set svn:eol-style=native for **.cpp

git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@1442 8ced0084-cf51-0410-be5f-012b33b47a6e

12 files changed, 3494 insertions, 3494 deletions

diff --git a/Source/Core/VideoCommon/Src/BPMemory.cpp b/Source/Core/VideoCommon/Src/BPMemory.cpp
index da7083da4c..4fdbf25306 100644
--- a/Source/Core/VideoCommon/Src/BPMemory.cpp
+++ b/Source/Core/VideoCommon/Src/BPMemory.cpp
@@ -1,24 +1,24 @@
-// Copyright (C) 2003-2008 Dolphin Project.

-

-// This program is free software: you can redistribute it and/or modify

-// it under the terms of the GNU General Public License as published by

-// the Free Software Foundation, version 2.0.

-

-// This program is distributed in the hope that it will be useful,

-// but WITHOUT ANY WARRANTY; without even the implied warranty of

-// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

-// GNU General Public License 2.0 for more details.

-

-// A copy of the GPL 2.0 should have been included with the program.

-// If not, see http://www.gnu.org/licenses/

-

-// Official SVN repository and contact information can be found at

-// http://code.google.com/p/dolphin-emu/

-

-#include "Common.h"

-

-#include "BPMemory.h"

-

-//BP state

-// STATE_TO_SAVE

+// Copyright (C) 2003-2008 Dolphin Project.
+
+// This program is free software: you can redistribute it and/or modify
+// it under the terms of the GNU General Public License as published by
+// the Free Software Foundation, version 2.0.
+
+// This program is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU General Public License 2.0 for more details.
+
+// A copy of the GPL 2.0 should have been included with the program.
+// If not, see http://www.gnu.org/licenses/
+
+// Official SVN repository and contact information can be found at
+// http://code.google.com/p/dolphin-emu/
+
+#include "Common.h"
+
+#include "BPMemory.h"
+
+//BP state
+// STATE_TO_SAVE
 BPMemory bpmem;
diff --git a/Source/Core/VideoCommon/Src/CPMemory.cpp b/Source/Core/VideoCommon/Src/CPMemory.cpp
index 837dde70ac..32cd4f0b32 100644
--- a/Source/Core/VideoCommon/Src/CPMemory.cpp
+++ b/Source/Core/VideoCommon/Src/CPMemory.cpp
@@ -1,29 +1,29 @@
-// Copyright (C) 2003-2008 Dolphin Project.

-

-// This program is free software: you can redistribute it and/or modify

-// it under the terms of the GNU General Public License as published by

-// the Free Software Foundation, version 2.0.

-

-// This program is distributed in the hope that it will be useful,

-// but WITHOUT ANY WARRANTY; without even the implied warranty of

-// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

-// GNU General Public License 2.0 for more details.

-

-// A copy of the GPL 2.0 should have been included with the program.

-// If not, see http://www.gnu.org/licenses/

-

-// Official SVN repository and contact information can be found at

-// http://code.google.com/p/dolphin-emu/

-

-#include "Common.h"

-#include "CPMemory.h"

-

-// CP state

-// STATE_TO_SAVE

-u32 arraybases[16];

-u32 arraystrides[16];

-TMatrixIndexA MatrixIndexA;

-TMatrixIndexB MatrixIndexB;

-TVtxDesc g_VtxDesc;

-// Most games only use the first VtxAttr and simply reconfigure it all the time as needed.

-VAT g_VtxAttr[8];

+// Copyright (C) 2003-2008 Dolphin Project.
+
+// This program is free software: you can redistribute it and/or modify
+// it under the terms of the GNU General Public License as published by
+// the Free Software Foundation, version 2.0.
+
+// This program is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU General Public License 2.0 for more details.
+
+// A copy of the GPL 2.0 should have been included with the program.
+// If not, see http://www.gnu.org/licenses/
+
+// Official SVN repository and contact information can be found at
+// http://code.google.com/p/dolphin-emu/
+
+#include "Common.h"
+#include "CPMemory.h"
+
+// CP state
+// STATE_TO_SAVE
+u32 arraybases[16];
+u32 arraystrides[16];
+TMatrixIndexA MatrixIndexA;
+TMatrixIndexB MatrixIndexB;
+TVtxDesc g_VtxDesc;
+// Most games only use the first VtxAttr and simply reconfigure it all the time as needed.
+VAT g_VtxAttr[8];
diff --git a/Source/Core/VideoCommon/Src/Fifo.cpp b/Source/Core/VideoCommon/Src/Fifo.cpp
index c1a59e4918..8ebe2dc720 100644
--- a/Source/Core/VideoCommon/Src/Fifo.cpp
+++ b/Source/Core/VideoCommon/Src/Fifo.cpp
@@ -1,167 +1,167 @@
-// Copyright (C) 2003-2008 Dolphin Project.

-

-// This program is free software: you can redistribute it and/or modify

-// it under the terms of the GNU General Public License as published by

-// the Free Software Foundation, version 2.0.

-

-// This program is distributed in the hope that it will be useful,

-// but WITHOUT ANY WARRANTY; without even the implied warranty of

-// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

-// GNU General Public License 2.0 for more details.

-

-// A copy of the GPL 2.0 should have been included with the program.

-// If not, see http://www.gnu.org/licenses/

-

-// Official SVN repository and contact information can be found at

-// http://code.google.com/p/dolphin-emu/

-

-#include <string.h>

-

-#include "MemoryUtil.h"

-#include "Thread.h"

-#include "OpcodeDecoding.h"

-

-#include "Fifo.h"

-

-extern u8* g_pVideoData;

-

-bool fifoStateRun = true;

-

-// STATE_TO_SAVE

-static u8 *videoBuffer;

-static int size = 0;

-

-void Fifo_DoState(PointerWrap &p) 

-{

-    p.DoArray(videoBuffer, FIFO_SIZE);

-    p.Do(size);

-	int pos = (int)(g_pVideoData-videoBuffer); // get offset

-	p.Do(pos); // read or write offset (depends on the mode afaik)

-	g_pVideoData = &videoBuffer[pos]; // overwrite g_pVideoData -> expected no change when load ss and change when save ss

-}

-

-void Fifo_Init()

-{

-    videoBuffer = (u8*)AllocateMemoryPages(FIFO_SIZE);

-    fifoStateRun = true;

-}

-

-void Fifo_Shutdown()

-{

-    FreeMemoryPages(videoBuffer, FIFO_SIZE);

-    fifoStateRun = false;

-}

-

-void Fifo_Stop() 

-{

-    fifoStateRun = false;

-}

-

-u8* FAKE_GetFifoStartPtr()

-{

-    return videoBuffer;

-}

-

-u8* FAKE_GetFifoEndPtr()

-{

-	return &videoBuffer[size];

-}

-

-void Video_SendFifoData(u8* _uData, u32 len)

-{

-    if (size + len >= FIFO_SIZE)

-    {

-		int pos = (int)(g_pVideoData-videoBuffer);

-        if (size-pos > pos)

-        {

-            PanicAlert("FIFO out of bounds (sz = %i, at %08x)", size, pos);

-        }

-        memmove(&videoBuffer[0], &videoBuffer[pos], size - pos );

-        size -= pos;

-		g_pVideoData = FAKE_GetFifoStartPtr();

-    }

-    memcpy(videoBuffer + size, _uData, len);

-    size += len;

-    OpcodeDecoder_Run();

-}

-

-void Fifo_EnterLoop(const SVideoInitialize &video_initialize)

-{

-    SCPFifoStruct &_fifo = *video_initialize.pCPFifo;

-	u32 distToSend;

-

-#ifdef _WIN32

-    // TODO(ector): Don't peek so often!

-    while (video_initialize.pPeekMessages())

-#else 

-    while (fifoStateRun)

-#endif

-    {

-        if (_fifo.CPReadWriteDistance < _fifo.CPLoWatermark)

-			Common::SleepCurrentThread(1);

-        //etc...

-

-        // check if we are able to run this buffer

-        if ((_fifo.bFF_GPReadEnable) && _fifo.CPReadWriteDistance && !(_fifo.bFF_BPEnable && _fifo.bFF_Breakpoint))

-        {

-			Common::SyncInterlockedExchange((LONG*)&_fifo.CPReadIdle, 0);

-			//video_initialize.pLog("RUN...........................",FALSE);

-			int peek_counter = 0;

-            while (_fifo.bFF_GPReadEnable && (_fifo.CPReadWriteDistance > 0))

-			{

-				peek_counter++;

-				if (peek_counter == 50) {

-					video_initialize.pPeekMessages();

-					peek_counter = 0;

-				}

-                // read the data and send it to the VideoPlugin

-				u32 readPtr = _fifo.CPReadPointer;

-                u8 *uData = video_initialize.pGetMemoryPointer(readPtr);

-

-				// if we are on BP mode we must send 32B chunks to Video plugin for BP checking

-				// TODO (mb2): test & check if MP1/MP2 realy need this now.

-				if (_fifo.bFF_BPEnable)

-                {

-					if (readPtr == _fifo.CPBreakpoint)

-                    {

-						video_initialize.pLog("!!! BP irq raised",FALSE);

-						Common::SyncInterlockedExchange((LONG*)&_fifo.bFF_Breakpoint, 1);

-

-                        video_initialize.pUpdateInterrupts();

-                        break;

-                    }

-					distToSend = 32;

-					readPtr += 32;

-					if ( readPtr >= _fifo.CPEnd) 

-						readPtr = _fifo.CPBase;

-                }

-				else

-				{

-#if 0 // ugly random GP slowdown for testing DC robustness... TODO: remove when completly sure DC is ok

-				int r=rand();if ((r&0xF)==r) Common::SleepCurrentThread(r);

-				distToSend = 32;

-				readPtr += 32;

-				if ( readPtr >= _fifo.CPEnd) 

-					readPtr = _fifo.CPBase;

-#else

-					// sending the whole CPReadWriteDistance

-					distToSend = _fifo.CPReadWriteDistance;

-					if ( (distToSend+readPtr) >= _fifo.CPEnd) // TODO: better?

-					{

-						distToSend =_fifo.CPEnd - readPtr;

-						readPtr = _fifo.CPBase;

-					}

-					else

-						readPtr += distToSend; 

-#endif

-				}

-				Video_SendFifoData(uData, distToSend);

-                Common::SyncInterlockedExchange((LONG*)&_fifo.CPReadPointer, readPtr);

-                Common::SyncInterlockedExchangeAdd((LONG*)&_fifo.CPReadWriteDistance, -distToSend);

-			}

-			//video_initialize.pLog("..........................IDLE",FALSE);

-			Common::SyncInterlockedExchange((LONG*)&_fifo.CPReadIdle, 1);

-        }

-    }

-}

-

+// Copyright (C) 2003-2008 Dolphin Project.
+
+// This program is free software: you can redistribute it and/or modify
+// it under the terms of the GNU General Public License as published by
+// the Free Software Foundation, version 2.0.
+
+// This program is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU General Public License 2.0 for more details.
+
+// A copy of the GPL 2.0 should have been included with the program.
+// If not, see http://www.gnu.org/licenses/
+
+// Official SVN repository and contact information can be found at
+// http://code.google.com/p/dolphin-emu/
+
+#include <string.h>
+
+#include "MemoryUtil.h"
+#include "Thread.h"
+#include "OpcodeDecoding.h"
+
+#include "Fifo.h"
+
+extern u8* g_pVideoData;
+
+bool fifoStateRun = true;
+
+// STATE_TO_SAVE
+static u8 *videoBuffer;
+static int size = 0;
+
+void Fifo_DoState(PointerWrap &p) 
+{
+    p.DoArray(videoBuffer, FIFO_SIZE);
+    p.Do(size);
+	int pos = (int)(g_pVideoData-videoBuffer); // get offset
+	p.Do(pos); // read or write offset (depends on the mode afaik)
+	g_pVideoData = &videoBuffer[pos]; // overwrite g_pVideoData -> expected no change when load ss and change when save ss
+}
+
+void Fifo_Init()
+{
+    videoBuffer = (u8*)AllocateMemoryPages(FIFO_SIZE);
+    fifoStateRun = true;
+}
+
+void Fifo_Shutdown()
+{
+    FreeMemoryPages(videoBuffer, FIFO_SIZE);
+    fifoStateRun = false;
+}
+
+void Fifo_Stop() 
+{
+    fifoStateRun = false;
+}
+
+u8* FAKE_GetFifoStartPtr()
+{
+    return videoBuffer;
+}
+
+u8* FAKE_GetFifoEndPtr()
+{
+	return &videoBuffer[size];
+}
+
+void Video_SendFifoData(u8* _uData, u32 len)
+{
+    if (size + len >= FIFO_SIZE)
+    {
+		int pos = (int)(g_pVideoData-videoBuffer);
+        if (size-pos > pos)
+        {
+            PanicAlert("FIFO out of bounds (sz = %i, at %08x)", size, pos);
+        }
+        memmove(&videoBuffer[0], &videoBuffer[pos], size - pos );
+        size -= pos;
+		g_pVideoData = FAKE_GetFifoStartPtr();
+    }
+    memcpy(videoBuffer + size, _uData, len);
+    size += len;
+    OpcodeDecoder_Run();
+}
+
+void Fifo_EnterLoop(const SVideoInitialize &video_initialize)
+{
+    SCPFifoStruct &_fifo = *video_initialize.pCPFifo;
+	u32 distToSend;
+
+#ifdef _WIN32
+    // TODO(ector): Don't peek so often!
+    while (video_initialize.pPeekMessages())
+#else 
+    while (fifoStateRun)
+#endif
+    {
+        if (_fifo.CPReadWriteDistance < _fifo.CPLoWatermark)
+			Common::SleepCurrentThread(1);
+        //etc...
+
+        // check if we are able to run this buffer
+        if ((_fifo.bFF_GPReadEnable) && _fifo.CPReadWriteDistance && !(_fifo.bFF_BPEnable && _fifo.bFF_Breakpoint))
+        {
+			Common::SyncInterlockedExchange((LONG*)&_fifo.CPReadIdle, 0);
+			//video_initialize.pLog("RUN...........................",FALSE);
+			int peek_counter = 0;
+            while (_fifo.bFF_GPReadEnable && (_fifo.CPReadWriteDistance > 0))
+			{
+				peek_counter++;
+				if (peek_counter == 50) {
+					video_initialize.pPeekMessages();
+					peek_counter = 0;
+				}
+                // read the data and send it to the VideoPlugin
+				u32 readPtr = _fifo.CPReadPointer;
+                u8 *uData = video_initialize.pGetMemoryPointer(readPtr);
+
+				// if we are on BP mode we must send 32B chunks to Video plugin for BP checking
+				// TODO (mb2): test & check if MP1/MP2 realy need this now.
+				if (_fifo.bFF_BPEnable)
+                {
+					if (readPtr == _fifo.CPBreakpoint)
+                    {
+						video_initialize.pLog("!!! BP irq raised",FALSE);
+						Common::SyncInterlockedExchange((LONG*)&_fifo.bFF_Breakpoint, 1);
+
+                        video_initialize.pUpdateInterrupts();
+                        break;
+                    }
+					distToSend = 32;
+					readPtr += 32;
+					if ( readPtr >= _fifo.CPEnd) 
+						readPtr = _fifo.CPBase;
+                }
+				else
+				{
+#if 0 // ugly random GP slowdown for testing DC robustness... TODO: remove when completly sure DC is ok
+				int r=rand();if ((r&0xF)==r) Common::SleepCurrentThread(r);
+				distToSend = 32;
+				readPtr += 32;
+				if ( readPtr >= _fifo.CPEnd) 
+					readPtr = _fifo.CPBase;
+#else
+					// sending the whole CPReadWriteDistance
+					distToSend = _fifo.CPReadWriteDistance;
+					if ( (distToSend+readPtr) >= _fifo.CPEnd) // TODO: better?
+					{
+						distToSend =_fifo.CPEnd - readPtr;
+						readPtr = _fifo.CPBase;
+					}
+					else
+						readPtr += distToSend; 
+#endif
+				}
+				Video_SendFifoData(uData, distToSend);
+                Common::SyncInterlockedExchange((LONG*)&_fifo.CPReadPointer, readPtr);
+                Common::SyncInterlockedExchangeAdd((LONG*)&_fifo.CPReadWriteDistance, -distToSend);
+			}
+			//video_initialize.pLog("..........................IDLE",FALSE);
+			Common::SyncInterlockedExchange((LONG*)&_fifo.CPReadIdle, 1);
+        }
+    }
+}
+
diff --git a/Source/Core/VideoCommon/Src/LookUpTables.cpp b/Source/Core/VideoCommon/Src/LookUpTables.cpp
index 99453f0ffc..3bb7cc5695 100644
--- a/Source/Core/VideoCommon/Src/LookUpTables.cpp
+++ b/Source/Core/VideoCommon/Src/LookUpTables.cpp
@@ -1,45 +1,45 @@
-// Copyright (C) 2003-2008 Dolphin Project.

