](int v){irr_driver->setSSAOSigma(v / 10.f); }
         );
-#endif
     }
     break;
     case DEBUG_ADJUST_LIGHTS:
     {
-#if !defined(__APPLE__)
         // Some sliders use multipliers because the spinner widget
         // only supports integers
         DebugSliderDialog *dsd = new DebugSliderDialog();
@@ -635,7 +634,6 @@ bool handleContextMenuAction(s32 cmd_id)
             [](int v){        findNearestLight()->setRadius(float(v)); }
         );
         dsd->changeLabel("SSAO Sigma", "[None]");
-#endif
         break;
     }
     case DEBUG_SCRIPT_CONSOLE:
@@ -711,6 +709,12 @@ bool handleContextMenuAction(s32 cmd_id)
                 return false;
             });
         break;
+        case DEBUG_START_RECORDING:
+            irr_driver->setRecording(true);
+        break;
+        case DEBUG_STOP_RECORDING:
+            irr_driver->setRecording(false);
+        break;
     }   // switch
     return false;
 }
@@ -793,8 +797,13 @@ bool onEvent(const SEvent &event)
             sub->addItem(L"Toggle smooth camera", DEBUG_GUI_CAM_SMOOTH);
             sub->addItem(L"Attach fps camera to kart", DEBUG_GUI_CAM_ATTACH);
 
-            mnu->addItem(L"Change camera target >",-1,true, true);
+            mnu->addItem(L"Recording >",-1,true, true);
             sub = mnu->getSubMenu(4);
+            sub->addItem(L"Start recording", DEBUG_START_RECORDING);
+            sub->addItem(L"Stop recording", DEBUG_STOP_RECORDING);
+
+            mnu->addItem(L"Change camera target >",-1,true, true);
+            sub = mnu->getSubMenu(5);
             sub->addItem(L"To kart one", DEBUG_VIEW_KART_ONE);
             sub->addItem(L"To kart two", DEBUG_VIEW_KART_TWO);
             sub->addItem(L"To kart three", DEBUG_VIEW_KART_THREE);
@@ -805,7 +814,7 @@ bool onEvent(const SEvent &event)
             sub->addItem(L"To kart eight", DEBUG_VIEW_KART_EIGHT);
 
             mnu->addItem(L"Font >",-1,true, true);
-            sub = mnu->getSubMenu(5);
+            sub = mnu->getSubMenu(6);
             sub->addItem(L"Dump glyph pages of fonts", DEBUG_FONT_DUMP_GLYPH_PAGE);
             sub->addItem(L"Reload all fonts", DEBUG_FONT_RELOAD);
 
diff --git a/src/utils/vs.hpp b/src/utils/vs.hpp
index 503c93959..68c8b4763 100644
--- a/src/utils/vs.hpp
+++ b/src/utils/vs.hpp
@@ -40,6 +40,10 @@
 #  include <windows.h>
 #endif
 
+#if defined(__linux__) && defined(__GLIBC__) && defined(__GLIBC_MINOR__)
+#  include <pthread.h>
+#endif
+
 namespace VS
 {
 #if defined(_MSC_VER) && defined(DEBUG)
@@ -77,6 +81,13 @@ namespace VS
         {
         }
 
+    }   // setThreadName
+#elif defined(__linux__) && defined(__GLIBC__) && defined(__GLIBC_MINOR__)
+    static void setThreadName(const char* name)
+    {
+#if __GLIBC__ > 2 || __GLIBC_MINOR__ > 11
+        pthread_setname_np(pthread_self(), name);
+#endif
     }   // setThreadName
 #else
     static void setThreadName(const char* name)