Create 2.0.4 release.

Select swizzle codes in nvtt instead of nvimage.

ChangeLog

@@ -1,3 +1,31 @@
+NVIDIA Texture Tools version 2.0.4
+ * Fix error in RGB format output; reported by jonsoh. See issue 49.
+ * Added support RGB format dithering by jonsoh. Fixes issue 50 and 51.
+ * Prevent infinite loop in indexMirror when width equal 1. Fixes issue 65.
+ * Implement general scale filter, including upsampling.
+
+NVIDIA Texture Tools version 2.0.3
+ * More accurate DXT3 compressor. Fixes issue 38.
+ * Remove legacy compressors. Fix issue 34.
+ * Check for single color in all compressors. Fixes issue 43.
+ * Fix error in fast downsample filter, reported by Noel Llopis.
+
+NVIDIA Texture Tools version 2.0.2
+ * Fix copy ctor error reported by Richard Sim.
+ * Fix indexMirror error reported by Chris Lambert.
+ * Fix vc8 post build command, reported by Richard Sim.
+ * Fix RGBA modes with less than 32 bpp by Viktor Linder.
+ * Fix alpha decompression by Amorilia. See issue 40.
+ * Avoid default-initialized constructors for POD types, reported by Jim Tilander.
+ * Add single color compresor for DXT1a.
+ * Set swizzle code to ATI2 files. See issue 41.
+
+NVIDIA Texture Tools version 2.0.1
+ * Fix memory leaks.
+ * Pre-allocate device memory for CUDA compressor.
+ * Add single color compressor. Thanks to Amir Ebrahimi.
+ * Better CUDA error checking.
+
 NVIDIA Texture Tools version 2.0.0
  * Fixed PSNR formula in nvimgdiff.
  * Added support for arbitrary RGB formats.

NVIDIA_Texture_Tools_README.txt

@@ -2,7 +2,7 @@
 --------------------------------------------------------------------------------
 NVIDIA Texture Tools
 README.txt
-Version 2.0.0
+Version 2.0
 --------------------------------------------------------------------------------
 --------------------------------------------------------------------------------
 

VERSION

@@ -1 +1 @@
-2.0.0
+2.0.4

cmake/FindCUDA.cmake

@@ -46,9 +46,9 @@ FIND_LIBRARY (CUDA_RUNTIME_LIBRARY
 	DOC "The CUDA runtime library")
 
 IF (CUDA_INCLUDE_PATH AND CUDA_RUNTIME_LIBRARY)
-	SET (CUDA_FOUND 1 CACHE STRING "Set to 1 if CUDA is found, 0 otherwise")
+	SET (CUDA_FOUND TRUE)
 ELSE (CUDA_INCLUDE_PATH AND CUDA_RUNTIME_LIBRARY)
-	SET (CUDA_FOUND 0 CACHE STRING "Set to 1 if CUDA is found, 0 otherwise")
+	SET (CUDA_FOUND FALSE)
 ENDIF (CUDA_INCLUDE_PATH AND CUDA_RUNTIME_LIBRARY)
 
 SET (CUDA_LIBRARIES ${CUDA_RUNTIME_LIBRARY})

project/vc8/nvassemble/nvassemble.vcproj

@@ -278,6 +278,7 @@
 				AdditionalDependencies="libpng.lib jpeg.lib tiff.lib"
 				OutputFile="$(SolutionDir)\$(ConfigurationName).$(PlatformName)\bin\$(ProjectName).exe"
 				AdditionalLibraryDirectories="..\..\..\gnuwin32\lib"
+				LinkTimeCodeGeneration="1"
 				TargetMachine="17"
 			/>
 			<Tool

project/vc8/nvcore/nvcore.vcproj

@@ -327,6 +327,10 @@
 				RelativePath="..\..\..\src\nvcore\nvcore.h"
 				>
 			</File>
+			<File
+				RelativePath="..\..\..\src\nvcore\Ptr.h"
+				>
+			</File>
 			<File
 				RelativePath="..\..\..\src\nvcore\StrLib.h"
 				>

project/vc8/nvddsinfo/nvddsinfo.vcproj

@@ -277,6 +277,7 @@
 				AdditionalDependencies="libpng.lib jpeg.lib tiff.lib"
 				OutputFile="$(SolutionDir)\$(ConfigurationName).$(PlatformName)\bin\$(ProjectName).exe"
 				AdditionalLibraryDirectories="..\..\..\gnuwin32\lib"
+				LinkTimeCodeGeneration="1"
 				TargetMachine="17"
 			/>
 			<Tool

project/vc8/nvimage/nvimage.vcproj

@@ -355,6 +355,10 @@
 				RelativePath="..\..\..\src\nvimage\nvimage.h"
 				>
 			</File>
+			<File
+				RelativePath="..\..\..\src\nvimage\PixelFormat.h"
+				>
+			</File>
 			<File
 				RelativePath="..\..\..\src\nvimage\PsdFile.h"
 				>

project/vc8/nvmath/nvmath.vcproj

@@ -278,11 +278,7 @@
 			UniqueIdentifier="{4FC737F1-C7A5-4376-A066-2A32D752A2FF}"
 			>
 			<File
-				RelativePath="..\..\..\src\nvmath\Eigen.cpp"
-				>
-			</File>
-			<File
-				RelativePath="..\..\..\src\nvmath\Fitting.cpp"
+				RelativePath="..\..\..\src\nvmath\Plane.cpp"
 				>
 			</File>
 		</Filter>
@@ -299,18 +295,14 @@
 				RelativePath="..\..\..\src\nvmath\Color.h"
 				>
 			</File>
-			<File
-				RelativePath="..\..\..\src\nvmath\Eigen.h"
-				>
-			</File>
-			<File
-				RelativePath="..\..\..\src\nvmath\Fitting.h"
-				>
-			</File>
 			<File
 				RelativePath="..\..\..\src\nvmath\Matrix.h"
 				>
 			</File>
+			<File
+				RelativePath="..\..\..\src\nvmath\Plane.h"
+				>
+			</File>
 			<File
 				RelativePath="..\..\..\src\nvmath\Vector.h"
 				>

project/vc8/nvtt/nvtt.rc

@@ -53,8 +53,8 @@ END
 //
 
 VS_VERSION_INFO VERSIONINFO
- FILEVERSION 2,0,0,0
- PRODUCTVERSION 2,0,0,0
+ FILEVERSION 2,0,4,0
+ PRODUCTVERSION 2,0,4,0
  FILEFLAGSMASK 0x17L
 #ifdef _DEBUG
  FILEFLAGS 0x1L
@@ -71,12 +71,12 @@ BEGIN
         BEGIN
             VALUE "CompanyName", "NVIDIA Corporation"
             VALUE "FileDescription", "NVIDIA Texture Tools Dynamic Link Library"
-            VALUE "FileVersion", "2, 0, 0, 0"
+            VALUE "FileVersion", "2, 0, 4, 0"
             VALUE "InternalName", "nvtt"
             VALUE "LegalCopyright", "Copyright (C) 2007"
             VALUE "OriginalFilename", "nvtt.dll"
             VALUE "ProductName", "NVIDIA Texture Tools Dynamic Link Library"
-            VALUE "ProductVersion", "2, 0, 0, 0"
+            VALUE "ProductVersion", "2, 0, 4, 0"
         END
     END
     BLOCK "VarFileInfo"

project/vc8/nvtt/nvtt.vcproj

@@ -96,6 +96,8 @@
 			/>
 			<Tool
 				Name="VCPostBuildEventTool"
+				Description="Copying header files..."
+				CommandLine="xcopy /y /f /i &quot;$(SolutionDir)\..\..\src\nvtt\nvtt*.h&quot; &quot;$(SolutionDir)\$(ConfigurationName).$(PlatformName)\include\nvtt\&quot;"
 			/>
 		</Configuration>
 		<Configuration
@@ -258,6 +260,8 @@
 			/>
 			<Tool
 				Name="VCPostBuildEventTool"
+				Description="Copying header files..."
+				CommandLine="xcopy /y /f /i &quot;$(SolutionDir)\..\..\src\nvtt\nvtt*.h&quot; &quot;$(SolutionDir)\$(ConfigurationName).$(PlatformName)\include\nvtt\&quot;"
 			/>
 		</Configuration>
 		<Configuration
@@ -420,6 +424,8 @@
 			/>
 			<Tool
 				Name="VCPostBuildEventTool"
+				Description="Copying header files..."
+				CommandLine="xcopy /y /f /i &quot;$(SolutionDir)\..\..\src\nvtt\nvtt*.h&quot; &quot;$(SolutionDir)\$(ConfigurationName).$(PlatformName)\include\nvtt\&quot;"
 			/>
 		</Configuration>
 		<Configuration
@@ -578,6 +584,8 @@
 			/>
 			<Tool
 				Name="VCPostBuildEventTool"
+				Description="Copying header files..."
+				CommandLine="xcopy /y /f /i &quot;$(SolutionDir)\..\..\src\nvtt\nvtt*.h&quot; &quot;$(SolutionDir)\$(ConfigurationName).$(PlatformName)\include\nvtt\&quot;"
 			/>
 		</Configuration>
 		<Configuration
@@ -683,7 +691,7 @@
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="&quot;$(CUDA_BIN_PATH)\nvcc.exe&quot; -keep -ccbin &quot;$(VCInstallDir)bin&quot; -c -D_DEBUG -DWIN32 -D_CONSOLE -D_MBCS -Xcompiler /EHsc,/W3,/nologo,/Wp64,/Od,/Zi,/RTC1,/MDd -I&quot;$(CUDA_INC_PATH)&quot; -I./ -o $(IntDir)\$(InputName).obj ..\\..\\..\\src\\nvtt\\cuda\\CompressKernel.cu&#x0D;&#x0A;"
+						CommandLine="&quot;$(CUDA_BIN_PATH)\nvcc.exe&quot; -m32 -ccbin &quot;$(VCInstallDir)bin&quot; -c -D_DEBUG -DWIN32 -D_CONSOLE -D_MBCS -Xcompiler /EHsc,/W3,/nologo,/Wp64,/Od,/Zi,/RTC1,/MDd -I&quot;$(CUDA_INC_PATH)&quot; -I./ -o $(IntDir)\$(InputName).obj ..\\..\\..\\src\\nvtt\\cuda\\CompressKernel.cu&#x0D;&#x0A;"
 						AdditionalDependencies="CudaMath.h"
 						Outputs="$(IntDir)\$(InputName).obj"
 					/>
@@ -693,7 +701,7 @@
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="&quot;$(CUDA_BIN_PATH)\nvcc.exe&quot; -keep -ccbin &quot;$(VCInstallDir)bin&quot; -c -D_DEBUG -DWIN32 -D_CONSOLE -D_MBCS -Xcompiler /EHsc,/W3,/nologo,/Wp64,/Od,/Zi,/RTC1,/MDd -I&quot;$(CUDA_INC_PATH)&quot; -I./ -o $(IntDir)\$(InputName).obj ..\\..\\..\\src\\nvtt\\cuda\\CompressKernel.cu&#x0D;&#x0A;"
+						CommandLine="&quot;$(CUDA_BIN_PATH)\nvcc.exe&quot; -m64 -ccbin &quot;$(VCInstallDir)bin&quot; -c -D_DEBUG -DWIN32 -D_CONSOLE -D_MBCS -Xcompiler /EHsc,/W3,/nologo,/Wp64,/Od,/Zi,/RTC1,/MDd -I&quot;$(CUDA_INC_PATH)&quot; -I./ -o $(IntDir)\$(InputName).obj ..\\..\\..\\src\\nvtt\\cuda\\CompressKernel.cu&#x0D;&#x0A;"
 						AdditionalDependencies="CudaMath.h"
 						Outputs="$(IntDir)\$(InputName).obj"
 					/>
@@ -703,7 +711,7 @@
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="&quot;$(CUDA_BIN_PATH)\nvcc.exe&quot; -keep -ccbin &quot;$(VCInstallDir)bin&quot; -c -DNDEBUG -DWIN32 -D_CONSOLE -D_MBCS -Xcompiler /EHsc,/W3,/nologo,/Wp64,/O2,/Zi,/MD -I&quot;$(CUDA_INC_PATH)&quot; -I./ -o $(IntDir)\$(InputName).obj ..\\..\\..\\src\\nvtt\\cuda\\CompressKernel.cu&#x0D;&#x0A;"
+						CommandLine="&quot;$(CUDA_BIN_PATH)\nvcc.exe&quot; -m32 -ccbin &quot;$(VCInstallDir)bin&quot; -c -DNDEBUG -DWIN32 -D_CONSOLE -D_MBCS -Xcompiler /EHsc,/W3,/nologo,/Wp64,/O2,/Zi,/MD -I&quot;$(CUDA_INC_PATH)&quot; -I./ -o $(IntDir)\$(InputName).obj ..\\..\\..\\src\\nvtt\\cuda\\CompressKernel.cu&#x0D;&#x0A;"
 						AdditionalDependencies="CudaMath.h"
 						Outputs="$(IntDir)\$(InputName).obj"
 					/>
@@ -713,7 +721,7 @@
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="&quot;$(CUDA_BIN_PATH)\nvcc.exe&quot; -keep -ccbin &quot;$(VCInstallDir)bin&quot; -c -DNDEBUG -DWIN32 -D_CONSOLE -D_MBCS -Xcompiler /EHsc,/W3,/nologo,/Wp64,/O2,/Zi,/MD -I&quot;$(CUDA_INC_PATH)&quot; -I./ -o $(IntDir)\$(InputName).obj ..\\..\\..\\src\\nvtt\\cuda\\CompressKernel.cu&#x0D;&#x0A;"
+						CommandLine="&quot;$(CUDA_BIN_PATH)\nvcc.exe&quot; -m64 -ccbin &quot;$(VCInstallDir)bin&quot; -c -DNDEBUG -DWIN32 -D_CONSOLE -D_MBCS -Xcompiler /EHsc,/W3,/nologo,/Wp64,/O2,/Zi,/MD -I&quot;$(CUDA_INC_PATH)&quot; -I./ -o $(IntDir)\$(InputName).obj ..\\..\\..\\src\\nvtt\\cuda\\CompressKernel.cu&#x0D;&#x0A;"
 						AdditionalDependencies="CudaMath.h"
 						Outputs="$(IntDir)\$(InputName).obj"
 					/>
@@ -841,10 +849,6 @@
 				RelativePath="..\..\..\src\nvtt\cuda\CudaUtils.cpp"
 				>
 			</File>
-			<File
-				RelativePath="..\..\..\src\nvtt\FastCompressDXT.cpp"
-				>
-			</File>
 			<File
 				RelativePath="..\..\..\src\nvtt\InputOptions.cpp"
 				>
@@ -857,6 +861,10 @@
 				RelativePath="..\..\..\src\nvtt\nvtt_wrapper.cpp"
 				>
 			</File>
+			<File
+				RelativePath="..\..\..\src\nvtt\OptimalCompressDXT.cpp"
+				>
+			</File>
 			<File
 				RelativePath="..\..\..\src\nvtt\OutputOptions.cpp"
 				>
@@ -903,10 +911,6 @@
 				RelativePath="..\..\..\src\nvtt\cuda\CudaUtils.h"
 				>
 			</File>
-			<File
-				RelativePath="..\..\..\src\nvtt\FastCompressDXT.h"
-				>
-			</File>
 			<File
 				RelativePath="..\..\..\src\nvtt\InputOptions.h"
 				>
@@ -919,6 +923,10 @@
 				RelativePath="..\..\..\src\nvtt\nvtt_wrapper.h"
 				>
 			</File>
+			<File
+				RelativePath="..\..\..\src\nvtt\OptimalCompressDXT.h"
+				>
+			</File>
 			<File
 				RelativePath="..\..\..\src\nvtt\OutputOptions.h"
 				>

src/CMakeLists.txt

@@ -50,6 +50,7 @@ ENDIF(CG_FOUND)
 # CUDA
 INCLUDE(${NV_CMAKE_DIR}/FindCUDA.cmake)
 IF(CUDA_FOUND)
+	SET(HAVE_CUDA ${CUDA_FOUND} CACHE BOOL "Set to TRUE if CUDA is found, FALSE otherwise")
 	MESSAGE(STATUS "Looking for CUDA - found")
 ELSE(CUDA_FOUND)
 	MESSAGE(STATUS "Looking for CUDA - not found")
@@ -58,7 +59,7 @@ ENDIF(CUDA_FOUND)
 # Maya
 INCLUDE(${NV_CMAKE_DIR}/FindMaya.cmake)
 IF(MAYA_FOUND)
-	SET(HAVE_MAYA MAYA_FOUND)
+	SET(HAVE_MAYA ${MAYA_FOUND} CACHE BOOL "Set to TRUE if Maya is found, FALSE otherwise")
 	MESSAGE(STATUS "Looking for Maya - found")
 ELSE(MAYA_FOUND)
 	MESSAGE(STATUS "Looking for Maya - not found")
@@ -67,7 +68,7 @@ ENDIF(MAYA_FOUND)
 # JPEG
 INCLUDE(FindJPEG)
 IF(JPEG_FOUND)
-	SET(HAVE_JPEG JPEG_FOUND)
+	SET(HAVE_JPEG ${JPEG_FOUND} CACHE BOOL "Set to TRUE if JPEG is found, FALSE otherwise")
 	MESSAGE(STATUS "Looking for JPEG - found")
 ELSE(JPEG_FOUND)
 	MESSAGE(STATUS "Looking for JPEG - not found")
@@ -76,7 +77,7 @@ ENDIF(JPEG_FOUND)
 # PNG
 INCLUDE(FindPNG)
 IF(PNG_FOUND)
-	SET(HAVE_PNG PNG_FOUND)
+	SET(HAVE_PNG ${PNG_FOUND} CACHE BOOL "Set to TRUE if PNG is found, FALSE otherwise")
 	MESSAGE(STATUS "Looking for PNG - found")
 ELSE(PNG_FOUND)
 	MESSAGE(STATUS "Looking for PNG - not found")
@@ -85,7 +86,7 @@ ENDIF(PNG_FOUND)
 # TIFF
 INCLUDE(FindTIFF)
 IF(TIFF_FOUND)
-	SET(HAVE_TIFF TIFF_FOUND)
+	SET(HAVE_TIFF ${TIFF_FOUND} CACHE BOOL "Set to TRUE if TIFF is found, FALSE otherwise")
 	MESSAGE(STATUS "Looking for TIFF - found")
 ELSE(TIFF_FOUND)
 	MESSAGE(STATUS "Looking for TIFF - not found")
@@ -94,7 +95,7 @@ ENDIF(TIFF_FOUND)
 # OpenEXR
 INCLUDE(${NV_CMAKE_DIR}/FindOpenEXR.cmake)
 IF(OPENEXR_FOUND)
-	SET(HAVE_OPENEXR OPENEXR_FOUND)
+	SET(HAVE_OPENEXR ${OPENEXR_FOUND} CACHE BOOL "Set to TRUE if OpenEXR is found, FALSE otherwise")
 	MESSAGE(STATUS "Looking for OpenEXR - found")
 ELSE(OPENEXR_FOUND)
 	MESSAGE(STATUS "Looking for OpenEXR - not found")

src/nvcore/CMakeLists.txt

@@ -18,8 +18,6 @@ SET(CORE_SRCS
 	TextReader.cpp
 	TextWriter.h
 	TextWriter.cpp
-	Tokenizer.h
-	Tokenizer.cpp
 	Radix.h
 	Radix.cpp)
 

src/nvcore/Containers.h

@@ -446,7 +446,7 @@ namespace nv
 			
 			// Call default constructors
 			for( i = old_size; i < new_size; i++ ) {
-				new(m_buffer+i) T();	// placement new
+				new(m_buffer+i) T;	// placement new
 			}
 		}
 	

src/nvcore/Debug.cpp

@@ -74,7 +74,9 @@ namespace
 
 	// TODO write minidump
 	
-	static LONG WINAPI nvTopLevelFilter( struct _EXCEPTION_POINTERS *pExceptionInfo ) {
+	static LONG WINAPI nvTopLevelFilter( struct _EXCEPTION_POINTERS * pExceptionInfo)
+	{
+		NV_UNUSED(pExceptionInfo);
 	/*	BOOL (WINAPI * Dump) (HANDLE, DWORD, HANDLE, MINIDUMP_TYPE, PMINIDUMP_EXCEPTION_INFORMATION, PMINIDUMP_USER_STREAM_INFORMATION, PMINIDUMP_CALLBACK_INFORMATION );
 	
 		AutoString dbghelp_path(512);

src/nvcore/Memory.cpp

@@ -18,6 +18,8 @@ void * nv::mem::malloc(size_t size)
 
 void * nv::mem::malloc(size_t size, const char * file, int line)
 {
+	NV_UNUSED(file);
+	NV_UNUSED(line);
 	return ::malloc(size);
 }
 

src/nvimage/ColorBlock.cpp

@@ -110,6 +110,19 @@ void ColorBlock::splatY()
 	}
 }
 
+/// Returns true if the block has a single color.
+bool ColorBlock::isSingleColor() const
+{
+	for(int i = 1; i < 16; i++)
+	{
+		if (m_color[0] != m_color[i])
+		{
+			return false;
+		}
+	}
+	
+	return true;
+}
 
 /// Count number of unique colors in this color block.
 uint ColorBlock::countUniqueColors() const
@@ -294,15 +307,6 @@ void ColorBlock::boundsRangeAlpha(Color32 * start, Color32 * end) const
 }
 
 
-void ColorBlock::bestFitRange(Color32 * start, Color32 * end) const
-{
-	nvDebugCheck(start != NULL);
-	nvDebugCheck(end != NULL);
-
-	Vector3 axis = bestFitLine().direction();
-	computeRange(axis, start, end);
-}
-
 /// Sort colors by abosolute value in their 16 bit representation.
 void ColorBlock::sortColorsByAbsoluteValue()
 {
@@ -380,19 +384,6 @@ void ColorBlock::sortColors(const Vector3 & axis)
 }
 
 
-/// Get least squares line that best approxiamtes the points of the color block.
-Line3 ColorBlock::bestFitLine() const
-{
-	Array<Vector3> pointArray(16);
-	
-	for(int i = 0; i < 16; i++) {
-		pointArray.append(Vector3(m_color[i].r, m_color[i].g, m_color[i].b));
-	}
-	
-	return Fit::bestLine(pointArray);
-}
-
-
 /// Get the volume of the color block.
 float ColorBlock::volume() const
 {

src/nvimage/ColorBlock.h

@@ -4,7 +4,6 @@
 #define NV_IMAGE_COLORBLOCK_H
 
 #include <nvmath/Color.h>
-#include <nvmath/Fitting.h>	// Line3
 
 namespace nv
 {
@@ -24,6 +23,7 @@ namespace nv
 		void splatX();
 		void splatY();
 		
+		bool isSingleColor() const;
 		uint countUniqueColors() const;
 		Color32 averageColor() const;
 		bool hasAlpha() const;
@@ -32,16 +32,13 @@ namespace nv
 		void luminanceRange(Color32 * start, Color32 * end) const;
 		void boundsRange(Color32 * start, Color32 * end) const;
 		void boundsRangeAlpha(Color32 * start, Color32 * end) const;
-		void bestFitRange(Color32 * start, Color32 * end) const;
 		
 		void sortColorsByAbsoluteValue();
 		
 		void computeRange(const Vector3 & axis, Color32 * start, Color32 * end) const;
 		void sortColors(const Vector3 & axis);
 		
-		Line3 bestFitLine() const;
 		float volume() const;
-		Line3 diameterLine() const;
 		
 		// Accessors
 		const Color32 * colors() const;

src/nvimage/DirectDrawSurface.cpp

@@ -54,6 +54,10 @@ namespace
 	static const uint FOURCC_ATI1 = MAKEFOURCC('A', 'T', 'I', '1');
 	static const uint FOURCC_ATI2 = MAKEFOURCC('A', 'T', 'I', '2');
 
+	static const uint FOURCC_A2XY = MAKEFOURCC('A', '2', 'X', 'Y');
+	
+	static const uint FOURCC_DX10 = MAKEFOURCC('D', 'X', '1', '0');
+
 	// 32 bit RGB formats.
 	static const uint D3DFMT_R8G8B8 = 20;
 	static const uint D3DFMT_A8R8G8B8 = 21;
@@ -253,6 +257,144 @@ namespace
 		D3D10_RESOURCE_DIMENSION_TEXTURE3D = 4,
 	};
 