-

-// This program is free software: you can redistribute it and/or modify

-// it under the terms of the GNU General Public License as published by

-// the Free Software Foundation, version 2.0.

-

-// This program is distributed in the hope that it will be useful,

-// but WITHOUT ANY WARRANTY; without even the implied warranty of

-// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

-// GNU General Public License 2.0 for more details.

-

-// A copy of the GPL 2.0 should have been included with the program.

-// If not, see http://www.gnu.org/licenses/

-

-// Official SVN repository and contact information can be found at

-// http://code.google.com/p/dolphin-emu/

-

-#include "LookUpTables.h"

-

-const int lut3to8[] = { 0x00,0x24,0x48,0x6D,0x91,0xB6,0xDA,0xFF};

-const int lut4to8[] = { 0x00,0x11,0x22,0x33,0x44,0x55,0x66,0x77,

-                        0x88,0x99,0xAA,0xBB,0xCC,0xDD,0xEE,0xFF};

-const int lut5to8[] = { 0x00,0x08,0x10,0x18,0x20,0x29,0x31,0x39,

-                        0x41,0x4A,0x52,0x5A,0x62,0x6A,0x73,0x7B,

-                        0x83,0x8B,0x94,0x9C,0xA4,0xAC,0xB4,0xBD,

-                        0xC5,0xCD,0xD5,0xDE,0xE6,0xEE,0xF6,0xFF};

-int lut6to8[64];

-float lutu8tosfloat[256];

-float lutu8toufloat[256];

-float luts8tosfloat[256];

-float shiftLookup[32];

-

-void InitLUTs()

-{

-	for (int i = 0; i < 32; i++)

-		shiftLookup[i] = 1.0f / float(1 << i);

-	for (int i = 0; i < 64; i++)

-		lut6to8[i] = (i*255) / 63;

-	for (int i = 0; i < 256; i++)

-	{

-		lutu8tosfloat[i] = (float)(i - 128) / 127.0f;

-		lutu8toufloat[i] = (float)(i) / 255.0f;

-		luts8tosfloat[i] = ((float)(signed char)(char)i) / 127.0f;

-	}

-}

+// Copyright (C) 2003-2008 Dolphin Project.
+
+// This program is free software: you can redistribute it and/or modify
+// it under the terms of the GNU General Public License as published by
+// the Free Software Foundation, version 2.0.
+
+// This program is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU General Public License 2.0 for more details.
+
+// A copy of the GPL 2.0 should have been included with the program.
+// If not, see http://www.gnu.org/licenses/
+
+// Official SVN repository and contact information can be found at
+// http://code.google.com/p/dolphin-emu/
+
+#include "LookUpTables.h"
+
+const int lut3to8[] = { 0x00,0x24,0x48,0x6D,0x91,0xB6,0xDA,0xFF};
+const int lut4to8[] = { 0x00,0x11,0x22,0x33,0x44,0x55,0x66,0x77,
+                        0x88,0x99,0xAA,0xBB,0xCC,0xDD,0xEE,0xFF};
+const int lut5to8[] = { 0x00,0x08,0x10,0x18,0x20,0x29,0x31,0x39,
+                        0x41,0x4A,0x52,0x5A,0x62,0x6A,0x73,0x7B,
+                        0x83,0x8B,0x94,0x9C,0xA4,0xAC,0xB4,0xBD,
+                        0xC5,0xCD,0xD5,0xDE,0xE6,0xEE,0xF6,0xFF};
+int lut6to8[64];
+float lutu8tosfloat[256];
+float lutu8toufloat[256];
+float luts8tosfloat[256];
+float shiftLookup[32];
+
+void InitLUTs()
+{
+	for (int i = 0; i < 32; i++)
+		shiftLookup[i] = 1.0f / float(1 << i);
+	for (int i = 0; i < 64; i++)
+		lut6to8[i] = (i*255) / 63;
+	for (int i = 0; i < 256; i++)
+	{
+		lutu8tosfloat[i] = (float)(i - 128) / 127.0f;
+		lutu8toufloat[i] = (float)(i) / 255.0f;
+		luts8tosfloat[i] = ((float)(signed char)(char)i) / 127.0f;
+	}
+}
diff --git a/Source/Core/VideoCommon/Src/PixelShader.cpp b/Source/Core/VideoCommon/Src/PixelShader.cpp
index 799335735d..c036ef7a83 100644
--- a/Source/Core/VideoCommon/Src/PixelShader.cpp
+++ b/Source/Core/VideoCommon/Src/PixelShader.cpp
@@ -1,846 +1,846 @@
-// Copyright (C) 2003-2008 Dolphin Project.

-

-// This program is free software: you can redistribute it and/or modify

-// it under the terms of the GNU General Public License as published by

-// the Free Software Foundation, version 2.0.

-

-// This program is distributed in the hope that it will be useful,

-// but WITHOUT ANY WARRANTY; without even the implied warranty of

-// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

-// GNU General Public License 2.0 for more details.

-

-// A copy of the GPL 2.0 should have been included with the program.

-// If not, see http://www.gnu.org/licenses/

-

-// Official SVN repository and contact information can be found at

-// http://code.google.com/p/dolphin-emu/

-

-#include <stdio.h>

-#include <cmath>

-#include <assert.h>

-

-#include "Profiler.h"

-#include "PixelShader.h"

-#include "XFMemory.h"  // for texture projection mode

-#include "BPMemory.h"

-

-//   old tev->pixelshader notes

-//

-//   color for this stage (alpha, color) is given by bpmem.tevorders[0].colorchan0

-//   konstant for this stage (alpha, color) is given by bpmem.tevksel   

-//   inputs are given by bpmem.combiners[0].colorC.a/b/c/d     << could be current chan color

-//   according to GXTevColorArg table above

-//   output is given by .outreg

-//   tevtemp is set according to swapmodetables and 

-

-static void WriteStage(char *&p, int n, u32 texture_mask);

-static void WrapNonPow2Tex(char* &p, const char* var, int texmap, u32 texture_mask);

-static void WriteAlphaCompare(char *&p, int num, int comp);

-static bool WriteAlphaTest(char *&p);

-

-const float epsilon8bit = 1.0f / 255.0f;

-

-static const char *tevKSelTableC[] = // KCSEL

-{

-    "1.0f,1.0f,1.0f",            //1   = 0x00

-    "0.875,0.875,0.875",//7_8 = 0x01

-    "0.75,0.75,0.75",	//3_4 = 0x02

-    "0.625,0.625,0.625",//5_8 = 0x03

-    "0.5,0.5,0.5",      //1_2 = 0x04

-    "0.375,0.375,0.375",//3_8 = 0x05

-    "0.25,0.25,0.25",   //1_4 = 0x06

-    "0.125,0.125,0.125",//1_8 = 0x07

-    "ERROR", //0x08

-    "ERROR", //0x09

-    "ERROR", //0x0a

-    "ERROR", //0x0b

-    I_KCOLORS"[0].rgb",//K0 = 0x0C

-    I_KCOLORS"[1].rgb",//K1 = 0x0D

-    I_KCOLORS"[2].rgb",//K2 = 0x0E

-    I_KCOLORS"[3].rgb",//K3 = 0x0F

-    I_KCOLORS"[0].rrr",//K0_R = 0x10

-    I_KCOLORS"[1].rrr",//K1_R = 0x11

-    I_KCOLORS"[2].rrr",//K2_R = 0x12

-    I_KCOLORS"[3].rrr",//K3_R = 0x13

-    I_KCOLORS"[0].ggg",//K0_G = 0x14 

-    I_KCOLORS"[1].ggg",//K1_G = 0x15

-    I_KCOLORS"[2].ggg",//K2_G = 0x16

-    I_KCOLORS"[3].ggg",//K3_G = 0x17

-    I_KCOLORS"[0].bbb",//K0_B = 0x18

-    I_KCOLORS"[1].bbb",//K1_B = 0x19

-    I_KCOLORS"[2].bbb",//K2_B = 0x1A

-    I_KCOLORS"[3].bbb",//K3_B = 0x1B

-    I_KCOLORS"[0].aaa",//K0_A = 0x1C

-    I_KCOLORS"[1].aaa",//K1_A = 0x1D

-    I_KCOLORS"[2].aaa",//K2_A = 0x1E

-    I_KCOLORS"[3].aaa",//K3_A = 0x1F

-};

-

-static const char *tevKSelTableA[] = // KASEL

-{

-    "1.0f",    //1   = 0x00

-    "0.875f",//7_8 = 0x01

-    "0.75f",	//3_4 = 0x02

-    "0.625f",//5_8 = 0x03

-    "0.5f",  //1_2 = 0x04

-    "0.375f",//3_8 = 0x05

-    "0.25f", //1_4 = 0x06

-    "0.125f",//1_8 = 0x07

-    "ERROR", //0x08

-    "ERROR", //0x09

-    "ERROR", //0x0a

-    "ERROR", //0x0b

-    "ERROR", //0x0c

-    "ERROR", //0x0d

-    "ERROR", //0x0e

-    "ERROR", //0x0f

-    I_KCOLORS"[0].r",//K0_R = 0x10

-    I_KCOLORS"[1].r",//K1_R = 0x11

-    I_KCOLORS"[2].r",//K2_R = 0x12

-    I_KCOLORS"[3].r",//K3_R = 0x13

-    I_KCOLORS"[0].g",//K0_G = 0x14

-    I_KCOLORS"[1].g",//K1_G = 0x15

-    I_KCOLORS"[2].g",//K2_G = 0x16

-    I_KCOLORS"[3].g",//K3_G = 0x17

-    I_KCOLORS"[0].b",//K0_B = 0x18

-    I_KCOLORS"[1].b",//K1_B = 0x19

-    I_KCOLORS"[2].b",//K2_B = 0x1A

-    I_KCOLORS"[3].b",//K3_B = 0x1B

-    I_KCOLORS"[0].a",//K0_A = 0x1C

-    I_KCOLORS"[1].a",//K1_A = 0x1D

-    I_KCOLORS"[2].a",//K2_A = 0x1E

-    I_KCOLORS"[3].a",//K3_A = 0x1F

-};

-

-static const char *tevScaleTable[] = // CS

-{

-    "1.0f",   //SCALE_1

-    "2.0f", //SCALE_2

-    "4.0f", //SCALE_4

-    "0.5f",//DIVIDE_2

-};

-

-static const char *tevBiasTable[] = // TB

-{

-    "",      //ZERO,

-    "+0.5f",  //ADDHALF,

-    "-0.5f",  //SUBHALF,

-    "",

-};

-

-static const char *tevOpTable[] = { // TEV

-    "+",      //TEVOP_ADD = 0,

-    "-",      //TEVOP_SUB = 1,

-};

-

-//static const char *tevCompOpTable[] = { ">", "==" };

-

-#define TEVCMP_R8    0

-#define TEVCMP_GR16  1

-#define TEVCMP_BGR24 2

-#define TEVCMP_RGB8  3

-

-static const char *tevCInputTable[] = // CC

-{

-    "prev.rgb",         //CPREV,

-    "prev.aaa",           //APREV,

-    "c0.rgb",           //C0,

-    "c0.aaa",             //A0,

-    "c1.rgb",           //C1,

-    "c1.aaa",             //A1,

-    "c2.rgb",           //C2,

-    "c2.aaa",             //A2,

-    "textemp.rgb",      //TEXC,

-    "textemp.aaa",        //TEXA,

-    "rastemp.rgb",    //RASC,

-    "rastemp.aaa",      //RASA,

-    "float3(1.0f,1.0f,1.0f)",              //ONE,

-    "float3(.5f,.5f,.5f)",              //HALF,

-    "konsttemp.rgb",    //KONST,

-    "float3(0.0f,0.0f,0.0f)",              //ZERO

-    "PADERROR",	"PADERROR",	"PADERROR",	"PADERROR",

-    "PADERROR",	"PADERROR",	"PADERROR",	"PADERROR",

-    "PADERROR",	"PADERROR",	"PADERROR",	"PADERROR",

-    "PADERROR",	"PADERROR",	"PADERROR",	"PADERROR",

-    "PADERROR",	"PADERROR",	"PADERROR",	"PADERROR",

-};

-

-static const char *tevCInputTable2[] = // CC

-{

-    "prev",         //CPREV,

-    "(prev.aaa)",           //APREV,

-    "c0",           //C0,

-    "(c0.aaa)",             //A0,

-    "c1",           //C1,

-    "(c1.aaa)",             //A1,

-    "c2",           //C2,

-    "(c2.aaa)",             //A2,

-    "textemp",      //TEXC,

-    "(textemp.aaa)",        //TEXA,

-    "rastemp",    //RASC,

-    "(rastemp.aaa)",      //RASA,

-    "float3(1.0f,1.0f,1.0f)",              //ONE,

-    "float3(.5f,.5f,.5f)",              //HALF,

-    "konsttemp", //"konsttemp.rgb",    //KONST,

-    "float3(0.0f,0.0f,0.0f)",              //ZERO

-    "PADERROR",	"PADERROR",	"PADERROR",	"PADERROR",

-    "PADERROR",	"PADERROR",	"PADERROR",	"PADERROR",

-    "PADERROR",	"PADERROR",	"PADERROR",	"PADERROR",

-    "PADERROR",	"PADERROR",	"PADERROR",	"PADERROR",

-    "PADERROR",	"PADERROR",	"PADERROR",	"PADERROR",

-};

-

-static const char *tevAInputTable[] = // CA

-{

-    "prev.a",           //APREV,

-    "c0.a",             //A0,

-    "c1.a",             //A1,

-    "c2.a",             //A2,

-    "textemp.a",        //TEXA,

-    "rastemp.a",        //RASA,

-    "konsttemp.a",      //KONST

-    "0.0", //ZERO

-    "PADERROR", "PADERROR", "PADERROR", "PADERROR",

-    "PADERROR", "PADERROR", "PADERROR", "PADERROR",

-    "PADERROR", "PADERROR", "PADERROR", "PADERROR",

-    "PADERROR", "PADERROR", "PADERROR",

-};	

-

-static const char *tevAInputTable2[] = // CA

-{

-    "prev",           //APREV,

-    "c0",             //A0,

-    "c1",             //A1,

-    "c2",             //A2,

-    "textemp",        //TEXA,

-    "rastemp",        //RASA,

-    "konsttemp",      //KONST,  (hw1 had quarter)

-    "float4(0,0,0,0)", //ZERO

-    "PADERROR", "PADERROR", "PADERROR", "PADERROR",

-    "PADERROR", "PADERROR", "PADERROR", "PADERROR",

-    "PADERROR", "PADERROR", "PADERROR", "PADERROR",

-    "PADERROR", "PADERROR", "PADERROR", "PADERROR",

-};	

-

-static const char *tevRasTable[] =

-{

-    "colors[0]",

-    "colors[1]",

-    "ERROR", //2

-    "ERROR", //3

-    "ERROR", //4

-    "alphabump", // use bump alpha

-    "(alphabump*(255.0f/248.0f))", //normalized

-    "float4(0,0,0,0)", // zero

-};

-

-static const char *alphaRef[2] = 

-{

-    I_ALPHA"[0].x",

-    I_ALPHA"[0].y"

-};

-

-//static const char *tevTexFunc[] = { "tex2D", "texRECT" };

-

-static const char *tevCOutputTable[]  = { "prev.rgb", "c0.rgb", "c1.rgb", "c2.rgb" };

-static const char *tevAOutputTable[]  = { "prev.a", "c0.a", "c1.a", "c2.a" };

-static const char* tevIndAlphaSel[]   = {"", "x", "y", "z"};

-static const char* tevIndAlphaScale[] = {"", "*32","*16","*8"};

-static const char* tevIndBiasField[]  = {"", "x", "y", "xy", "z", "xz", "yz", "xyz"}; // indexed by bias

-static const char* tevIndBiasAdd[]    = {"-128.0f", "1.0f", "1.0f", "1.0f" }; // indexed by fmt

-static const char* tevIndWrapStart[]  = {"0", "256", "128", "64", "32", "16", "0.001" };

-static const char* tevIndFmtScale[]   = {"255.0f", "31.0f", "15.0f", "8.0f" };

-

-#define WRITE p+=sprintf

-

-static const char *swapColors = "rgba";

-static char swapModeTable[4][5];

-

-static char text[16384];

-

-static void BuildSwapModeTable()

-{

-    //bpmem.tevregs[0].

-    for (int i = 0; i < 4; i++)

-    {

-        swapModeTable[i][0] = swapColors[bpmem.tevksel[i*2].swap1];

-        swapModeTable[i][1] = swapColors[bpmem.tevksel[i*2].swap2];

-        swapModeTable[i][2] = swapColors[bpmem.tevksel[i*2+1].swap1];

-        swapModeTable[i][3] = swapColors[bpmem.tevksel[i*2+1].swap2];

-        swapModeTable[i][4] = 0;

-    }

-}

-

-char *GeneratePixelShader(u32 texture_mask, bool has_zbuffer_target, bool bRenderZToCol0)

-{

-	text[sizeof(text) - 1] = 0x7C;  // canary

-    DVSTARTPROFILE();

-

-    BuildSwapModeTable();

-    int numStages = bpmem.genMode.numtevstages + 1;

-    int numTexgen = bpmem.genMode.numtexgens;

-

-    char *p = text;

-    WRITE(p, "//Pixel Shader for TEV stages\n");

-    WRITE(p, "//%i TEV stages, %i texgens, %i IND stages\n",

-		numStages, numTexgen, bpmem.genMode.numindstages);

-

-    bool bRenderZ = has_zbuffer_target && bpmem.zmode.updateenable;

-    bool bOutputZ = bpmem.ztex2.op != ZTEXTURE_DISABLE;

-    bool bInputZ = bpmem.ztex2.op==ZTEXTURE_ADD || bRenderZ;

-

-    // bool bRenderZToCol0 = ; // output z and alpha to color0

-    assert( !bRenderZToCol0 || bRenderZ );

-

-    int ztexcoord = -1;

-    if (bInputZ)

-        ztexcoord = numTexgen == 0 ? 0 : numTexgen-1;

-

-    int nIndirectStagesUsed = 0;

-    if (bpmem.genMode.numindstages > 0) {

-        for (int i = 0; i < numStages; ++i) {

-            if (bpmem.tevind[i].IsActive() && bpmem.tevind[i].bt < bpmem.genMode.numindstages) {

-                nIndirectStagesUsed |= 1<<bpmem.tevind[i].bt;

-            }

-        }

-    }

-

-    // Declare samplers

-    if (texture_mask) {

-        WRITE(p, "uniform samplerRECT ");

-        bool bfirst = true;

-        for (int i = 0; i < 8; ++i) {

-            if (texture_mask & (1<<i)) {

-                WRITE(p, "%s samp%d : register(s%d)", bfirst?"":",", i, i);

-                bfirst = false;

-            }

-        }

-        WRITE(p, ";\n");

-    }

-

-    if (texture_mask != 0xff) {

-        WRITE(p, "uniform sampler2D ");

-        bool bfirst = true;

-        for (int i = 0; i < 8; ++i) {

-            if (!(texture_mask & (1<<i))) {

-                WRITE(p, "%s samp%d : register(s%d)", bfirst?"":",",i, i);

-                bfirst = false;

-            }

-        }

-        WRITE(p, ";\n");

-    }

-

-    WRITE(p, "\n");

-

-    WRITE(p, "uniform float4 "I_COLORS"[4] : register(c%d);\n", C_COLORS);

-    WRITE(p, "uniform float4 "I_KCOLORS"[4] : register(c%d);\n", C_KCOLORS);

-    WRITE(p, "uniform float4 "I_ALPHA"[1] : register(c%d);\n", C_ALPHA);

-    WRITE(p, "uniform float4 "I_TEXDIMS"[8] : register(c%d);\n", C_TEXDIMS);

-    WRITE(p, "uniform float4 "I_ZBIAS"[2] : register(c%d);\n", C_ZBIAS);

-    WRITE(p, "uniform float4 "I_INDTEXSCALE"[2] : register(c%d);\n", C_INDTEXSCALE);

-    WRITE(p, "uniform float4 "I_INDTEXMTX"[6] : register(c%d);\n", C_INDTEXMTX);

-

-    WRITE(p, "void main(\n");

-

-    WRITE(p, "out half4 ocol0 : COLOR0,\n");

-    if (bRenderZ && !bRenderZToCol0 )

-        WRITE(p, "out half4 ocol1 : COLOR1,\n");

-

-    if (bOutputZ )

-        WRITE(p, "  out float depth : DEPTH,\n");

-

-    // if zcoord might come from vertex shader in texcoord

-    if (bInputZ) {

-        if (numTexgen) {

-            for (int i = 0; i < numTexgen; ++i)

-                WRITE(p, "  in float%d uv%d : TEXCOORD%d, \n", i==ztexcoord?4:3, i,i);

-        }

-        else

-            WRITE(p, "  in float4 uv0 : TEXCOORD0,"); //HACK

-    }

-    else {

-        if (numTexgen) {

-            for (int i = 0; i < numTexgen; ++i)

-                WRITE(p, "  in float3 uv%d : TEXCOORD%d,\n",i,i);

-        }

-        else

-            WRITE(p, "  in float3 uv0 : TEXCOORD0,\n"); //HACK

-    }

-

-    WRITE(p, "  in float4 colors[2] : COLOR0){\n");

-

-    char* pmainstart = p;

-

-    WRITE(p, "float4 c0="I_COLORS"[1],c1="I_COLORS"[2],c2="I_COLORS"[3],prev=float4(0.0f,0.0f,0.0f,0.0f),textemp,rastemp,konsttemp=float4(0.0f,0.0f,0.0f,0.0f);\n"

-            "float3 comp16 = float3(1,255,0), comp24 = float3(1,255,255*255);\n"

-            "float4 alphabump=0;\n"

-            "float3 tevcoord;\n"

-            "float2 wrappedcoord, tempcoord;\n");

-

-    //if (bOutputZ ) WRITE(p, "  float depth;\n");

-//    WRITE(p, "return 1;}\n");

-//    return PixelShaderMngr::CompilePixelShader(ps, text);

-

-    // indirect texture map lookup

-    for(u32 i = 0; i < bpmem.genMode.numindstages; ++i) {

-        if (nIndirectStagesUsed & (1<<i)) {

-            // perform indirect texture map lookup

-            // note that we have to scale by the regular texture map's coordinates since this is a texRECT call

-            // (and we have to match with the game's texscale calls)

-            int texcoord = bpmem.tevindref.getTexCoord(i);

-            if (texture_mask & (1<<bpmem.tevindref.getTexMap(i))) {

-                // TODO: I removed a superfluous argument, please check that the resulting expression is correct. (mthuurne 2008-08-27)

-                WRITE(p, "float2 induv%d=uv%d.xy * "I_INDTEXSCALE"[%d].%s;\n", i, texcoord, i/2, (i&1)?"zw":"xy"); //, bpmem.tevindref.getTexMap(i)

-

-                char str[16];

-                sprintf(str, "induv%d", i);

-                WrapNonPow2Tex(p, str, bpmem.tevindref.getTexMap(i), texture_mask);

-                WRITE(p, "float3 indtex%d=texRECT(samp%d,induv%d.xy).abg;\n", i, bpmem.tevindref.getTexMap(i), i);

-            }

-            else {

-                WRITE(p, "float3 indtex%d=tex2D(samp%d,uv%d.xy*"I_INDTEXSCALE"[%d].%s).abg;\n", i, bpmem.tevindref.getTexMap(i), texcoord, i/2, (i&1)?"zw":"xy");

-            }

-        }

-    }

-

-    for (int i = 0; i < numStages; i++)

-        WriteStage(p, i, texture_mask); //build the equation for this stage

-

-    if (bOutputZ) {

-        // use the texture input of the last texture stage (textemp), hopefully this has been read and is in correct format...

-        if (bpmem.ztex2.op == ZTEXTURE_ADD) {

-            WRITE(p, "depth = frac(dot("I_ZBIAS"[0].xyzw, textemp.xyzw) + "I_ZBIAS"[1].w + uv%d.w);\n", ztexcoord);

-        }

-        else {

-            _assert_(bpmem.ztex2.op == ZTEXTURE_REPLACE);

-            WRITE(p, "depth = frac(dot("I_ZBIAS"[0].xyz, textemp.xyz) + "I_ZBIAS"[0].w);\n");

-        }

-    }

-

-    //if (bpmem.genMode.numindstages ) WRITE(p, "prev.rg = indtex0.xy;\nprev.b = 0;\n");

-

-    if (!WriteAlphaTest(p)) {

-        // alpha test will always fail, so restart the shader and just make it an empty function

-        p = pmainstart;

-		WRITE(p, "discard;\n");

-        WRITE(p, "ocol0 = 0;\n");

-    }

-    else {

-        if (!bRenderZToCol0) {

-            /* donkopunchstania: NEEDS FIX - dstalpha does not change how fragments are blended with the EFB

-			   once the blending is done, the dstalpha is written to the EFB in place of the

-			   fragment alpha if dstalpha is enabled. this only matters if the EFB supports alpha.

-               Commenting this out fixed Metroids but causes glitches in Super Mario Sunshine.

-

-			if (bpmem.dstalpha.enable)

-                WRITE(p, "  ocol0 = float4(prev.rgb,"I_ALPHA"[0].w);\n");

-            else

-			   */

-            	WRITE(p, "  ocol0 = prev;\n");

-        }

-    }

-

-    if (bRenderZ) {

-        // write depth as color

-        if (bRenderZToCol0) {

-            if (bOutputZ )

-                WRITE(p, "ocol0.xyz = frac(float3(256.0f*256.0f, 256.0f, 1.0f) * depth);\n");

-            else

-                WRITE(p, "ocol0.xyz = frac(float3(256.0f*256.0f, 256.0f, 1.0f) * uv%d.w);\n", ztexcoord);

-            WRITE(p, "ocol0.w = prev.w;\n");

-        }

-        else {

-            if (bOutputZ)

-                WRITE(p, "ocol1 = frac(float4(256.0f*256.0f, 256.0f, 1.0f, 0.0f) * depth);\n");

-            else

-                WRITE(p, "ocol1 = frac(float4(256.0f*256.0f, 256.0f, 1.0f, 0.0f) * uv%d.w);\n", ztexcoord);

-        }

-    }

-    WRITE(p, "}\n");

-	if (text[sizeof(text) - 1] != 0x7C)

-		PanicAlert("PixelShader generator - buffer too small, canary has been eaten!");

-    return text;

-}

-

-static void WriteStage(char *&p, int n, u32 texture_mask)

-{

-    char *rasswap = swapModeTable[bpmem.combiners[n].alphaC.rswap];

-    char *texswap = swapModeTable[bpmem.combiners[n].alphaC.tswap];

-

-

-    int texcoord = bpmem.tevorders[n/2].getTexCoord(n&1);

-    int texfun = xfregs.texcoords[texcoord].texmtxinfo.projection;

-    bool bHasIndStage = bpmem.tevind[n].IsActive() && bpmem.tevind[n].bt < bpmem.genMode.numindstages;

-

-    if (bHasIndStage) {

-        // perform the indirect op on the incoming regular coordinates using indtex%d as the offset coords

-        bHasIndStage = true;

-        int texmap = bpmem.tevorders[n/2].getEnable(n&1) ? bpmem.tevorders[n/2].getTexMap(n&1) : bpmem.tevindref.getTexMap(bpmem.tevind[n].bt);

-

-		if (bpmem.tevind[n].bs != ITBA_OFF) {

-            // write the bump alpha

-

-			if (bpmem.tevind[n].fmt == ITF_8) {

-				WRITE(p, "alphabump = indtex%d.%s %s;\n", bpmem.tevind[n].bt, 

-					tevIndAlphaSel[bpmem.tevind[n].bs], tevIndAlphaScale[bpmem.tevind[n].fmt]);

-			}

-			else {			

-				// donkopunchstania: really bad way to do this

-				// cannot always use fract because fract(1.0) is 0.0 when it needs to be 1.0

-				// omitting fract seems to work as well

-				WRITE(p, "if (indtex%d.%s >= 1.0f )\n", bpmem.tevind[n].bt, 

-					tevIndAlphaSel[bpmem.tevind[n].bs]);

-				WRITE(p, "   alphabump = 1.0f;\n");

-				WRITE(p, "else\n");

-				WRITE(p, "   alphabump = fract ( indtex%d.%s %s );\n", bpmem.tevind[n].bt, 

-					tevIndAlphaSel[bpmem.tevind[n].bs], tevIndAlphaScale[bpmem.tevind[n].fmt]);

-			}

-		}

-

-        // bias

-        WRITE(p, "float3 indtevcrd%d = indtex%d;\n", n, bpmem.tevind[n].bt);

-        WRITE(p, "indtevcrd%d.xyz *= %s;\n", n, tevIndFmtScale[bpmem.tevind[n].fmt]);

-        if (bpmem.tevind[n].bias != ITB_NONE )

-            WRITE(p, "indtevcrd%d.%s += %s;\n", n, tevIndBiasField[bpmem.tevind[n].bias], tevIndBiasAdd[bpmem.tevind[n].fmt]);

-

-        // multiply by offset matrix and scale

-        if (bpmem.tevind[n].mid != 0) {

-            if (bpmem.tevind[n].mid <= 3) {

-                int mtxidx = 2*(bpmem.tevind[n].mid-1);

-                WRITE(p, "float2 indtevtrans%d = float2(dot("I_INDTEXMTX"[%d].xyz, indtevcrd%d), dot("I_INDTEXMTX"[%d].xyz, indtevcrd%d));\n",

-                    n, mtxidx, n, mtxidx+1, n);

-            }

-            else if (bpmem.tevind[n].mid <= 5) { // s matrix

-                int mtxidx = 2*(bpmem.tevind[n].mid-5);

-                WRITE(p, "float2 indtevtrans%d = "I_INDTEXMTX"[%d].ww * uv%d.xy * indtevcrd%d.xx;\n", n, mtxidx, texcoord, n);

-            }

-            else if (bpmem.tevind[n].mid <= 9) { // t matrix

-                int mtxidx = 2*(bpmem.tevind[n].mid-9);

-                WRITE(p, "float2 indtevtrans%d = "I_INDTEXMTX"[%d].ww * uv%d.xy * indtevcrd%d.yy;\n", n, mtxidx, texcoord, n);

-            }

-            else {

-                // TODO: I removed a superfluous argument, please check that the resulting expression is correct. (mthuurne 2008-08-27)

-                WRITE(p, "float2 indtevtrans%d = 0;\n", n); //, n

-            }

-        }

-        else {

-            // TODO: I removed a superfluous argument, please check that the resulting expression is correct. (mthuurne 2008-08-27)

-            WRITE(p, "float2 indtevtrans%d = 0;\n", n); //, n

-        }

-

-        // wrapping

-        if (!bpmem.tevorders[n/2].getEnable(n&1) || (texture_mask & (1<<texmap))) {

-            // non pow2

-

-            if (bpmem.tevind[n].sw != ITW_OFF || bpmem.tevind[n].tw != ITW_OFF) {

-                if (bpmem.tevind[n].sw == ITW_0) {

-                    if (bpmem.tevind[n].tw == ITW_0) {

-                        // zero out completely

-                        WRITE(p, "wrappedcoord = float2(0.0f,0.0f);\n");

-                    }

-                    else {

-                        WRITE(p, "wrappedcoord.x = fmod( (uv%d.x+%s)*"I_TEXDIMS"[%d].x*"I_TEXDIMS"[%d].z, %s);\n"

-                            "wrappedcoord.y = 0;\n", texcoord, tevIndWrapStart[bpmem.tevind[n].sw], texmap, texmap, tevIndWrapStart[bpmem.tevind[n].sw]);

-                    }

-                }

-                else if (bpmem.tevind[n].tw == ITW_0) {

-                    WRITE(p, "wrappedcoord.y = fmod( (uv%d.y+%s)*"I_TEXDIMS"[%d].y*"I_TEXDIMS"[%d].w, %s);\n"