+
+	const char * getDxgiFormatString(DXGI_FORMAT dxgiFormat)
+	{
+#define CASE(format) case DXGI_FORMAT_##format: return #format
+		switch(dxgiFormat)
+		{
+			CASE(UNKNOWN);
+			
+			CASE(R32G32B32A32_TYPELESS);
+			CASE(R32G32B32A32_FLOAT);
+			CASE(R32G32B32A32_UINT);
+			CASE(R32G32B32A32_SINT);
+			
+			CASE(R32G32B32_TYPELESS);
+			CASE(R32G32B32_FLOAT);
+			CASE(R32G32B32_UINT);
+			CASE(R32G32B32_SINT);
+			
+			CASE(R16G16B16A16_TYPELESS);
+			CASE(R16G16B16A16_FLOAT);
+			CASE(R16G16B16A16_UNORM);
+			CASE(R16G16B16A16_UINT);
+			CASE(R16G16B16A16_SNORM);
+			CASE(R16G16B16A16_SINT);
+			
+			CASE(R32G32_TYPELESS);
+			CASE(R32G32_FLOAT);
+			CASE(R32G32_UINT);
+			CASE(R32G32_SINT);
+			
+			CASE(R32G8X24_TYPELESS);
+			CASE(D32_FLOAT_S8X24_UINT);
+			CASE(R32_FLOAT_X8X24_TYPELESS);
+			CASE(X32_TYPELESS_G8X24_UINT);
+			
+			CASE(R10G10B10A2_TYPELESS);
+			CASE(R10G10B10A2_UNORM);
+			CASE(R10G10B10A2_UINT);
+			
+			CASE(R11G11B10_FLOAT);
+			
+			CASE(R8G8B8A8_TYPELESS);
+			CASE(R8G8B8A8_UNORM);
+			CASE(R8G8B8A8_UNORM_SRGB);
+			CASE(R8G8B8A8_UINT);
+			CASE(R8G8B8A8_SNORM);
+			CASE(R8G8B8A8_SINT);
+			
+			CASE(R16G16_TYPELESS);
+			CASE(R16G16_FLOAT);
+			CASE(R16G16_UNORM);
+			CASE(R16G16_UINT);
+			CASE(R16G16_SNORM);
+			CASE(R16G16_SINT);
+			
+			CASE(R32_TYPELESS);
+			CASE(D32_FLOAT);
+			CASE(R32_FLOAT);
+			CASE(R32_UINT);
+			CASE(R32_SINT);
+			
+			CASE(R24G8_TYPELESS);
+			CASE(D24_UNORM_S8_UINT);
+			CASE(R24_UNORM_X8_TYPELESS);
+			CASE(X24_TYPELESS_G8_UINT);
+			
+			CASE(R8G8_TYPELESS);
+			CASE(R8G8_UNORM);
+			CASE(R8G8_UINT);
+			CASE(R8G8_SNORM);
+			CASE(R8G8_SINT);
+			
+			CASE(R16_TYPELESS);
+			CASE(R16_FLOAT);
+			CASE(D16_UNORM);
+			CASE(R16_UNORM);
+			CASE(R16_UINT);
+			CASE(R16_SNORM);
+			CASE(R16_SINT);
+			
+			CASE(R8_TYPELESS);
+			CASE(R8_UNORM);
+			CASE(R8_UINT);
+			CASE(R8_SNORM);
+			CASE(R8_SINT);
+			CASE(A8_UNORM);
+
+			CASE(R1_UNORM);
+		
+			CASE(R9G9B9E5_SHAREDEXP);
+			
+			CASE(R8G8_B8G8_UNORM);
+			CASE(G8R8_G8B8_UNORM);
+
+			CASE(BC1_TYPELESS);
+			CASE(BC1_UNORM);
+			CASE(BC1_UNORM_SRGB);
+		
+			CASE(BC2_TYPELESS);
+			CASE(BC2_UNORM);
+			CASE(BC2_UNORM_SRGB);
+		
+			CASE(BC3_TYPELESS);
+			CASE(BC3_UNORM);
+			CASE(BC3_UNORM_SRGB);
+		
+			CASE(BC4_TYPELESS);
+			CASE(BC4_UNORM);
+			CASE(BC4_SNORM);
+		
+			CASE(BC5_TYPELESS);
+			CASE(BC5_UNORM);
+			CASE(BC5_SNORM);
+
+			CASE(B5G6R5_UNORM);
+			CASE(B5G5R5A1_UNORM);
+			CASE(B8G8R8A8_UNORM);
+			CASE(B8G8R8X8_UNORM);
+
+			default: 
+				return "UNKNOWN";
+		}
+#undef CASE
+	}
+	
+	const char * getD3d10ResourceDimensionString(D3D10_RESOURCE_DIMENSION resourceDimension)
+	{
+		switch(resourceDimension)
+		{
+			default:
+			case D3D10_RESOURCE_DIMENSION_UNKNOWN: return "UNKNOWN";
+			case D3D10_RESOURCE_DIMENSION_BUFFER: return "BUFFER";
+			case D3D10_RESOURCE_DIMENSION_TEXTURE1D: return "TEXTURE1D";
+			case D3D10_RESOURCE_DIMENSION_TEXTURE2D: return "TEXTURE2D";
+			case D3D10_RESOURCE_DIMENSION_TEXTURE3D: return "TEXTURE3D";
+		}
+	}
+
 } // namespace
 
 namespace nv
@@ -390,7 +532,7 @@ DDSHeader::DDSHeader()
 
 	// Store version information on the reserved header attributes.
 	this->reserved[9] = MAKEFOURCC('N', 'V', 'T', 'T');
-	this->reserved[10] = (0 << 16) | (9 << 8) | (5);	// major.minor.revision
+	this->reserved[10] = (2 << 16) | (0 << 8) | (4);	// major.minor.revision
 
 	this->pf.size = 32;
 	this->pf.flags = 0;
@@ -494,7 +636,16 @@ void DDSHeader::setFourCC(uint8 c0, uint8 c1, uint8 c2, uint8 c3)
 	// set fourcc pixel format.
 	this->pf.flags = DDPF_FOURCC;
 	this->pf.fourcc = MAKEFOURCC(c0, c1, c2, c3);
-	this->pf.bitcount = 0;
+	
+	if (this->pf.fourcc == FOURCC_ATI2)
+	{
+		this->pf.bitcount = FOURCC_A2XY;
+	}
+	else
+	{
+		this->pf.bitcount = 0;
+	}
+	
 	this->pf.rmask = 0;
 	this->pf.gmask = 0;
 	this->pf.bmask = 0;
@@ -530,9 +681,9 @@ void DDSHeader::setPixelFormat(uint bitcount, uint rmask, uint gmask, uint bmask
 	nvCheck(bitcount > 0 && bitcount <= 32);
 
 	// Align to 8.
-	if (bitcount < 8) bitcount = 8;
-	else if (bitcount < 16) bitcount = 16;
-	else if (bitcount < 24) bitcount = 24;
+	if (bitcount <= 8) bitcount = 8;
+	else if (bitcount <= 16) bitcount = 16;
+	else if (bitcount <= 24) bitcount = 24;
 	else bitcount = 32;
 
 	this->pf.fourcc = 0; //findD3D9Format(bitcount, rmask, gmask, bmask, amask);
@@ -545,7 +696,8 @@ void DDSHeader::setPixelFormat(uint bitcount, uint rmask, uint gmask, uint bmask
 
 void DDSHeader::setDX10Format(uint format)
 {
-	this->pf.flags = 0;
+	//this->pf.flags = 0;
+	this->pf.fourcc = FOURCC_DX10;
 	this->header10.dxgiFormat = format;
 }
 
@@ -593,7 +745,8 @@ void DDSHeader::swapBytes()
 
 bool DDSHeader::hasDX10Header() const
 {
-	return this->pf.flags == 0;
+	return this->pf.fourcc == FOURCC_DX10;  // @@ This is according to AMD
+	//return this->pf.flags == 0;             // @@ This is according to MS
 }
 
 
@@ -623,7 +776,7 @@ bool DirectDrawSurface::isValid() const
 		return false;
 	}
 	
-	const uint required = (DDSD_WIDTH|DDSD_HEIGHT|DDSD_CAPS|DDSD_PIXELFORMAT);
+	const uint required = (DDSD_WIDTH|DDSD_HEIGHT/*|DDSD_CAPS|DDSD_PIXELFORMAT*/);
 	if( (header.flags & required) != required ) {
 		return false;
 	}
@@ -643,40 +796,46 @@ bool DirectDrawSurface::isSupported() const
 {
 	nvDebugCheck(isValid());
 	
-	if (header.pf.flags & DDPF_FOURCC)
+	if (header.hasDX10Header())
 	{
-		if (header.pf.fourcc != FOURCC_DXT1 &&
-		    header.pf.fourcc != FOURCC_DXT2 &&
-		    header.pf.fourcc != FOURCC_DXT3 &&
-		    header.pf.fourcc != FOURCC_DXT4 &&
-		    header.pf.fourcc != FOURCC_DXT5 &&
-		    header.pf.fourcc != FOURCC_RXGB &&
-		    header.pf.fourcc != FOURCC_ATI1 &&
-		    header.pf.fourcc != FOURCC_ATI2)
-		{
-			// Unknown fourcc code.
-			return false;
-		}
-	}
-	else if (header.pf.flags & DDPF_RGB)
-	{
-		// All RGB formats are supported now.
 	}
 	else
 	{
-		return false;
-	}
+		if (header.pf.flags & DDPF_FOURCC)
+		{
+			if (header.pf.fourcc != FOURCC_DXT1 &&
+				header.pf.fourcc != FOURCC_DXT2 &&
+				header.pf.fourcc != FOURCC_DXT3 &&
+				header.pf.fourcc != FOURCC_DXT4 &&
+				header.pf.fourcc != FOURCC_DXT5 &&
+				header.pf.fourcc != FOURCC_RXGB &&
+				header.pf.fourcc != FOURCC_ATI1 &&
+				header.pf.fourcc != FOURCC_ATI2)
+			{
+				// Unknown fourcc code.
+				return false;
+			}
+		}
+		else if (header.pf.flags & DDPF_RGB)
+		{
+			// All RGB formats are supported now.
+		}
+		else
+		{
+			return false;
+		}
 		
-	if (isTextureCube() && (header.caps.caps2 & DDSCAPS2_CUBEMAP_ALL_FACES) != DDSCAPS2_CUBEMAP_ALL_FACES)
-	{
-		// Cubemaps must contain all faces.
-		return false;
-	}
+		if (isTextureCube() && (header.caps.caps2 & DDSCAPS2_CUBEMAP_ALL_FACES) != DDSCAPS2_CUBEMAP_ALL_FACES)
+		{
+			// Cubemaps must contain all faces.
+			return false;
+		}
 		
-	if (isTexture3D())
-	{
-		// @@ 3D textures not supported yet.
-		return false;
+		if (isTexture3D())
+		{
+			// @@ 3D textures not supported yet.
+			return false;
+		}
 	}
 	
 	return true;
@@ -712,16 +871,40 @@ uint DirectDrawSurface::depth() const
 	else return 1;
 }
 
+bool DirectDrawSurface::isTexture1D() const
+{
+	nvDebugCheck(isValid());
+	if (header.hasDX10Header())
+	{
+		return header.header10.resourceDimension == D3D10_RESOURCE_DIMENSION_TEXTURE1D;
+	}
+	return false;
+}
+
 bool DirectDrawSurface::isTexture2D() const
 {
 	nvDebugCheck(isValid());
-	return !isTexture3D() && !isTextureCube();
+	if (header.hasDX10Header())
+	{
+		return header.header10.resourceDimension == D3D10_RESOURCE_DIMENSION_TEXTURE2D;
+	}
+	else
+	{
+		return !isTexture3D() && !isTextureCube();
+	}
 }
 
 bool DirectDrawSurface::isTexture3D() const
 {
 	nvDebugCheck(isValid());
-	return (header.caps.caps2 & DDSCAPS2_VOLUME) != 0;
+	if (header.hasDX10Header())
+	{
+		return header.header10.resourceDimension == D3D10_RESOURCE_DIMENSION_TEXTURE3D;
+	}
+	else
+	{
+		return (header.caps.caps2 & DDSCAPS2_VOLUME) != 0;
+	}
 }
 
 bool DirectDrawSurface::isTextureCube() const
@@ -730,6 +913,12 @@ bool DirectDrawSurface::isTextureCube() const
 	return (header.caps.caps2 & DDSCAPS2_CUBEMAP) != 0;
 }
 
+void DirectDrawSurface::setNormalFlag(bool b)
+{
+	nvDebugCheck(isValid());
+	header.setNormalFlag(b);
+}
+
 void DirectDrawSurface::mipmap(Image * img, uint face, uint mipmap)
 {
 	nvDebugCheck(isValid());
@@ -780,7 +969,13 @@ void DirectDrawSurface::readLinearImage(Image * img)
 
 	uint byteCount = (header.pf.bitcount + 7) / 8;
 
-	if (header.pf.amask != 0)
+	// set image format: RGB or ARGB
+	// alpha channel exists if and only if the alpha mask is non-zero
+	if (header.pf.amask == 0)
+ 	{
+		img->setFormat(Image::Format_RGB);
+	}
+	else
 	{
 		img->setFormat(Image::Format_ARGB);
 	}
@@ -809,6 +1004,19 @@ void DirectDrawSurface::readBlockImage(Image * img)
 	nvDebugCheck(stream != NULL);
 	nvDebugCheck(img != NULL);
 
+	// set image format: RGB or ARGB
+	if (header.pf.fourcc == FOURCC_RXGB ||
+		header.pf.fourcc == FOURCC_ATI1 ||
+		header.pf.fourcc == FOURCC_ATI2 ||
+		header.pf.flags & DDPF_NORMAL)
+	{
+		img->setFormat(Image::Format_RGB);
+	}
+	else
+	{
+		img->setFormat(Image::Format_ARGB);
+	}
+
 	const uint w = img->width();
 	const uint h = img->height();
 	
@@ -1044,8 +1252,23 @@ void DirectDrawSurface::printInfo() const
 	if (header.pf.flags & DDPF_ALPHAPREMULT) printf("\t\tDDPF_ALPHAPREMULT\n");
 	if (header.pf.flags & DDPF_NORMAL) printf("\t\tDDPF_NORMAL\n");
 	
-	printf("\tFourCC: '%c%c%c%c'\n", ((header.pf.fourcc >> 0) & 0xFF), ((header.pf.fourcc >> 8) & 0xFF), ((header.pf.fourcc >> 16) & 0xFF), ((header.pf.fourcc >> 24) & 0xFF));
-	printf("\tBit count: %d\n", header.pf.bitcount);
+	printf("\tFourCC: '%c%c%c%c'\n",
+		((header.pf.fourcc >> 0) & 0xFF),
+		((header.pf.fourcc >> 8) & 0xFF),
+		((header.pf.fourcc >> 16) & 0xFF),
+		((header.pf.fourcc >> 24) & 0xFF));
+	if ((header.pf.fourcc & DDPF_FOURCC) && (header.pf.bitcount != 0))
+	{
+		printf("\tSwizzle: '%c%c%c%c'\n", 
+			(header.pf.bitcount >> 0) & 0xFF,
+			(header.pf.bitcount >> 8) & 0xFF,
+			(header.pf.bitcount >> 16) & 0xFF,
+			(header.pf.bitcount >> 24) & 0xFF);
+	}
+	else
+	{
+		printf("\tBit count: %d\n", header.pf.bitcount);
+	}
 	printf("\tRed mask: 0x%.8X\n", header.pf.rmask);
 	printf("\tGreen mask: 0x%.8X\n", header.pf.gmask);
 	printf("\tBlue mask: 0x%.8X\n", header.pf.bmask);
@@ -1076,11 +1299,11 @@ void DirectDrawSurface::printInfo() const
 	printf("\tCaps 3: 0x%.8X\n", header.caps.caps3);
 	printf("\tCaps 4: 0x%.8X\n", header.caps.caps4);
 
-	if (header.pf.flags == 0)
+	if (header.hasDX10Header())
 	{
 		printf("DX10 Header:\n");
-		printf("\tDXGI Format: %u\n", header.header10.dxgiFormat);
-		printf("\tResource dimension: %u\n", header.header10.resourceDimension);
+		printf("\tDXGI Format: %u (%s)\n", header.header10.dxgiFormat, getDxgiFormatString((DXGI_FORMAT)header.header10.dxgiFormat));
+		printf("\tResource dimension: %u (%s)\n", header.header10.resourceDimension, getD3d10ResourceDimensionString((D3D10_RESOURCE_DIMENSION)header.header10.resourceDimension));
 		printf("\tMisc flag: %u\n", header.header10.miscFlag);
 		printf("\tArray size: %u\n", header.header10.arraySize);
 	}

src/nvimage/DirectDrawSurface.h

@@ -119,10 +119,13 @@ namespace nv
 		uint width() const;
 		uint height() const;
 		uint depth() const;
+		bool isTexture1D() const;
 		bool isTexture2D() const;
 		bool isTexture3D() const;
 		bool isTextureCube() const;
 
+		void setNormalFlag(bool b);
+		
 		void mipmap(Image * img, uint f, uint m);
 		//	void mipmap(FloatImage * img, uint f, uint m);
 		

src/nvimage/Filter.cpp

@@ -244,7 +244,7 @@ SincFilter::SincFilter(float w) : Filter(w) {}
 
 float SincFilter::evaluate(float x) const
 {
-	return 0.0f;
+	return sincf(PI * x);
 }
 
 
@@ -541,12 +541,17 @@ void Kernel2::initBlendedSobel(const Vector4 & scale)
 
 PolyphaseKernel::PolyphaseKernel(const Filter & f, uint srcLength, uint dstLength, int samples/*= 32*/)
 {
-	nvCheck(srcLength >= dstLength);	// @@ Upsampling not implemented!
 	nvDebugCheck(samples > 0);
 
-	const float scale = float(dstLength) / float(srcLength);
+	float scale = float(dstLength) / float(srcLength);
 	const float iscale = 1.0f / scale;
 
+	if (scale > 1) {
+		// Upsampling.
+		samples = 1;
+		scale = 1;
+	}
+
 	m_length = dstLength;
 	m_width = f.width() * iscale;
 	m_windowSize = (int)ceilf(m_width * 2) + 1;
@@ -577,6 +582,7 @@ PolyphaseKernel::PolyphaseKernel(const Filter & f, uint srcLength, uint dstLengt
 			m_data[i * m_windowSize + j] /= total;
 		}
 	}
+
 }
 
 PolyphaseKernel::~PolyphaseKernel()

src/nvimage/FloatImage.cpp

@@ -376,7 +376,7 @@ FloatImage * FloatImage::fastDownSample() const
 	{
 		const uint n = w * h;
 		
-		if (n & 1)
+		if ((m_width * m_height) & 1)
 		{
 			const float scale = 1.0f / (2 * n + 1);
 			
@@ -540,73 +540,18 @@ FloatImage * FloatImage::fastDownSample() const
 	return dst_image.release();
 }
 
-/*
-/// Downsample applying a 1D kernel separately in each dimension.
-FloatImage * FloatImage::downSample(const Kernel1 & kernel, WrapMode wm) const
-{
-	const uint w = max(1, m_width / 2);
-	const uint h = max(1, m_height / 2);
-	
-	return downSample(kernel, w, h, wm);
-}
-
-
-/// Downsample applying a 1D kernel separately in each dimension.
-FloatImage * FloatImage::downSample(const Kernel1 & kernel, uint w, uint h, WrapMode wm) const
-{
-	nvCheck(!(kernel.windowSize() & 1));	// Make sure that kernel m_width is even.
-
-	AutoPtr<FloatImage> tmp_image( new FloatImage() );
-	tmp_image->allocate(m_componentNum, w, m_height);
-	
-	AutoPtr<FloatImage> dst_image( new FloatImage() );	
-	dst_image->allocate(m_componentNum, w, h);
-	
-	const float xscale = float(m_width) / float(w);
-	const float yscale = float(m_height) / float(h);
-	
-	for(uint c = 0; c < m_componentNum; c++) {
-		float * tmp_channel = tmp_image->channel(c);
-		
-		for(uint y = 0; y < m_height; y++) {
-			for(uint x = 0; x < w; x++) {
-				
-				float sum = this->applyKernelHorizontal(&kernel, uint(x*xscale), y, c, wm);
-				
-				const uint tmp_index = tmp_image->index(x, y);
-				tmp_channel[tmp_index] = sum;
-			}
-		}
-		
-		float * dst_channel = dst_image->channel(c);
-		
-		for(uint y = 0; y < h; y++) {
-			for(uint x = 0; x < w; x++) {
-				
-				float sum = tmp_image->applyKernelVertical(&kernel, uint(x*xscale), uint(y*yscale), c, wm);
-				
-				const uint dst_index = dst_image->index(x, y);
-				dst_channel[dst_index] = sum;
-			}
-		}
-	}
-	
-	return dst_image.release();
-}
-*/
-
 /// Downsample applying a 1D kernel separately in each dimension.
 FloatImage * FloatImage::downSample(const Filter & filter, WrapMode wm) const
 {
 	const uint w = max(1, m_width / 2);
 	const uint h = max(1, m_height / 2);
 
-	return downSample(filter, w, h, wm);
+	return resize(filter, w, h, wm);
 }
 
 
 /// Downsample applying a 1D kernel separately in each dimension.
-FloatImage * FloatImage::downSample(const Filter & filter, uint w, uint h, WrapMode wm) const
+FloatImage * FloatImage::resize(const Filter & filter, uint w, uint h, WrapMode wm) const
 {
 	// @@ Use monophase filters when frac(m_width / w) == 0
 

src/nvimage/FloatImage.h

@@ -63,7 +63,7 @@ public:
 	
 	NVIMAGE_API FloatImage * fastDownSample() const;
 	NVIMAGE_API FloatImage * downSample(const Filter & filter, WrapMode wm) const;
-	NVIMAGE_API FloatImage * downSample(const Filter & filter, uint w, uint h, WrapMode wm) const;
+	NVIMAGE_API FloatImage * resize(const Filter & filter, uint w, uint h, WrapMode wm) const;
 
 	//NVIMAGE_API FloatImage * downSample(const Kernel1 & filter, WrapMode wm) const;
 	//NVIMAGE_API FloatImage * downSample(const Kernel1 & filter, uint w, uint h, WrapMode wm) const;
@@ -226,14 +226,18 @@ inline uint FloatImage::indexRepeat(int x, int y) const
 
 inline uint FloatImage::indexMirror(int x, int y) const
 {
+	if (m_width == 1) x = 0;
+
 	x = abs(x);
 	while (x >= m_width) {
-		x = m_width + m_width - x - 2;
+		x = abs(m_width + m_width - x - 2);
 	}
 
+	if (m_height == 1) y = 0;
+	
 	y = abs(y);
 	while (y >= m_height) {
-		y = m_height + m_height - y - 2;
+		y = abs(m_height + m_height - y - 2);
 	}
 
 	return index(x, y);

src/nvimage/HoleFilling.cpp

@@ -296,7 +296,7 @@ static bool downsample(const FloatImage * src, const BitMap * srcMask, const Flo
 	return true;
 }
 
-// This is the filter used in the Lumigraph paper. The Unreal engine uses something similar.
+// This is the filter used in the Lumigraph paper.
 void nv::fillPullPush(FloatImage * img, const BitMap * bmap)
 {
 	nvCheck(img != NULL);
@@ -644,8 +644,8 @@ struct LocalPixels
 
 
 
-// This is a cubic extrapolation filter from Charles Bloom (DoPixelSeamFix).
-void nv::fillCubicExtrapolate(int passCount, FloatImage * img, BitMap * bmap, int coverageIndex /*= -1*/)
+// This is a quadratic extrapolation filter from Charles Bloom (DoPixelSeamFix). Used with his permission.
+void nv::fillQuadraticExtrapolate(int passCount, FloatImage * img, BitMap * bmap, int coverageIndex /*= -1*/)
 {
 	nvCheck(passCount > 0);
 	nvCheck(img != NULL);

src/nvimage/HoleFilling.h

@@ -89,7 +89,7 @@ namespace nv
 	NVIMAGE_API void fillPullPush(FloatImage * img, const BitMap * bmap);
 	
 	NVIMAGE_API void fillExtrapolate(int passCount, FloatImage * img, BitMap * bmap);
-	NVIMAGE_API void fillCubicExtrapolate(int passCount, FloatImage * img, BitMap * bmap, int coverageIndex = -1);
+	NVIMAGE_API void fillQuadraticExtrapolate(int passCount, FloatImage * img, BitMap * bmap, int coverageIndex = -1);
 	
 } // nv namespace
 

src/nvimage/Image.cpp

@@ -15,7 +15,7 @@ Image::Image() : m_width(0), m_height(0), m_format(Format_RGB), m_data(NULL)
 {
 }
 
-Image::Image(const Image & img)
+Image::Image(const Image & img) : m_data(NULL)
 {
 	allocate(img.m_width, img.m_height);
 	m_format = img.m_format;

src/nvimage/Quantize.cpp

@@ -16,6 +16,7 @@ http://www.efg2.com/Lab/Library/ImageProcessing/DHALF.TXT
 
 #include <nvimage/Image.h>
 #include <nvimage/Quantize.h>
+#include <nvimage/PixelFormat.h>
 
 using namespace nv;
 
@@ -47,94 +48,20 @@ void nv::Quantize::BinaryAlpha( Image * image, int alpha_threshold /*= 127*/ )
 // Simple quantization.
 void nv::Quantize::RGB16( Image * image )
 {
-	nvCheck(image != NULL);
-	
-	const uint w = image->width();
-	const uint h = image->height();
-	
-	for(uint y = 0; y < h; y++) {
-		for(uint x = 0; x < w; x++) {
-			
-			Color32 pixel32 = image->pixel(x, y);
-			
-			// Convert to 16 bit and back to 32 using regular bit expansion.
-			Color32 pixel16 = toColor32( toColor16(pixel32) );
-			
-			// Store color.
-			image->pixel(x, y) = pixel16;
-		}
-	}
+	Truncate(image, 5, 6, 5, 8);
 }
 
 // Alpha quantization.
 void nv::Quantize::Alpha4( Image * image )
 {
-	nvCheck(image != NULL);
-	
-	const uint w = image->width();
-	const uint h = image->height();
-	
-	for(uint y = 0; y < h; y++) {
-		for(uint x = 0; x < w; x++) {
-			
-			Color32 pixel = image->pixel(x, y);
-			
-			// Convert to 4 bit using regular bit expansion.
-			pixel.a = (pixel.a & 0xF0) | ((pixel.a & 0xF0) >> 4);
-			
-			// Store color.
-			image->pixel(x, y) = pixel;
-		}
-	}
+	Truncate(image, 8, 8, 8, 4);
 }
 
 
 // Error diffusion. Floyd Steinberg.
 void nv::Quantize::FloydSteinberg_RGB16( Image * image )
 {
-	nvCheck(image != NULL);
-	
-	const uint w = image->width();
-	const uint h = image->height();
-	
-	// @@ Use fixed point?
-	Vector3 * row0 = new Vector3[w+2];
-	Vector3 * row1 = new Vector3[w+2];
-	memset(row0, 0, sizeof(Vector3)*(w+2));
-	memset(row1, 0, sizeof(Vector3)*(w+2));
-	
-	for(uint y = 0; y < h; y++) {
-		for(uint x = 0; x < w; x++) {
-			
-			Color32 pixel32 = image->pixel(x, y);
-			
-			// Add error.	// @@ We shouldn't clamp here!
-			pixel32.r = clamp(int(pixel32.r) + int(row0[1+x].x()), 0, 255);
-			pixel32.g = clamp(int(pixel32.g) + int(row0[1+x].y()), 0, 255);
-			pixel32.b = clamp(int(pixel32.b) + int(row0[1+x].z()), 0, 255);
-			
-			// Convert to 16 bit. @@ Use regular clamp?
-			Color32 pixel16 = toColor32( toColor16(pixel32) );
-			
-			// Store color.
-			image->pixel(x, y) = pixel16;
-			
-			// Compute new error.
-			Vector3 diff(float(pixel32.r - pixel16.r), float(pixel32.g - pixel16.g), float(pixel32.b - pixel16.b));
-			
-			// Propagate new error.
-			row0[1+x+1] += 7.0f / 16.0f * diff;
-			row1[1+x-1] += 3.0f / 16.0f * diff;
-			row1[1+x+0] += 5.0f / 16.0f * diff;
-			row1[1+x+1] += 1.0f / 16.0f * diff;
-		}
-		
-		swap(row0, row1);
-		memset(row1, 0, sizeof(Vector3)*(w+2));
-	}
-	
-	delete [] row0;
-	delete [] row1;
+	FloydSteinberg(image, 5, 6, 5, 8);
 }
 
 
@@ -188,34 +115,90 @@ void nv::Quantize::FloydSteinberg_BinaryAlpha( Image * image, int alpha_threshol
 
 // Error diffusion. Floyd Steinberg.
 void nv::Quantize::FloydSteinberg_Alpha4( Image * image )
+{
+	FloydSteinberg(image, 8, 8, 8, 4);
+}
+
+
+void nv::Quantize::Truncate(Image * image, uint rsize, uint gsize, uint bsize, uint asize)
 {
 	nvCheck(image != NULL);
 	
 	const uint w = image->width();
 	const uint h = image->height();
 	