-                        "wrappedcoord.x = 0;\n", texcoord, tevIndWrapStart[bpmem.tevind[n].tw], texmap, texmap, tevIndWrapStart[bpmem.tevind[n].tw]);

-                }

-                else {

-                    WRITE(p, "wrappedcoord = fmod( (uv%d.xy+float2(%s,%s))*"I_TEXDIMS"[%d].xy*"I_TEXDIMS"[%d].zw, float2(%s,%s));\n", texcoord,

-                        tevIndWrapStart[bpmem.tevind[n].sw], tevIndWrapStart[bpmem.tevind[n].tw],texmap,texmap,

-                        tevIndWrapStart[bpmem.tevind[n].sw], tevIndWrapStart[bpmem.tevind[n].tw]);

-                }

-            }

-            else {

-                WRITE(p, "wrappedcoord = uv%d.xy*"I_TEXDIMS"[%d].xy;\n", texcoord, texmap);

-            }

-        }

-        else {

-            // pow of 2

-            WRITE(p, "indtevtrans%d.xy *= "I_TEXDIMS"[%d].xy * "I_TEXDIMS"[%d].zw;\n", n, texmap, texmap);

-

-            // mult by bitdepth / tex dimensions

-            if (bpmem.tevind[n].sw != ITW_OFF || bpmem.tevind[n].tw != ITW_OFF) {

-                if (bpmem.tevind[n].sw == ITW_0) {

-                    if (bpmem.tevind[n].tw == ITW_0) {

-                        // zero out completely

-                        WRITE(p, "wrappedcoord = float2(0.0f,0.0f);\n");

-                    }

-                    else {

-                        WRITE(p, "wrappedcoord.x = "I_TEXDIMS"[%d].x * fmod( uv%d.x+%s, "I_TEXDIMS"[%d].z*%s);\n"

-                            "wrappedcoord.y = 0;\n", texmap, texcoord, tevIndWrapStart[bpmem.tevind[n].sw], texmap, tevIndWrapStart[bpmem.tevind[n].sw]);

-                    }

-                }

-                else if (bpmem.tevind[n].tw == ITW_0) {

-                    WRITE(p, "wrappedcoord.y = "I_TEXDIMS"[%d].y * fmod( uv%d.y+%s, "I_TEXDIMS"[%d].w*%s);\n"

-                        "wrappedcoord.x = 0;\n", texmap, texcoord, tevIndWrapStart[bpmem.tevind[n].tw], texmap, tevIndWrapStart[bpmem.tevind[n].tw]);

-                }

-                else {

-                    // have to add an offset or else might get negative values!

-                    WRITE(p, "wrappedcoord = "I_TEXDIMS"[%d].xy * fmod( uv%d.xy+float2(%s,%s), "I_TEXDIMS"[%d].zw*float2(%s,%s));\n", texmap, texcoord,

-                        tevIndWrapStart[bpmem.tevind[n].sw], tevIndWrapStart[bpmem.tevind[n].tw], texmap, 

-                        tevIndWrapStart[bpmem.tevind[n].sw], tevIndWrapStart[bpmem.tevind[n].tw]);

-                }

-            }

-            else {

-                WRITE(p, "wrappedcoord = uv%d.xy;\n", texcoord);

-            }

-        }

-

-        if (bpmem.tevind[n].fb_addprev) {

-            // add previous tevcoord

-

-            if (texfun == XF_TEXPROJ_STQ) {

-                WRITE(p, "tevcoord.xy += wrappedcoord/uv%d.z + indtevtrans%d;\n", texcoord, n);

-                //WRITE(p, "tevcoord.z += uv%d.z;\n", texcoord);

-            }

-            else {

-                WRITE(p, "tevcoord.xy += wrappedcoord + indtevtrans%d;\n", n);

-            }

-        }

-        else {

-            WRITE(p, "tevcoord.xy = wrappedcoord/uv%d.z + indtevtrans%d;\n", texcoord, n);

-            //if (texfun == XF_TEXPROJ_STQ )

-            //    WRITE(p, "tevcoord.z = uv%d.z;\n", texcoord);

-        }

-    }

-

-    WRITE(p, "rastemp=%s.%s;\n",tevRasTable[bpmem.tevorders[n/2].getColorChan(n&1)],rasswap);

-

-    if (bpmem.tevorders[n/2].getEnable(n&1)) {

-        int texmap = bpmem.tevorders[n/2].getTexMap(n&1);

-        if(!bHasIndStage) {

-            // calc tevcord

-            //tevcoord.xy = texdim[1].xy * uv1.xy / uv1.z;

-            int OurTexCoord = 0;

-            if(bpmem.genMode.numtexgens)

-                OurTexCoord = texcoord;

-            else

-                OurTexCoord = 0;

-            if (texture_mask & (1<<texmap)) {

-                // nonpow2

-                if (texfun == XF_TEXPROJ_STQ )

-                    WRITE(p, "tevcoord.xy = uv%d.xy / uv%d.z;\n", texcoord, OurTexCoord);

-                else

-                    WRITE(p, "tevcoord.xy = uv%d.xy;\n", OurTexCoord);

-                WrapNonPow2Tex(p, "tevcoord", texmap, texture_mask);

-            }

-            else {

-                if (texfun == XF_TEXPROJ_STQ ) 

-                    WRITE(p, "tevcoord.xy = "I_TEXDIMS"[%d].xy * uv%d.xy / uv%d.z;\n", texmap, OurTexCoord , OurTexCoord );

-                else 

-                    WRITE(p, "tevcoord.xy = "I_TEXDIMS"[%d].xy * uv%d.xy;\n", texmap, OurTexCoord);

-

-            }

-        }

-        else if (texture_mask & (1<<texmap)) {

-            // if non pow 2, have to manually repeat

-            //WrapNonPow2Tex(p, "tevcoord", texmap);

-            bool bwraps = !!(texture_mask & (0x100<<texmap));

-            bool bwrapt = !!(texture_mask & (0x10000<<texmap));

-                

-            if (bwraps || bwrapt) {

-                const char* field = bwraps ? (bwrapt ? "xy" : "x") : "y";

-                WRITE(p, "tevcoord.%s = fmod(tevcoord.%s+32*"I_TEXDIMS"[%d].%s,"I_TEXDIMS"[%d].%s);\n", field, field, texmap, field, texmap, field);

-            }

-        }

-

-        if (texture_mask & (1<<texmap) )

-            WRITE(p, "textemp=texRECT(samp%d,tevcoord.xy).%s;\n", texmap, texswap);

-        else

-            WRITE(p, "textemp=tex2D(samp%d,tevcoord.xy).%s;\n", texmap, texswap);

-    }

-    else

-        WRITE(p, "textemp=float4(1,1,1,1);\n");

-

-    int kc = bpmem.tevksel[n/2].getKC(n&1);

-    int ka = bpmem.tevksel[n/2].getKA(n&1);

-

-    TevStageCombiner::ColorCombiner &cc = bpmem.combiners[n].colorC;

-    TevStageCombiner::AlphaCombiner &ac = bpmem.combiners[n].alphaC;

-

-    bool bCKonst = cc.a == TEVCOLORARG_KONST || cc.b == TEVCOLORARG_KONST || cc.c == TEVCOLORARG_KONST || cc.d == TEVCOLORARG_KONST;

-    bool bAKonst = ac.a == TEVALPHAARG_KONST || ac.b == TEVALPHAARG_KONST || ac.c == TEVALPHAARG_KONST || ac.d == TEVALPHAARG_KONST;

-    if (bCKonst || bAKonst )

-        WRITE(p, "konsttemp=float4(%s,%s);\n",tevKSelTableC[kc],tevKSelTableA[ka]);  

-

-    WRITE(p, "%s= ", tevCOutputTable[cc.dest]); 

-

-    // combine the color channel

-    if (cc.bias != 3) {  // if not compare

-        //normal color combiner goes here

-        WRITE(p, "   %s*(%s%s",tevScaleTable[cc.shift],tevCInputTable[cc.d],tevOpTable[cc.op]);

-        WRITE(p, "lerp(%s,%s,%s) %s);\n",

-              tevCInputTable[cc.a], tevCInputTable[cc.b],

-              tevCInputTable[cc.c], tevBiasTable[cc.bias]);

-    }

-    else {

-        int cmp = (cc.shift<<1)|cc.op|8; // comparemode stored here

-        switch(cmp) {

-        case TEVCMP_R8_GT:

-        case TEVCMP_RGB8_GT: // per component compares

-            WRITE(p, "   %s + ((%s.%s > %s.%s) ? %s : float3(0.0f,0.0f,0.0f));\n",

-                tevCInputTable[cc.d], tevCInputTable2[cc.a], cmp==TEVCMP_R8_GT?"r":"rgb", tevCInputTable2[cc.b], cmp==TEVCMP_R8_GT?"r":"rgb", tevCInputTable[cc.c]);

-            break;

-        case TEVCMP_R8_EQ:

-        case TEVCMP_RGB8_EQ:

-            WRITE(p, "   %s + (abs(%s.r - %s.r)<%f ? %s : float3(0.0f,0.0f,0.0f));\n",

-                tevCInputTable[cc.d], tevCInputTable2[cc.a], tevCInputTable2[cc.b], epsilon8bit, tevCInputTable[cc.c]);

-            break;

-        

-        case TEVCMP_GR16_GT: // 16 bit compares: 255*g+r (probably used for ztextures, so make sure in ztextures, g is the most significant byte)

-        case TEVCMP_BGR24_GT: // 24 bit compares: 255*255*b+255*g+r

-            WRITE(p, "   %s + (( dot(%s.rgb-%s.rgb, comp%s) > 0) ? %s : float3(0.0f,0.0f,0.0f));\n",

-                tevCInputTable[cc.d], tevCInputTable2[cc.a], tevCInputTable2[cc.b], cmp==TEVCMP_GR16_GT?"16":"24", tevCInputTable[cc.c]);

-            break;

-        case TEVCMP_GR16_EQ:

-        case TEVCMP_BGR24_EQ:

-            WRITE(p, "   %s + (abs(dot(%s.rgb - %s.rgb, comp%s))<%f ? %s : float3(0.0f,0.0f,0.0f));\n",

-                tevCInputTable[cc.d], tevCInputTable2[cc.a], tevCInputTable2[cc.b], cmp==TEVCMP_GR16_GT?"16":"24", epsilon8bit, tevCInputTable[cc.c]);

-            break;

-        default:

-            WRITE(p, "float3(0.0f,0.0f,0.0f);\n");

-            break;

-        }

-    }

-    

-    if (cc.clamp)

-        WRITE(p, "%s = clamp(%s,0.0f,1.0f);\n", tevCOutputTable[cc.dest],tevCOutputTable[cc.dest]);

-

-    // combine the alpha channel

-    WRITE(p, "%s= ", tevAOutputTable[ac.dest]);

-

-    if (ac.bias != 3) { // if not compare

-        //normal alpha combiner goes here

-        WRITE(p, "   %s*(%s%s",tevScaleTable[ac.shift],tevAInputTable[ac.d],tevOpTable[ac.op]);

-        WRITE(p, "lerp(%s,%s,%s) %s)\n",

-            tevAInputTable[ac.a],tevAInputTable[ac.b],

-            tevAInputTable[ac.c],tevBiasTable[ac.bias]);

-    }

-    else {

-        //compare alpha combiner goes here

-        int cmp = (ac.shift<<1)|ac.op|8; // comparemode stored here

-        switch(cmp) {

-        case TEVCMP_R8_GT:

-        case TEVCMP_A8_GT:

-            WRITE(p, "   %s + ((%s.%s > %s.%s) ? %s : 0)\n",

-                tevAInputTable[ac.d],tevAInputTable2[ac.a], cmp==TEVCMP_R8_GT?"r":"a", tevAInputTable2[ac.b], cmp==TEVCMP_R8_GT?"r":"a", tevAInputTable[ac.c]);

-            break;

-        case TEVCMP_R8_EQ:

-        case TEVCMP_A8_EQ:

-            WRITE(p, "   %s + (abs(%s.r - %s.r)<%f ? %s : 0)\n",

-                tevAInputTable[ac.d],tevAInputTable2[ac.a], tevAInputTable2[ac.b],epsilon8bit,tevAInputTable[ac.c]);

-            break;

-        

-        case TEVCMP_GR16_GT: // 16 bit compares: 255*g+r (probably used for ztextures, so make sure in ztextures, g is the most significant byte)

-        case TEVCMP_BGR24_GT: // 24 bit compares: 255*255*b+255*g+r

-            WRITE(p, "   %s + (( dot(%s.rgb-%s.rgb, comp%s) > 0) ? %s : 0)\n",

-                tevAInputTable[ac.d],tevAInputTable2[ac.a], tevAInputTable2[ac.b], cmp==TEVCMP_GR16_GT?"16":"24", tevAInputTable[ac.c]);

-            break;

-        case TEVCMP_GR16_EQ:

-        case TEVCMP_BGR24_EQ:

-            WRITE(p, "   %s + (abs(dot(%s.rgb - %s.rgb, comp%s))<%f ? %s : 0)\n",

-                tevAInputTable[ac.d],tevAInputTable2[ac.a], tevAInputTable2[ac.b],cmp==TEVCMP_GR16_GT?"16":"24",epsilon8bit,tevAInputTable[ac.c]);

-            break;

-        default:

-            WRITE(p, "0)\n");

-            break;

-        }

-    }

-

-    WRITE(p, ";\n");

-

-    if (ac.clamp)

-        WRITE(p, "%s = clamp(%s,0.0f,1.0f);\n", tevAOutputTable[ac.dest],tevAOutputTable[ac.dest]);

-    WRITE(p, "\n");

-}

-

-void WrapNonPow2Tex(char* &p, const char* var, int texmap, u32 texture_mask)

-{

-    _assert_(texture_mask & (1<<texmap));

-    bool bwraps = !!(texture_mask & (0x100<<texmap));

-    bool bwrapt = !!(texture_mask & (0x10000<<texmap));

-        

-    if (bwraps || bwrapt) {

-        const char* field = bwraps ? (bwrapt ? "xy" : "x") : "y";

-        const char* wrapfield = bwraps ? (bwrapt ? "zw" : "z") : "w";

-        WRITE(p, "%s.%s = "I_TEXDIMS"[%d].%s*frac(%s.%s*"I_TEXDIMS"[%d].%s+32);\n", var, field, texmap, field, var, field, texmap, wrapfield);

-

-        if (!bwraps )

-            WRITE(p, "%s.x *= "I_TEXDIMS"[%d].x * "I_TEXDIMS"[%d].z;\n", var, texmap, texmap);

-        if (!bwrapt )

-            WRITE(p, "%s.y *= "I_TEXDIMS"[%d].y * "I_TEXDIMS"[%d].w;\n", var, texmap, texmap);

-    }

-    else {

-        WRITE(p, "%s.xy *= "I_TEXDIMS"[%d].xy * "I_TEXDIMS"[%d].zw;\n", var, texmap, texmap);

-    }

-}

-

-static void WriteAlphaCompare(char *&p, int num, int comp)

-{

-    switch(comp) {

-    case ALPHACMP_ALWAYS:  WRITE(p, "(false)");	break;

-    case ALPHACMP_NEVER:   WRITE(p, "(true)");	break;

-    case ALPHACMP_LEQUAL:  WRITE(p, "(prev.a > %s)",alphaRef[num]);	break;

-    case ALPHACMP_LESS:    WRITE(p, "(prev.a >= %s - %f)",alphaRef[num],epsilon8bit*0.5f);break;

-    case ALPHACMP_GEQUAL:  WRITE(p, "(prev.a < %s)",alphaRef[num]);	break;

-    case ALPHACMP_GREATER: WRITE(p, "(prev.a <= %s + %f)",alphaRef[num],epsilon8bit*0.5f);break;

-    case ALPHACMP_EQUAL:   WRITE(p, "(abs(prev.a-%s)>%f)",alphaRef[num],epsilon8bit*2); break;

-    case ALPHACMP_NEQUAL:  WRITE(p, "(abs(prev.a-%s)<%f)",alphaRef[num],epsilon8bit*2); break;

-    }

-}

-

-static bool WriteAlphaTest(char *&p)