-	// @@ Use fixed point?
-	float * row0 = new float[(w+2)];
-	float * row1 = new float[(w+2)];
-	memset(row0, 0, sizeof(float)*(w+2));
-	memset(row1, 0, sizeof(float)*(w+2));
-	
 	for(uint y = 0; y < h; y++) {
 		for(uint x = 0; x < w; x++) {
 			
 			Color32 pixel = image->pixel(x, y);
 
-			// Add error.
-			int alpha = int(pixel.a) + int(row0[1+x]);
+			// Convert to our desired size, and reconstruct.
+			pixel.r = PixelFormat::convert(pixel.r, 8, rsize);
+			pixel.r = PixelFormat::convert(pixel.r, rsize, 8);
 
-			// Convert to 4 bit using regular bit expansion.
-			pixel.a = (pixel.a & 0xF0) | ((pixel.a & 0xF0) >> 4);
+			pixel.g = PixelFormat::convert(pixel.g, 8, gsize);
+			pixel.g = PixelFormat::convert(pixel.g, gsize, 8);
+
+			pixel.b = PixelFormat::convert(pixel.b, 8, bsize);
+			pixel.b = PixelFormat::convert(pixel.b, bsize, 8);
+
+			pixel.a = PixelFormat::convert(pixel.a, 8, asize);
+			pixel.a = PixelFormat::convert(pixel.a, asize, 8);
 
 			// Store color.
 			image->pixel(x, y) = pixel;
+		}
+	}
+}
+
+
+// Error diffusion. Floyd Steinberg.
+void nv::Quantize::FloydSteinberg(Image * image, uint rsize, uint gsize, uint bsize, uint asize)
+{
+	nvCheck(image != NULL);
+	
+	const uint w = image->width();
+	const uint h = image->height();
+	
+	Vector4 * row0 = new Vector4[w+2];
+	Vector4 * row1 = new Vector4[w+2];
+	memset(row0, 0, sizeof(Vector4)*(w+2));
+	memset(row1, 0, sizeof(Vector4)*(w+2));
+	
+	for (uint y = 0; y < h; y++) {
+		for (uint x = 0; x < w; x++) {
+			
+			Color32 pixel = image->pixel(x, y);
+
+			// Add error.
+			pixel.r = clamp(int(pixel.r) + int(row0[1+x].x()), 0, 255);
+			pixel.g = clamp(int(pixel.g) + int(row0[1+x].y()), 0, 255);
+			pixel.b = clamp(int(pixel.b) + int(row0[1+x].z()), 0, 255);
+			pixel.a = clamp(int(pixel.a) + int(row0[1+x].w()), 0, 255);
+			
+			int r = pixel.r;
+			int g = pixel.g;
+			int b = pixel.b;
+			int a = pixel.a;
+
+			// Convert to our desired size, and reconstruct.
+			r = PixelFormat::convert(r, 8, rsize);
+			r = PixelFormat::convert(r, rsize, 8);
+
+			g = PixelFormat::convert(g, 8, gsize);
+			g = PixelFormat::convert(g, gsize, 8);
+
+			b = PixelFormat::convert(b, 8, bsize);
+			b = PixelFormat::convert(b, bsize, 8);
+
+			a = PixelFormat::convert(a, 8, asize);
+			a = PixelFormat::convert(a, asize, 8);
+
+			// Store color.
+			image->pixel(x, y) = Color32(r, g, b, a);
 			
 			// Compute new error.
-			float diff = float(alpha - pixel.a);
+			Vector4 diff(float(int(pixel.r) - r), float(int(pixel.g) - g), float(int(pixel.b) - b), float(int(pixel.a) - a));
 			
 			// Propagate new error.
 			row0[1+x+1] += 7.0f / 16.0f * diff;
@@ -225,10 +208,9 @@ void nv::Quantize::FloydSteinberg_Alpha4( Image * image )
 		}
 		
 		swap(row0, row1);
-		memset(row1, 0, sizeof(float)*(w+2));
+		memset(row1, 0, sizeof(Vector4)*(w+2));
 	}
 	
 	delete [] row0;
 	delete [] row1;
 }
-

src/nvimage/Quantize.h

@@ -17,6 +17,9 @@ namespace nv
 		void FloydSteinberg_BinaryAlpha(Image * img, int alpha_threshold = 127);
 		void FloydSteinberg_Alpha4(Image * img);
 
+		void Truncate(Image * image, uint rsize, uint gsize, uint bsize, uint asize);
+		void FloydSteinberg(Image * image, uint rsize, uint gsize, uint bsize, uint asize);
+
 		// @@ Add palette quantization algorithms!
 	}
 }

src/nvmath/CMakeLists.txt

@@ -7,8 +7,6 @@ SET(MATH_SRCS
 	Quaternion.h
 	Box.h
 	Color.h
-	Eigen.h Eigen.cpp
-	Fitting.h Fitting.cpp
 	Montecarlo.h Montecarlo.cpp
 	Random.h Random.cpp
 	SphericalHarmonic.h SphericalHarmonic.cpp

src/nvmath/Eigen.cpp

@@ -1,533 +0,0 @@
-// This code is in the public domain -- castanyo@yahoo.es
-
-#include "Eigen.h"
-
-using namespace nv;
-
-static const float EPS = 0.00001f;
-static const int MAX_ITER = 100;
-
-static void semi_definite_symmetric_eigen(const float *mat, int n, float *eigen_vec, float *eigen_val);
-
-
-// Use power method to find the first eigenvector.
-// http://www.miislita.com/information-retrieval-tutorial/matrix-tutorial-3-eigenvalues-eigenvectors.html
-Vector3 nv::firstEigenVector(float matrix[6])
-{
-	// Number of iterations. @@ Use a variable number of iterations.
-	const int NUM = 8;
-
-	Vector3 v(1, 1, 1);
-	for(int i = 0; i < NUM; i++) {
-		float x = v.x() * matrix[0] + v.y() * matrix[1] + v.z() * matrix[2];
-		float y = v.x() * matrix[1] + v.y() * matrix[3] + v.z() * matrix[4];
-		float z = v.x() * matrix[2] + v.y() * matrix[4] + v.z() * matrix[5];
-		
-		float norm = max(max(x, y), z);
-		float iv = 1.0f / norm;
-		if (norm == 0.0f) {
-			return Vector3(zero);
-		}
-		
-		v.set(x*iv, y*iv, z*iv);
-	}
-
-	return v;
-}
-
-
-/// Solve eigen system.
-void Eigen::solve() {
-	semi_definite_symmetric_eigen(matrix, N, eigen_vec, eigen_val);
-}
-
-/// Solve eigen system.
-void Eigen3::solve() {
-	// @@ Use lengyel code that seems to be more optimized.
-#if 1
-	float v[3*3];
-	semi_definite_symmetric_eigen(matrix, 3, v, eigen_val);
-
-	eigen_vec[0].set(v[0], v[1], v[2]);
-	eigen_vec[1].set(v[3], v[4], v[5]);
-	eigen_vec[2].set(v[6], v[7], v[8]);
-#else
-	const int maxSweeps = 32;
-	const float epsilon = 1.0e-10f;
-	
-	float m11 = matrix[0]; // m(0,0);
-	float m12 = matrix[1]; // m(0,1);
-	float m13 = matrix[2]; // m(0,2);
-	float m22 = matrix[3]; // m(1,1);
-	float m23 = matrix[4]; // m(1,2);
-	float m33 = matrix[5]; // m(2,2);
-	
-	//r.SetIdentity();	
-	eigen_vec[0].set(1, 0, 0);
-	eigen_vec[1].set(0, 1, 0);
-	eigen_vec[2].set(0, 0, 1);
-	
-	for (int a = 0; a < maxSweeps; a++)
-	{
-		// Exit if off-diagonal entries small enough
-		if ((fabs(m12) < epsilon) && (fabs(m13) < epsilon) && (fabs(m23) < epsilon))
-		{
-			break;
-		}
-		
-		// Annihilate (1,2) entry
-		if (m12 != 0.0f)
-		{
-			float u = (m22 - m11) * 0.5f / m12;
-			float u2 = u * u;
-			float u2p1 = u2 + 1.0f;
-			float t = (u2p1 != u2) ? ((u < 0.0f) ? -1.0f : 1.0f) * (sqrt(u2p1) - fabs(u)) : 0.5f / u;
-			float c = 1.0f / sqrt(t * t + 1.0f);
-			float s = c * t;
-			
-			m11 -= t * m12;
-			m22 += t * m12;
-			m12 = 0.0f;
-			
-			float temp = c * m13 - s * m23;
-			m23 = s * m13 + c * m23;
-			m13 = temp;
-			
-			for (int i = 0; i < 3; i++)
-			{
-				float temp = c * eigen_vec[i].x - s * eigen_vec[i].y;
-				eigen_vec[i].y = s * eigen_vec[i].x + c * eigen_vec[i].y;
-				eigen_vec[i].x = temp;
-			}
-		}
-		
-		// Annihilate (1,3) entry
-		if (m13 != 0.0f)
-		{
-			float u = (m33 - m11) * 0.5f / m13;
-			float u2 = u * u;
-			float u2p1 = u2 + 1.0f;
-			float t = (u2p1 != u2) ? ((u < 0.0f) ? -1.0f : 1.0f) * (sqrt(u2p1) - fabs(u)) : 0.5f / u;
-			float c = 1.0f / sqrt(t * t + 1.0f);
-			float s = c * t;
-			
-			m11 -= t * m13;
-			m33 += t * m13;
-			m13 = 0.0f;
-			
-			float temp = c * m12 - s * m23;
-			m23 = s * m12 + c * m23;
-			m12 = temp;
-			
-			for (int i = 0; i < 3; i++)
-			{
-				float temp = c * eigen_vec[i].x - s * eigen_vec[i].z;
-				eigen_vec[i].z = s * eigen_vec[i].x + c * eigen_vec[i].z;
-				eigen_vec[i].x = temp;
-			}
-		}
-		
-		// Annihilate (2,3) entry
-		if (m23 != 0.0f)
-		{
-			float u = (m33 - m22) * 0.5f / m23;
-			float u2 = u * u;
-			float u2p1 = u2 + 1.0f;
-			float t = (u2p1 != u2) ? ((u < 0.0f) ? -1.0f : 1.0f) * (sqrt(u2p1) - fabs(u)) : 0.5f / u;
-			float c = 1.0f / sqrt(t * t + 1.0f);
-			float s = c * t;
-			
-			m22 -= t * m23;
-			m33 += t * m23;
-			m23 = 0.0f;
-			
-			float temp = c * m12 - s * m13;
-			m13 = s * m12 + c * m13;
-			m12 = temp;
-			
-			for (int i = 0; i < 3; i++)
-			{
-				float temp = c * eigen_vec[i].y - s * eigen_vec[i].z;
-				eigen_vec[i].z = s * eigen_vec[i].y + c * eigen_vec[i].z;
-				eigen_vec[i].y = temp;
-			}
-		}
-	}
-	
-	eigen_val[0] = m11;
-	eigen_val[1] = m22;
-	eigen_val[2] = m33;
-#endif
-}
-
-
-/*---------------------------------------------------------------------------
-	Functions
----------------------------------------------------------------------------*/
-    
-
-/** @@ I don't remember where did I get this function.
- * computes the eigen values and eigen vectors   
- * of a semi definite symmetric matrix
- *
- * -  matrix is stored in column symmetric storage, i.e.
- *     matrix = { m11, m12, m22, m13, m23, m33, m14, m24, m34, m44 ... }
- *     size = n(n+1)/2
- *
- * - eigen_vectors (return) = { v1, v2, v3, ..., vn } where vk = vk0, vk1, ..., vkn
- *     size = n^2, must be allocated by caller
- *
- * - eigen_values  (return) are in decreasing order
- *     size = n,   must be allocated by caller
- */
-
-void semi_definite_symmetric_eigen(
-    const float *mat, int n, float *eigen_vec, float *eigen_val
-) {
-    float *a,*v;
-    float a_norm,a_normEPS,thr,thr_nn;
-    int nb_iter = 0;
-    int jj;
-    int i,j,k,ij,ik,l,m,lm,mq,lq,ll,mm,imv,im,iq,ilv,il,nn;
-    int *index;
-    float a_ij,a_lm,a_ll,a_mm,a_im,a_il;
-    float a_lm_2;
-    float v_ilv,v_imv;
-    float x;
-    float sinx,sinx_2,cosx,cosx_2,sincos;
-    float delta;
-        
-    // Number of entries in mat
-        
-    nn = (n*(n+1))/2;
-        
-    // Step 1: Copy mat to a
-        
-    a = new float[nn];
-        
-    for( ij=0; ij<nn; ij++ ) {
-        a[ij] = mat[ij];
-    }
-        
-    // Ugly Fortran-porting trick: indices for a are between 1 and n
-    a--;
-        
-    // Step 2 : Init diagonalization matrix as the unit matrix
-    v = new float[n*n];
-        
-    ij = 0;
-    for( i=0; i<n; i++ ) {
-        for( j=0; j<n; j++ ) {
-            if( i==j ) {
-                v[ij++] = 1.0;
-            } else {
-                v[ij++] = 0.0;
-            }
-        }
-    }
-        
-    // Ugly Fortran-porting trick: indices for v are between 1 and n
-    v--;
-        
-    // Step 3 : compute the weight of the non diagonal terms 
-    ij     = 1  ;
-    a_norm = 0.0;
-    for( i=1; i<=n; i++ ) {
-        for( j=1; j<=i; j++ ) {
-            if( i!=j ) {
-                a_ij    = a[ij];
-                a_norm += a_ij*a_ij;
-            }
-            ij++;
-        }
-    }
-        
-    if( a_norm != 0.0 ) {
-        
-        a_normEPS = a_norm*EPS;
-        thr       = a_norm    ;
-    
-        // Step 4 : rotations
-        while( thr > a_normEPS   &&   nb_iter < MAX_ITER ) {
-        
-            nb_iter++;
-            thr_nn = thr / nn;
-            
-            for( l=1  ; l< n; l++ ) {
-                for( m=l+1; m<=n; m++ ) {
-                    
-                    // compute sinx and cosx 
-                    
-                    lq   = (l*l-l)/2;
-                    mq   = (m*m-m)/2;
-                    
-                    lm     = l+mq;
-                    a_lm   = a[lm];
-                    a_lm_2 = a_lm*a_lm;
-                    
-                    if( a_lm_2 < thr_nn ) {
-                        continue ;
-                    }
-                    
-                    ll   = l+lq;
-                    mm   = m+mq;
-                    a_ll = a[ll];
-                    a_mm = a[mm];
-                    
-                    delta = a_ll - a_mm;
-                    
-                    if( delta == 0.0f ) {
-                        x = - PI/4 ; 
-                    } else {
-                        x = - atanf( (a_lm+a_lm) / delta ) / 2.0f ;
-                    }
-
-                    sinx    = sinf(x);
-                    cosx    = cosf(x);
-                    sinx_2  = sinx*sinx;
-                    cosx_2  = cosx*cosx;
-                    sincos  = sinx*cosx;
-                    
-                    // rotate L and M columns 
-       
-                    ilv = n*(l-1);
-                    imv = n*(m-1);
-                    
-                    for( i=1; i<=n;i++ ) {
-                        if( (i!=l) && (i!=m) ) {
-                            iq = (i*i-i)/2;
-                            
-                            if( i<m )  { 
-                                im = i + mq; 
-                            } else {
-                                im = m + iq;
-                            }
-                            a_im = a[im];
-                            
-                            if( i<l ) {
-                                il = i + lq; 
-                            } else {
-                                il = l + iq;
-                            }
-                            a_il = a[il];
-                            
-                            a[il] =  a_il*cosx - a_im*sinx;
-                            a[im] =  a_il*sinx + a_im*cosx;
-                        }
-                        
-                        ilv++;
-                        imv++;
-                        
-                        v_ilv = v[ilv];
-                        v_imv = v[imv];
-                        
-                        v[ilv] = cosx*v_ilv - sinx*v_imv;
-                        v[imv] = sinx*v_ilv + cosx*v_imv;
-                    } 
-                    
-                    x = a_lm*sincos; x+=x;
-                    
-                    a[ll] =  a_ll*cosx_2 + a_mm*sinx_2 - x;
-                    a[mm] =  a_ll*sinx_2 + a_mm*cosx_2 + x;
-                    a[lm] =  0.0;
-                    
-                    thr = fabs( thr - a_lm_2 );
-                }
-            }
-        }         
-    }
-        
-    // Step 5: index conversion and copy eigen values 
-        
-    // back from Fortran to C++
-    a++;
-        
-    for( i=0; i<n; i++ ) {
-        k = i + (i*(i+1))/2;
-        eigen_val[i] = a[k];
-    }
-     
-    delete[] a;
-        
-    // Step 6: sort the eigen values and eigen vectors 
-        
-    index = new int[n];
-    for( i=0; i<n; i++ ) {
-        index[i] = i;
-    }
-        
-    for( i=0; i<(n-1); i++ ) {
-        x = eigen_val[i];
-        k = i;
-            
-        for( j=i+1; j<n; j++ ) {
-            if( x < eigen_val[j] ) {
-                k = j;
-                x = eigen_val[j];
-            }
-        }
-            
-        eigen_val[k] = eigen_val[i];
-        eigen_val[i] = x;
-          
-        jj       = index[k];
-        index[k] = index[i];
-        index[i] = jj;
-    }
-
-
-    // Step 7: save the eigen vectors 
-    
-    v++; // back from Fortran to to C++
-        
-    ij = 0;
-    for( k=0; k<n; k++ ) {
-        ik = index[k]*n;
-        for( i=0; i<n; i++ ) {
-            eigen_vec[ij++] = v[ik++];
-        }
-    }
-    
-    delete[] v    ;
-    delete[] index;
-    return;
-}
-    
-//_________________________________________________________
-
-
-// Eric Lengyel code:
-// http://www.terathon.com/code/linear.html
-#if 0 
-
-const float epsilon = 1.0e-10F;
-const int maxSweeps = 32;
-
-
-struct Matrix3D
-{
-	float n[3][3];
-	
-	float& operator()(int i, int j)
-	{
-		return (n[j][i]);
-	}
-	
-	const float& operator()(int i, int j) const
-	{
-		return (n[j][i]);
-	}
-	
-	void SetIdentity(void)
-	{
-		n[0][0] = n[1][1] = n[2][2] = 1.0F;
-		n[0][1] = n[0][2] = n[1][0] = n[1][2] = n[2][0] = n[2][1] = 0.0F;
-	}
-};
-
-
-void CalculateEigensystem(const Matrix3D& m, float *lambda, Matrix3D& r)
-{
-	float m11 = m(0,0);
-	float m12 = m(0,1);
-	float m13 = m(0,2);
-	float m22 = m(1,1);
-	float m23 = m(1,2);
-	float m33 = m(2,2);
-	
-	r.SetIdentity();
-	for (int a = 0; a < maxSweeps; a++)
-	{
-		// Exit if off-diagonal entries small enough
-		if ((Fabs(m12) < epsilon) && (Fabs(m13) < epsilon) &&
-			(Fabs(m23) < epsilon)) break;
-		
-		// Annihilate (1,2) entry
-		if (m12 != 0.0F)
-		{
-			float u = (m22 - m11) * 0.5F / m12;
-			float u2 = u * u;
-			float u2p1 = u2 + 1.0F;
-			float t = (u2p1 != u2) ?
-					((u < 0.0F) ? -1.0F : 1.0F) * (sqrt(u2p1) - fabs(u)) : 0.5F / u;
-			float c = 1.0F / sqrt(t * t + 1.0F);
-			float s = c * t;
-			
-			m11 -= t * m12;
-			m22 += t * m12;
-			m12 = 0.0F;
-			
-			float temp = c * m13 - s * m23;
-			m23 = s * m13 + c * m23;
-			m13 = temp;
-			
-			for (int i = 0; i < 3; i++)
-			{
-				float temp = c * r(i,0) - s * r(i,1);
-				r(i,1) = s * r(i,0) + c * r(i,1);
-				r(i,0) = temp;
-			}
-		}
-		
-		// Annihilate (1,3) entry
-		if (m13 != 0.0F)
-		{
-			float u = (m33 - m11) * 0.5F / m13;
-			float u2 = u * u;
-			float u2p1 = u2 + 1.0F;
-			float t = (u2p1 != u2) ?
-				((u < 0.0F) ? -1.0F : 1.0F) * (sqrt(u2p1) - fabs(u)) : 0.5F / u;
-			float c = 1.0F / sqrt(t * t + 1.0F);
-			float s = c * t;
-			
-			m11 -= t * m13;
-			m33 += t * m13;
-			m13 = 0.0F;
-			
-			float temp = c * m12 - s * m23;
-			m23 = s * m12 + c * m23;
-			m12 = temp;
-			
-			for (int i = 0; i < 3; i++)
-			{
-				float temp = c * r(i,0) - s * r(i,2);
-				r(i,2) = s * r(i,0) + c * r(i,2);
-				r(i,0) = temp;
-			}
-		}
-		
-		// Annihilate (2,3) entry
-		if (m23 != 0.0F)
-		{
-			float u = (m33 - m22) * 0.5F / m23;
-			float u2 = u * u;
-			float u2p1 = u2 + 1.0F;
-			float t = (u2p1 != u2) ?
-				((u < 0.0F) ? -1.0F : 1.0F) * (sqrt(u2p1) - fabs(u)) : 0.5F / u;
-			float c = 1.0F / sqrt(t * t + 1.0F);
-			float s = c * t;
-			
-			m22 -= t * m23;
-			m33 += t * m23;
-			m23 = 0.0F;
-			
-			float temp = c * m12 - s * m13;
-			m13 = s * m12 + c * m13;
-			m12 = temp;
-			
-			for (int i = 0; i < 3; i++)
-			{
-				float temp = c * r(i,1) - s * r(i,2);
-				r(i,2) = s * r(i,1) + c * r(i,2);
-				r(i,1) = temp;
-			}
-		}
-	}
-	
-	lambda[0] = m11;
-	lambda[1] = m22;
-	lambda[2] = m33;
-}
-
-
-#endif

src/nvmath/Eigen.h

@@ -1,140 +0,0 @@
-// This code is in the public domain -- castanyo@yahoo.es
-
-#ifndef NV_MATH_EIGEN_H
-#define NV_MATH_EIGEN_H
-
-#include <nvcore/Containers.h> // swap
-#include <nvmath/nvmath.h>
-#include <nvmath/Vector.h>
-
-namespace nv
-{
-
-	// Compute first eigen vector using the power method.
-	Vector3 firstEigenVector(float matrix[6]);
-	
-	/// Generic eigen-solver.
-	class Eigen
-	{
-	public:
-		
-		/// Ctor.
-		Eigen(uint n) : N(n)
-		{
-			uint size = n * (n + 1) / 2;
-			matrix = new float[size];
-			eigen_vec = new float[N*N];
-			eigen_val = new float[N];
-		}
-		
-		/// Dtor.
-		~Eigen()
-		{
-			delete [] matrix;
-			delete [] eigen_vec;
-			delete [] eigen_val;
-		}
-		
-		NVMATH_API void solve();
-		
-		/// Matrix accesor.
-		float & operator()(uint x, uint y)
-		{
-			if( x > y ) {
-				swap(x, y);
-			}
-			return matrix[y * (y + 1) / 2 + x];
-		}
-		
-		/// Matrix const accessor.
-		float operator()(uint x, uint y) const
-		{
-			if( x > y ) {
-				swap(x, y);
-			}
-			return matrix[y * (y + 1) / 2 + x];
-		}
-		
-		Vector3 eigenVector3(uint i) const
-		{
-			nvCheck(3 == N);
-			nvCheck(i < N);
-			return Vector3(eigen_vec[i*N+0], eigen_vec[i*N+1], eigen_vec[i*N+2]);
-		}
-		
-		Vector4 eigenVector4(uint i) const
-		{
-			nvCheck(4 == N);
-			nvCheck(i < N);
-			return Vector4(eigen_vec[i*N+0], eigen_vec[i*N+1], eigen_vec[i*N+2], eigen_vec[i*N+3]);
-		}
-		
-		float eigenValue(uint i) const
-		{
-			nvCheck(i < N);
-			return eigen_val[i];
-		}
-	
-	private:
-		const uint N;
-		float * matrix;
-		float * eigen_vec;
-		float * eigen_val;
-	};
-	
-	
-	/// 3x3 eigen-solver. 
-	/// Based on Eric Lengyel's code:
-	/// http://www.terathon.com/code/linear.html
-	class Eigen3
-	{
-	public:
-		
-		/** Ctor. */
-		Eigen3() {}
-		
-		NVMATH_API void solve();
-		
-		/// Matrix accesor.
-		float & operator()(uint x, uint y)
-		{
-			nvDebugCheck( x < 3 && y < 3 );
-			if( x > y ) {
-				swap(x, y);
-			}
-			return matrix[y * (y + 1) / 2 + x];
-		}
-		
-		/// Matrix const accessor.
-		float operator()(uint x, uint y) const
-		{
-			nvDebugCheck( x < 3 && y < 3 );
-			if( x > y ) {
-				swap(x, y);
-			}
-			return matrix[y * (y + 1) / 2 + x];
-		}
-		
-		/// Get ith eigen vector.
-		Vector3 eigenVector(uint i) const
-		{
-			nvCheck(i < 3);
-			return eigen_vec[i];
-		}
-		
-		/** Get ith eigen value. */
-		float eigenValue(uint i) const
-		{
-			nvCheck(i < 3);
-			return eigen_val[i];
-		}
-	
-	private:
-		float matrix[3+2+1];
-		Vector3 eigen_vec[3];
-		float eigen_val[3];
-	};
-
-} // nv namespace
-
-#endif // NV_MATH_EIGEN_H

src/nvmath/Fitting.cpp

@@ -1,134 +0,0 @@
-// License: Wild Magic License Version 3
-// http://geometrictools.com/License/WildMagic3License.pdf
-
-#include "Fitting.h"
-#include "Eigen.h"
-
-using namespace nv;
-
-
-/** Fit a 3d line to the given set of points. 
- *
- * Based on code from:
- * http://geometrictools.com/
- */
-Line3 Fit::bestLine(const Array<Vector3> & pointArray)
-{
-	nvDebugCheck(pointArray.count() > 0);
-	
-	Line3 line;
-	
-	const uint pointCount = pointArray.count();
-	const float inv_num = 1.0f / pointCount;
-
-	// compute the mean of the points
-	Vector3 center(zero);
-	for(uint i = 0; i < pointCount; i++) {
-		center += pointArray[i];
-	}
-	line.setOrigin(center * inv_num);
-
-	// compute the covariance matrix of the points
-	float covariance[6] = {0, 0, 0, 0, 0, 0};
-	for(uint i = 0; i < pointCount; i++) {
-		Vector3 diff = pointArray[i] - line.origin();
-		covariance[0] += diff.x() * diff.x();
-		covariance[1] += diff.x() * diff.y();
-		covariance[2] += diff.x() * diff.z();
-		covariance[3] += diff.y() * diff.y();
-		covariance[4] += diff.y() * diff.z();
-		covariance[5] += diff.z() * diff.z();
-	}
-	
-	line.setDirection(normalizeSafe(firstEigenVector(covariance), Vector3(zero), 0.0f));
-	
-	// @@ This variant is from David Eberly... I'm not sure how that works.
-	/*sum_xx *= inv_num;
-	sum_xy *= inv_num;
-	sum_xz *= inv_num;
-	sum_yy *= inv_num;
-	sum_yz *= inv_num;
-	sum_zz *= inv_num;
-
-    // set up the eigensolver
-	Eigen3 ES;
-	ES(0,0) = sum_yy + sum_zz;
-	ES(0,1) = -sum_xy;
-	ES(0,2) = -sum_xz;
-	ES(1,1) = sum_xx + sum_zz;
-	ES(1,2) = -sum_yz;
-	ES(2,2) = sum_xx + sum_yy;
-
-    // compute eigenstuff, smallest eigenvalue is in last position
-	ES.solve();
-
-	line.setDirection(ES.eigenVector(2));
-	
-	nvCheck( isNormalized(line.direction()) );
-	*/
-	return line;
-}
-
-
-/** Fit a 3d plane to the given set of points. 
- *
- * Based on code from:
- * http://geometrictools.com/
- */
-Vector4 Fit::bestPlane(const Array<Vector3> & pointArray)
-{
-	Vector3 center(zero);
-	
-	const uint pointCount = pointArray.count();
-	const float inv_num = 1.0f / pointCount;
-
-    // compute the mean of the points
-	for(uint i = 0; i < pointCount; i++) {
-		center += pointArray[i];
-	}
-	center *= inv_num;
-
-    // compute the covariance matrix of the points
-	float sum_xx = 0.0f;
-	float sum_xy = 0.0f;
-	float sum_xz = 0.0f;
-	float sum_yy = 0.0f;
-	float sum_yz = 0.0f;
-	float sum_zz = 0.0f;
-	
-	for(uint i = 0; i < pointCount; i++) {
-		Vector3 diff = pointArray[i] - center;
-		sum_xx += diff.x() * diff.x();
-		sum_xy += diff.x() * diff.y();
-		sum_xz += diff.x() * diff.z();
-		sum_yy += diff.y() * diff.y();
-		sum_yz += diff.y() * diff.z();
-		sum_zz += diff.z() * diff.z();
-	}
-	
-	sum_xx *= inv_num;
-	sum_xy *= inv_num;
-	sum_xz *= inv_num;
-	sum_yy *= inv_num;
-	sum_yz *= inv_num;
-	sum_zz *= inv_num;
-
-    // set up the eigensolver
-	Eigen3 ES;
-	ES(0,0) = sum_yy + sum_zz;
-	ES(0,1) = -sum_xy;
-	ES(0,2) = -sum_xz;
-	ES(1,1) = sum_xx + sum_zz;
-	ES(1,2) = -sum_yz;
-	ES(2,2) = sum_xx + sum_yy;
-
-    // compute eigenstuff, greatest eigenvalue is in first position
-	ES.solve();
-
-	Vector3 normal = ES.eigenVector(0);
-	nvCheck(isNormalized(normal));
-	
-	float offset = dot(normal, center);
-
-	return Vector4(normal, offset);
-}

src/nvmath/Fitting.h

@@ -1,78 +0,0 @@
-// This code is in the public domain -- castanyo@yahoo.es
-
-#ifndef NV_MATH_FITTING_H
-#define NV_MATH_FITTING_H
-
-#include <nvmath/Vector.h>
-
-namespace nv
-{
-
-	/// 3D Line.
-	struct Line3
-	{
-		/// Ctor.
-		Line3() : m_origin(zero), m_direction(zero)
-		{
-		}
-		
-		/// Copy ctor.
-		Line3(const Line3 & l) : m_origin(l.m_origin), m_direction(l.m_direction)
-		{
-		}
-		
-		/// Ctor.
-		Line3(Vector3::Arg o, Vector3::Arg d) : m_origin(o), m_direction(d)
-		{
-		}
-	
-		/// Normalize the line.
-		void normalize()
-		{
-			m_direction = nv::normalize(m_direction);
-		}
-		
-		/// Project a point onto the line.
-		Vector3 projectPoint(Vector3::Arg point) const
-		{
-			nvDebugCheck(isNormalized(m_direction));
-			
-			Vector3 v = point - m_origin;
-			return m_origin + m_direction * dot(m_direction, v);
-		}
-		
-		/// Compute distance to line.
-		float distanceToPoint(Vector3::Arg point) const
-		{
-			nvDebugCheck(isNormalized(m_direction));
-			
-			Vector3 v = point - m_origin;
-			Vector3 l = v - m_direction * dot(m_direction, v);
-			
-			return length(l);
-		}
-		
-		const Vector3 & origin() const { return m_origin; }
-		void setOrigin(Vector3::Arg value) { m_origin = value; }
-	
-		const Vector3 & direction() const { return m_direction; }
-		void setDirection(Vector3::Arg value) { m_direction = value; }
-		
-		
-	private:	
-		Vector3 m_origin;
-		Vector3 m_direction;
-	};
-
-	
-	namespace Fit
-	{
-		
-		NVMATH_API Line3 bestLine(const Array<Vector3> & pointArray);
-		NVMATH_API Vector4 bestPlane(const Array<Vector3> & pointArray);
-		
-	} // Fit namespace
-
-} // nv namespace
-
-#endif // _PI_MATHLIB_FITTING_H_

src/nvmath/Plane.cpp

@@ -0,0 +1,17 @@
+// This code is in the public domain -- castanyo@yahoo.es
+
+#include "Plane.h"
+#include "Matrix.h"
+
+namespace nv
+{
+	Plane transformPlane(const Matrix& m, Plane::Arg p)
+	{
+		Vector3 newVec = transformVector(m, p.vector());
+
+		Vector3 ptInPlane = p.offset() * p.vector();
+		ptInPlane = transformPoint(m, ptInPlane);
+
+		return Plane(newVec, ptInPlane);
+	}
+}

src/nvmath/Plane.h

@@ -0,0 +1,77 @@
+// This code is in the public domain -- castanyo@yahoo.es
+
+#ifndef NV_MATH_PLANE_H
+#define NV_MATH_PLANE_H
+
+#include <nvmath/nvmath.h>
+#include <nvmath/Vector.h>
+
+namespace nv
+{
+	class Matrix;
+
+
+	class NVMATH_CLASS Plane
+	{
+	public:
+		typedef Plane const & Arg;
+		
+		Plane();
+		Plane(float x, float y, float z, float w);
+		Plane(Vector4::Arg v);
+		Plane(Vector3::Arg v, float d);
+		Plane(Vector3::Arg normal, Vector3::Arg point);
+		
+		const Plane & operator=(Plane::Arg v);
+		
+		Vector3 vector() const;
+		scalar offset() const;
+		
+		const Vector4 & asVector() const;
+		Vector4 & asVector();
+		
+		void operator*=(scalar s);
+
+	private:
+		Vector4 p;
+	};
+
+	inline Plane::Plane() {}
+	inline Plane::Plane(float x, float y, float z, float w) : p(x, y, z, w) {}
+	inline Plane::Plane(Vector4::Arg v) : p(v) {}
+	inline Plane::Plane(Vector3::Arg v, float d) : p(v, d) {}
+	inline Plane::Plane(Vector3::Arg normal, Vector3::Arg point) : p(normal, dot(normal, point)) {}
+	
+	inline const Plane & Plane::operator=(Plane::Arg v) { p = v.p; return *this; }
+	
+	inline Vector3 Plane::vector() const { return p.xyz(); }
+	inline scalar Plane::offset() const { return p.w(); }
+
+	inline const Vector4 & Plane::asVector() const { return p; }
+	inline Vector4 & Plane::asVector() { return p; }
+
+	// Normalize plane.
+	inline Plane normalize(Plane::Arg plane, float epsilon = NV_EPSILON)
+	{
+		const float len = length(plane.vector());
+		nvDebugCheck(!isZero(len, epsilon));
+		const float inv = 1.0f / len;
+		return Plane(plane.asVector() * inv);
+	}
+
+	// Get the distance from the given point to this plane.
+	inline float distance(Plane::Arg plane, Vector3::Arg point)
+	{
+		return dot(plane.vector(), point) - plane.offset();
+	}
+
+	inline void Plane::operator*=(scalar s)
+	{
+		scale(p, s);
+	}
+
+	Plane transformPlane(const Matrix&, Plane::Arg);
+	
+} // nv namespace
+
+#endif // NV_MATH_PLANE_H

src/nvtt/CMakeLists.txt

@@ -13,10 +13,10 @@ SET(NVTT_SRCS
 	CompressDXT.cpp
 	CompressRGB.h
 	CompressRGB.cpp
-	FastCompressDXT.h
-	FastCompressDXT.cpp
 	QuickCompressDXT.h
 	QuickCompressDXT.cpp
+	OptimalCompressDXT.h
+	OptimalCompressDXT.cpp
 	SingleColorLookup.h
 	CompressionOptions.h
 	CompressionOptions.cpp

src/nvtt/CompressDXT.cpp

@@ -29,8 +29,8 @@
 
 #include "nvtt.h"
 #include "CompressDXT.h"
-#include "FastCompressDXT.h"
 #include "QuickCompressDXT.h"
+#include "OptimalCompressDXT.h"
 #include "CompressionOptions.h"
 #include "OutputOptions.h"
 
@@ -57,17 +57,31 @@ using namespace nv;
 using namespace nvtt;
 
 
-void nv::fastCompressDXT1(const Image * image, const OutputOptions::Private & outputOptions)
+nv::FastCompressor::FastCompressor() : m_image(NULL), m_alphaMode(AlphaMode_None)
 {
-	const uint w = image->width();
-	const uint h = image->height();
+}
+
+nv::FastCompressor::~FastCompressor()
+{
+}
+
+void nv::FastCompressor::setImage(const Image * image, nvtt::AlphaMode alphaMode)
+{
+	m_image = image;
+	m_alphaMode = alphaMode;
+}
+
+void nv::FastCompressor::compressDXT1(const OutputOptions::Private & outputOptions)
+{
+	const uint w = m_image->width();
+	const uint h = m_image->height();
 	
 	ColorBlock rgba;
 	BlockDXT1 block;
 
 	for (uint y = 0; y < h; y += 4) {
 		for (uint x = 0; x < w; x += 4) {
-			rgba.init(image, x, y);
+			rgba.init(m_image, x, y);
 			
 			QuickCompress::compressDXT1(rgba, &block);
 			
@@ -79,17 +93,17 @@ void nv::fastCompressDXT1(const Image * image, const OutputOptions::Private & ou
 }
 
 
-void nv::fastCompressDXT1a(const Image * image, const OutputOptions::Private & outputOptions)
+void nv::FastCompressor::compressDXT1a(const OutputOptions::Private & outputOptions)
 {
-	const uint w = image->width();
-	const uint h = image->height();
+	const uint w = m_image->width();
+	const uint h = m_image->height();
 	
 	ColorBlock rgba;
 	BlockDXT1 block;
 
 	for (uint y = 0; y < h; y += 4) {
 		for (uint x = 0; x < w; x += 4) {
-			rgba.init(image, x, y);
+			rgba.init(m_image, x, y);
 			
 			QuickCompress::compressDXT1a(rgba, &block);
 			
@@ -101,18 +115,19 @@ void nv::fastCompressDXT1a(const Image * image, const OutputOptions::Private & o
 }
 
 
-void nv::fastCompressDXT3(const Image * image, const nvtt::OutputOptions::Private & outputOptions)
+void nv::FastCompressor::compressDXT3(const nvtt::OutputOptions::Private & outputOptions)
 {
-	const uint w = image->width();
-	const uint h = image->height();
+	const uint w = m_image->width();
+	const uint h = m_image->height();
 	
 	ColorBlock rgba;
 	BlockDXT3 block;
 
 	for (uint y = 0; y < h; y += 4) {
 		for (uint x = 0; x < w; x += 4) {
-			rgba.init(image, x, y);
-			compressBlock_BoundsRange(rgba, &block);
+			rgba.init(m_image, x, y);
+
+			QuickCompress::compressDXT3(rgba, &block);
 			
 			if (outputOptions.outputHandler != NULL) {
 				outputOptions.outputHandler->writeData(&block, sizeof(block));
@@ -122,18 +137,19 @@ void nv::fastCompressDXT3(const Image * image, const nvtt::OutputOptions::Privat
 }
 
 
-void nv::fastCompressDXT5(const Image * image, const nvtt::OutputOptions::Private & outputOptions)
+void nv::FastCompressor::compressDXT5(const nvtt::OutputOptions::Private & outputOptions)
 {
-	const uint w = image->width();
-	const uint h = image->height();
+	const uint w = m_image->width();
+	const uint h = m_image->height();
 	
 	ColorBlock rgba;
 	BlockDXT5 block;
 
 	for (uint y = 0; y < h; y += 4) {
 		for (uint x = 0; x < w; x += 4) {
-			rgba.init(image, x, y);
-			compressBlock_BoundsRange(rgba, &block);
+			rgba.init(m_image, x, y);
+			
+			QuickCompress::compressDXT5(rgba, &block, 0);
 			
 			if (outputOptions.outputHandler != NULL) {
 				outputOptions.outputHandler->writeData(&block, sizeof(block));
@@ -143,22 +159,21 @@ void nv::fastCompressDXT5(const Image * image, const nvtt::OutputOptions::Privat
 }
 
 
-void nv::fastCompressDXT5n(const Image * image, const nvtt::OutputOptions::Private & outputOptions)
+void nv::FastCompressor::compressDXT5n(const nvtt::OutputOptions::Private & outputOptions)
 {
-	const uint w = image->width();
-	const uint h = image->height();
+	const uint w = m_image->width();
+	const uint h = m_image->height();
 	
 	ColorBlock rgba;
 	BlockDXT5 block;
 
 	for (uint y = 0; y < h; y += 4) {
 		for (uint x = 0; x < w; x += 4) {
-			rgba.init(image, x, y);
+			rgba.init(m_image, x, y);
 			
-			// copy X coordinate to alpha channel and Y coordinate to green channel.
 			rgba.swizzleDXT5n();
 
-			compressBlock_BoundsRange(rgba, &block);
+			QuickCompress::compressDXT5(rgba, &block, 0);
 			
 			if (outputOptions.outputHandler != NULL) {
 				outputOptions.outputHandler->writeData(&block, sizeof(block));
@@ -168,42 +183,28 @@ void nv::fastCompressDXT5n(const Image * image, const nvtt::OutputOptions::Priva
 }
 
 
-void nv::fastCompressBC4(const Image * image, const nvtt::OutputOptions::Private & outputOptions)
+nv::SlowCompressor::SlowCompressor() : m_image(NULL), m_alphaMode(AlphaMode_None)
 {
-	// @@ TODO
-	// compress red channel (X)
 }
 
-
-void nv::fastCompressBC5(const Image * image, const nvtt::OutputOptions::Private & outputOptions)
+nv::SlowCompressor::~SlowCompressor()
 {
-	// @@ TODO
-	// compress red, green channels (X,Y)
 }
 
-
-void nv::doPrecomputation()
+void nv::SlowCompressor::setImage(const Image * image, nvtt::AlphaMode alphaMode)
 {
-	static bool done = false;	// @@ Stop using statics for reentrancy.
-	
-	if (!done)
-	{
-		done = true;
-		squish::FastClusterFit::DoPrecomputation();
-	}
+	m_image = image;
+	m_alphaMode = alphaMode;
 }
 
-
-void nv::compressDXT1(const Image * image, const OutputOptions::Private & outputOptions, const CompressionOptions::Private & compressionOptions)
+void nv::SlowCompressor::compressDXT1(const CompressionOptions::Private & compressionOptions, const OutputOptions::Private & outputOptions)
 {
-	const uint w = image->width();
-	const uint h = image->height();
+	const uint w = m_image->width();
+	const uint h = m_image->height();
 	
 	ColorBlock rgba;
 	BlockDXT1 block;
 
-	doPrecomputation();
-
 	//squish::WeightedClusterFit fit;
 	//squish::ClusterFit fit;
 	squish::FastClusterFit fit;
@@ -212,12 +213,18 @@ void nv::compressDXT1(const Image * image, const OutputOptions::Private & output
 	for (uint y = 0; y < h; y += 4) {
 		for (uint x = 0; x < w; x += 4) {
 			
-			rgba.init(image, x, y);
+			rgba.init(m_image, x, y);
 			
-			// Compress color.
-			squish::ColourSet colours((uint8 *)rgba.colors(), 0);
-			fit.SetColourSet(&colours, squish::kDxt1);
-			fit.Compress(&block);
+			if (rgba.isSingleColor())
+			{
+				OptimalCompress::compressDXT1(rgba.color(0), &block);
+			}
+			else
+			{
+				squish::ColourSet colours((uint8 *)rgba.colors(), 0);
+				fit.SetColourSet(&colours, squish::kDxt1);
+				fit.Compress(&block);
+			}
 			
 			if (outputOptions.outputHandler != NULL) {
 				outputOptions.outputHandler->writeData(&block, sizeof(block));
@@ -227,10 +234,10 @@ void nv::compressDXT1(const Image * image, const OutputOptions::Private & output
 }
 
 
-void nv::compressDXT1a(const Image * image, const OutputOptions::Private & outputOptions, const CompressionOptions::Private & compressionOptions)
+void nv::SlowCompressor::compressDXT1a(const CompressionOptions::Private & compressionOptions, const OutputOptions::Private & outputOptions)
 {
-	const uint w = image->width();
-	const uint h = image->height();
+	const uint w = m_image->width();
+	const uint h = m_image->height();
 	
 	ColorBlock rgba;
 	BlockDXT1 block;
@@ -241,12 +248,27 @@ void nv::compressDXT1a(const Image * image, const OutputOptions::Private & outpu
 	for (uint y = 0; y < h; y += 4) {
 		for (uint x = 0; x < w; x += 4) {
 			
-			rgba.init(image, x, y);
+			rgba.init(m_image, x, y);
 			
-			// Compress color.
-			squish::ColourSet colours((uint8 *)rgba.colors(), squish::kDxt1|squish::kWeightColourByAlpha);
-			fit.SetColourSet(&colours, squish::kDxt1);
-			fit.Compress(&block);
+			bool anyAlpha = false;
+			bool allAlpha = true;
+			
+			for (uint i = 0; i < 16; i++)
+			{
+				if (rgba.color(i).a < 128) anyAlpha = true;
+				else allAlpha = false;
+			}
+			
+			if ((!anyAlpha && rgba.isSingleColor() || allAlpha))
+			{
+				OptimalCompress::compressDXT1a(rgba.color(0), &block);
+			}
+			else
+			{
+				squish::ColourSet colours((uint8 *)rgba.colors(), squish::kDxt1|squish::kWeightColourByAlpha);
+				fit.SetColourSet(&colours, squish::kDxt1);
+				fit.Compress(&block);
+			}
 			
 			if (outputOptions.outputHandler != NULL) {
 				outputOptions.outputHandler->writeData(&block, sizeof(block));
@@ -256,29 +278,37 @@ void nv::compressDXT1a(const Image * image, const OutputOptions::Private & outpu
 }
 
 
-void nv::compressDXT3(const Image * image, const OutputOptions::Private & outputOptions, const CompressionOptions::Private & compressionOptions)
+void nv::SlowCompressor::compressDXT3(const CompressionOptions::Private & compressionOptions, const OutputOptions::Private & outputOptions)
 {
-	const uint w = image->width();
-	const uint h = image->height();
+	const uint w = m_image->width();
+	const uint h = m_image->height();
 	
 	ColorBlock rgba;
 	BlockDXT3 block;
 	
 	squish::WeightedClusterFit fit;
+	//squish::FastClusterFit fit;
 	fit.SetMetric(compressionOptions.colorWeight.x(), compressionOptions.colorWeight.y(), compressionOptions.colorWeight.z());
 
 	for (uint y = 0; y < h; y += 4) {
 		for (uint x = 0; x < w; x += 4) {
 			
-			rgba.init(image, x, y);
+			rgba.init(m_image, x, y);
 			
 			// Compress explicit alpha.
-			compressBlock(rgba, &block.alpha);
+			OptimalCompress::compressDXT3A(rgba, &block.alpha);
 
 			// Compress color.
-			squish::ColourSet colours((uint8 *)rgba.colors(), squish::kWeightColourByAlpha);
-			fit.SetColourSet(&colours, 0);
-			fit.Compress(&block.color);
+			if (rgba.isSingleColor())
+			{
+				OptimalCompress::compressDXT1(rgba.color(0), &block.color);
+			}
+			else
+			{
+				squish::ColourSet colours((uint8 *)rgba.colors(), squish::kWeightColourByAlpha);
+				fit.SetColourSet(&colours, 0);
+				fit.Compress(&block.color);
+			}
 			
 			if (outputOptions.outputHandler != NULL) {
 				outputOptions.outputHandler->writeData(&block, sizeof(block));
@@ -287,10 +317,10 @@ void nv::compressDXT3(const Image * image, const OutputOptions::Private & output
 	}
 }
 
-void nv::compressDXT5(const Image * image, const OutputOptions::Private & outputOptions, const CompressionOptions::Private & compressionOptions)
+void nv::SlowCompressor::compressDXT5(const CompressionOptions::Private & compressionOptions, const OutputOptions::Private & outputOptions)
 {
-	const uint w = image->width();
-	const uint h = image->height();
+	const uint w = m_image->width();
+	const uint h = m_image->height();
 	
 	ColorBlock rgba;
 	BlockDXT5 block;
@@ -301,23 +331,29 @@ void nv::compressDXT5(const Image * image, const OutputOptions::Private & output
 	for (uint y = 0; y < h; y += 4) {
 		for (uint x = 0; x < w; x += 4) {
 			
-			rgba.init(image, x, y);
+			rgba.init(m_image, x, y);
 
 			// Compress alpha.
-			uint error;
 			if (compressionOptions.quality == Quality_Highest)
 			{
-				error = compressBlock_BruteForce(rgba, &block.alpha);
+				OptimalCompress::compressDXT5A(rgba, &block.alpha);
 			}
 			else
 			{
-				error = compressBlock_Iterative(rgba, &block.alpha);
+				QuickCompress::compressDXT5A(rgba, &block.alpha);
 			}
 		
 			// Compress color.
-			squish::ColourSet colours((uint8 *)rgba.colors(), squish::kWeightColourByAlpha);
-			fit.SetColourSet(&colours, 0);
-			fit.Compress(&block.color);
+			if (rgba.isSingleColor())
+			{
+				OptimalCompress::compressDXT1(rgba.color(0), &block.color);
+			}
+			else
+			{
+				squish::ColourSet colours((uint8 *)rgba.colors(), squish::kWeightColourByAlpha);
+				fit.SetColourSet(&colours, 0);
+				fit.Compress(&block.color);
+			}
 			
 			if (outputOptions.outputHandler != NULL) {
 				outputOptions.outputHandler->writeData(&block, sizeof(block));
@@ -327,33 +363,33 @@ void nv::compressDXT5(const Image * image, const OutputOptions::Private & output
 }
 
 
-void nv::compressDXT5n(const Image * image, const OutputOptions::Private & outputOptions, const CompressionOptions::Private & compressionOptions)
+void nv::SlowCompressor::compressDXT5n(const CompressionOptions::Private & compressionOptions, const OutputOptions::Private & outputOptions)
 {
-	const uint w = image->width();
-	const uint h = image->height();
+	const uint w = m_image->width();
+	const uint h = m_image->height();
 	
 	ColorBlock rgba;
 	BlockDXT5 block;
 	
-	doPrecomputation();
-
 	for (uint y = 0; y < h; y += 4) {
 		for (uint x = 0; x < w; x += 4) {
 			
-			rgba.init(image, x, y);
+			rgba.init(m_image, x, y);
 			
-			// copy X coordinate to green channel and Y coordinate to alpha channel.
 			rgba.swizzleDXT5n();			
 			
 			// Compress X.
-			uint error = compressBlock_Iterative(rgba, &block.alpha);
 			if (compressionOptions.quality == Quality_Highest)
 			{
-				error = compressBlock_BruteForce(rgba, &block.alpha);
+				OptimalCompress::compressDXT5A(rgba, &block.alpha);
+			}
+			else
+			{
+				QuickCompress::compressDXT5A(rgba, &block.alpha);
 			}
 			
 			// Compress Y.
-			QuickCompress::compressDXT1G(rgba, &block.color);
+			OptimalCompress::compressDXT1G(rgba, &block.color);
 			
 			if (outputOptions.outputHandler != NULL) {
 				outputOptions.outputHandler->writeData(&block, sizeof(block));
@@ -363,31 +399,27 @@ void nv::compressDXT5n(const Image * image, const OutputOptions::Private & outpu
 }
 
 
-void nv::compressBC4(const Image * image, const nvtt::OutputOptions::Private & outputOptions, const CompressionOptions::Private & compressionOptions)
+void nv::SlowCompressor::compressBC4(const CompressionOptions::Private & compressionOptions, const nvtt::OutputOptions::Private & outputOptions)
 {
-	const uint w = image->width();
-	const uint h = image->height();
+	const uint w = m_image->width();
+	const uint h = m_image->height();
 	
 	ColorBlock rgba;
 	AlphaBlockDXT5 block;
 	
-	uint totalError = 0;
-	
 	for (uint y = 0; y < h; y += 4) {
 		for (uint x = 0; x < w; x += 4) {
 			
-			rgba.init(image, x, y);
-
-			//error = compressBlock_BoundsRange(rgba, &block);
-			uint error = compressBlock_Iterative(rgba, &block);
+			rgba.init(m_image, x, y);
 
 			if (compressionOptions.quality == Quality_Highest)
 			{
-				// Try brute force algorithm.
-				error = compressBlock_BruteForce(rgba, &block);
+				OptimalCompress::compressDXT5A(rgba, &block);
+			}
+			else
+			{
+				QuickCompress::compressDXT5A(rgba, &block);
 			}
-
-			totalError += error;
 
 			if (outputOptions.outputHandler != NULL) {
 				outputOptions.outputHandler->writeData(&block, sizeof(block));
@@ -397,10 +429,10 @@ void nv::compressBC4(const Image * image, const nvtt::OutputOptions::Private & o
 }
 
 
-void nv::compressBC5(const Image * image, const nvtt::OutputOptions::Private & outputOptions, const CompressionOptions::Private & compressionOptions)
+void nv::SlowCompressor::compressBC5(const CompressionOptions::Private & compressionOptions, const nvtt::OutputOptions::Private & outputOptions)
 {
-	const uint w = image->width();
-	const uint h = image->height();
+	const uint w = m_image->width();
+	const uint h = m_image->height();
 
 	ColorBlock xcolor;
 	ColorBlock ycolor;
@@ -410,24 +442,21 @@ void nv::compressBC5(const Image * image, const nvtt::OutputOptions::Private & o
 	for (uint y = 0; y < h; y += 4) {
 		for (uint x = 0; x < w; x += 4) {
 			