-{

-    u32 op = bpmem.alphaFunc.logic;

-    u32 comp[2] = {bpmem.alphaFunc.comp0,bpmem.alphaFunc.comp1};

-

-    //first kill all the simple cases

-    switch(op) {

-    case 0: // and

-        if (comp[0] == ALPHACMP_ALWAYS && comp[1] == ALPHACMP_ALWAYS) return true;

-        if (comp[0] == ALPHACMP_NEVER || comp[1] == ALPHACMP_NEVER) {

-            WRITE(p, "discard;\n");

-            return false;

-        }

-        break;

-    case 1: // or

-        if (comp[0] == ALPHACMP_ALWAYS || comp[1] == ALPHACMP_ALWAYS) return true;

-        if (comp[0] == ALPHACMP_NEVER && comp[1] == ALPHACMP_NEVER) {

-            WRITE(p, "discard;\n");

-            return false;

-        }

-        break;

-    case 2: // xor

-        if ( (comp[0] == ALPHACMP_ALWAYS && comp[1] == ALPHACMP_NEVER) || (comp[0] == ALPHACMP_NEVER && comp[1] == ALPHACMP_ALWAYS) ) return true;

-        if ( (comp[0] == ALPHACMP_ALWAYS && comp[1] == ALPHACMP_ALWAYS) || (comp[0] == ALPHACMP_NEVER && comp[1] == ALPHACMP_NEVER)) {

-            WRITE(p, "discard;\n");

-            return false;

-        }

-        break;

-    case 3: // xnor

-        if ( (comp[0] == ALPHACMP_ALWAYS && comp[1] == ALPHACMP_NEVER) || (comp[0] == ALPHACMP_NEVER && comp[1] == ALPHACMP_ALWAYS)) {

-            WRITE(p, "discard;\n");

-            return false;

-        }

-        if ( (comp[0] == ALPHACMP_ALWAYS && comp[1] == ALPHACMP_ALWAYS) || (comp[0] == ALPHACMP_NEVER && comp[1] == ALPHACMP_NEVER) )

-            return true;

-        break;

-    }

-

-    WRITE(p, "discard( ");

-    WriteAlphaCompare(p, 0, bpmem.alphaFunc.comp0);

-    

-    // negated because testing the inverse condition

-    switch(bpmem.alphaFunc.logic) {

-    case 0: WRITE(p, " || "); break; // and

-    case 1: WRITE(p, " && "); break; // or

-    case 2: WRITE(p, " == "); break; // xor

-    case 3: WRITE(p, " != "); break; // xnor

-    }

-    WriteAlphaCompare(p, 1, bpmem.alphaFunc.comp1);

-    WRITE(p, ");\n");

-    return true;

-}

+// Copyright (C) 2003-2008 Dolphin Project.
+
+// This program is free software: you can redistribute it and/or modify
+// it under the terms of the GNU General Public License as published by
+// the Free Software Foundation, version 2.0.
+
+// This program is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU General Public License 2.0 for more details.
+
+// A copy of the GPL 2.0 should have been included with the program.
+// If not, see http://www.gnu.org/licenses/
+
+// Official SVN repository and contact information can be found at
+// http://code.google.com/p/dolphin-emu/
+
+#include <stdio.h>
+#include <cmath>
+#include <assert.h>
+
+#include "Profiler.h"
+#include "PixelShader.h"
+#include "XFMemory.h"  // for texture projection mode
+#include "BPMemory.h"
+
+//   old tev->pixelshader notes
+//
+//   color for this stage (alpha, color) is given by bpmem.tevorders[0].colorchan0
+//   konstant for this stage (alpha, color) is given by bpmem.tevksel   
+//   inputs are given by bpmem.combiners[0].colorC.a/b/c/d     << could be current chan color
+//   according to GXTevColorArg table above
+//   output is given by .outreg
+//   tevtemp is set according to swapmodetables and 
+
+static void WriteStage(char *&p, int n, u32 texture_mask);
+static void WrapNonPow2Tex(char* &p, const char* var, int texmap, u32 texture_mask);
+static void WriteAlphaCompare(char *&p, int num, int comp);
+static bool WriteAlphaTest(char *&p);
+
+const float epsilon8bit = 1.0f / 255.0f;
+
+static const char *tevKSelTableC[] = // KCSEL
+{
+    "1.0f,1.0f,1.0f",            //1   = 0x00
+    "0.875,0.875,0.875",//7_8 = 0x01
+    "0.75,0.75,0.75",	//3_4 = 0x02
+    "0.625,0.625,0.625",//5_8 = 0x03
+    "0.5,0.5,0.5",      //1_2 = 0x04
+    "0.375,0.375,0.375",//3_8 = 0x05
+    "0.25,0.25,0.25",   //1_4 = 0x06
+    "0.125,0.125,0.125",//1_8 = 0x07
+    "ERROR", //0x08
+    "ERROR", //0x09
+    "ERROR", //0x0a
+    "ERROR", //0x0b
+    I_KCOLORS"[0].rgb",//K0 = 0x0C
+    I_KCOLORS"[1].rgb",//K1 = 0x0D
+    I_KCOLORS"[2].rgb",//K2 = 0x0E
+    I_KCOLORS"[3].rgb",//K3 = 0x0F
+    I_KCOLORS"[0].rrr",//K0_R = 0x10
+    I_KCOLORS"[1].rrr",//K1_R = 0x11
+    I_KCOLORS"[2].rrr",//K2_R = 0x12
+    I_KCOLORS"[3].rrr",//K3_R = 0x13
+    I_KCOLORS"[0].ggg",//K0_G = 0x14 
+    I_KCOLORS"[1].ggg",//K1_G = 0x15
+    I_KCOLORS"[2].ggg",//K2_G = 0x16
+    I_KCOLORS"[3].ggg",//K3_G = 0x17
+    I_KCOLORS"[0].bbb",//K0_B = 0x18
+    I_KCOLORS"[1].bbb",//K1_B = 0x19
+    I_KCOLORS"[2].bbb",//K2_B = 0x1A
+    I_KCOLORS"[3].bbb",//K3_B = 0x1B
+    I_KCOLORS"[0].aaa",//K0_A = 0x1C
+    I_KCOLORS"[1].aaa",//K1_A = 0x1D
+    I_KCOLORS"[2].aaa",//K2_A = 0x1E
+    I_KCOLORS"[3].aaa",//K3_A = 0x1F
+};
+
+static const char *tevKSelTableA[] = // KASEL
+{
+    "1.0f",    //1   = 0x00
+    "0.875f",//7_8 = 0x01
+    "0.75f",	//3_4 = 0x02
+    "0.625f",//5_8 = 0x03
+    "0.5f",  //1_2 = 0x04
+    "0.375f",//3_8 = 0x05
+    "0.25f", //1_4 = 0x06
+    "0.125f",//1_8 = 0x07
+    "ERROR", //0x08
+    "ERROR", //0x09
+    "ERROR", //0x0a
+    "ERROR", //0x0b
+    "ERROR", //0x0c
+    "ERROR", //0x0d
+    "ERROR", //0x0e
+    "ERROR", //0x0f
+    I_KCOLORS"[0].r",//K0_R = 0x10
+    I_KCOLORS"[1].r",//K1_R = 0x11
+    I_KCOLORS"[2].r",//K2_R = 0x12
+    I_KCOLORS"[3].r",//K3_R = 0x13
+    I_KCOLORS"[0].g",//K0_G = 0x14
+    I_KCOLORS"[1].g",//K1_G = 0x15
+    I_KCOLORS"[2].g",//K2_G = 0x16
+    I_KCOLORS"[3].g",//K3_G = 0x17
+    I_KCOLORS"[0].b",//K0_B = 0x18
+    I_KCOLORS"[1].b",//K1_B = 0x19
+    I_KCOLORS"[2].b",//K2_B = 0x1A
+    I_KCOLORS"[3].b",//K3_B = 0x1B
+    I_KCOLORS"[0].a",//K0_A = 0x1C
+    I_KCOLORS"[1].a",//K1_A = 0x1D
+    I_KCOLORS"[2].a",//K2_A = 0x1E
+    I_KCOLORS"[3].a",//K3_A = 0x1F
+};
+
+static const char *tevScaleTable[] = // CS
+{
+    "1.0f",   //SCALE_1
+    "2.0f", //SCALE_2
+    "4.0f", //SCALE_4
+    "0.5f",//DIVIDE_2
+};
+
+static const char *tevBiasTable[] = // TB
+{
+    "",      //ZERO,
+    "+0.5f",  //ADDHALF,
+    "-0.5f",  //SUBHALF,
+    "",
+};
+
+static const char *tevOpTable[] = { // TEV
+    "+",      //TEVOP_ADD = 0,
+    "-",      //TEVOP_SUB = 1,
+};
+
+//static const char *tevCompOpTable[] = { ">", "==" };
+
+#define TEVCMP_R8    0
+#define TEVCMP_GR16  1
+#define TEVCMP_BGR24 2
+#define TEVCMP_RGB8  3
+
+static const char *tevCInputTable[] = // CC
+{
+    "prev.rgb",         //CPREV,
+    "prev.aaa",           //APREV,
+    "c0.rgb",           //C0,
+    "c0.aaa",             //A0,
+    "c1.rgb",           //C1,
+    "c1.aaa",             //A1,
+    "c2.rgb",           //C2,
+    "c2.aaa",             //A2,
+    "textemp.rgb",      //TEXC,
+    "textemp.aaa",        //TEXA,
+    "rastemp.rgb",    //RASC,
+    "rastemp.aaa",      //RASA,
+    "float3(1.0f,1.0f,1.0f)",              //ONE,
+    "float3(.5f,.5f,.5f)",              //HALF,
+    "konsttemp.rgb",    //KONST,
+    "float3(0.0f,0.0f,0.0f)",              //ZERO
+    "PADERROR",	"PADERROR",	"PADERROR",	"PADERROR",
+    "PADERROR",	"PADERROR",	"PADERROR",	"PADERROR",
+    "PADERROR",	"PADERROR",	"PADERROR",	"PADERROR",
+    "PADERROR",	"PADERROR",	"PADERROR",	"PADERROR",
+    "PADERROR",	"PADERROR",	"PADERROR",	"PADERROR",
+};
+
+static const char *tevCInputTable2[] = // CC
+{
+    "prev",         //CPREV,
+    "(prev.aaa)",           //APREV,
+    "c0",           //C0,
+    "(c0.aaa)",             //A0,
+    "c1",           //C1,
+    "(c1.aaa)",             //A1,
+    "c2",           //C2,
+    "(c2.aaa)",             //A2,
+    "textemp",      //TEXC,
+    "(textemp.aaa)",        //TEXA,
+    "rastemp",    //RASC,
+    "(rastemp.aaa)",      //RASA,
+    "float3(1.0f,1.0f,1.0f)",              //ONE,
+    "float3(.5f,.5f,.5f)",              //HALF,
+    "konsttemp", //"konsttemp.rgb",    //KONST,
+    "float3(0.0f,0.0f,0.0f)",              //ZERO
+    "PADERROR",	"PADERROR",	"PADERROR",	"PADERROR",
+    "PADERROR",	"PADERROR",	"PADERROR",	"PADERROR",
+    "PADERROR",	"PADERROR",	"PADERROR",	"PADERROR",
+    "PADERROR",	"PADERROR",	"PADERROR",	"PADERROR",
+    "PADERROR",	"PADERROR",	"PADERROR",	"PADERROR",
+};
+
+static const char *tevAInputTable[] = // CA
+{
+    "prev.a",           //APREV,
+    "c0.a",             //A0,
+    "c1.a",             //A1,
+    "c2.a",             //A2,
+    "textemp.a",        //TEXA,
+    "rastemp.a",        //RASA,
+    "konsttemp.a",      //KONST
+    "0.0", //ZERO
+    "PADERROR", "PADERROR", "PADERROR", "PADERROR",
+    "PADERROR", "PADERROR", "PADERROR", "PADERROR",
+    "PADERROR", "PADERROR", "PADERROR", "PADERROR",
+    "PADERROR", "PADERROR", "PADERROR",
+};	
+
+static const char *tevAInputTable2[] = // CA
+{
+    "prev",           //APREV,
+    "c0",             //A0,
+    "c1",             //A1,
+    "c2",             //A2,
+    "textemp",        //TEXA,
+    "rastemp",        //RASA,
+    "konsttemp",      //KONST,  (hw1 had quarter)
+    "float4(0,0,0,0)", //ZERO
+    "PADERROR", "PADERROR", "PADERROR", "PADERROR",
+    "PADERROR", "PADERROR", "PADERROR", "PADERROR",
+    "PADERROR", "PADERROR", "PADERROR", "PADERROR",
+    "PADERROR", "PADERROR", "PADERROR", "PADERROR",
+};	
+
+static const char *tevRasTable[] =
+{
+    "colors[0]",
+    "colors[1]",
+    "ERROR", //2
+    "ERROR", //3
+    "ERROR", //4
+    "alphabump", // use bump alpha
+    "(alphabump*(255.0f/248.0f))", //normalized
+    "float4(0,0,0,0)", // zero
+};
+
+static const char *alphaRef[2] = 
+{
+    I_ALPHA"[0].x",
+    I_ALPHA"[0].y"
+};
+
+//static const char *tevTexFunc[] = { "tex2D", "texRECT" };
+
+static const char *tevCOutputTable[]  = { "prev.rgb", "c0.rgb", "c1.rgb", "c2.rgb" };
+static const char *tevAOutputTable[]  = { "prev.a", "c0.a", "c1.a", "c2.a" };
+static const char* tevIndAlphaSel[]   = {"", "x", "y", "z"};
+static const char* tevIndAlphaScale[] = {"", "*32","*16","*8"};
+static const char* tevIndBiasField[]  = {"", "x", "y", "xy", "z", "xz", "yz", "xyz"}; // indexed by bias
+static const char* tevIndBiasAdd[]    = {"-128.0f", "1.0f", "1.0f", "1.0f" }; // indexed by fmt
+static const char* tevIndWrapStart[]  = {"0", "256", "128", "64", "32", "16", "0.001" };
+static const char* tevIndFmtScale[]   = {"255.0f", "31.0f", "15.0f", "8.0f" };
+
+#define WRITE p+=sprintf
+
+static const char *swapColors = "rgba";
+static char swapModeTable[4][5];
+
+static char text[16384];
+
+static void BuildSwapModeTable()
+{
+    //bpmem.tevregs[0].
+    for (int i = 0; i < 4; i++)
+    {
+        swapModeTable[i][0] = swapColors[bpmem.tevksel[i*2].swap1];
+        swapModeTable[i][1] = swapColors[bpmem.tevksel[i*2].swap2];
+        swapModeTable[i][2] = swapColors[bpmem.tevksel[i*2+1].swap1];
+        swapModeTable[i][3] = swapColors[bpmem.tevksel[i*2+1].swap2];
+        swapModeTable[i][4] = 0;
+    }
+}
+
+char *GeneratePixelShader(u32 texture_mask, bool has_zbuffer_target, bool bRenderZToCol0)
+{
+	text[sizeof(text) - 1] = 0x7C;  // canary
+    DVSTARTPROFILE();
+
+    BuildSwapModeTable();
+    int numStages = bpmem.genMode.numtevstages + 1;
+    int numTexgen = bpmem.genMode.numtexgens;
+
+    char *p = text;
+    WRITE(p, "//Pixel Shader for TEV stages\n");
+    WRITE(p, "//%i TEV stages, %i texgens, %i IND stages\n",
+		numStages, numTexgen, bpmem.genMode.numindstages);
+
+    bool bRenderZ = has_zbuffer_target && bpmem.zmode.updateenable;
+    bool bOutputZ = bpmem.ztex2.op != ZTEXTURE_DISABLE;
+    bool bInputZ = bpmem.ztex2.op==ZTEXTURE_ADD || bRenderZ;
+
+    // bool bRenderZToCol0 = ; // output z and alpha to color0
+    assert( !bRenderZToCol0 || bRenderZ );
+
+    int ztexcoord = -1;
+    if (bInputZ)
+        ztexcoord = numTexgen == 0 ? 0 : numTexgen-1;
+
+    int nIndirectStagesUsed = 0;
+    if (bpmem.genMode.numindstages > 0) {
+        for (int i = 0; i < numStages; ++i) {
+            if (bpmem.tevind[i].IsActive() && bpmem.tevind[i].bt < bpmem.genMode.numindstages) {
+                nIndirectStagesUsed |= 1<<bpmem.tevind[i].bt;
+            }
+        }
+    }
+
+    // Declare samplers
+    if (texture_mask) {
+        WRITE(p, "uniform samplerRECT ");
+        bool bfirst = true;
+        for (int i = 0; i < 8; ++i) {
+            if (texture_mask & (1<<i)) {
+                WRITE(p, "%s samp%d : register(s%d)", bfirst?"":",", i, i);
+                bfirst = false;
+            }
+        }
+        WRITE(p, ";\n");
+    }
+
+    if (texture_mask != 0xff) {
+        WRITE(p, "uniform sampler2D ");
+        bool bfirst = true;
+        for (int i = 0; i < 8; ++i) {
+            if (!(texture_mask & (1<<i))) {
+                WRITE(p, "%s samp%d : register(s%d)", bfirst?"":",",i, i);
+                bfirst = false;
+            }
+        }
+        WRITE(p, ";\n");
+    }
+
+    WRITE(p, "\n");
+
+    WRITE(p, "uniform float4 "I_COLORS"[4] : register(c%d);\n", C_COLORS);
+    WRITE(p, "uniform float4 "I_KCOLORS"[4] : register(c%d);\n", C_KCOLORS);
+    WRITE(p, "uniform float4 "I_ALPHA"[1] : register(c%d);\n", C_ALPHA);
+    WRITE(p, "uniform float4 "I_TEXDIMS"[8] : register(c%d);\n", C_TEXDIMS);
+    WRITE(p, "uniform float4 "I_ZBIAS"[2] : register(c%d);\n", C_ZBIAS);
+    WRITE(p, "uniform float4 "I_INDTEXSCALE"[2] : register(c%d);\n", C_INDTEXSCALE);
+    WRITE(p, "uniform float4 "I_INDTEXMTX"[6] : register(c%d);\n", C_INDTEXMTX);
+
+    WRITE(p, "void main(\n");
+
+    WRITE(p, "out half4 ocol0 : COLOR0,\n");
+    if (bRenderZ && !bRenderZToCol0 )
+        WRITE(p, "out half4 ocol1 : COLOR1,\n");
+
+    if (bOutputZ )
+        WRITE(p, "  out float depth : DEPTH,\n");
+
+    // if zcoord might come from vertex shader in texcoord
+    if (bInputZ) {
+        if (numTexgen) {
+            for (int i = 0; i < numTexgen; ++i)
+                WRITE(p, "  in float%d uv%d : TEXCOORD%d, \n", i==ztexcoord?4:3, i,i);
+        }
+        else
+            WRITE(p, "  in float4 uv0 : TEXCOORD0,"); //HACK
+    }
+    else {
+        if (numTexgen) {
+            for (int i = 0; i < numTexgen; ++i)
+                WRITE(p, "  in float3 uv%d : TEXCOORD%d,\n",i,i);
+        }
+        else
+            WRITE(p, "  in float3 uv0 : TEXCOORD0,\n"); //HACK
+    }
+
+    WRITE(p, "  in float4 colors[2] : COLOR0){\n");
+
+    char* pmainstart = p;
+
+    WRITE(p, "float4 c0="I_COLORS"[1],c1="I_COLORS"[2],c2="I_COLORS"[3],prev=float4(0.0f,0.0f,0.0f,0.0f),textemp,rastemp,konsttemp=float4(0.0f,0.0f,0.0f,0.0f);\n"
+            "float3 comp16 = float3(1,255,0), comp24 = float3(1,255,255*255);\n"
+            "float4 alphabump=0;\n"
+            "float3 tevcoord;\n"
+            "float2 wrappedcoord, tempcoord;\n");
+
+    //if (bOutputZ ) WRITE(p, "  float depth;\n");
+//    WRITE(p, "return 1;}\n");
+//    return PixelShaderMngr::CompilePixelShader(ps, text);
+
+    // indirect texture map lookup
+    for(u32 i = 0; i < bpmem.genMode.numindstages; ++i) {
+        if (nIndirectStagesUsed & (1<<i)) {
+            // perform indirect texture map lookup
+            // note that we have to scale by the regular texture map's coordinates since this is a texRECT call
+            // (and we have to match with the game's texscale calls)
+            int texcoord = bpmem.tevindref.getTexCoord(i);
+            if (texture_mask & (1<<bpmem.tevindref.getTexMap(i))) {
+                // TODO: I removed a superfluous argument, please check that the resulting expression is correct. (mthuurne 2008-08-27)
+                WRITE(p, "float2 induv%d=uv%d.xy * "I_INDTEXSCALE"[%d].%s;\n", i, texcoord, i/2, (i&1)?"zw":"xy"); //, bpmem.tevindref.getTexMap(i)
+
+                char str[16];
+                sprintf(str, "induv%d", i);
+                WrapNonPow2Tex(p, str, bpmem.tevindref.getTexMap(i), texture_mask);
+                WRITE(p, "float3 indtex%d=texRECT(samp%d,induv%d.xy).abg;\n", i, bpmem.tevindref.getTexMap(i), i);
+            }
+            else {
+                WRITE(p, "float3 indtex%d=tex2D(samp%d,uv%d.xy*"I_INDTEXSCALE"[%d].%s).abg;\n", i, bpmem.tevindref.getTexMap(i), texcoord, i/2, (i&1)?"zw":"xy");
+            }
+        }
+    }
+
+    for (int i = 0; i < numStages; i++)
+        WriteStage(p, i, texture_mask); //build the equation for this stage
+
+    if (bOutputZ) {
+        // use the texture input of the last texture stage (textemp), hopefully this has been read and is in correct format...
+        if (bpmem.ztex2.op == ZTEXTURE_ADD) {
+            WRITE(p, "depth = frac(dot("I_ZBIAS"[0].xyzw, textemp.xyzw) + "I_ZBIAS"[1].w + uv%d.w);\n", ztexcoord);
+        }
+        else {
+            _assert_(bpmem.ztex2.op == ZTEXTURE_REPLACE);
+            WRITE(p, "depth = frac(dot("I_ZBIAS"[0].xyz, textemp.xyz) + "I_ZBIAS"[0].w);\n");
+        }
+    }
+
+    //if (bpmem.genMode.numindstages ) WRITE(p, "prev.rg = indtex0.xy;\nprev.b = 0;\n");
+
+    if (!WriteAlphaTest(p)) {
+        // alpha test will always fail, so restart the shader and just make it an empty function
+        p = pmainstart;
+		WRITE(p, "discard;\n");
+        WRITE(p, "ocol0 = 0;\n");
+    }
+    else {
+        if (!bRenderZToCol0) {
+            /* donkopunchstania: NEEDS FIX - dstalpha does not change how fragments are blended with the EFB
+			   once the blending is done, the dstalpha is written to the EFB in place of the
+			   fragment alpha if dstalpha is enabled. this only matters if the EFB supports alpha.
+               Commenting this out fixed Metroids but causes glitches in Super Mario Sunshine.
+
+			if (bpmem.dstalpha.enable)
+                WRITE(p, "  ocol0 = float4(prev.rgb,"I_ALPHA"[0].w);\n");
+            else
+			   */
+            	WRITE(p, "  ocol0 = prev;\n");
+        }
+    }
+
+    if (bRenderZ) {
+        // write depth as color
+        if (bRenderZToCol0) {
+            if (bOutputZ )
+                WRITE(p, "ocol0.xyz = frac(float3(256.0f*256.0f, 256.0f, 1.0f) * depth);\n");
+            else
+                WRITE(p, "ocol0.xyz = frac(float3(256.0f*256.0f, 256.0f, 1.0f) * uv%d.w);\n", ztexcoord);
+            WRITE(p, "ocol0.w = prev.w;\n");
+        }
+        else {
+            if (bOutputZ)
+                WRITE(p, "ocol1 = frac(float4(256.0f*256.0f, 256.0f, 1.0f, 0.0f) * depth);\n");
+            else
+                WRITE(p, "ocol1 = frac(float4(256.0f*256.0f, 256.0f, 1.0f, 0.0f) * uv%d.w);\n", ztexcoord);
+        }
+    }
+    WRITE(p, "}\n");
+	if (text[sizeof(text) - 1] != 0x7C)
+		PanicAlert("PixelShader generator - buffer too small, canary has been eaten!");
+    return text;
+}
+
+static void WriteStage(char *&p, int n, u32 texture_mask)
+{
+    char *rasswap = swapModeTable[bpmem.combiners[n].alphaC.rswap];
+    char *texswap = swapModeTable[bpmem.combiners[n].alphaC.tswap];
+
+
+    int texcoord = bpmem.tevorders[n/2].getTexCoord(n&1);
+    int texfun = xfregs.texcoords[texcoord].texmtxinfo.projection;
+    bool bHasIndStage = bpmem.tevind[n].IsActive() && bpmem.tevind[n].bt < bpmem.genMode.numindstages;
+
+    if (bHasIndStage) {
+        // perform the indirect op on the incoming regular coordinates using indtex%d as the offset coords
+        bHasIndStage = true;
+        int texmap = bpmem.tevorders[n/2].getEnable(n&1) ? bpmem.tevorders[n/2].getTexMap(n&1) : bpmem.tevindref.getTexMap(bpmem.tevind[n].bt);
+
+		if (bpmem.tevind[n].bs != ITBA_OFF) {
+            // write the bump alpha
+
+			if (bpmem.tevind[n].fmt == ITF_8) {
+				WRITE(p, "alphabump = indtex%d.%s %s;\n", bpmem.tevind[n].bt, 
+					tevIndAlphaSel[bpmem.tevind[n].bs], tevIndAlphaScale[bpmem.tevind[n].fmt]);
+			}
+			else {			
+				// donkopunchstania: really bad way to do this