-			xcolor.init(image, x, y);
+			xcolor.init(m_image, x, y);
 			xcolor.splatX();
 			
-			ycolor.init(image, x, y);
+			ycolor.init(m_image, x, y);
 			ycolor.splatY();
 
-			// @@ Compute normal error, instead of separate xy errors.
-			uint xerror, yerror;
-			
 			if (compressionOptions.quality == Quality_Highest)
 			{
-				xerror = compressBlock_BruteForce(xcolor, &block.x);
-				yerror = compressBlock_BruteForce(ycolor, &block.y);
+				OptimalCompress::compressDXT5A(xcolor, &block.x);
+				OptimalCompress::compressDXT5A(ycolor, &block.y);
 			}
 			else
 			{
-				xerror = compressBlock_Iterative(xcolor, &block.x);
-				yerror = compressBlock_Iterative(ycolor, &block.y);
+				QuickCompress::compressDXT5A(xcolor, &block.x);
+				QuickCompress::compressDXT5A(ycolor, &block.y);
 			}
 
 			if (outputOptions.outputHandler != NULL) {

src/nvtt/CompressDXT.h

@@ -32,25 +32,45 @@ namespace nv
 	class Image;
 	class FloatImage;
 
-	void doPrecomputation();
+	class FastCompressor
+	{
+	public:
+		FastCompressor();
+		~FastCompressor();
 
-	// Fast compressors.
-	void fastCompressDXT1(const Image * image, const nvtt::OutputOptions::Private & outputOptions);
-	void fastCompressDXT1a(const Image * image, const nvtt::OutputOptions::Private & outputOptions);
-	void fastCompressDXT3(const Image * image, const nvtt::OutputOptions::Private & outputOptions);
-	void fastCompressDXT5(const Image * image, const nvtt::OutputOptions::Private & outputOptions);
-	void fastCompressDXT5n(const Image * image, const nvtt::OutputOptions::Private & outputOptions);
-	void fastCompressBC4(const Image * image, const nvtt::OutputOptions::Private & outputOptions);
-	void fastCompressBC5(const Image * image, const nvtt::OutputOptions::Private & outputOptions);
+		void setImage(const Image * image, nvtt::AlphaMode alphaMode);
 
-	// Normal compressors.
-	void compressDXT1(const Image * image, const nvtt::OutputOptions::Private & outputOptions, const nvtt::CompressionOptions::Private & compressionOptions);
-	void compressDXT1a(const Image * image, const nvtt::OutputOptions::Private & outputOptions, const nvtt::CompressionOptions::Private & compressionOptions);
-	void compressDXT3(const Image * image, const nvtt::OutputOptions::Private & outputOptions, const nvtt::CompressionOptions::Private & compressionOptions);
-	void compressDXT5(const Image * image, const nvtt::OutputOptions::Private & outputOptions, const nvtt::CompressionOptions::Private & compressionOptions);
-	void compressDXT5n(const Image * image, const nvtt::OutputOptions::Private & outputOptions, const nvtt::CompressionOptions::Private & compressionOptions);
-	void compressBC4(const Image * image, const nvtt::OutputOptions::Private & outputOptions, const nvtt::CompressionOptions::Private & compressionOptions);
-	void compressBC5(const Image * image, const nvtt::OutputOptions::Private & outputOptions, const nvtt::CompressionOptions::Private & compressionOptions);
+		void compressDXT1(const nvtt::OutputOptions::Private & outputOptions);
+		void compressDXT1a(const nvtt::OutputOptions::Private & outputOptions);
+		void compressDXT3(const nvtt::OutputOptions::Private & outputOptions);
+		void compressDXT5(const nvtt::OutputOptions::Private & outputOptions);
+		void compressDXT5n(const nvtt::OutputOptions::Private & outputOptions);
+
+	private:
+		const Image * m_image;
+		nvtt::AlphaMode m_alphaMode;
+	};
+
+	class SlowCompressor
+	{
+	public:
+		SlowCompressor();
+		~SlowCompressor();
+
+		void setImage(const Image * image, nvtt::AlphaMode alphaMode);
+
+		void compressDXT1(const nvtt::CompressionOptions::Private & compressionOptions, const nvtt::OutputOptions::Private & outputOptions);
+		void compressDXT1a(const nvtt::CompressionOptions::Private & compressionOptions, const nvtt::OutputOptions::Private & outputOptions);
+		void compressDXT3(const nvtt::CompressionOptions::Private & compressionOptions, const nvtt::OutputOptions::Private & outputOptions);
+		void compressDXT5(const nvtt::CompressionOptions::Private & compressionOptions, const nvtt::OutputOptions::Private & outputOptions);
+		void compressDXT5n(const nvtt::CompressionOptions::Private & compressionOptions, const nvtt::OutputOptions::Private & outputOptions);
+		void compressBC4(const nvtt::CompressionOptions::Private & compressionOptions, const nvtt::OutputOptions::Private & outputOptions);
+		void compressBC5(const nvtt::CompressionOptions::Private & compressionOptions, const nvtt::OutputOptions::Private & outputOptions);
+
+	private:
+		const Image * m_image;
+		nvtt::AlphaMode m_alphaMode;
+	};
 
 	// External compressors.
 #if defined(HAVE_S3QUANT)

src/nvtt/CompressRGB.cpp

@@ -115,12 +115,18 @@ void nv::compressRGB(const Image * image, const OutputOptions::Private & outputO
 				c |= PixelFormat::convert(src[x].b, 8, bsize) << bshift;
 				c |= PixelFormat::convert(src[x].a, 8, asize) << ashift;
 				
-				// Output one byte at a time. @@ Not tested... Does this work on LE and BE?
+				// Output one byte at a time.
 				for (uint i = 0; i < byteCount; i++)
 				{
-					*(dst + x * byteCount) = (c >> (i * 8)) & 0xFF;
+					*(dst + x * byteCount + i) = (c >> (i * 8)) & 0xFF;
 				}
 			}
+			
+			// Zero padding.
+			for (uint x = w * byteCount; x < pitch; x++)
+			{
+				*(dst + x) = 0;
+			}
 		}
 
 		if (outputOptions.outputHandler != NULL)

src/nvtt/Compressor.cpp

@@ -34,6 +34,7 @@
 #include <nvimage/Filter.h>
 #include <nvimage/Quantize.h>
 #include <nvimage/NormalMap.h>
+#include <nvimage/PixelFormat.h>
 
 #include "Compressor.h"
 #include "InputOptions.h"
@@ -41,7 +42,6 @@
 #include "OutputOptions.h"
 
 #include "CompressDXT.h"
-#include "FastCompressDXT.h"
 #include "CompressRGB.h"
 #include "cuda/CudaUtils.h"
 #include "cuda/CudaCompressDXT.h"
@@ -200,21 +200,30 @@ namespace nvtt
 		AutoPtr<FloatImage> m_floatImage;
 	};
 
-}
+} // nvtt namespace
 
 
 Compressor::Compressor() : m(*new Compressor::Private())
 {
+	// CUDA initialization.
 	m.cudaSupported = cuda::isHardwarePresent();
 	m.cudaEnabled = m.cudaSupported;
 
-	// @@ Do CUDA initialization here.
+	if (m.cudaEnabled)
+	{
+		m.cuda = new CudaCompressor();
 
+		if (!m.cuda->isValid())
+		{
+			m.cudaEnabled = false;
+			m.cuda = NULL;
+		}
+	}
 }
 
 Compressor::~Compressor()
 {
-	// @@ Free CUDA resources here.
+	delete &m;
 }
 
 
@@ -225,6 +234,17 @@ void Compressor::enableCudaAcceleration(bool enable)
 	{
 		m.cudaEnabled = enable;
 	}
+
+	if (m.cudaEnabled && m.cuda == NULL)
+	{
+		m.cuda = new CudaCompressor();
+		
+		if (!m.cuda->isValid())
+		{
+			m.cudaEnabled = false;
+			m.cuda = NULL;
+		}
+	}
 }
 
 /// Check if CUDA acceleration is enabled.
@@ -318,7 +338,7 @@ bool Compressor::Private::outputHeader(const InputOptions::Private & inputOption
 	
 	if (compressionOptions.format == Format_RGBA)
 	{
-		header.setPitch(4 * inputOptions.targetWidth);
+		header.setPitch(computePitch(inputOptions.targetWidth, compressionOptions.bitcount));
 		header.setPixelFormat(compressionOptions.bitcount, compressionOptions.rmask, compressionOptions.gmask, compressionOptions.bmask, compressionOptions.amask);
 	}
 	else
@@ -402,7 +422,7 @@ bool Compressor::Private::compressMipmaps(uint f, const InputOptions::Private &
 		
 		quantizeMipmap(mipmap, compressionOptions);
 
-		compressMipmap(mipmap, compressionOptions, outputOptions);
+		compressMipmap(mipmap, inputOptions, compressionOptions, outputOptions);
 
 		// Compute extents of next mipmap:
 		w = max(1U, w / 2);
@@ -551,7 +571,7 @@ void Compressor::Private::scaleMipmap(Mipmap & mipmap, const InputOptions::Priva
 
 	// Resize image. 
 	BoxFilter boxFilter;
-	mipmap.setImage(mipmap.asFloatImage()->downSample(boxFilter, w, h, (FloatImage::WrapMode)inputOptions.wrapMode));
+	mipmap.setImage(mipmap.asFloatImage()->resize(boxFilter, w, h, (FloatImage::WrapMode)inputOptions.wrapMode));
 }
 
 
@@ -598,13 +618,6 @@ void Compressor::Private::quantizeMipmap(Mipmap & mipmap, const CompressionOptio
 {
 	nvDebugCheck(mipmap.asFixedImage() != NULL);
 
-	if (compressionOptions.enableColorDithering)
-	{
-		if (compressionOptions.format >= Format_DXT1 && compressionOptions.format <= Format_DXT5)
-		{
-			Quantize::FloydSteinberg_RGB16(mipmap.asMutableFixedImage());
-		}
-	}
 	if (compressionOptions.binaryAlpha)
 	{
 		if (compressionOptions.enableAlphaDithering)
@@ -616,30 +629,67 @@ void Compressor::Private::quantizeMipmap(Mipmap & mipmap, const CompressionOptio
 			Quantize::BinaryAlpha(mipmap.asMutableFixedImage(), compressionOptions.alphaThreshold);
 		}
 	}
-	else
+
+	if (compressionOptions.enableColorDithering || compressionOptions.enableAlphaDithering)
 	{
+		uint rsize = 8;
+		uint gsize = 8;
+		uint bsize = 8;
+		uint asize = 8;
+
+		if (compressionOptions.enableColorDithering)
+		{
+			if (compressionOptions.format >= Format_DXT1 && compressionOptions.format <= Format_DXT5)
+			{
+				rsize = 5;
+				gsize = 6;
+				bsize = 5;
+			}
+			else if (compressionOptions.format == Format_RGB)
+			{
+				uint rshift, gshift, bshift;
+				PixelFormat::maskShiftAndSize(compressionOptions.rmask, &rshift, &rsize);
+				PixelFormat::maskShiftAndSize(compressionOptions.gmask, &gshift, &gsize);
+				PixelFormat::maskShiftAndSize(compressionOptions.bmask, &bshift, &bsize);
+			}
+		}
+
 		if (compressionOptions.enableAlphaDithering)
 		{
 			if (compressionOptions.format == Format_DXT3)
 			{
-				Quantize::Alpha4(mipmap.asMutableFixedImage());
+				asize = 4;
 			}
-			else if (compressionOptions.format == Format_DXT1a)
+			else if (compressionOptions.format == Format_RGB)
 			{
-				Quantize::BinaryAlpha(mipmap.asMutableFixedImage(), compressionOptions.alphaThreshold);
+				uint ashift;
+				PixelFormat::maskShiftAndSize(compressionOptions.amask, &ashift, &asize);
 			}
 		}
+
+		if (compressionOptions.binaryAlpha)
+		{
+			asize = 8; // Already quantized.
+		}
+
+		Quantize::FloydSteinberg(mipmap.asMutableFixedImage(), rsize, gsize, bsize, asize);
 	}
 }
 
 
 // Compress the given mipmap.
-bool Compressor::Private::compressMipmap(const Mipmap & mipmap, const CompressionOptions::Private & compressionOptions, const OutputOptions::Private & outputOptions) const
+bool Compressor::Private::compressMipmap(const Mipmap & mipmap, const InputOptions::Private & inputOptions, const CompressionOptions::Private & compressionOptions, const OutputOptions::Private & outputOptions) const
 {
 	const Image * image = mipmap.asFixedImage();
-
 	nvDebugCheck(image != NULL);
 
+	FastCompressor fast;
+	fast.setImage(image, inputOptions.alphaMode);
+
+	SlowCompressor slow;
+	slow.setImage(image, inputOptions.alphaMode);
+
+
 	if (compressionOptions.format == Format_RGBA || compressionOptions.format == Format_RGB)
 	{
 		compressRGB(image, outputOptions, compressionOptions);
@@ -663,18 +713,19 @@ bool Compressor::Private::compressMipmap(const Mipmap & mipmap, const Compressio
 #endif
 		if (compressionOptions.quality == Quality_Fastest)
 		{
-			fastCompressDXT1(image, outputOptions);
+			fast.compressDXT1(outputOptions);
 		}
 		else
 		{
 			if (cudaEnabled)
 			{
 				nvDebugCheck(cudaSupported);
-				cudaCompressDXT1(image, outputOptions, compressionOptions);
+				cuda->setImage(image, inputOptions.alphaMode);
+				cuda->compressDXT1(compressionOptions, outputOptions);
 			}
 			else
 			{
-				compressDXT1(image, outputOptions, compressionOptions);
+				slow.compressDXT1(compressionOptions, outputOptions);
 			}
 		}
 	}
@@ -682,18 +733,18 @@ bool Compressor::Private::compressMipmap(const Mipmap & mipmap, const Compressio
 	{
 		if (compressionOptions.quality == Quality_Fastest)
 		{
-			fastCompressDXT1a(image, outputOptions);
+			fast.compressDXT1a(outputOptions);
 		}
 		else
 		{
 			if (cudaEnabled)
 			{
 				nvDebugCheck(cudaSupported);
-				/*cuda*/compressDXT1a(image, outputOptions, compressionOptions);
+				/*cuda*/slow.compressDXT1a(compressionOptions, outputOptions);
 			}
 			else
 			{
-				compressDXT1a(image, outputOptions, compressionOptions);
+				slow.compressDXT1a(compressionOptions, outputOptions);
 			}
 		}
 	}
@@ -701,18 +752,19 @@ bool Compressor::Private::compressMipmap(const Mipmap & mipmap, const Compressio
 	{
 		if (compressionOptions.quality == Quality_Fastest)
 		{
-			fastCompressDXT3(image, outputOptions);
+			fast.compressDXT3(outputOptions);
 		}
 		else
 		{
 			if (cudaEnabled)
 			{
 				nvDebugCheck(cudaSupported);
-				cudaCompressDXT3(image, outputOptions, compressionOptions);
+				cuda->setImage(image, inputOptions.alphaMode);
+				cuda->compressDXT3(compressionOptions, outputOptions);
 			}
 			else
 			{
-				compressDXT3(image, outputOptions, compressionOptions);
+				slow.compressDXT3(compressionOptions, outputOptions);
 			}
 		}
 	}
@@ -720,18 +772,19 @@ bool Compressor::Private::compressMipmap(const Mipmap & mipmap, const Compressio
 	{
 		if (compressionOptions.quality == Quality_Fastest)
 		{
-			fastCompressDXT5(image, outputOptions);
+			fast.compressDXT5(outputOptions);
 		}
 		else
 		{
 			if (cudaEnabled)
 			{
 				nvDebugCheck(cudaSupported);
-				cudaCompressDXT5(image, outputOptions, compressionOptions);
+				cuda->setImage(image, inputOptions.alphaMode);
+				cuda->compressDXT5(compressionOptions, outputOptions);
 			}
 			else
 			{
-				compressDXT5(image, outputOptions, compressionOptions);
+				slow.compressDXT5(compressionOptions, outputOptions);
 			}
 		}
 	}
@@ -739,20 +792,20 @@ bool Compressor::Private::compressMipmap(const Mipmap & mipmap, const Compressio
 	{
 		if (compressionOptions.quality == Quality_Fastest)
 		{
-			fastCompressDXT5n(image, outputOptions);
+			fast.compressDXT5n(outputOptions);
 		}
 		else
 		{
-			compressDXT5n(image, outputOptions, compressionOptions);
+			slow.compressDXT5n(compressionOptions, outputOptions);
 		}
 	}
 	else if (compressionOptions.format == Format_BC4)
 	{
-		compressBC4(image, outputOptions, compressionOptions);
+		slow.compressBC4(compressionOptions, outputOptions);
 	}
 	else if (compressionOptions.format == Format_BC5)
 	{
-		compressBC5(image, outputOptions, compressionOptions);
+		slow.compressBC5(compressionOptions, outputOptions);
 	}
 
 	return true;

src/nvtt/Compressor.h

@@ -24,6 +24,10 @@
 #ifndef NV_TT_COMPRESSOR_H
 #define NV_TT_COMPRESSOR_H
 
+#include <nvcore/Ptr.h>
+
+#include <nvtt/cuda/CudaCompressDXT.h>
+
 #include "nvtt.h"
 
 namespace nv
@@ -56,13 +60,16 @@ namespace nvtt
 		void scaleMipmap(Mipmap & mipmap, const InputOptions::Private & inputOptions, uint w, uint h, uint d) const;
 		void processInputImage(Mipmap & mipmap, const InputOptions::Private & inputOptions) const;
 		void quantizeMipmap(Mipmap & mipmap, const CompressionOptions::Private & compressionOptions) const;
-		bool compressMipmap(const Mipmap & mipmap, const CompressionOptions::Private & compressionOptions, const OutputOptions::Private & outputOptions) const;
+		bool compressMipmap(const Mipmap & mipmap, const InputOptions::Private & inputOptions, const CompressionOptions::Private & compressionOptions, const OutputOptions::Private & outputOptions) const;
 
 
 	public:
 
 		bool cudaSupported;
 		bool cudaEnabled;
+
+		nv::AutoPtr<nv::CudaCompressor> cuda;
+
 	};
 
 } // nvtt namespace

src/nvtt/FastCompressDXT.h

@@ -1,87 +0,0 @@
-// Copyright NVIDIA Corporation 2007 -- Ignacio Castano <icastano@nvidia.com>
-// 
-// 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.
-
-#ifndef NV_TT_FASTCOMPRESSDXT_H
-#define NV_TT_FASTCOMPRESSDXT_H
-
-#include <nvimage/nvimage.h>
-
-namespace nv
-{
-	struct ColorBlock;
-	struct BlockDXT1;
-	struct BlockDXT3;
-	struct BlockDXT5;
-	struct AlphaBlockDXT3;
-	struct AlphaBlockDXT5;
-
-	// Color compression:
-
-	// Compressor that uses the extremes of the luminance axis.
-	void compressBlock_DiameterAxis(const ColorBlock & rgba, BlockDXT1 * block);
-
-	// Compressor that uses the extremes of the luminance axis.
-	void compressBlock_LuminanceAxis(const ColorBlock & rgba, BlockDXT1 * block);
-
-	// Compressor that uses bounding box.
-	void compressBlock_BoundsRange(const ColorBlock & rgba, BlockDXT1 * block);
-
-	// Compressor that uses bounding box and takes alpha into account.
-	void compressBlock_BoundsRangeAlpha(const ColorBlock & rgba, BlockDXT1 * block);
-
-	// Compressor that uses the best fit axis.
-	void compressBlock_BestFitAxis(const ColorBlock & rgba, BlockDXT1 * block);
-
-
-	// Simple, but slow compressor that tests all color pairs.
-	void compressBlock_TestAllPairs(const ColorBlock & rgba, BlockDXT1 * block);
-	
-	// Brute force 6d search along the best fit axis.
-	void compressBlock_AnalyzeBestFitAxis(const ColorBlock & rgba, BlockDXT1 * block);
-
-	// Spatial greedy search.
-	void refineSolution_1dSearch(const ColorBlock & rgba, BlockDXT1 * block);
-	void refineSolution_3dSearch(const ColorBlock & rgba, BlockDXT1 * block);
-	void refineSolution_6dSearch(const ColorBlock & rgba, BlockDXT1 * block);
-	
-	// Brute force compressor for DXT5n
-	void compressGreenBlock_BruteForce(const ColorBlock & rgba, BlockDXT1 * block);
-
-	// Minimize error of the endpoints.
-	void optimizeEndPoints(const ColorBlock & rgba, BlockDXT1 * block);
-	
-	uint blockError(const ColorBlock & rgba, const BlockDXT1 & block);
-	uint blockError(const ColorBlock & rgba, const AlphaBlockDXT5 & block);
-
-	// Alpha compression:
-	void compressBlock(const ColorBlock & rgba, AlphaBlockDXT3 * block);
-	void compressBlock_BoundsRange(const ColorBlock & rgba, BlockDXT3 * block);
-	void compressBlock_BoundsRange(const ColorBlock & rgba, BlockDXT5 * block);
-
-	uint compressBlock_BoundsRange(const ColorBlock & rgba, AlphaBlockDXT5 * block);
-	uint compressBlock_BruteForce(const ColorBlock & rgba, AlphaBlockDXT5 * block);
-	uint compressBlock_Iterative(const ColorBlock & rgba, AlphaBlockDXT5 * block);
-
-} // nv namespace
-
-#endif // NV_TT_FASTCOMPRESSDXT_H