+				// cannot always use fract because fract(1.0) is 0.0 when it needs to be 1.0
+				// omitting fract seems to work as well
+				WRITE(p, "if (indtex%d.%s >= 1.0f )\n", bpmem.tevind[n].bt, 
+					tevIndAlphaSel[bpmem.tevind[n].bs]);
+				WRITE(p, "   alphabump = 1.0f;\n");
+				WRITE(p, "else\n");
+				WRITE(p, "   alphabump = fract ( indtex%d.%s %s );\n", bpmem.tevind[n].bt, 
+					tevIndAlphaSel[bpmem.tevind[n].bs], tevIndAlphaScale[bpmem.tevind[n].fmt]);
+			}
+		}
+
+        // bias
+        WRITE(p, "float3 indtevcrd%d = indtex%d;\n", n, bpmem.tevind[n].bt);
+        WRITE(p, "indtevcrd%d.xyz *= %s;\n", n, tevIndFmtScale[bpmem.tevind[n].fmt]);
+        if (bpmem.tevind[n].bias != ITB_NONE )
+            WRITE(p, "indtevcrd%d.%s += %s;\n", n, tevIndBiasField[bpmem.tevind[n].bias], tevIndBiasAdd[bpmem.tevind[n].fmt]);
+
+        // multiply by offset matrix and scale
+        if (bpmem.tevind[n].mid != 0) {
+            if (bpmem.tevind[n].mid <= 3) {
+                int mtxidx = 2*(bpmem.tevind[n].mid-1);
+                WRITE(p, "float2 indtevtrans%d = float2(dot("I_INDTEXMTX"[%d].xyz, indtevcrd%d), dot("I_INDTEXMTX"[%d].xyz, indtevcrd%d));\n",
+                    n, mtxidx, n, mtxidx+1, n);
+            }
+            else if (bpmem.tevind[n].mid <= 5) { // s matrix
+                int mtxidx = 2*(bpmem.tevind[n].mid-5);
+                WRITE(p, "float2 indtevtrans%d = "I_INDTEXMTX"[%d].ww * uv%d.xy * indtevcrd%d.xx;\n", n, mtxidx, texcoord, n);
+            }
+            else if (bpmem.tevind[n].mid <= 9) { // t matrix
+                int mtxidx = 2*(bpmem.tevind[n].mid-9);
+                WRITE(p, "float2 indtevtrans%d = "I_INDTEXMTX"[%d].ww * uv%d.xy * indtevcrd%d.yy;\n", n, mtxidx, texcoord, n);
+            }
+            else {
+                // TODO: I removed a superfluous argument, please check that the resulting expression is correct. (mthuurne 2008-08-27)
+                WRITE(p, "float2 indtevtrans%d = 0;\n", n); //, n
+            }
+        }
+        else {
+            // TODO: I removed a superfluous argument, please check that the resulting expression is correct. (mthuurne 2008-08-27)
+            WRITE(p, "float2 indtevtrans%d = 0;\n", n); //, n
+        }
+
+        // wrapping
+        if (!bpmem.tevorders[n/2].getEnable(n&1) || (texture_mask & (1<<texmap))) {
+            // non pow2
+
+            if (bpmem.tevind[n].sw != ITW_OFF || bpmem.tevind[n].tw != ITW_OFF) {
+                if (bpmem.tevind[n].sw == ITW_0) {
+                    if (bpmem.tevind[n].tw == ITW_0) {
+                        // zero out completely
+                        WRITE(p, "wrappedcoord = float2(0.0f,0.0f);\n");
+                    }
+                    else {
+                        WRITE(p, "wrappedcoord.x = fmod( (uv%d.x+%s)*"I_TEXDIMS"[%d].x*"I_TEXDIMS"[%d].z, %s);\n"
+                            "wrappedcoord.y = 0;\n", texcoord, tevIndWrapStart[bpmem.tevind[n].sw], texmap, texmap, tevIndWrapStart[bpmem.tevind[n].sw]);
+                    }
+                }
+                else if (bpmem.tevind[n].tw == ITW_0) {
+                    WRITE(p, "wrappedcoord.y = fmod( (uv%d.y+%s)*"I_TEXDIMS"[%d].y*"I_TEXDIMS"[%d].w, %s);\n"
+                        "wrappedcoord.x = 0;\n", texcoord, tevIndWrapStart[bpmem.tevind[n].tw], texmap, texmap, tevIndWrapStart[bpmem.tevind[n].tw]);
+                }
+                else {
+                    WRITE(p, "wrappedcoord = fmod( (uv%d.xy+float2(%s,%s))*"I_TEXDIMS"[%d].xy*"I_TEXDIMS"[%d].zw, float2(%s,%s));\n", texcoord,
+                        tevIndWrapStart[bpmem.tevind[n].sw], tevIndWrapStart[bpmem.tevind[n].tw],texmap,texmap,
+                        tevIndWrapStart[bpmem.tevind[n].sw], tevIndWrapStart[bpmem.tevind[n].tw]);
+                }
+            }
+            else {
+                WRITE(p, "wrappedcoord = uv%d.xy*"I_TEXDIMS"[%d].xy;\n", texcoord, texmap);
+            }
+        }
+        else {
+            // pow of 2
+            WRITE(p, "indtevtrans%d.xy *= "I_TEXDIMS"[%d].xy * "I_TEXDIMS"[%d].zw;\n", n, texmap, texmap);
+
+            // mult by bitdepth / tex dimensions
+            if (bpmem.tevind[n].sw != ITW_OFF || bpmem.tevind[n].tw != ITW_OFF) {
+                if (bpmem.tevind[n].sw == ITW_0) {
+                    if (bpmem.tevind[n].tw == ITW_0) {
+                        // zero out completely
+                        WRITE(p, "wrappedcoord = float2(0.0f,0.0f);\n");
+                    }
+                    else {
+                        WRITE(p, "wrappedcoord.x = "I_TEXDIMS"[%d].x * fmod( uv%d.x+%s, "I_TEXDIMS"[%d].z*%s);\n"
+                            "wrappedcoord.y = 0;\n", texmap, texcoord, tevIndWrapStart[bpmem.tevind[n].sw], texmap, tevIndWrapStart[bpmem.tevind[n].sw]);
+                    }
+                }
+                else if (bpmem.tevind[n].tw == ITW_0) {
+                    WRITE(p, "wrappedcoord.y = "I_TEXDIMS"[%d].y * fmod( uv%d.y+%s, "I_TEXDIMS"[%d].w*%s);\n"
+                        "wrappedcoord.x = 0;\n", texmap, texcoord, tevIndWrapStart[bpmem.tevind[n].tw], texmap, tevIndWrapStart[bpmem.tevind[n].tw]);
+                }
+                else {
+                    // have to add an offset or else might get negative values!
+                    WRITE(p, "wrappedcoord = "I_TEXDIMS"[%d].xy * fmod( uv%d.xy+float2(%s,%s), "I_TEXDIMS"[%d].zw*float2(%s,%s));\n", texmap, texcoord,
+                        tevIndWrapStart[bpmem.tevind[n].sw], tevIndWrapStart[bpmem.tevind[n].tw], texmap, 
+                        tevIndWrapStart[bpmem.tevind[n].sw], tevIndWrapStart[bpmem.tevind[n].tw]);
+                }
+            }
+            else {
+                WRITE(p, "wrappedcoord = uv%d.xy;\n", texcoord);
+            }
+        }
+
+        if (bpmem.tevind[n].fb_addprev) {
+            // add previous tevcoord
+
+            if (texfun == XF_TEXPROJ_STQ) {
+                WRITE(p, "tevcoord.xy += wrappedcoord/uv%d.z + indtevtrans%d;\n", texcoord, n);
+                //WRITE(p, "tevcoord.z += uv%d.z;\n", texcoord);
+            }
+            else {
+                WRITE(p, "tevcoord.xy += wrappedcoord + indtevtrans%d;\n", n);
+            }
+        }
+        else {
+            WRITE(p, "tevcoord.xy = wrappedcoord/uv%d.z + indtevtrans%d;\n", texcoord, n);
+            //if (texfun == XF_TEXPROJ_STQ )
+            //    WRITE(p, "tevcoord.z = uv%d.z;\n", texcoord);
+        }
+    }
+
+    WRITE(p, "rastemp=%s.%s;\n",tevRasTable[bpmem.tevorders[n/2].getColorChan(n&1)],rasswap);
+
+    if (bpmem.tevorders[n/2].getEnable(n&1)) {
+        int texmap = bpmem.tevorders[n/2].getTexMap(n&1);
+        if(!bHasIndStage) {
+            // calc tevcord
+            //tevcoord.xy = texdim[1].xy * uv1.xy / uv1.z;
+            int OurTexCoord = 0;
+            if(bpmem.genMode.numtexgens)
+                OurTexCoord = texcoord;
+            else
+                OurTexCoord = 0;
+            if (texture_mask & (1<<texmap)) {
+                // nonpow2
+                if (texfun == XF_TEXPROJ_STQ )
+                    WRITE(p, "tevcoord.xy = uv%d.xy / uv%d.z;\n", texcoord, OurTexCoord);
+                else
+                    WRITE(p, "tevcoord.xy = uv%d.xy;\n", OurTexCoord);
+                WrapNonPow2Tex(p, "tevcoord", texmap, texture_mask);
+            }
+            else {
+                if (texfun == XF_TEXPROJ_STQ ) 
+                    WRITE(p, "tevcoord.xy = "I_TEXDIMS"[%d].xy * uv%d.xy / uv%d.z;\n", texmap, OurTexCoord , OurTexCoord );
+                else 
+                    WRITE(p, "tevcoord.xy = "I_TEXDIMS"[%d].xy * uv%d.xy;\n", texmap, OurTexCoord);
+
+            }
+        }
+        else if (texture_mask & (1<<texmap)) {
+            // if non pow 2, have to manually repeat
+            //WrapNonPow2Tex(p, "tevcoord", texmap);
+            bool bwraps = !!(texture_mask & (0x100<<texmap));
+            bool bwrapt = !!(texture_mask & (0x10000<<texmap));
+                
+            if (bwraps || bwrapt) {
+                const char* field = bwraps ? (bwrapt ? "xy" : "x") : "y";
+                WRITE(p, "tevcoord.%s = fmod(tevcoord.%s+32*"I_TEXDIMS"[%d].%s,"I_TEXDIMS"[%d].%s);\n", field, field, texmap, field, texmap, field);
+            }
+        }
+
+        if (texture_mask & (1<<texmap) )
+            WRITE(p, "textemp=texRECT(samp%d,tevcoord.xy).%s;\n", texmap, texswap);
+        else
+            WRITE(p, "textemp=tex2D(samp%d,tevcoord.xy).%s;\n", texmap, texswap);
+    }
+    else
+        WRITE(p, "textemp=float4(1,1,1,1);\n");
+
+    int kc = bpmem.tevksel[n/2].getKC(n&1);
+    int ka = bpmem.tevksel[n/2].getKA(n&1);
+
+    TevStageCombiner::ColorCombiner &cc = bpmem.combiners[n].colorC;
+    TevStageCombiner::AlphaCombiner &ac = bpmem.combiners[n].alphaC;
+
+    bool bCKonst = cc.a == TEVCOLORARG_KONST || cc.b == TEVCOLORARG_KONST || cc.c == TEVCOLORARG_KONST || cc.d == TEVCOLORARG_KONST;
+    bool bAKonst = ac.a == TEVALPHAARG_KONST || ac.b == TEVALPHAARG_KONST || ac.c == TEVALPHAARG_KONST || ac.d == TEVALPHAARG_KONST;
+    if (bCKonst || bAKonst )
+        WRITE(p, "konsttemp=float4(%s,%s);\n",tevKSelTableC[kc],tevKSelTableA[ka]);  
+
+    WRITE(p, "%s= ", tevCOutputTable[cc.dest]); 
+
+    // combine the color channel
+    if (cc.bias != 3) {  // if not compare
+        //normal color combiner goes here
+        WRITE(p, "   %s*(%s%s",tevScaleTable[cc.shift],tevCInputTable[cc.d],tevOpTable[cc.op]);
+        WRITE(p, "lerp(%s,%s,%s) %s);\n",
+              tevCInputTable[cc.a], tevCInputTable[cc.b],
+              tevCInputTable[cc.c], tevBiasTable[cc.bias]);
+    }
+    else {
+        int cmp = (cc.shift<<1)|cc.op|8; // comparemode stored here
+        switch(cmp) {
+        case TEVCMP_R8_GT:
+        case TEVCMP_RGB8_GT: // per component compares
+            WRITE(p, "   %s + ((%s.%s > %s.%s) ? %s : float3(0.0f,0.0f,0.0f));\n",
+                tevCInputTable[cc.d], tevCInputTable2[cc.a], cmp==TEVCMP_R8_GT?"r":"rgb", tevCInputTable2[cc.b], cmp==TEVCMP_R8_GT?"r":"rgb", tevCInputTable[cc.c]);
+            break;
+        case TEVCMP_R8_EQ:
+        case TEVCMP_RGB8_EQ:
+            WRITE(p, "   %s + (abs(%s.r - %s.r)<%f ? %s : float3(0.0f,0.0f,0.0f));\n",
+                tevCInputTable[cc.d], tevCInputTable2[cc.a], tevCInputTable2[cc.b], epsilon8bit, tevCInputTable[cc.c]);
+            break;
+        
+        case TEVCMP_GR16_GT: // 16 bit compares: 255*g+r (probably used for ztextures, so make sure in ztextures, g is the most significant byte)
+        case TEVCMP_BGR24_GT: // 24 bit compares: 255*255*b+255*g+r
+            WRITE(p, "   %s + (( dot(%s.rgb-%s.rgb, comp%s) > 0) ? %s : float3(0.0f,0.0f,0.0f));\n",
+                tevCInputTable[cc.d], tevCInputTable2[cc.a], tevCInputTable2[cc.b], cmp==TEVCMP_GR16_GT?"16":"24", tevCInputTable[cc.c]);
+            break;
+        case TEVCMP_GR16_EQ:
+        case TEVCMP_BGR24_EQ:
+            WRITE(p, "   %s + (abs(dot(%s.rgb - %s.rgb, comp%s))<%f ? %s : float3(0.0f,0.0f,0.0f));\n",
+                tevCInputTable[cc.d], tevCInputTable2[cc.a], tevCInputTable2[cc.b], cmp==TEVCMP_GR16_GT?"16":"24", epsilon8bit, tevCInputTable[cc.c]);
+            break;
+        default:
+            WRITE(p, "float3(0.0f,0.0f,0.0f);\n");
+            break;
+        }
+    }
+    
+    if (cc.clamp)
+        WRITE(p, "%s = clamp(%s,0.0f,1.0f);\n", tevCOutputTable[cc.dest],tevCOutputTable[cc.dest]);
+
+    // combine the alpha channel
+    WRITE(p, "%s= ", tevAOutputTable[ac.dest]);
+
+    if (ac.bias != 3) { // if not compare
+        //normal alpha combiner goes here
+        WRITE(p, "   %s*(%s%s",tevScaleTable[ac.shift],tevAInputTable[ac.d],tevOpTable[ac.op]);
+        WRITE(p, "lerp(%s,%s,%s) %s)\n",
+            tevAInputTable[ac.a],tevAInputTable[ac.b],
+            tevAInputTable[ac.c],tevBiasTable[ac.bias]);
+    }
+    else {
+        //compare alpha combiner goes here
+        int cmp = (ac.shift<<1)|ac.op|8; // comparemode stored here
+        switch(cmp) {
+        case TEVCMP_R8_GT:
+        case TEVCMP_A8_GT:
+            WRITE(p, "   %s + ((%s.%s > %s.%s) ? %s : 0)\n",
+                tevAInputTable[ac.d],tevAInputTable2[ac.a], cmp==TEVCMP_R8_GT?"r":"a", tevAInputTable2[ac.b], cmp==TEVCMP_R8_GT?"r":"a", tevAInputTable[ac.c]);
+            break;
+        case TEVCMP_R8_EQ:
+        case TEVCMP_A8_EQ:
+            WRITE(p, "   %s + (abs(%s.r - %s.r)<%f ? %s : 0)\n",
+                tevAInputTable[ac.d],tevAInputTable2[ac.a], tevAInputTable2[ac.b],epsilon8bit,tevAInputTable[ac.c]);
+            break;
+        
+        case TEVCMP_GR16_GT: // 16 bit compares: 255*g+r (probably used for ztextures, so make sure in ztextures, g is the most significant byte)
+        case TEVCMP_BGR24_GT: // 24 bit compares: 255*255*b+255*g+r
+            WRITE(p, "   %s + (( dot(%s.rgb-%s.rgb, comp%s) > 0) ? %s : 0)\n",
+                tevAInputTable[ac.d],tevAInputTable2[ac.a], tevAInputTable2[ac.b], cmp==TEVCMP_GR16_GT?"16":"24", tevAInputTable[ac.c]);
+            break;
+        case TEVCMP_GR16_EQ:
+        case TEVCMP_BGR24_EQ:
+            WRITE(p, "   %s + (abs(dot(%s.rgb - %s.rgb, comp%s))<%f ? %s : 0)\n",
+                tevAInputTable[ac.d],tevAInputTable2[ac.a], tevAInputTable2[ac.b],cmp==TEVCMP_GR16_GT?"16":"24",epsilon8bit,tevAInputTable[ac.c]);
+            break;
+        default:
+            WRITE(p, "0)\n");
+            break;
+        }
+    }
+
+    WRITE(p, ";\n");
+
+    if (ac.clamp)
+        WRITE(p, "%s = clamp(%s,0.0f,1.0f);\n", tevAOutputTable[ac.dest],tevAOutputTable[ac.dest]);
+    WRITE(p, "\n");
+}
+
+void WrapNonPow2Tex(char* &p, const char* var, int texmap, u32 texture_mask)
+{
+    _assert_(texture_mask & (1<<texmap));
+    bool bwraps = !!(texture_mask & (0x100<<texmap));
+    bool bwrapt = !!(texture_mask & (0x10000<<texmap));
+        
+    if (bwraps || bwrapt) {
+        const char* field = bwraps ? (bwrapt ? "xy" : "x") : "y";
+        const char* wrapfield = bwraps ? (bwrapt ? "zw" : "z") : "w";
+        WRITE(p, "%s.%s = "I_TEXDIMS"[%d].%s*frac(%s.%s*"I_TEXDIMS"[%d].%s+32);\n", var, field, texmap, field, var, field, texmap, wrapfield);
+
+        if (!bwraps )
+            WRITE(p, "%s.x *= "I_TEXDIMS"[%d].x * "I_TEXDIMS"[%d].z;\n", var, texmap, texmap);
+        if (!bwrapt )
+            WRITE(p, "%s.y *= "I_TEXDIMS"[%d].y * "I_TEXDIMS"[%d].w;\n", var, texmap, texmap);
+    }
+    else {
+        WRITE(p, "%s.xy *= "I_TEXDIMS"[%d].xy * "I_TEXDIMS"[%d].zw;\n", var, texmap, texmap);
+    }
+}
+
+static void WriteAlphaCompare(char *&p, int num, int comp)
+{
+    switch(comp) {
+    case ALPHACMP_ALWAYS:  WRITE(p, "(false)");	break;
+    case ALPHACMP_NEVER:   WRITE(p, "(true)");	break;
+    case ALPHACMP_LEQUAL:  WRITE(p, "(prev.a > %s)",alphaRef[num]);	break;
+    case ALPHACMP_LESS:    WRITE(p, "(prev.a >= %s - %f)",alphaRef[num],epsilon8bit*0.5f);break;
+    case ALPHACMP_GEQUAL:  WRITE(p, "(prev.a < %s)",alphaRef[num]);	break;
+    case ALPHACMP_GREATER: WRITE(p, "(prev.a <= %s + %f)",alphaRef[num],epsilon8bit*0.5f);break;
+    case ALPHACMP_EQUAL:   WRITE(p, "(abs(prev.a-%s)>%f)",alphaRef[num],epsilon8bit*2); break;
+    case ALPHACMP_NEQUAL:  WRITE(p, "(abs(prev.a-%s)<%f)",alphaRef[num],epsilon8bit*2); break;
+    }
+}
+
+static bool WriteAlphaTest(char *&p)
+{
+    u32 op = bpmem.alphaFunc.logic;
+    u32 comp[2] = {bpmem.alphaFunc.comp0,bpmem.alphaFunc.comp1};
+
+    //first kill all the simple cases
+    switch(op) {
+    case 0: // and
+        if (comp[0] == ALPHACMP_ALWAYS && comp[1] == ALPHACMP_ALWAYS) return true;
+        if (comp[0] == ALPHACMP_NEVER || comp[1] == ALPHACMP_NEVER) {
+            WRITE(p, "discard;\n");
+            return false;
+        }
+        break;
+    case 1: // or
+        if (comp[0] == ALPHACMP_ALWAYS || comp[1] == ALPHACMP_ALWAYS) return true;
+        if (comp[0] == ALPHACMP_NEVER && comp[1] == ALPHACMP_NEVER) {
+            WRITE(p, "discard;\n");
+            return false;
+        }
+        break;
+    case 2: // xor
+        if ( (comp[0] == ALPHACMP_ALWAYS && comp[1] == ALPHACMP_NEVER) || (comp[0] == ALPHACMP_NEVER && comp[1] == ALPHACMP_ALWAYS) ) return true;
+        if ( (comp[0] == ALPHACMP_ALWAYS && comp[1] == ALPHACMP_ALWAYS) || (comp[0] == ALPHACMP_NEVER && comp[1] == ALPHACMP_NEVER)) {
+            WRITE(p, "discard;\n");
+            return false;
+        }
+        break;
+    case 3: // xnor
+        if ( (comp[0] == ALPHACMP_ALWAYS && comp[1] == ALPHACMP_NEVER) || (comp[0] == ALPHACMP_NEVER && comp[1] == ALPHACMP_ALWAYS)) {
+            WRITE(p, "discard;\n");
+            return false;
+        }
+        if ( (comp[0] == ALPHACMP_ALWAYS && comp[1] == ALPHACMP_ALWAYS) || (comp[0] == ALPHACMP_NEVER && comp[1] == ALPHACMP_NEVER) )
+            return true;
+        break;
+    }
+
+    WRITE(p, "discard( ");
+    WriteAlphaCompare(p, 0, bpmem.alphaFunc.comp0);
+    
+    // negated because testing the inverse condition
+    switch(bpmem.alphaFunc.logic) {
+    case 0: WRITE(p, " || "); break; // and
+    case 1: WRITE(p, " && "); break; // or
+    case 2: WRITE(p, " == "); break; // xor
+    case 3: WRITE(p, " != "); break; // xnor
+    }
+    WriteAlphaCompare(p, 1, bpmem.alphaFunc.comp1);
+    WRITE(p, ");\n");
+    return true;
+}
diff --git a/Source/Core/VideoCommon/Src/Profiler.cpp b/Source/Core/VideoCommon/Src/Profiler.cpp
index b66a4d32d2..d18c6a0ba9 100644
--- a/Source/Core/VideoCommon/Src/Profiler.cpp
+++ b/Source/Core/VideoCommon/Src/Profiler.cpp
@@ -1,291 +1,291 @@
-// Copyright (C) 2003-2008 Dolphin Project.