src/nvtt/OptimalCompressDXT.cpp

@@ -0,0 +1,368 @@
+// Copyright NVIDIA Corporation 2007 -- Ignacio Castano <icastano@nvidia.com>
+// 
+// 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.
+
+#include <nvcore/Containers.h> // swap
+
+#include <nvmath/Color.h>
+
+#include <nvimage/ColorBlock.h>
+#include <nvimage/BlockDXT.h>
+
+#include "OptimalCompressDXT.h"
+#include "SingleColorLookup.h"
+
+
+using namespace nv;
+using namespace OptimalCompress;
+
+
+
+namespace
+{
+	static int computeGreenError(const ColorBlock & rgba, const BlockDXT1 * block)
+	{
+		nvDebugCheck(block != NULL);
+
+		int palette[4];
+		palette[0] = (block->col0.g << 2) | (block->col0.g >> 4);
+		palette[1] = (block->col1.g << 2) | (block->col1.g >> 4);
+		palette[2] = (2 * palette[0] + palette[1]) / 3;
+		palette[3] = (2 * palette[1] + palette[0]) / 3;
+
+		int totalError = 0;
+
+		for (int i = 0; i < 16; i++)
+		{
+			const int green = rgba.color(i).g;
+			
+			int error = abs(green - palette[0]);
+			error = min(error, abs(green - palette[1]));
+			error = min(error, abs(green - palette[2]));
+			error = min(error, abs(green - palette[3]));
+			
+			totalError += error;
+		}
+
+		return totalError;
+	}
+
+	static uint computeGreenIndices(const ColorBlock & rgba, const Color32 palette[4])
+	{
+		const int color0 = palette[0].g;
+		const int color1 = palette[1].g;
+		const int color2 = palette[2].g;
+		const int color3 = palette[3].g;
+		
+		uint indices = 0;
+		for (int i = 0; i < 16; i++)
+		{
+			const int color = rgba.color(i).g;
+			
+			uint d0 = abs(color0 - color);
+			uint d1 = abs(color1 - color);
+			uint d2 = abs(color2 - color);
+			uint d3 = abs(color3 - color);
+			
+			uint b0 = d0 > d3;
+			uint b1 = d1 > d2;
+			uint b2 = d0 > d2;
+			uint b3 = d1 > d3;
+			uint b4 = d2 > d3;
+			
+			uint x0 = b1 & b2;
+			uint x1 = b0 & b3;
+			uint x2 = b0 & b4;
+			
+			indices |= (x2 | ((x0 | x1) << 1)) << (2 * i);
+		}
+
+		return indices;
+	}
+
+	// Choose quantized color that produces less error. Used by DXT3 compressor.
+	inline static uint quantize4(uint8 a)
+	{
+		int q0 = (a >> 4) - 1;
+		int q1 = (a >> 4);
+		int q2 = (a >> 4) + 1;
+		
+		q0 = (q0 << 4) | q0;
+		q1 = (q1 << 4) | q1;
+		q2 = (q2 << 4) | q2;
+		
+		int d0 = abs(q0 - a);
+		int d1 = abs(q1 - a);
+		int d2 = abs(q2 - a);
+
+		if (d0 < d1 && d0 < d2) return q0 >> 4;
+		if (d1 < d2) return q1 >> 4;
+		return q2 >> 4;
+	}
+	
+	static uint computeAlphaError(const ColorBlock & rgba, const AlphaBlockDXT5 * block)
+	{
+		uint8 alphas[8];
+		block->evaluatePalette(alphas);
+
+		uint totalError = 0;
+
+		for (uint i = 0; i < 16; i++)
+		{
+			uint8 alpha = rgba.color(i).a;
+
+			uint besterror = 256*256;
+			uint best;
+			for (uint p = 0; p < 8; p++)
+			{
+				int d = alphas[p] - alpha;
+				uint error = d * d;
+
+				if (error < besterror)
+				{
+					besterror = error;
+					best = p;
+				}
+			}
+
+			totalError += besterror;
+		}
+
+		return totalError;
+	}
+	
+	static void computeAlphaIndices(const ColorBlock & rgba, AlphaBlockDXT5 * block)
+	{
+		uint8 alphas[8];
+		block->evaluatePalette(alphas);
+
+		for (uint i = 0; i < 16; i++)
+		{
+			uint8 alpha = rgba.color(i).a;
+
+			uint besterror = 256*256;
+			uint best = 8;
+			for(uint p = 0; p < 8; p++)
+			{
+				int d = alphas[p] - alpha;
+				uint error = d * d;
+
+				if (error < besterror)
+				{
+					besterror = error;
+					best = p;
+				}
+			}
+			nvDebugCheck(best < 8);
+
+			block->setIndex(i, best);
+		}
+	}
+
+} // namespace
+
+
+
+
+
+// Single color compressor, based on:
+// https://mollyrocket.com/forums/viewtopic.php?t=392
+void OptimalCompress::compressDXT1(Color32 c, BlockDXT1 * dxtBlock)
+{
+	dxtBlock->col0.r = OMatch5[c.r][0];
+	dxtBlock->col0.g = OMatch6[c.g][0];
+	dxtBlock->col0.b = OMatch5[c.b][0];
+	dxtBlock->col1.r = OMatch5[c.r][1];
+	dxtBlock->col1.g = OMatch6[c.g][1];
+	dxtBlock->col1.b = OMatch5[c.b][1];
+	dxtBlock->indices = 0xaaaaaaaa;
+
+	if (dxtBlock->col0.u < dxtBlock->col1.u)
+	{
+		swap(dxtBlock->col0.u, dxtBlock->col1.u);
+		dxtBlock->indices ^= 0x55555555;
+	}
+}
+
+void OptimalCompress::compressDXT1a(Color32 rgba, BlockDXT1 * dxtBlock)
+{
+	if (rgba.a < 128)
+	{
+		dxtBlock->col0.u = 0;
+		dxtBlock->col1.u = 0;
+		dxtBlock->indices = 0xFFFFFFFF;
+	}
+	else
+	{
+		compressDXT1(rgba, dxtBlock);
+	}
+}
+
+
+// Brute force green channel compressor
+void OptimalCompress::compressDXT1G(const ColorBlock & rgba, BlockDXT1 * block)
+{
+	nvDebugCheck(block != NULL);
+	
+	uint8 ming = 63;
+	uint8 maxg = 0;
+	
+	// Get min/max green.
+	for (uint i = 0; i < 16; i++)
+	{
+		uint8 green = rgba.color(i).g >> 2;
+		ming = min(ming, green);
+		maxg = max(maxg, green);
+	}
+
+	block->col0.r = 31;
+	block->col1.r = 31;
+	block->col0.g = maxg;
+	block->col1.g = ming;
+	block->col0.b = 0;
+	block->col1.b = 0;
+
+	if (maxg - ming > 4)
+	{
+		int besterror = computeGreenError(rgba, block);
+		int bestg0 = maxg;
+		int bestg1 = ming;
+		
+		for (int g0 = ming+5; g0 < maxg; g0++)
+		{
+			for (int g1 = ming; g1 < g0-4; g1++)
+			{
+				if ((maxg-g0) + (g1-ming) > besterror)
+					continue;
+				
+				block->col0.g = g0;
+				block->col1.g = g1;
+				int error = computeGreenError(rgba, block);
+				
+				if (error < besterror)
+				{
+					besterror = error;
+					bestg0 = g0;
+					bestg1 = g1;
+				}
+			}
+		}
+		
+		block->col0.g = bestg0;
+		block->col1.g = bestg1;
+	}
+	
+	Color32 palette[4];
+	block->evaluatePalette(palette);
+	block->indices = computeGreenIndices(rgba, palette);
+}
+
+void OptimalCompress::compressDXT3A(const ColorBlock & rgba, AlphaBlockDXT3 * dxtBlock)
+{
+	dxtBlock->alpha0 = quantize4(rgba.color(0).a);
+	dxtBlock->alpha1 = quantize4(rgba.color(1).a);
+	dxtBlock->alpha2 = quantize4(rgba.color(2).a);
+	dxtBlock->alpha3 = quantize4(rgba.color(3).a);
+	dxtBlock->alpha4 = quantize4(rgba.color(4).a);
+	dxtBlock->alpha5 = quantize4(rgba.color(5).a);
+	dxtBlock->alpha6 = quantize4(rgba.color(6).a);
+	dxtBlock->alpha7 = quantize4(rgba.color(7).a);
+	dxtBlock->alpha8 = quantize4(rgba.color(8).a);
+	dxtBlock->alpha9 = quantize4(rgba.color(9).a);
+	dxtBlock->alphaA = quantize4(rgba.color(10).a);
+	dxtBlock->alphaB = quantize4(rgba.color(11).a);
+	dxtBlock->alphaC = quantize4(rgba.color(12).a);
+	dxtBlock->alphaD = quantize4(rgba.color(13).a);
+	dxtBlock->alphaE = quantize4(rgba.color(14).a);
+	dxtBlock->alphaF = quantize4(rgba.color(15).a);
+}
+
+
+void OptimalCompress::compressDXT5A(const ColorBlock & rgba, AlphaBlockDXT5 * dxtBlock)
+{
+	uint8 mina = 255;
+	uint8 maxa = 0;
+
+	// Get min/max alpha.
+	for (uint i = 0; i < 16; i++)
+	{
+		uint8 alpha = rgba.color(i).a;
+		mina = min(mina, alpha);
+		maxa = max(maxa, alpha);
+	}
+
+	dxtBlock->alpha0 = maxa;
+	dxtBlock->alpha1 = mina;
+
+	/*int centroidDist = 256;
+	int centroid;
+
+	// Get the closest to the centroid.
+	for (uint i = 0; i < 16; i++)
+	{
+		uint8 alpha = rgba.color(i).a;
+		int dist = abs(alpha - (maxa + mina) / 2);
+		if (dist < centroidDist)
+		{
+			centroidDist = dist;
+			centroid = alpha;
+		}
+	}*/
+
+	if (maxa - mina > 8)
+	{
+		int besterror = computeAlphaError(rgba, dxtBlock);
+		int besta0 = maxa;
+		int besta1 = mina;
+
+		for (int a0 = mina+9; a0 < maxa; a0++)
+		{
+			for (int a1 = mina; a1 < a0-8; a1++)
+			//for (int a1 = mina; a1 < maxa; a1++)
+			{
+				//nvCheck(abs(a1-a0) > 8);
+
+				//if (abs(a0 - a1) < 8) continue;
+				//if ((maxa-a0) + (a1-mina) + min(abs(centroid-a0), abs(centroid-a1)) > besterror)
+				if ((maxa-a0) + (a1-mina) > besterror)
+					continue;
+
+				dxtBlock->alpha0 = a0;
+				dxtBlock->alpha1 = a1;
+				int error = computeAlphaError(rgba, dxtBlock);
+
+				if (error < besterror)
+				{
+					besterror = error;
+					besta0 = a0;
+					besta1 = a1;
+				}
+			}
+		}
+
+		dxtBlock->alpha0 = besta0;
+		dxtBlock->alpha1 = besta1;
+	}
+
+	computeAlphaIndices(rgba, dxtBlock);
+}
+

src/nvtt/OptimalCompressDXT.h

@@ -0,0 +1,49 @@
+// Copyright NVIDIA Corporation 2008 -- Ignacio Castano <icastano@nvidia.com>
+// 
+// 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.
+
+#ifndef NV_TT_OPTIMALCOMPRESSDXT_H
+#define NV_TT_OPTIMALCOMPRESSDXT_H
+
+#include <nvimage/nvimage.h>
+
+namespace nv
+{
+	struct ColorBlock;
+	struct BlockDXT1;
+	struct BlockDXT3;
+	struct BlockDXT5;
+	struct AlphaBlockDXT3;
+	struct AlphaBlockDXT5;
+
+	namespace OptimalCompress
+	{
+		void compressDXT1(Color32 rgba, BlockDXT1 * dxtBlock);
+		void compressDXT1a(Color32 rgba, BlockDXT1 * dxtBlock);
+		
+		void compressDXT1G(const ColorBlock & rgba, BlockDXT1 * block);
+		void compressDXT3A(const ColorBlock & rgba, AlphaBlockDXT3 * dxtBlock);
+		void compressDXT5A(const ColorBlock & rgba, AlphaBlockDXT5 * dxtBlock);
+	}
+} // nv namespace
+
+#endif // NV_TT_OPTIMALCOMPRESSDXT_H

src/nvtt/QuickCompressDXT.cpp

@@ -27,7 +27,7 @@
 #include <nvimage/BlockDXT.h>
 
 #include "QuickCompressDXT.h"
-#include "SingleColorLookup.h"
+#include "OptimalCompressDXT.h"
 
 
 using namespace nv;
@@ -288,116 +288,213 @@ static void optimizeEndPoints4(Vector3 block[16], BlockDXT1 * dxtBlock)
 	dxtBlock->indices = computeIndices3(block, a, b);
 }*/
 
-
-static void optimizeAlpha8(const ColorBlock & rgba, AlphaBlockDXT5 * block)
+namespace
 {
-	float alpha2_sum = 0;
-	float beta2_sum = 0;
-	float alphabeta_sum = 0;
-	float alphax_sum = 0;
-	float betax_sum = 0;
 
-	for (int i = 0; i < 16; i++)
+	static uint computeAlphaIndices(const ColorBlock & rgba, AlphaBlockDXT5 * block)
 	{
-		uint idx = block->index(i);
-		float alpha;
-		if (idx < 2) alpha = 1.0f - idx;
-		else alpha = (8.0f - idx) / 7.0f;
+		uint8 alphas[8];
+		block->evaluatePalette(alphas);
 
-		float beta = 1 - alpha;
+		uint totalError = 0;
 
-		alpha2_sum += alpha * alpha;
-		beta2_sum += beta * beta;
-		alphabeta_sum += alpha * beta;
-		alphax_sum += alpha * rgba.color(i).a;
-		betax_sum += beta * rgba.color(i).a;
+		for (uint i = 0; i < 16; i++)
+		{
+			uint8 alpha = rgba.color(i).a;
+
+			uint besterror = 256*256;
+			uint best = 8;
+			for(uint p = 0; p < 8; p++)
+			{
+				int d = alphas[p] - alpha;
+				uint error = d * d;
+
+				if (error < besterror)
+				{
+					besterror = error;
+					best = p;
+				}
+			}
+			nvDebugCheck(best < 8);
+
+			totalError += besterror;
+			block->setIndex(i, best);
+		}
+
+		return totalError;
 	}
 
-	const float factor = 1.0f / (alpha2_sum * beta2_sum - alphabeta_sum * alphabeta_sum);
-
-	float a = (alphax_sum * beta2_sum - betax_sum * alphabeta_sum) * factor;
-	float b = (betax_sum * alpha2_sum - alphax_sum * alphabeta_sum) * factor;
-
-	uint alpha0 = uint(min(max(a, 0.0f), 255.0f));
-	uint alpha1 = uint(min(max(b, 0.0f), 255.0f));
-
-	if (alpha0 < alpha1)
+	static void optimizeAlpha8(const ColorBlock & rgba, AlphaBlockDXT5 * block)
 	{
-		swap(alpha0, alpha1);
+		float alpha2_sum = 0;
+		float beta2_sum = 0;
+		float alphabeta_sum = 0;
+		float alphax_sum = 0;
+		float betax_sum = 0;
 
-		// Flip indices:
 		for (int i = 0; i < 16; i++)
 		{
 			uint idx = block->index(i);
-			if (idx < 2) block->setIndex(i, 1 - idx);
-			else block->setIndex(i, 9 - idx);
+			float alpha;
+			if (idx < 2) alpha = 1.0f - idx;
+			else alpha = (8.0f - idx) / 7.0f;
+			
+			float beta = 1 - alpha;
+
+			alpha2_sum += alpha * alpha;
+			beta2_sum += beta * beta;
+			alphabeta_sum += alpha * beta;
+			alphax_sum += alpha * rgba.color(i).a;
+			betax_sum += beta * rgba.color(i).a;
 		}
+
+		const float factor = 1.0f / (alpha2_sum * beta2_sum - alphabeta_sum * alphabeta_sum);
+
+		float a = (alphax_sum * beta2_sum - betax_sum * alphabeta_sum) * factor;
+		float b = (betax_sum * alpha2_sum - alphax_sum * alphabeta_sum) * factor;
+
+		uint alpha0 = uint(min(max(a, 0.0f), 255.0f));
+		uint alpha1 = uint(min(max(b, 0.0f), 255.0f));
+
+		if (alpha0 < alpha1)
+		{
+			swap(alpha0, alpha1);
+
+			// Flip indices:
+			for (int i = 0; i < 16; i++)
+			{
+				uint idx = block->index(i);
+				if (idx < 2) block->setIndex(i, 1 - idx);
+				else block->setIndex(i, 9 - idx);
+			}
+		}
+		else if (alpha0 == alpha1)
+		{
+			for (int i = 0; i < 16; i++)
+			{
+				block->setIndex(i, 0);
+			}
+		}
+
+		block->alpha0 = alpha0;
+		block->alpha1 = alpha1;
 	}
-	else if (alpha0 == alpha1)
+
+	/*
+	static void optimizeAlpha6(const ColorBlock & rgba, AlphaBlockDXT5 * block)
 	{
+		float alpha2_sum = 0;
+		float beta2_sum = 0;
+		float alphabeta_sum = 0;
+		float alphax_sum = 0;
+		float betax_sum = 0;
+
 		for (int i = 0; i < 16; i++)
 		{
-			block->setIndex(i, 0);
+			uint8 x = rgba.color(i).a;
+			if (x == 0 || x == 255) continue;
+
+			uint bits = block->index(i);
+			if (bits == 6 || bits == 7) continue;
+
+			float alpha;
+			if (bits == 0) alpha = 1.0f;
+			else if (bits == 1) alpha = 0.0f;
+			else alpha = (6.0f - block->index(i)) / 5.0f;
+			
+			float beta = 1 - alpha;
+
+			alpha2_sum += alpha * alpha;
+			beta2_sum += beta * beta;
+			alphabeta_sum += alpha * beta;
+			alphax_sum += alpha * x;
+			betax_sum += beta * x;
 		}
+
+		const float factor = 1.0f / (alpha2_sum * beta2_sum - alphabeta_sum * alphabeta_sum);
+
+		float a = (alphax_sum * beta2_sum - betax_sum * alphabeta_sum) * factor;
+		float b = (betax_sum * alpha2_sum - alphax_sum * alphabeta_sum) * factor;
+
+		uint alpha0 = uint(min(max(a, 0.0f), 255.0f));
+		uint alpha1 = uint(min(max(b, 0.0f), 255.0f));
+
+		if (alpha0 > alpha1)
+		{
+			swap(alpha0, alpha1);
+		}
+
+		block->alpha0 = alpha0;
+		block->alpha1 = alpha1;
+	}
+	*/
+
+	static bool sameIndices(const AlphaBlockDXT5 & block0, const AlphaBlockDXT5 & block1)
+	{
+		const uint64 mask = ~uint64(0xFFFF);
+		return (block0.u | mask) == (block1.u | mask);
 	}
 
-	block->alpha0 = alpha0;
-	block->alpha1 = alpha1;
-}
+} // namespace
 
 
 
-
-// Single color compressor, based on:
-// https://mollyrocket.com/forums/viewtopic.php?t=392
-void QuickCompress::compressDXT1(Color32 c, BlockDXT1 * dxtBlock)
-{
-	dxtBlock->col0.r = OMatch5[c.r][0];
-	dxtBlock->col0.g = OMatch5[c.g][0];
-	dxtBlock->col0.b = OMatch5[c.b][0];
-	dxtBlock->col1.r = OMatch5[c.r][1];
-	dxtBlock->col1.g = OMatch5[c.g][1];
-	dxtBlock->col1.b = OMatch5[c.b][1];
-	dxtBlock->indices = 0xaaaaaaaa;
-}
-
 void QuickCompress::compressDXT1(const ColorBlock & rgba, BlockDXT1 * dxtBlock)
 {
-	// read block
-	Vector3 block[16];
-	extractColorBlockRGB(rgba, block);
-	
-	// find min and max colors
-	Vector3 maxColor, minColor;
-	findMinMaxColorsBox(block, 16, &maxColor, &minColor);
-	
-	selectDiagonal(block, 16, &maxColor, &minColor);
-	
-	insetBBox(&maxColor, &minColor);
-	
-	uint16 color0 = roundAndExpand(&maxColor);
-	uint16 color1 = roundAndExpand(&minColor);
-
-	if (color0 < color1)
+	if (rgba.isSingleColor())
 	{
-		swap(maxColor, minColor);
-		swap(color0, color1);
+		OptimalCompress::compressDXT1(rgba.color(0), dxtBlock);
 	}
+	else
+	{
+		// read block
+		Vector3 block[16];
+		extractColorBlockRGB(rgba, block);
 		
-	dxtBlock->col0 = Color16(color0);
-	dxtBlock->col1 = Color16(color1);
-	dxtBlock->indices = computeIndices4(block, maxColor, minColor);
+		// find min and max colors
+		Vector3 maxColor, minColor;
+		findMinMaxColorsBox(block, 16, &maxColor, &minColor);
 		
-	optimizeEndPoints4(block, dxtBlock);
+		selectDiagonal(block, 16, &maxColor, &minColor);
+		
+		insetBBox(&maxColor, &minColor);
+		
+		uint16 color0 = roundAndExpand(&maxColor);
+		uint16 color1 = roundAndExpand(&minColor);
+
+		if (color0 < color1)
+		{
+			swap(maxColor, minColor);
+			swap(color0, color1);
+		}
+
+		dxtBlock->col0 = Color16(color0);
+		dxtBlock->col1 = Color16(color1);
+		dxtBlock->indices = computeIndices4(block, maxColor, minColor);
+
+		optimizeEndPoints4(block, dxtBlock);
+	}
 }
 
 
 void QuickCompress::compressDXT1a(const ColorBlock & rgba, BlockDXT1 * dxtBlock)
 {
-	if (!rgba.hasAlpha())
+	bool hasAlpha = false;
+	
+	for (uint i = 0; i < 16; i++)
+	{
+		if (rgba.color(i).a < 128) {
+			hasAlpha = true;
+			break;
+		}
+	}
+	
+	if (!hasAlpha)
 	{
 		compressDXT1(rgba, dxtBlock);
 	}
+	// @@ Handle single RGB, with varying alpha? We need tables for single color compressor in 3 color mode.
+	//else if (rgba.isSingleColorNoAlpha()) { ... }
 	else 
 	{
 		// read block
@@ -430,160 +527,59 @@ void QuickCompress::compressDXT1a(const ColorBlock & rgba, BlockDXT1 * dxtBlock)
 }
 
 
-static int computeGreenError(const ColorBlock & rgba, const BlockDXT1 * block)
-{
-	nvDebugCheck(block != NULL);
-
-	int palette[4];
-	palette[0] = (block->col0.g << 2) | (block->col0.g >> 4);
-	palette[1] = (block->col1.g << 2) | (block->col1.g >> 4);
-	palette[2] = (2 * palette[0] + palette[1]) / 3;
-	palette[3] = (2 * palette[1] + palette[0]) / 3;
-
-	int totalError = 0;
-
-	for (int i = 0; i < 16; i++)
-	{
-		const int green = rgba.color(i).g;
-		
-		int error = abs(green - palette[0]);
-		error = min(error, abs(green - palette[1]));
-		error = min(error, abs(green - palette[2]));
-		error = min(error, abs(green - palette[3]));
-		
-		totalError += error;
-	}
-
-	return totalError;
-}
-
-static uint computeGreenIndices(const ColorBlock & rgba, const Color32 palette[4])
-{
-	const int color0 = palette[0].g;
-	const int color1 = palette[1].g;
-	const int color2 = palette[2].g;
-	const int color3 = palette[3].g;
-	
-	uint indices = 0;
-	for (int i = 0; i < 16; i++)
-	{
-		const int color = rgba.color(i).g;
-		
-		uint d0 = abs(color0 - color);
-		uint d1 = abs(color1 - color);
-		uint d2 = abs(color2 - color);
-		uint d3 = abs(color3 - color);
-		
-		uint b0 = d0 > d3;
-		uint b1 = d1 > d2;
-		uint b2 = d0 > d2;
-		uint b3 = d1 > d3;
-		uint b4 = d2 > d3;
-		
-		uint x0 = b1 & b2;
-		uint x1 = b0 & b3;
-		uint x2 = b0 & b4;
-		
-		indices |= (x2 | ((x0 | x1) << 1)) << (2 * i);
-	}
-
-	return indices;
-}
-
-// Brute force green channel compressor
-void QuickCompress::compressDXT1G(const ColorBlock & rgba, BlockDXT1 * block)
-{
-	nvDebugCheck(block != NULL);
-	
-	uint8 ming = 63;
-	uint8 maxg = 0;
-	
-	// Get min/max green.
-	for (uint i = 0; i < 16; i++)
-	{
-		uint8 green = rgba.color(i).g >> 2;
-		ming = min(ming, green);
-		maxg = max(maxg, green);
-	}
-
-	block->col0.r = 31;
-	block->col1.r = 31;
-	block->col0.g = maxg;
-	block->col1.g = ming;
-	block->col0.b = 0;
-	block->col1.b = 0;
-
-	if (maxg - ming > 4)
-	{
-		int besterror = computeGreenError(rgba, block);
-		int bestg0 = maxg;
-		int bestg1 = ming;
-		
-		for (int g0 = ming+5; g0 < maxg; g0++)
-		{
-			for (int g1 = ming; g1 < g0-4; g1++)
-			{
-				if ((maxg-g0) + (g1-ming) > besterror)
-					continue;
-				
-				block->col0.g = g0;
-				block->col1.g = g1;
-				int error = computeGreenError(rgba, block);
-				
-				if (error < besterror)
-				{
-					besterror = error;
-					bestg0 = g0;
-					bestg1 = g1;
-				}
-			}
-		}
-		
-		block->col0.g = bestg0;
-		block->col1.g = bestg1;
-	}
-	
-	Color32 palette[4];
-	block->evaluatePalette(palette);
-	block->indices = computeGreenIndices(rgba, palette);
-}
-
-
-
-void QuickCompress::compressDXT3A(const ColorBlock & rgba, AlphaBlockDXT3 * dxtBlock)
-{
-	dxtBlock->alpha0 = rgba.color(0).a >> 4;
-	dxtBlock->alpha1 = rgba.color(1).a >> 4;
-	dxtBlock->alpha2 = rgba.color(2).a >> 4;
-	dxtBlock->alpha3 = rgba.color(3).a >> 4;
-	dxtBlock->alpha4 = rgba.color(4).a >> 4;
-	dxtBlock->alpha5 = rgba.color(5).a >> 4;
-	dxtBlock->alpha6 = rgba.color(6).a >> 4;
-	dxtBlock->alpha7 = rgba.color(7).a >> 4;
-	dxtBlock->alpha8 = rgba.color(8).a >> 4;
-	dxtBlock->alpha9 = rgba.color(9).a >> 4;
-	dxtBlock->alphaA = rgba.color(10).a >> 4;
-	dxtBlock->alphaB = rgba.color(11).a >> 4;
-	dxtBlock->alphaC = rgba.color(12).a >> 4;
-	dxtBlock->alphaD = rgba.color(13).a >> 4;
-	dxtBlock->alphaE = rgba.color(14).a >> 4;
-	dxtBlock->alphaF = rgba.color(15).a >> 4;
-}
-
 void QuickCompress::compressDXT3(const ColorBlock & rgba, BlockDXT3 * dxtBlock)
 {
 	compressDXT1(rgba, &dxtBlock->color);
-	compressDXT3A(rgba, &dxtBlock->alpha);
+	OptimalCompress::compressDXT3A(rgba, &dxtBlock->alpha);
 }
 
-void QuickCompress::compressDXT5A(const ColorBlock & rgba, AlphaBlockDXT5 * dxtBlock)
+
+void QuickCompress::compressDXT5A(const ColorBlock & rgba, AlphaBlockDXT5 * dxtBlock, int iterationCount/*=8*/)
 {
-	// @@ TODO
+	uint8 alpha0 = 0;
+	uint8 alpha1 = 255;
+	
+	// Get min/max alpha.
+	for (uint i = 0; i < 16; i++)
+	{
+		uint8 alpha = rgba.color(i).a;
+		alpha0 = max(alpha0, alpha);
+		alpha1 = min(alpha1, alpha);
+	}
+	
+	AlphaBlockDXT5 block;
+	block.alpha0 = alpha0 - (alpha0 - alpha1) / 34;
+	block.alpha1 = alpha1 + (alpha0 - alpha1) / 34;
+	uint besterror = computeAlphaIndices(rgba, &block);
+	
+	AlphaBlockDXT5 bestblock = block;
+
+	for (int i = 0; i < iterationCount; i++)
+	{
+		optimizeAlpha8(rgba, &block);
+		uint error = computeAlphaIndices(rgba, &block);
+		
+		if (error >= besterror)
+		{
+			// No improvement, stop.
+			break;
+		}
+		if (sameIndices(block, bestblock))
+		{
+			bestblock = block;
+			break;
+		}
+		
+		besterror = error;
+		bestblock = block;
+	};
+	
+	// Copy best block to result;
+	*dxtBlock = bestblock;
 }
 
-void QuickCompress::compressDXT5(const ColorBlock & rgba, BlockDXT5 * dxtBlock)
+void QuickCompress::compressDXT5(const ColorBlock & rgba, BlockDXT5 * dxtBlock, int iterationCount/*=8*/)
 {
 	compressDXT1(rgba, &dxtBlock->color);
-	compressDXT5A(rgba, &dxtBlock->alpha);
+	compressDXT5A(rgba, &dxtBlock->alpha, iterationCount);
 }
-

src/nvtt/QuickCompressDXT.h

@@ -37,16 +37,13 @@ namespace nv
 
 	namespace QuickCompress
 	{
-		void compressDXT1(const Color32 rgba, BlockDXT1 * dxtBlock);
 		void compressDXT1(const ColorBlock & rgba, BlockDXT1 * dxtBlock);
 		void compressDXT1a(const ColorBlock & rgba, BlockDXT1 * dxtBlock);
-		void compressDXT1G(const ColorBlock & rgba, BlockDXT1 * block);
 		
-		void compressDXT3A(const ColorBlock & rgba, AlphaBlockDXT3 * dxtBlock);
 		void compressDXT3(const ColorBlock & rgba, BlockDXT3 * dxtBlock);
 		
-		void compressDXT5A(const ColorBlock & rgba, AlphaBlockDXT5 * dxtBlock);
-		void compressDXT5(const ColorBlock & rgba, BlockDXT5 * dxtBlock);
+		void compressDXT5A(const ColorBlock & rgba, AlphaBlockDXT5 * dxtBlock, int iterationCount=8);
+		void compressDXT5(const ColorBlock & rgba, BlockDXT5 * dxtBlock, int iterationCount=8);
 	}
 } // nv namespace
 

src/nvtt/SingleColorLookup.h

@@ -48,7 +48,12 @@ void initTables()
 };
 */
 
-const static uint8 OMatch5[256][2] =
+#if __CUDACC__
+__constant__ unsigned short
+#else
+const static uint8 
+#endif
+OMatch5[256][2] =
 {
 	{0x00, 0x00},
 	{0x00, 0x00},
@@ -308,7 +313,12 @@ const static uint8 OMatch5[256][2] =
 	{0x1F, 0x1F},
 };
 
-const static uint8 OMatch6[256][2] =
+#if __CUDACC__
+__constant__ unsigned short
+#else
+const static uint8
+#endif
+OMatch6[256][2] =
 {
 	{0x00, 0x00},
 	{0x00, 0x01},

src/nvtt/cuda/Bitmaps.h

@@ -122,7 +122,7 @@ static void doPrecomputation()
 */
 
 
-const static uint bitmaps[992] =
+const static uint s_bitmapTable[992] =
 {
 	0x80000000,
 	0x40000000,

src/nvtt/cuda/CompressKernel.cu

@@ -28,6 +28,8 @@
 
 #include "CudaMath.h"
 
+#include "../SingleColorLookup.h"
+
 #define NUM_THREADS 64		// Number of threads per block.
 
 #if __DEVICE_EMULATION__
@@ -60,6 +62,7 @@ __device__ void sortColors(const float * values, int * cmp)
 {
 	int tid = threadIdx.x;
 
+#if 1
 	cmp[tid] = (values[0] < values[tid]);
 	cmp[tid] += (values[1] < values[tid]);
 	cmp[tid] += (values[2] < values[tid]);
@@ -93,13 +96,30 @@ __device__ void sortColors(const float * values, int * cmp)
 	if (tid > 12 && cmp[tid] == cmp[12]) ++cmp[tid];
 	if (tid > 13 && cmp[tid] == cmp[13]) ++cmp[tid];
 	if (tid > 14 && cmp[tid] == cmp[14]) ++cmp[tid];
+#else
+
+	cmp[tid] = 0;
+
+	#pragma unroll
+	for (int i = 0; i < 16; i++)
+	{
+		cmp[tid] += (values[i] < values[tid]);
+	}
+
+	// Resolve elements with the same index.
+	#pragma unroll
+	for (int i = 0; i < 15; i++)
+	{
+		if (tid > 0 && cmp[tid] == cmp[i]) ++cmp[tid];
+	}
+#endif
 }
 
 
 ////////////////////////////////////////////////////////////////////////////////
 // Load color block to shared mem
 ////////////////////////////////////////////////////////////////////////////////
-__device__ void loadColorBlock(const uint * image, float3 colors[16], float3 sums[16], int xrefs[16])
+__device__ void loadColorBlock(const uint * image, float3 colors[16], float3 sums[16], int xrefs[16], int * sameColor)
 {
 	const int bid = blockIdx.x;
 	const int idx = threadIdx.x;
@@ -124,6 +144,8 @@ __device__ void loadColorBlock(const uint * image, float3 colors[16], float3 sum
 		colorSums(colors, sums);
 		float3 axis = bestFitLine(colors, sums[0], kColorMetric);
 		
+		*sameColor = (axis == make_float3(0, 0, 0));
+		
 		dps[idx] = dot(colors[idx], axis);
 		
 #if __DEVICE_EMULATION__
@@ -137,7 +159,7 @@ __device__ void loadColorBlock(const uint * image, float3 colors[16], float3 sum
 	}
 }
 
-__device__ void loadColorBlock(const uint * image, float3 colors[16], float3 sums[16], float weights[16], int xrefs[16])
+__device__ void loadColorBlock(const uint * image, float3 colors[16], float3 sums[16], float weights[16], int xrefs[16], int * sameColor)
 {
 	const int bid = blockIdx.x;
 	const int idx = threadIdx.x;
@@ -167,6 +189,8 @@ __device__ void loadColorBlock(const uint * image, float3 colors[16], float3 sum
 		colorSums(colors, sums);
 		float3 axis = bestFitLine(colors, sums[0], kColorMetric);
 
+		*sameColor = (axis == make_float3(0, 0, 0));
+
 		dps[idx] = dot(rawColors[idx], axis);
 		
 #if __DEVICE_EMULATION__
@@ -187,7 +211,7 @@ __device__ void loadColorBlock(const uint * image, float3 colors[16], float3 sum
 ////////////////////////////////////////////////////////////////////////////////
 // Round color to RGB565 and expand
 ////////////////////////////////////////////////////////////////////////////////
-inline __device__ float3 roundAndExpand(float3 v, ushort * w)
+inline __device__ float3 roundAndExpand565(float3 v, ushort * w)
 {
 	v.x = rintf(__saturatef(v.x) * 31.0f);
 	v.y = rintf(__saturatef(v.y) * 63.0f);
@@ -234,8 +258,8 @@ __device__ float evalPermutation4(const float3 * colors, uint permutation, ushor
 	float3 b = (betax_sum * alpha2_sum - alphax_sum * alphabeta_sum) * factor;
 	
 	// Round a, b to the closest 5-6-5 color and expand...
-	a = roundAndExpand(a, start);
-	b = roundAndExpand(b, end);
+	a = roundAndExpand565(a, start);
+	b = roundAndExpand565(b, end);
 
 	// compute the error
 	float3 e = a * a * alpha2_sum + b * b * beta2_sum + 2.0f * (a * b * alphabeta_sum - a * alphax_sum - b * betax_sum);
@@ -274,8 +298,8 @@ __device__ float evalPermutation3(const float3 * colors, uint permutation, ushor
 	float3 b = (betax_sum * alpha2_sum - alphax_sum * alphabeta_sum) * factor;
 	
 	// Round a, b to the closest 5-6-5 color and expand...
-	a = roundAndExpand(a, start);
-	b = roundAndExpand(b, end);
+	a = roundAndExpand565(a, start);
+	b = roundAndExpand565(b, end);
 
 	// compute the error
 	float3 e = a * a * alpha2_sum + b * b * beta2_sum + 2.0f * (a * b * alphabeta_sum - a * alphax_sum - b * betax_sum);
@@ -315,8 +339,8 @@ __device__ float evalPermutation4(const float3 * colors, float3 color_sum, uint
 	float3 b = (betax_sum * alpha2_sum - alphax_sum * alphabeta_sum) * factor;
 	
 	// Round a, b to the closest 5-6-5 color and expand...
-	a = roundAndExpand(a, start);
-	b = roundAndExpand(b, end);
+	a = roundAndExpand565(a, start);
+	b = roundAndExpand565(b, end);
 
 	// compute the error
 	float3 e = a * a * alpha2_sum + b * b * beta2_sum + 2.0f * (a * b * alphabeta_sum - a * alphax_sum - b * betax_sum);
@@ -351,8 +375,8 @@ __device__ float evalPermutation3(const float3 * colors, float3 color_sum, uint
 	float3 b = (betax_sum * alpha2_sum - alphax_sum * alphabeta_sum) * factor;
 	
 	// Round a, b to the closest 5-6-5 color and expand...
-	a = roundAndExpand(a, start);
-	b = roundAndExpand(b, end);
+	a = roundAndExpand565(a, start);
+	b = roundAndExpand565(b, end);
 
 	// compute the error
 	float3 e = a * a * alpha2_sum + b * b * beta2_sum + 2.0f * (a * b * alphabeta_sum - a * alphax_sum - b * betax_sum);
@@ -391,8 +415,8 @@ __device__ float evalPermutation4(const float3 * colors, const float * weights,
 	float3 b = (betax_sum * alpha2_sum - alphax_sum * alphabeta_sum) * factor;
 	
 	// Round a, b to the closest 5-6-5 color and expand...
-	a = roundAndExpand(a, start);
-	b = roundAndExpand(b, end);
+	a = roundAndExpand565(a, start);
+	b = roundAndExpand565(b, end);
 
 	// compute the error
 	float3 e = a * a * alpha2_sum + b * b * beta2_sum + 2.0f * (a * b * alphabeta_sum - a * alphax_sum - b * betax_sum);
@@ -432,8 +456,8 @@ __device__ float evalPermutation3(const float3 * colors, const float * weights,
 	float3 b = (betax_sum * alpha2_sum - alphax_sum * alphabeta_sum) * factor;
 
 	// Round a, b to the closest 5-6-5 color and expand...
-	a = roundAndExpand(a, start);
-	b = roundAndExpand(b, end);
+	a = roundAndExpand565(a, start);
+	b = roundAndExpand565(b, end);
 
 	// compute the error
 	float3 e = a * a * alpha2_sum + b * b * beta2_sum + 2.0f * (a * b * alphabeta_sum - a * alphax_sum - b * betax_sum);
@@ -570,6 +594,40 @@ __device__ void evalAllPermutations(const float3 * colors, const float * weights
 }
 */
 
+__device__ void evalLevel4Permutations(const float3 * colors, float3 colorSum, const uint * permutations, ushort & bestStart, ushort & bestEnd, uint & bestPermutation, float * errors)
+{
+	const int idx = threadIdx.x;
+	
+	float bestError = FLT_MAX;
+	
+	for(int i = 0; i < 16; i++)
+	{
+		int pidx = idx + NUM_THREADS * i;
+		if (pidx >= 992) break;
+		
+		ushort start, end;
+		uint permutation = permutations[pidx];
+
+		float error = evalPermutation4(colors, colorSum, permutation, &start, &end);
+		
+		if (error < bestError)
+		{
+			bestError = error;
+			bestPermutation = permutation;
+			bestStart = start;
+			bestEnd = end;
+		}
+	}
+
+	if (bestStart < bestEnd)
+	{
+		swap(bestEnd, bestStart);
+		bestPermutation ^= 0x55555555;	// Flip indices.
+	}
+
+	errors[idx] = bestError;
+}
+
 __device__ void evalLevel4Permutations(const float3 * colors, const float * weights, float3 colorSum, const uint * permutations, ushort & bestStart, ushort & bestEnd, uint & bestPermutation, float * errors)
 {
 	const int idx = threadIdx.x;
@@ -605,7 +663,6 @@ __device__ void evalLevel4Permutations(const float3 * colors, const float * weig
 }
 
 
-
 ////////////////////////////////////////////////////////////////////////////////
 // Find index with minimum error
 ////////////////////////////////////////////////////////////////////////////////
@@ -715,22 +772,50 @@ __device__ void saveBlockDXT1(ushort start, ushort end, uint permutation, int xr
 	result[bid].y = indices;
 }
 
+__device__ void saveSingleColorBlockDXT1(float3 color, uint2 * result)
+{
+	const int bid = blockIdx.x;
 
+	int r = color.x * 255;
+	int g = color.y * 255;
+	int b = color.z * 255;
+
+	ushort color0 = (OMatch5[r][0] << 11) | (OMatch6[g][0] << 5) | OMatch5[b][0];
+	ushort color1 = (OMatch5[r][1] << 11) | (OMatch6[g][1] << 5) | OMatch5[b][1];
+
+	if (color0 < color1)
+	{
+		result[bid].x = (color0 << 16) | color1;
+		result[bid].y = 0xffffffff;
+	}
+	else
+	{
+		result[bid].x = (color1 << 16) | color0;
+		result[bid].y = 0xaaaaaaaa;
+	}
+}
 
 
 ////////////////////////////////////////////////////////////////////////////////
 // Compress color block
 ////////////////////////////////////////////////////////////////////////////////
-__global__ void compress(const uint * permutations, const uint * image, uint2 * result)
+__global__ void compressDXT1(const uint * permutations, const uint * image, uint2 * result)
 {
 	__shared__ float3 colors[16];
 	__shared__ float3 sums[16];
 	__shared__ int xrefs[16];
+	__shared__ int sameColor;
 	
-	loadColorBlock(image, colors, sums, xrefs);
+	loadColorBlock(image, colors, sums, xrefs, &sameColor);
 
 	__syncthreads();
 
+	if (sameColor)
+	{
+		if (threadIdx.x == 0) saveSingleColorBlockDXT1(colors[0], result);
+		return;
+	}
+
 	ushort bestStart, bestEnd;
 	uint bestPermutation;
 
@@ -748,18 +833,58 @@ __global__ void compress(const uint * permutations, const uint * image, uint2 *
 	}
 }
 
+__global__ void compressLevel4DXT1(const uint * permutations, const uint * image, uint2 * result)
+{
+	__shared__ float3 colors[16];
+	__shared__ float3 sums[16];
+	__shared__ int xrefs[16];
+	__shared__ int sameColor;
 	
-__global__ void compressWeighted(const uint * permutations, const uint * image, uint2 * result)
+	loadColorBlock(image, colors, sums, xrefs, &sameColor);
+
+	__syncthreads();
+
+	if (sameColor)
+	{
+		if (threadIdx.x == 0) saveSingleColorBlockDXT1(colors[0], result);
+		return;
+	}
+
+	ushort bestStart, bestEnd;
+	uint bestPermutation;
+
+	__shared__ float errors[NUM_THREADS];
+
+	evalLevel4Permutations(colors, sums[0], permutations, bestStart, bestEnd, bestPermutation, errors);
+	
+	// Use a parallel reduction to find minimum error.
+	const int minIdx = findMinError(errors);
+	
+	// Only write the result of the winner thread.
+	if (threadIdx.x == minIdx)
+	{
+		saveBlockDXT1(bestStart, bestEnd, bestPermutation, xrefs, result);
+	}
+}
+
+__global__ void compressWeightedDXT1(const uint * permutations, const uint * image, uint2 * result)
 {
 	__shared__ float3 colors[16];
 	__shared__ float3 sums[16];
 	__shared__ float weights[16];
 	__shared__ int xrefs[16];
+	__shared__ int sameColor;
 	
-	loadColorBlock(image, colors, sums, weights, xrefs);
+	loadColorBlock(image, colors, sums, weights, xrefs, &sameColor);
 	
 	__syncthreads();
 
+	if (sameColor)
+	{
+		if (threadIdx.x == 0) saveSingleColorBlockDXT1(colors[0], result);
+		return;
+	}
+
 	ushort bestStart, bestEnd;
 	uint bestPermutation;
 
@@ -845,8 +970,8 @@ __device__ void optimizeAlpha8(const float alphas[16], uchar & a0, uchar & a1)
 	float a = (alphax_sum * beta2_sum - betax_sum * alphabeta_sum) * factor;
 	float b = (betax_sum * alpha2_sum - alphax_sum * alphabeta_sum) * factor;
 
-	a0 = roundAndExpand(a);
-	a1 = roundAndExpand(b);
+	a0 = roundAndExpand8(a);
+	a1 = roundAndExpand8(b);
 }
 */
 /*
@@ -978,12 +1103,17 @@ extern "C" void setupCompressKernel(const float weights[3])
 // Launch kernel
 ////////////////////////////////////////////////////////////////////////////////
 
-extern "C" void compressKernel(uint blockNum, uint * d_data, uint * d_result, uint * d_bitmaps)
+extern "C" void compressKernelDXT1(uint blockNum, uint * d_data, uint * d_result, uint * d_bitmaps)
 {
-	compress<<<blockNum, NUM_THREADS>>>(d_bitmaps, d_data, (uint2 *)d_result);
+	compressDXT1<<<blockNum, NUM_THREADS>>>(d_bitmaps, d_data, (uint2 *)d_result);
 }
 
-extern "C" void compressWeightedKernel(uint blockNum, uint * d_data, uint * d_result, uint * d_bitmaps)
+extern "C" void compressKernelDXT1_Level4(uint blockNum, uint * d_data, uint * d_result, uint * d_bitmaps)
 {
-	compressWeighted<<<blockNum, NUM_THREADS>>>(d_bitmaps, d_data, (uint2 *)d_result);
+	compressLevel4DXT1<<<blockNum, NUM_THREADS>>>(d_bitmaps, d_data, (uint2 *)d_result);
+}
+
+extern "C" void compressWeightedKernelDXT1(uint blockNum, uint * d_data, uint * d_result, uint * d_bitmaps)
+{
+	compressWeightedDXT1<<<blockNum, NUM_THREADS>>>(d_bitmaps, d_data, (uint2 *)d_result);
 }

src/nvtt/cuda/CudaCompressDXT.cpp

@@ -24,13 +24,13 @@
 #include <nvcore/Debug.h>
 #include <nvcore/Containers.h>
 #include <nvmath/Color.h>
-#include <nvmath/Fitting.h>
 #include <nvimage/Image.h>
 #include <nvimage/ColorBlock.h>
 #include <nvimage/BlockDXT.h>
 #include <nvtt/CompressionOptions.h>
 #include <nvtt/OutputOptions.h>
-#include <nvtt/FastCompressDXT.h>
+#include <nvtt/QuickCompressDXT.h>
+#include <nvtt/OptimalCompressDXT.h>
 
 #include "CudaCompressDXT.h"
 #include "CudaUtils.h"
@@ -48,29 +48,15 @@ using namespace nvtt;
 
 #if defined HAVE_CUDA
 
+#define MAX_BLOCKS 8192U // 32768, 65535
+
+
 extern "C" void setupCompressKernel(const float weights[3]);
-extern "C" void compressKernel(uint blockNum, uint * d_data, uint * d_result, uint * d_bitmaps);
-extern "C" void compressWeightedKernel(uint blockNum, uint * d_data, uint * d_result, uint * d_bitmaps);
+extern "C" void compressKernelDXT1(uint blockNum, uint * d_data, uint * d_result, uint * d_bitmaps);
+extern "C" void compressKernelDXT1_Level4(uint blockNum, uint * d_data, uint * d_result, uint * d_bitmaps);
+extern "C" void compressWeightedKernelDXT1(uint blockNum, uint * d_data, uint * d_result, uint * d_bitmaps);
 
-#include "Bitmaps.h"
-
-// @@ Store this pointer in CompressionOptions. Allocate in ctor, free in dtor.
-static uint * d_bitmaps = NULL;
-
-static void doPrecomputation()
-{
-	if (d_bitmaps != NULL) {
-		return;
-	}
-
-    // Upload bitmaps.
-    cudaMalloc((void**) &d_bitmaps, 992 * sizeof(uint));
-    cudaMemcpy(d_bitmaps, bitmaps, 992 * sizeof(uint), cudaMemcpyHostToDevice);
-
-	// @@ Check for errors.
-
-	// @@ Free allocated memory.
-}
+#include "Bitmaps.h"	// @@ Rename to BitmapTable.h
 
 // Convert linear image to block linear.
 static void convertToBlockLinear(const Image * image, uint * blockLinearImage)
@@ -92,53 +78,85 @@ static void convertToBlockLinear(const Image * image, uint * blockLinearImage)
 	}
 }
 
-#endif // defined HAVE_CUDA
+#endif
+
+
+CudaCompressor::CudaCompressor() : m_bitmapTable(NULL), m_data(NULL), m_result(NULL)
+{
+#if defined HAVE_CUDA
+    // Allocate and upload bitmaps.
+    cudaMalloc((void**) &m_bitmapTable, 992 * sizeof(uint));
+	if (m_bitmapTable != NULL)
+	{
+		cudaMemcpy(m_bitmapTable, s_bitmapTable, 992 * sizeof(uint), cudaMemcpyHostToDevice);
+	}
+
+	// Allocate scratch buffers.
+    cudaMalloc((void**) &m_data, MAX_BLOCKS * 64U);
+    cudaMalloc((void**) &m_result, MAX_BLOCKS * 8U);
+#endif
+}
+
+CudaCompressor::~CudaCompressor()
+{
+#if defined HAVE_CUDA
+	// Free device mem allocations.
+	cudaFree(m_data);
+	cudaFree(m_result);
+	cudaFree(m_bitmapTable);
+#endif
+}
+
+bool CudaCompressor::isValid() const
+{
+#if defined HAVE_CUDA
+	if (cudaGetLastError() != cudaSuccess)
+   	{
+		return false;
+	}
+#endif
+	return m_data != NULL && m_result != NULL && m_bitmapTable != NULL;
+}
 
 // @@ This code is very repetitive and needs to be cleaned up.
 
+void CudaCompressor::setImage(const Image * image, nvtt::AlphaMode alphaMode)
+{
+	m_image = image;
+	m_alphaMode = alphaMode;
+}
 
 /// Compress image using CUDA.
-void nv::cudaCompressDXT1(const Image * image, const OutputOptions::Private & outputOptions, const CompressionOptions::Private & compressionOptions)
+void CudaCompressor::compressDXT1(const CompressionOptions::Private & compressionOptions, const OutputOptions::Private & outputOptions)
 {
 	nvDebugCheck(cuda::isHardwarePresent());
 #if defined HAVE_CUDA
 
-	doPrecomputation();
-
 	// Image size in blocks.
-	const uint w = (image->width() + 3) / 4;
-	const uint h = (image->height() + 3) / 4;
+	const uint w = (m_image->width() + 3) / 4;
+	const uint h = (m_image->height() + 3) / 4;
 
 	uint imageSize = w * h * 16 * sizeof(Color32);
     uint * blockLinearImage = (uint *) malloc(imageSize);
-	convertToBlockLinear(image, blockLinearImage);	// @@ Do this on the GPU!
+	convertToBlockLinear(m_image, blockLinearImage);	// @@ Do this in parallel with the GPU, or in the GPU!
 
 	const uint blockNum = w * h;
 	const uint compressedSize = blockNum * 8;
-	const uint blockMax = 32768; // 49152, 65535
 
 	clock_t start = clock();
 
-    // Allocate image in device memory.
-    uint * d_data = NULL;
-    cudaMalloc((void**) &d_data, min(imageSize, blockMax * 64U));
-
-	// Allocate result.
-    uint * d_result = NULL;
-    cudaMalloc((void**) &d_result, min(compressedSize, blockMax * 8U));
-
 	setupCompressKernel(compressionOptions.colorWeight.ptr());
 	
 	// TODO: Add support for multiple GPUs.
 	uint bn = 0;
 	while(bn != blockNum)
 	{
-		uint count = min(blockNum - bn, blockMax);
+		uint count = min(blockNum - bn, MAX_BLOCKS);
 
-	    cudaMemcpy(d_data, blockLinearImage + bn * 16, count * 64, cudaMemcpyHostToDevice);
+	    cudaMemcpy(m_data, blockLinearImage + bn * 16, count * 64, cudaMemcpyHostToDevice);
 
 		// Launch kernel.
-		compressKernel(count, d_data, d_result, d_bitmaps);
+		compressKernelDXT1(count, m_data, m_result, m_bitmapTable);
 
 		// Check for errors.
 		cudaError_t err = cudaGetLastError();
@@ -153,7 +171,7 @@ void nv::cudaCompressDXT1(const Image * image, const OutputOptions::Private & ou
 		}
 
 		// Copy result to host, overwrite swizzled image.
-		cudaMemcpy(blockLinearImage, d_result, count * 8, cudaMemcpyDeviceToHost);
+		cudaMemcpy(blockLinearImage, m_result, count * 8, cudaMemcpyDeviceToHost);
 
 		// Output result.
 		if (outputOptions.outputHandler != NULL)
@@ -165,11 +183,9 @@ void nv::cudaCompressDXT1(const Image * image, const OutputOptions::Private & ou
 	}
 
 	clock_t end = clock();
-	printf("\rCUDA time taken: %.3f seconds\n", float(end-start) / CLOCKS_PER_SEC);
+	//printf("\rCUDA time taken: %.3f seconds\n", float(end-start) / CLOCKS_PER_SEC);
 
 	free(blockLinearImage);
-	cudaFree(d_data);
-	cudaFree(d_result);
 
 #else
 	if (outputOptions.errorHandler != NULL)
@@ -181,35 +197,24 @@ void nv::cudaCompressDXT1(const Image * image, const OutputOptions::Private & ou
 
 
 /// Compress image using CUDA.
-void nv::cudaCompressDXT3(const Image * image, const OutputOptions::Private & outputOptions, const CompressionOptions::Private & compressionOptions)
+void CudaCompressor::compressDXT3(const CompressionOptions::Private & compressionOptions, const OutputOptions::Private & outputOptions)
 {
 	nvDebugCheck(cuda::isHardwarePresent());
 #if defined HAVE_CUDA
 
-	doPrecomputation();
-
 	// Image size in blocks.
-	const uint w = (image->width() + 3) / 4;
-	const uint h = (image->height() + 3) / 4;
+	const uint w = (m_image->width() + 3) / 4;
+	const uint h = (m_image->height() + 3) / 4;
 
 	uint imageSize = w * h * 16 * sizeof(Color32);
     uint * blockLinearImage = (uint *) malloc(imageSize);
-	convertToBlockLinear(image, blockLinearImage);
+	convertToBlockLinear(m_image, blockLinearImage);
 
 	const uint blockNum = w * h;
 	const uint compressedSize = blockNum * 8;
-	const uint blockMax = 32768; // 49152, 65535
-
-    // Allocate image in device memory.
-    uint * d_data = NULL;
-    cudaMalloc((void**) &d_data, min(imageSize, blockMax * 64U));
-
-	// Allocate result.
-    uint * d_result = NULL;
-    cudaMalloc((void**) &d_result, min(compressedSize, blockMax * 8U));
 
 	AlphaBlockDXT3 * alphaBlocks = NULL;
-	alphaBlocks = (AlphaBlockDXT3 *)malloc(min(compressedSize, blockMax * 8U));
+	alphaBlocks = (AlphaBlockDXT3 *)malloc(min(compressedSize, MAX_BLOCKS * 8U));
 
 	setupCompressKernel(compressionOptions.colorWeight.ptr());
 	
@@ -218,18 +223,25 @@ void nv::cudaCompressDXT3(const Image * image, const OutputOptions::Private & ou
 	uint bn = 0;
 	while(bn != blockNum)
 	{
-		uint count = min(blockNum - bn, blockMax);
+		uint count = min(blockNum - bn, MAX_BLOCKS);
 
-	    cudaMemcpy(d_data, blockLinearImage + bn * 16, count * 64, cudaMemcpyHostToDevice);
+	    cudaMemcpy(m_data, blockLinearImage + bn * 16, count * 64, cudaMemcpyHostToDevice);
 
 		// Launch kernel.
-		compressWeightedKernel(count, d_data, d_result, d_bitmaps);
+		if (m_alphaMode == AlphaMode_Transparency)
+		{
+			compressWeightedKernelDXT1(count, m_data, m_result, m_bitmapTable);
+		}
+		else
+		{
+			compressKernelDXT1_Level4(count, m_data, m_result, m_bitmapTable);
+		}
 
 		// Compress alpha in parallel with the GPU.
 		for (uint i = 0; i < count; i++)
 		{
 			ColorBlock rgba(blockLinearImage + (bn + i) * 16);
-			compressBlock(rgba, alphaBlocks + i);
+			OptimalCompress::compressDXT3A(rgba, alphaBlocks + i);
 		}
 
 		// Check for errors.
@@ -245,7 +257,7 @@ void nv::cudaCompressDXT3(const Image * image, const OutputOptions::Private & ou
 		}
 
 		// Copy result to host, overwrite swizzled image.
-		cudaMemcpy(blockLinearImage, d_result, count * 8, cudaMemcpyDeviceToHost);
+		cudaMemcpy(blockLinearImage, m_result, count * 8, cudaMemcpyDeviceToHost);
 
 		// Output result.
 		if (outputOptions.outputHandler != NULL)
@@ -261,12 +273,10 @@ void nv::cudaCompressDXT3(const Image * image, const OutputOptions::Private & ou
 	}
 
 	clock_t end = clock();
-	printf("\rCUDA time taken: %.3f seconds\n", float(end-start) / CLOCKS_PER_SEC);
+	//printf("\rCUDA time taken: %.3f seconds\n", float(end-start) / CLOCKS_PER_SEC);
 
 	free(alphaBlocks);
 	free(blockLinearImage);
-	cudaFree(d_data);
-	cudaFree(d_result);
 
 #else
 	if (outputOptions.errorHandler != NULL)
@@ -278,35 +288,24 @@ void nv::cudaCompressDXT3(const Image * image, const OutputOptions::Private & ou
 
 
 /// Compress image using CUDA.
-void nv::cudaCompressDXT5(const Image * image, const OutputOptions::Private & outputOptions, const CompressionOptions::Private & compressionOptions)
+void CudaCompressor::compressDXT5(const CompressionOptions::Private & compressionOptions, const OutputOptions::Private & outputOptions)
 {
 	nvDebugCheck(cuda::isHardwarePresent());
 #if defined HAVE_CUDA
 
-	doPrecomputation();
-
 	// Image size in blocks.
-	const uint w = (image->width() + 3) / 4;
-	const uint h = (image->height() + 3) / 4;
+	const uint w = (m_image->width() + 3) / 4;
+	const uint h = (m_image->height() + 3) / 4;
 
 	uint imageSize = w * h * 16 * sizeof(Color32);
     uint * blockLinearImage = (uint *) malloc(imageSize);
-	convertToBlockLinear(image, blockLinearImage);
+	convertToBlockLinear(m_image, blockLinearImage);
 
 	const uint blockNum = w * h;
 	const uint compressedSize = blockNum * 8;
-	const uint blockMax = 32768; // 49152, 65535
-
-    // Allocate image in device memory.
-    uint * d_data = NULL;
-    cudaMalloc((void**) &d_data, min(imageSize, blockMax * 64U));
-
-	// Allocate result.
-    uint * d_result = NULL;
-    cudaMalloc((void**) &d_result, min(compressedSize, blockMax * 8U));
 
 	AlphaBlockDXT5 * alphaBlocks = NULL;
-	alphaBlocks = (AlphaBlockDXT5 *)malloc(min(compressedSize, blockMax * 8U));
+	alphaBlocks = (AlphaBlockDXT5 *)malloc(min(compressedSize, MAX_BLOCKS * 8U));
 
 	setupCompressKernel(compressionOptions.colorWeight.ptr());
 	
@@ -315,18 +314,25 @@ void nv::cudaCompressDXT5(const Image * image, const OutputOptions::Private & ou
 	uint bn = 0;
 	while(bn != blockNum)
 	{
-		uint count = min(blockNum - bn, blockMax);
+		uint count = min(blockNum - bn, MAX_BLOCKS);
 
-	    cudaMemcpy(d_data, blockLinearImage + bn * 16, count * 64, cudaMemcpyHostToDevice);
+	    cudaMemcpy(m_data, blockLinearImage + bn * 16, count * 64, cudaMemcpyHostToDevice);
 
 		// Launch kernel.
-		compressWeightedKernel(count, d_data, d_result, d_bitmaps);
+		if (m_alphaMode == AlphaMode_Transparency)
+		{
+			compressWeightedKernelDXT1(count, m_data, m_result, m_bitmapTable);
+		}
+		else
+		{
+			compressKernelDXT1_Level4(count, m_data, m_result, m_bitmapTable);
+		}
 
 		// Compress alpha in parallel with the GPU.
 		for (uint i = 0; i < count; i++)
 		{
 			ColorBlock rgba(blockLinearImage + (bn + i) * 16);
-			compressBlock_Iterative(rgba, alphaBlocks + i);
+			QuickCompress::compressDXT5A(rgba, alphaBlocks + i);
 		}
 
 		// Check for errors.
@@ -342,7 +348,7 @@ void nv::cudaCompressDXT5(const Image * image, const OutputOptions::Private & ou
 		}
 
 		// Copy result to host, overwrite swizzled image.
-		cudaMemcpy(blockLinearImage, d_result, count * 8, cudaMemcpyDeviceToHost);
+		cudaMemcpy(blockLinearImage, m_result, count * 8, cudaMemcpyDeviceToHost);
 
 		// Output result.
 		if (outputOptions.outputHandler != NULL)
@@ -358,198 +364,10 @@ void nv::cudaCompressDXT5(const Image * image, const OutputOptions::Private & ou
 	}
 
 	clock_t end = clock();
-	printf("\rCUDA time taken: %.3f seconds\n", float(end-start) / CLOCKS_PER_SEC);
+	//printf("\rCUDA time taken: %.3f seconds\n", float(end-start) / CLOCKS_PER_SEC);
 
 	free(alphaBlocks);
 	free(blockLinearImage);
-	cudaFree(d_data);
-	cudaFree(d_result);
-
-#else
-	if (outputOptions.errorHandler != NULL)
-	{
-		outputOptions.errorHandler->error(Error_CudaError);
-	}
-#endif
-}
-
-
-
-#if defined HAVE_CUDA
-
-class Task
-{
-public:
-	explicit Task(uint numBlocks) : blockMaxCount(numBlocks), blockCount(0)
-	{
-		// System memory allocations.
-		blockLinearImage = new uint[blockMaxCount * 16];
-		xrefs = new uint[blockMaxCount * 16];
-		
-		// Device memory allocations.
-		cudaMalloc((void**) &d_blockLinearImage, blockMaxCount * 16 * sizeof(uint));
-		cudaMalloc((void**) &d_compressedImage, blockMaxCount * 8U);
-		
-		// @@ Check for allocation errors.
-	}
-	
-	~Task()
-	{
-		delete [] blockLinearImage;
-		delete [] xrefs;
-		
-		cudaFree(d_blockLinearImage);
-		cudaFree(d_compressedImage);
-	}
-	
-	
-	
-	void addColorBlock(const ColorBlock & rgba)
-	{
-		nvDebugCheck(!isFull());
-		
-		// @@ Count unique colors?
-		/*
-		// Convert colors to vectors.
-		Array<Vector3> pointArray(16);
-		
-		for(int i = 0; i < 16; i++) {
-			const Color32 color = rgba.color(i);
-			pointArray.append(Vector3(color.r, color.g, color.b));
-		}
-		
-		// Find best fit line.
-		const Vector3 axis = Fit::bestLine(pointArray).direction();
-		
-		// Project points to axis.
-		float dps[16];
-		uint * order = &xrefs[blockCount * 16];
-		
-		for (uint i = 0; i < 16; ++i)
-		{
-			dps[i] = dot(pointArray[i], axis);
-			order[i] = i;
-		}
-		
-		// Sort them.
-		for (uint i = 0; i < 16; ++i)
-		{
-			for (uint j = i; j > 0 && dps[j] < dps[j - 1]; --j)
-			{
-				swap(dps[j], dps[j - 1]);
-				swap(order[j], order[j - 1]);
-			}
-		}
-		*/
-		// Write sorted colors to blockLinearImage.
-		for(uint i = 0; i < 16; ++i)
-		{
-		//	blockLinearImage[blockCount * 16 + i] = rgba.color(order[i]);
-			blockLinearImage[blockCount * 16 + i] = rgba.color(i);
-		}
-		
-		++blockCount;
-	}
-	
-	bool isFull()
-	{
-		nvDebugCheck(blockCount <= blockMaxCount);
-		return blockCount == blockMaxCount;
-	}
-	
-	void flush(const OutputOptions::Private & outputOptions)
-	{
-		if (blockCount == 0)
-		{
-			// Nothing to do.
-			return;
-		}
-		
-		// Copy input color blocks.
-		cudaMemcpy(d_blockLinearImage, blockLinearImage, blockCount * 64, cudaMemcpyHostToDevice);
-		
-		// Launch kernel.
-		compressKernel(blockCount, d_blockLinearImage, d_compressedImage, d_bitmaps);
-		
-		// Check for errors.
-		cudaError_t err = cudaGetLastError();
-		if (err != cudaSuccess)
-		{
-			nvDebug("CUDA Error: %s\n", cudaGetErrorString(err));
-			
-			if (outputOptions.errorHandler != NULL)
-			{
-				outputOptions.errorHandler->error(Error_CudaError);
-			}
-		}
-		
-		// Copy result to host, overwrite swizzled image.
-		uint * compressedImage = blockLinearImage;
-		cudaMemcpy(compressedImage, d_compressedImage, blockCount * 8, cudaMemcpyDeviceToHost);
-		
-		// @@ Sort block indices.
-		
-		// Output result.
-		if (outputOptions.outputHandler != NULL)
-		{
-		//	outputOptions.outputHandler->writeData(compressedImage, blockCount * 8);
-		}
-
-		blockCount = 0;
-	}
-	
-private:
-	
-	const uint blockMaxCount;
-	uint blockCount;
-	
-	uint * blockLinearImage;
-	uint * xrefs;
-	
-	uint * d_blockLinearImage;
-	uint * d_compressedImage;
-	
-};
-
-#endif // defined HAVE_CUDA
-
-
-void nv::cudaCompressDXT1_2(const Image * image, const OutputOptions::Private & outputOptions, const CompressionOptions::Private & compressionOptions)
-{
-#if defined HAVE_CUDA	
-	const uint w = image->width();
-	const uint h = image->height();
-	
-	const uint blockNum = ((w + 3) / 4) * ((h + 3) / 4);
-	const uint blockMax = 32768; // 49152, 65535
-	
-	doPrecomputation();
-	
-	setupCompressKernel(compressionOptions.colorWeight.ptr());
-
-	ColorBlock rgba;
-	Task task(min(blockNum, blockMax));
-
-	clock_t start = clock();
-
-	for (uint y = 0; y < h; y += 4) {
-		for (uint x = 0; x < w; x += 4) {
-			
-			rgba.init(image, x, y);
-			
-			task.addColorBlock(rgba);
-			
-			if (task.isFull())
-			{
-				task.flush(outputOptions);
-			}
-		}
-	}
-	
-	task.flush(outputOptions);
-
-	clock_t end = clock();
-	printf("\rCUDA time taken: %.3f seconds\n", float(end-start) / CLOCKS_PER_SEC);
 
 #else
 	if (outputOptions.errorHandler != NULL)

src/nvtt/cuda/CudaCompressDXT.h

@@ -31,11 +31,29 @@ namespace nv
 {
 	class Image;
 
-	void cudaCompressDXT1(const Image * image, const nvtt::OutputOptions::Private & outputOptions, const nvtt::CompressionOptions::Private & compressionOptions);
-	void cudaCompressDXT3(const Image * image, const nvtt::OutputOptions::Private & outputOptions, const nvtt::CompressionOptions::Private & compressionOptions);
-	void cudaCompressDXT5(const Image * image, const nvtt::OutputOptions::Private & outputOptions, const nvtt::CompressionOptions::Private & compressionOptions);
+	class CudaCompressor
+	{
+	public:
+		CudaCompressor();
+		~CudaCompressor();
 
-	void cudaCompressDXT1_2(const Image * image, const nvtt::OutputOptions::Private & outputOptions, const nvtt::CompressionOptions::Private & compressionOptions);
+		bool isValid() const;
+
+		void setImage(const Image * image, nvtt::AlphaMode alphaMode);
+
+		void compressDXT1(const nvtt::CompressionOptions::Private & compressionOptions, const nvtt::OutputOptions::Private & outputOptions);
+		void compressDXT3(const nvtt::CompressionOptions::Private & compressionOptions, const nvtt::OutputOptions::Private & outputOptions);
+		void compressDXT5(const nvtt::CompressionOptions::Private & compressionOptions, const nvtt::OutputOptions::Private & outputOptions);
+
+	private:
+
+		uint * m_bitmapTable;
+		uint * m_data;
+		uint * m_result;
+		
+		const Image * m_image;
+		nvtt::AlphaMode m_alphaMode;
+	};
 
 } // nv namespace
 

src/nvtt/cuda/CudaMath.h

@@ -82,6 +82,10 @@ inline __device__ __host__ void operator /=(float3 & b, float f)
     b.z *= inv;
 }
 
+inline __device__ __host__ bool operator ==(float3 a, float3 b)
+{
+	return a.x == b.x && a.y == b.y && a.z == b.z;
+}
 
 inline __device__ __host__ float dot(float3 a, float3 b)
 {
@@ -131,15 +135,37 @@ inline __device__ __host__ float3 firstEigenVector( float matrix[6] )
 		float z = v.x * matrix[2] + v.y * matrix[4] + v.z * matrix[5];
 		float m = max(max(x, y), z);        
 		float iv = 1.0f / m;
-		#if __DEVICE_EMULATION__
 		if (m == 0.0f) iv = 0.0f;
-		#endif
 		v = make_float3(x*iv, y*iv, z*iv);
 	}
 
 	return v;
 }
 
+inline __device__ bool singleColor(const float3 * colors)
+{
+#if __DEVICE_EMULATION__
+	bool sameColor = false;
+	for (int i = 0; i < 16; i++)
+	{
+		sameColor &= (colors[idx] == colors[0]);
+	}
+	return sameColor;
+#else
+	__shared__ int sameColor[16];
+	
+	const int idx = threadIdx.x;
+	
+	sameColor[idx] = (colors[idx] == colors[0]);
+	sameColor[idx] &= sameColor[idx^8];
+	sameColor[idx] &= sameColor[idx^4];
+	sameColor[idx] &= sameColor[idx^2];
+	sameColor[idx] &= sameColor[idx^1];
+	
+	return sameColor[0];
+#endif
+}
+
 inline __device__ void colorSums(const float3 * colors, float3 * sums)
 {
 #if __DEVICE_EMULATION__

src/nvtt/cuda/CudaUtils.cpp

@@ -74,7 +74,7 @@ bool nv::cuda::isHardwarePresent()
 {
 #if defined HAVE_CUDA
 #if NV_OS_WIN32
-	if (isWindowsVista()) return false;
+	//if (isWindowsVista()) return false;
 	//if (isWindowsVista() || !isWow32()) return false;
 #endif
 	int count = deviceCount();

src/nvtt/nvtt.h

@@ -49,6 +49,14 @@
 
 #define NVTT_VERSION 200
 
+#define NVTT_DECLARE_PIMPL(Class) \
+	private: \
+		Class(const Class &); \
+		void operator=(const Class &); \
+	public: \
+		struct Private; \
+		Private & m
+
 
 // Public interface.
 namespace nvtt
@@ -89,6 +97,8 @@ namespace nvtt
 	/// Compression options. This class describes the desired compression format and other compression settings.
 	struct CompressionOptions
 	{
+		NVTT_DECLARE_PIMPL(CompressionOptions);
+
 		NVTT_API CompressionOptions();
 		NVTT_API ~CompressionOptions();
 		
@@ -104,10 +114,6 @@ namespace nvtt
 		NVTT_API void setPixelFormat(unsigned int bitcount, unsigned int rmask, unsigned int gmask, unsigned int bmask, unsigned int amask);
 
 		NVTT_API void setQuantization(bool colorDithering, bool alphaDithering, bool binaryAlpha, int alphaThreshold = 127);
-
-	//private:
-		struct Private;
-		Private & m;
 	};
 
 
@@ -170,6 +176,8 @@ namespace nvtt
 	/// Input options. Specify format and layout of the input texture.
 	struct InputOptions
 	{
+		NVTT_DECLARE_PIMPL(InputOptions);
+
 		NVTT_API InputOptions();
 		NVTT_API ~InputOptions();
 		
@@ -214,10 +222,6 @@ namespace nvtt
 		// Set resizing options.
 		NVTT_API void setMaxExtents(int d);
 		NVTT_API void setRoundMode(RoundMode mode);
-		
-	//private:
-		struct Private;
-		Private & m;
 	};
 	
 	
@@ -258,6 +262,8 @@ namespace nvtt
 	/// the compressor to the user.
 	struct OutputOptions
 	{
+		NVTT_DECLARE_PIMPL(OutputOptions);
+
 		NVTT_API OutputOptions();
 		NVTT_API ~OutputOptions();
 		
@@ -269,16 +275,14 @@ namespace nvtt
 		NVTT_API void setOutputHandler(OutputHandler * outputHandler);
 		NVTT_API void setErrorHandler(ErrorHandler * errorHandler);
 		NVTT_API void setOutputHeader(bool outputHeader);
-		
-	//private:
-		struct Private;
-		Private & m;
 	};
 
 
 	/// Texture compressor.
 	struct Compressor
 	{
+		NVTT_DECLARE_PIMPL(Compressor);
+
 		NVTT_API Compressor();
 		NVTT_API ~Compressor();
 
@@ -290,10 +294,6 @@ namespace nvtt
 		
 		// Estimate the size of compressing the input with the given options.
 		NVTT_API int estimateSize(const InputOptions & inputOptions, const CompressionOptions & compressionOptions) const;
-
-	//private:
-		struct Private;
-		Private & m;
 	};
 	
 	

src/nvtt/nvtt_wrapper.h

@@ -207,7 +207,6 @@ NVTT_API void nvttDestroyCompressionOptions(NvttCompressionOptions * compression
 NVTT_API void nvttSetCompressionOptionsFormat(NvttCompressionOptions * compressionOptions, NvttFormat format);
 NVTT_API void nvttSetCompressionOptionsQuality(NvttCompressionOptions * compressionOptions, NvttQuality quality);
 NVTT_API void nvttSetCompressionOptionsColorWeights(NvttCompressionOptions * compressionOptions, float red, float green, float blue, float alpha);
-NVTT_API void nvttEnableCompressionOptionsCudaCompression(NvttCompressionOptions * compressionOptions, NvttBoolean enable);
 NVTT_API void nvttSetCompressionOptionsPixelFormat(NvttCompressionOptions * compressionOptions, unsigned int bitcount, unsigned int rmask, unsigned int gmask, unsigned int bmask, unsigned int amask);
 NVTT_API void nvttSetCompressionOptionsQuantization(NvttCompressionOptions * compressionOptions, NvttBoolean colorDithering, NvttBoolean alphaDithering, NvttBoolean binaryAlpha, int alphaThreshold);
 

src/nvtt/squish/fastclusterfit.cpp

@@ -29,6 +29,8 @@
 #include "colourblock.h"
 #include <cfloat>
 
+#include "fastclusterlookup.inl"
+
 namespace squish {
 
 FastClusterFit::FastClusterFit()
@@ -97,91 +99,6 @@ void FastClusterFit::SetColourSet( ColourSet const* colours, int flags )
 }
 
 
-struct Precomp {
-	float alpha2_sum;
-	float beta2_sum;
-	float alphabeta_sum;
-	float factor;
-};
-
-static SQUISH_ALIGN_16 Precomp s_threeElement[153];
-static SQUISH_ALIGN_16 Precomp s_fourElement[969];
-
-void FastClusterFit::DoPrecomputation()
-{
-	int i = 0;
-	
-	// Three element clusters:
-	for( int c0 = 0; c0 <= 16; c0++)	// At least two clusters.
-	{
-		for( int c1 = 0; c1 <=  16-c0; c1++)
-		{
-			int c2 = 16 - c0 - c1;
-
-			/*if (c2 == 16) {
-				// a = b = x2 / 16
-				s_threeElement[i].alpha2_sum = 0;
-				s_threeElement[i].beta2_sum = 16;
-				s_threeElement[i].alphabeta_sum = -16;
-				s_threeElement[i].factor = 1.0f / 256.0f;
-			}
-			else if (c0 == 16) {
-				// a = b = x0 / 16
-				s_threeElement[i].alpha2_sum = 16;
-				s_threeElement[i].beta2_sum = 0;
-				s_threeElement[i].alphabeta_sum = -16;
-				s_threeElement[i].factor = 1.0f / 256.0f;
-			}
-			else*/ {
-				s_threeElement[i].alpha2_sum = c0 + c1 * 0.25f;
-				s_threeElement[i].beta2_sum = c2 + c1 * 0.25f;
-				s_threeElement[i].alphabeta_sum = c1 * 0.25f;
-				s_threeElement[i].factor = 1.0f / (s_threeElement[i].alpha2_sum * s_threeElement[i].beta2_sum - s_threeElement[i].alphabeta_sum * s_threeElement[i].alphabeta_sum);
-			}
-			
-			i++;
-		}
-	}
-	//printf("%d three cluster elements\n", i);
-	
-	// Four element clusters:
-	i = 0;
-	for( int c0 = 0; c0 <= 16; c0++)
-	{
-		for( int c1 = 0; c1 <=  16-c0; c1++)
-		{
-			for( int c2 = 0; c2 <=  16-c0-c1; c2++)
-			{
-				int c3 = 16 - c0 - c1 - c2;
-				
-				/*if (c3 == 16) {
-					// a = b = x3 / 16
-					s_fourElement[i].alpha2_sum = 16.0f;
-					s_fourElement[i].beta2_sum = 0.0f;
-					s_fourElement[i].alphabeta_sum = -16.0f;
-					s_fourElement[i].factor = 1.0f / 256.0f;					
-				}
-				else if (c0 == 16) {
-					// a = b = x0 / 16
-					s_fourElement[i].alpha2_sum = 0.0f;
-					s_fourElement[i].beta2_sum = 16.0f;
-					s_fourElement[i].alphabeta_sum = -16.0f;
-					s_fourElement[i].factor = 1.0f / 256.0f;					
-				}
-				else*/ {
-					s_fourElement[i].alpha2_sum = c0 + c1 * (4.0f/9.0f) + c2 * (1.0f/9.0f);
-					s_fourElement[i].beta2_sum = c3 + c2 * (4.0f/9.0f) + c1 * (1.0f/9.0f);
-					s_fourElement[i].alphabeta_sum = (c1 + c2) * (2.0f/9.0f);
-					s_fourElement[i].factor = 1.0f / (s_fourElement[i].alpha2_sum * s_fourElement[i].beta2_sum - s_fourElement[i].alphabeta_sum * s_fourElement[i].alphabeta_sum);
-				}
-
-				i++;
-			}
-		}
-	}
-	//printf("%d four cluster elements\n", i);
-}
-
 void FastClusterFit::SetMetric(float r, float g, float b)
 {
 #if SQUISH_USE_SIMD

src/nvtt/squish/fastclusterfit.h

@@ -44,8 +44,6 @@ public:
 	void SetMetric(float r, float g, float b);
 	float GetBestError() const;
 
-	static void DoPrecomputation();
-
 	// Make them public
 	virtual void Compress3( void* block );
 	virtual void Compress4( void* block );

src/nvtt/squish/simd_ve.h

@@ -50,6 +50,16 @@ public:
 		return *this;
 	}
 	
+	Vec4( const float * v )
+	{
+		union { vector float v; float c[4]; } u;
+		u.c[0] = v[0];
+		u.c[1] = v[1];
+		u.c[2] = v[2];
+		u.c[3] = v[3];
+		m_v = u.v;
+	}
+
 	Vec4( float x, float y, float z, float w )
 	{
 		union { vector float v; float c[4]; } u;

src/nvtt/tools/compress.cpp

@@ -42,11 +42,11 @@ struct MyOutputHandler : public nvtt::OutputHandler
 	MyOutputHandler(const char * name) : total(0), progress(0), percentage(0), stream(new nv::StdOutputStream(name)) {}
 	virtual ~MyOutputHandler() { delete stream; }
 	
-	virtual void setTotal(int64 t)
+	void setTotal(int64 t)
 	{
 		total = t + 128;
 	}
-	virtual void setDisplayProgress(bool b)
+	void setDisplayProgress(bool b)
 	{
 		verbose = b;
 	}
@@ -373,7 +373,6 @@ int main(int argc, char *argv[])
 		inputOptions.setMipmapGeneration(false);
 	}
 
-
 	nvtt::CompressionOptions compressionOptions;
 	compressionOptions.setFormat(format);
 	if (fast)
@@ -397,6 +396,21 @@ int main(int argc, char *argv[])
 		compressionOptions.setExternalCompressor(externalCompressor);
 	}
 
+	if (format == nvtt::Format_RGB)
+	{
+		compressionOptions.setQuantization(true, false, false);
+		//compressionOptions.setPixelFormat(16, 0xF000, 0x0F00, 0x00F0, 0x000F);
+		compressionOptions.setPixelFormat(16, 
+			0x0F00, 
+			0x00F0, 
+			0x000F, 
+			0xF000);
+		//	0x003F0000, 
+		//	0x00003F00, 
+		//	0x0000003F, 
+		//	0x3F000000);
+	}
+
 	
 	MyErrorHandler errorHandler;
 	MyOutputHandler outputHandler(output);

src/nvtt/tools/imgdiff.cpp

@@ -130,10 +130,13 @@ struct NormalError
 
 	void done()
 	{
-		ade /= samples;
-		mse /= samples * 3;
-		rmse = sqrt(mse);
-		psnr = (rmse == 0) ? 999.0f : 20.0f * log10(255.0f / rmse);
+		if (samples)
+		{
+			ade /= samples;
+			mse /= samples * 3;
+			rmse = sqrt(mse);
+			psnr = (rmse == 0) ? 999.0f : 20.0f * log10(255.0f / rmse);
+		}
 	}
 
 	void print()

src/nvtt/tools/resize.cpp

@@ -73,10 +73,12 @@ int main(int argc, char *argv[])
 
 	float scale = 0.5f;
 	float gamma = 2.2f;
-	nv::Filter * filter = NULL;
+	nv::AutoPtr<nv::Filter> filter;
 	nv::Path input;
 	nv::Path output;
 	
+	nv::FloatImage::WrapMode wrapMode = nv::FloatImage::WrapMode_Mirror;
+
 	// Parse arguments.
 	for (int i = 1; i < argc; i++)
 	{
@@ -108,9 +110,18 @@ int main(int argc, char *argv[])
 			else if (strcmp("lanczos", argv[i]) == 0) filter = new nv::LanczosFilter();
 			else if (strcmp("kaiser", argv[i]) == 0) {
 				filter = new nv::KaiserFilter(3);
-				((nv::KaiserFilter *)filter)->setParameters(4.0f, 1.0f);
+				((nv::KaiserFilter *)filter.ptr())->setParameters(4.0f, 1.0f);
 			}
 		}
+		else if (strcmp("-f", argv[i]) == 0)
+		{
+			if (i+1 == argc) break;
+			i++;
+
+			if (strcmp("mirror", argv[i]) == 0) wrapMode = nv::FloatImage::WrapMode_Mirror;
+			else if (strcmp("repeat", argv[i]) == 0) wrapMode = nv::FloatImage::WrapMode_Repeat;
+			else if (strcmp("clamp", argv[i]) == 0) wrapMode = nv::FloatImage::WrapMode_Clamp;
+		}
 		else if (argv[i][0] != '-')
 		{
 			input = argv[i];
@@ -140,6 +151,10 @@ int main(int argc, char *argv[])
 		printf("                * mitchell\n");
 		printf("                * lanczos\n");
 		printf("                * kaiser\n");
+		printf("  -w mode      One of the following: (default = 'mirror')\n");
+		printf("                * mirror\n");
+		printf("                * repeat\n");
+		printf("                * clamp\n");
 
 		return 1;
 	}
@@ -155,15 +170,14 @@ int main(int argc, char *argv[])
 	nv::FloatImage fimage(&image);
 	fimage.toLinear(0, 3, gamma);
 	
-	nv::AutoPtr<nv::FloatImage> fresult(fimage.downSample(*filter, uint(image.width() * scale), uint(image.height() * scale), nv::FloatImage::WrapMode_Mirror));
+	nv::AutoPtr<nv::FloatImage> fresult(fimage.resize(*filter, uint(image.width() * scale), uint(image.height() * scale), wrapMode));
 	
 	nv::AutoPtr<nv::Image> result(fresult->createImageGammaCorrect(gamma));
+	result->setFormat(nv::Image::Format_ARGB);
 
 	nv::StdOutputStream stream(output);
 	nv::ImageIO::saveTGA(stream, result.ptr());	// @@ Add generic save function. Add support for png too.
 	
-	delete filter;
-	
 	return 0;
 }