-

-// This program is free software: you can redistribute it and/or modify

-// it under the terms of the GNU General Public License as published by

-// the Free Software Foundation, version 2.0.

-

-// This program is distributed in the hope that it will be useful,

-// but WITHOUT ANY WARRANTY; without even the implied warranty of

-// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

-// GNU General Public License 2.0 for more details.

-

-// A copy of the GPL 2.0 should have been included with the program.

-// If not, see http://www.gnu.org/licenses/

-

-// Official SVN repository and contact information can be found at

-// http://code.google.com/p/dolphin-emu/

-

-// Simple profiler

-

-#include "Common.h"

-#include "Profiler.h"

-

-#include <list>

-#include <string>

-#include <map>

-

-#ifdef _WIN32

-

-#if defined (_MSC_VER) && _MSC_VER >= 1400

-#define _interlockedbittestandset workaround_ms_header_bug_platform_sdk6_set

-#define _interlockedbittestandreset workaround_ms_header_bug_platform_sdk6_reset

-#define _interlockedbittestandset64 workaround_ms_header_bug_platform_sdk6_set64

-#define _interlockedbittestandreset64 workaround_ms_header_bug_platform_sdk6_reset64

-#include <intrin.h>

-#undef _interlockedbittestandset

-#undef _interlockedbittestandreset

-#undef _interlockedbittestandset64

-#undef _interlockedbittestandreset64

-#pragma intrinsic(__rdtsc)

-#endif

-

-// Globals

-static u64 luPerfFreq = 0;

-#ifdef DVPROFILE

-int g_bWriteProfile = 1;

-#else

-int g_bWriteProfile = 1;

-#endif

-

-inline u64 GET_PROFILE_TIME()

-{

-#if defined (_MSC_VER) && _MSC_VER >= 1400

-    return __rdtsc();

-#else

-    LARGE_INTEGER lu;

-    QueryPerformanceCounter(&lu);

-    return lu.QuadPart;

-#endif

-}

-#else

-static u64 luPerfFreq = 1000000;

-#define GET_PROFILE_TIME() //GetCpuTick()

-#endif

-

-struct DVPROFSTRUCT;

-

-struct DVPROFSTRUCT

-{

-    struct DATA

-    {

-        DATA(u64 time, u32 user = 0) : dwTime(time), dwUserData(user) {}

-        DATA() : dwTime(0), dwUserData(0) {}

-

-        u64 dwTime;

-        u32 dwUserData;

-    };

-

-    ~DVPROFSTRUCT() {

-        std::list<DVPROFSTRUCT *>::iterator it = listpChild.begin();

-        while (it != listpChild.end()) {

-            delete *it;

-			*it = NULL;

-            ++it;

-        }

-    }

-

-	// before DVProfEnd is called, contains the global time it started

-	// after DVProfEnd is called, contains the time it lasted

-	// the list contains all the tracked times 

-    std::list<DATA> listTimes;

-                              

-    char pname[256];

-    std::list<DVPROFSTRUCT*> listpChild;     // other profilers called during this profiler period

-};

-

-struct DVPROFTRACK

-{

-    u32 dwUserData;

-    DVPROFSTRUCT::DATA* pdwTime;

-    DVPROFSTRUCT* pprof;

-};

-

-// the current profiling functions, the back element is the

-// one that will first get popped off the list when DVProfEnd is called

-// the pointer is an element in DVPROFSTRUCT::listTimes

-static std::list<DVPROFTRACK> g_listCurTracking;

-

-// the current profilers, note that these are the parents

-// any profiler started during the time of another is held in

-// DVPROFSTRUCT::listpChild

-static std::list<DVPROFSTRUCT> g_listProfilers; 

-

-// ignores the hierarchy, pointer to elements in g_listProfilers

-static std::list<DVPROFSTRUCT*> g_listAllProfilers;

-

-

-void DVProfRegister(const char *pname)

-{

-    if (!g_bWriteProfile)

-        return;

-

-#ifdef _WIN32

-    if (luPerfFreq <= 1) {

-#if defined (_MSC_VER) && _MSC_VER >= 1400

-        luPerfFreq = 1000000;

-#else

-        LARGE_INTEGER temp;

-        QueryPerformanceFrequency(&temp);

-        luPerfFreq = temp.QuadPart;

-#endif

-    }

-#endif

-

-    std::list<DVPROFSTRUCT*>::iterator it = g_listAllProfilers.begin();

-    

-//  while(it != g_listAllProfilers.end() ) {

-//

-//      if( _tcscmp(pname, (*it)->pname) == 0 ) {

-//          (*it)->listTimes.push_back(timeGetTime());

-//          DVPROFTRACK dvtrack;

-//          dvtrack.pdwTime = &(*it)->listTimes.back();

-//          dvtrack.pprof = *it;

-//          g_listCurTracking.push_back(dvtrack);

-//          return;

-//      }

-//

-//      ++it;

-//  }

-

-    // else add in a new profiler to the appropriate parent profiler

-    DVPROFSTRUCT* pprof = NULL;

-    

-    if (g_listCurTracking.size() > 0) {

-        _assert_( g_listCurTracking.back().pprof != NULL );

-        g_listCurTracking.back().pprof->listpChild.push_back(new DVPROFSTRUCT());

-        pprof = g_listCurTracking.back().pprof->listpChild.back();

-    }

-    else {

-        g_listProfilers.push_back(DVPROFSTRUCT());

-        pprof = &g_listProfilers.back();

-    }

-

-    strncpy(pprof->pname, pname, 256);

-

-    // setup the profiler for tracking

-    pprof->listTimes.push_back(DVPROFSTRUCT::DATA(GET_PROFILE_TIME()));

-

-    DVPROFTRACK dvtrack;

-    dvtrack.pdwTime = &pprof->listTimes.back();

-    dvtrack.pprof = pprof;

-    dvtrack.dwUserData = 0;

-

-    g_listCurTracking.push_back(dvtrack);

-

-    // add to all profiler list

-    g_listAllProfilers.push_back(pprof);

-}

-

-void DVProfEnd(u32 dwUserData)

-{

-    if (!g_bWriteProfile)

-        return;

-    if (g_listCurTracking.size() == 0)

-        return;

-

-    DVPROFTRACK dvtrack = g_listCurTracking.back();

-

-    _assert_( dvtrack.pdwTime != NULL && dvtrack.pprof != NULL );

-

-    dvtrack.pdwTime->dwTime = GET_PROFILE_TIME()- dvtrack.pdwTime->dwTime;

-    dvtrack.pdwTime->dwUserData= dwUserData;

-

-    g_listCurTracking.pop_back();

-}

-

-struct DVTIMEINFO

-{

-    DVTIMEINFO() : uInclusive(0), uExclusive(0) {}

-    u64 uInclusive, uExclusive;

-};

-

-std::map<std::string, DVTIMEINFO> mapAggregateTimes;

-

-u64 DVProfWriteStruct(FILE* f, const DVPROFSTRUCT* p, int ident)

-{

-    fprintf(f, "%*s%s - ", ident, "", p->pname);

-    std::list<DVPROFSTRUCT::DATA>::const_iterator ittime = p->listTimes.begin();

-    u64 utime = 0;

-    while (ittime != p->listTimes.end()) {

-        utime += ittime->dwTime;

-        if (ittime->dwUserData) 

-            fprintf(f, "time: %d, user: 0x%8.8x", (u32)ittime->dwTime, ittime->dwUserData);

-        else

-            fprintf(f, "time: %d", (u32)ittime->dwTime);

-        ++ittime;

-    }

-

-    // yes this is necessary, maps have problems with constructors on their type

-    std::map<std::string, DVTIMEINFO>::iterator ittimes = mapAggregateTimes.find(p->pname);

-    if (ittimes == mapAggregateTimes.end()) {

-        ittimes = mapAggregateTimes.insert(std::map<std::string, DVTIMEINFO>::value_type(p->pname, DVTIMEINFO())).first;

-        ittimes->second.uExclusive = 0;

-        ittimes->second.uInclusive = 0;

-    }

-

-    ittimes->second.uInclusive += utime;

-

-    fprintf(f, "\n");

-

-    std::list<DVPROFSTRUCT*>::const_iterator itprof = p->listpChild.begin();

-

-    u64 uex = utime;

-    while (itprof != p->listpChild.end()) {

-        uex -= DVProfWriteStruct(f, *itprof, ident+4);

-        ++itprof;

-    }

-

-    if (uex > utime) {

-        uex = 0;

-    }

-

-    ittimes->second.uExclusive += uex;

-    return utime;

-}

-

-void DVProfWrite(const char* pfilename, u32 frames)

-{

-    _assert_( pfilename != NULL );

-    FILE* f = fopen(pfilename, "w");

-

-    // pop back any unused

-    mapAggregateTimes.clear();

-    std::list<DVPROFSTRUCT>::iterator it = g_listProfilers.begin();

-

-    while (it != g_listProfilers.end() ) {

-        DVProfWriteStruct(f, &(*it), 0);

-        ++it;

-    }

-

-    std::map<std::string, DVTIMEINFO>::const_iterator iter;

-    fprintf(f, "\n\n-------------------------------------------------------------------\n\n");

-

-    u64 uTotal[2] = {0};

-    double fiTotalTime[2];

-

-    for (iter = mapAggregateTimes.begin(); iter != mapAggregateTimes.end(); ++iter) {

-        uTotal[0] += iter->second.uExclusive;

-        uTotal[1] += iter->second.uInclusive;

-    }

-

-    fprintf(f, "total times (%d): ex: %Lu ", frames, 1000000 * uTotal[0] / (luPerfFreq*(u64)frames));

-    fprintf(f, "inc: %Lu\n", 1000000 * uTotal[1]/(luPerfFreq*(u64)frames));

-

-    fiTotalTime[0] = 1.0 / (double)uTotal[0];

-    fiTotalTime[1] = 1.0 / (double)uTotal[1];

-

-    // output the combined times

-    for (iter = mapAggregateTimes.begin(); iter != mapAggregateTimes.end(); ++iter) {

-        fprintf(f, "%s - ex: %f inc: %f\n", iter->first.c_str(), (float)((double)iter->second.uExclusive * fiTotalTime[0]),

-            (float)((double)iter->second.uInclusive * fiTotalTime[1]));

-    }

-

-    fclose(f);

-}

-

-void DVProfClear()

-{

-    g_listCurTracking.clear();

-    g_listProfilers.clear();

-    g_listAllProfilers.clear();

-}

+// Copyright (C) 2003-2008 Dolphin Project.
+
+// This program is free software: you can redistribute it and/or modify
+// it under the terms of the GNU General Public License as published by
+// the Free Software Foundation, version 2.0.
+
+// This program is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU General Public License 2.0 for more details.
+
+// A copy of the GPL 2.0 should have been included with the program.
+// If not, see http://www.gnu.org/licenses/
+
+// Official SVN repository and contact information can be found at
+// http://code.google.com/p/dolphin-emu/
+
+// Simple profiler
+
+#include "Common.h"
+#include "Profiler.h"
+
+#include <list>
+#include <string>
+#include <map>
+
+#ifdef _WIN32
+
+#if defined (_MSC_VER) && _MSC_VER >= 1400
+#define _interlockedbittestandset workaround_ms_header_bug_platform_sdk6_set
+#define _interlockedbittestandreset workaround_ms_header_bug_platform_sdk6_reset
+#define _interlockedbittestandset64 workaround_ms_header_bug_platform_sdk6_set64
+#define _interlockedbittestandreset64 workaround_ms_header_bug_platform_sdk6_reset64
+#include <intrin.h>
+#undef _interlockedbittestandset
+#undef _interlockedbittestandreset
+#undef _interlockedbittestandset64
+#undef _interlockedbittestandreset64
+#pragma intrinsic(__rdtsc)
+#endif
+
+// Globals
+static u64 luPerfFreq = 0;
+#ifdef DVPROFILE
+int g_bWriteProfile = 1;
+#else
+int g_bWriteProfile = 1;
+#endif
+
+inline u64 GET_PROFILE_TIME()
+{
+#if defined (_MSC_VER) && _MSC_VER >= 1400
+    return __rdtsc();
+#else
+    LARGE_INTEGER lu;
+    QueryPerformanceCounter(&lu);
+    return lu.QuadPart;
+#endif
+}
+#else
+static u64 luPerfFreq = 1000000;
+#define GET_PROFILE_TIME() //GetCpuTick()
+#endif
+
+struct DVPROFSTRUCT;
+
+struct DVPROFSTRUCT
+{
+    struct DATA
+    {
+        DATA(u64 time, u32 user = 0) : dwTime(time), dwUserData(user) {}
+        DATA() : dwTime(0), dwUserData(0) {}
+
+        u64 dwTime;
+        u32 dwUserData;
+    };
+
+    ~DVPROFSTRUCT() {
+        std::list<DVPROFSTRUCT *>::iterator it = listpChild.begin();
+        while (it != listpChild.end()) {
+            delete *it;
+			*it = NULL;
+            ++it;
+        }
+    }
+
+	// before DVProfEnd is called, contains the global time it started
+	// after DVProfEnd is called, contains the time it lasted
+	// the list contains all the tracked times 
+    std::list<DATA> listTimes;
+                              
+    char pname[256];
+    std::list<DVPROFSTRUCT*> listpChild;     // other profilers called during this profiler period
+};
+
+struct DVPROFTRACK
+{
+    u32 dwUserData;
+    DVPROFSTRUCT::DATA* pdwTime;
+    DVPROFSTRUCT* pprof;
+};
+
+// the current profiling functions, the back element is the
+// one that will first get popped off the list when DVProfEnd is called
+// the pointer is an element in DVPROFSTRUCT::listTimes
+static std::list<DVPROFTRACK> g_listCurTracking;
+
+// the current profilers, note that these are the parents
+// any profiler started during the time of another is held in
+// DVPROFSTRUCT::listpChild
+static std::list<DVPROFSTRUCT> g_listProfilers; 
+
+// ignores the hierarchy, pointer to elements in g_listProfilers
+static std::list<DVPROFSTRUCT*> g_listAllProfilers;
+
+
+void DVProfRegister(const char *pname)
+{
+    if (!g_bWriteProfile)
+        return;
+
+#ifdef _WIN32
+    if (luPerfFreq <= 1) {
+#if defined (_MSC_VER) && _MSC_VER >= 1400
+        luPerfFreq = 1000000;
+#else
+        LARGE_INTEGER temp;
+        QueryPerformanceFrequency(&temp);
+        luPerfFreq = temp.QuadPart;
+#endif
+    }
+#endif
+
+    std::list<DVPROFSTRUCT*>::iterator it = g_listAllProfilers.begin();
+    
+//  while(it != g_listAllProfilers.end() ) {
+//
+//      if( _tcscmp(pname, (*it)->pname) == 0 ) {
+//          (*it)->listTimes.push_back(timeGetTime());
+//          DVPROFTRACK dvtrack;
+//          dvtrack.pdwTime = &(*it)->listTimes.back();
+//          dvtrack.pprof = *it;
+//          g_listCurTracking.push_back(dvtrack);
+//          return;
+//      }
+//
+//      ++it;
+//  }
+
+    // else add in a new profiler to the appropriate parent profiler
+    DVPROFSTRUCT* pprof = NULL;
+    
+    if (g_listCurTracking.size() > 0) {
+        _assert_( g_listCurTracking.back().pprof != NULL );
+        g_listCurTracking.back().pprof->listpChild.push_back(new DVPROFSTRUCT());
+        pprof = g_listCurTracking.back().pprof->listpChild.back();
+    }
+    else {
+        g_listProfilers.push_back(DVPROFSTRUCT());
+        pprof = &g_listProfilers.back();
+    }
+
+    strncpy(pprof->pname, pname, 256);
+
+    // setup the profiler for tracking
+    pprof->listTimes.push_back(DVPROFSTRUCT::DATA(GET_PROFILE_TIME()));
+
+    DVPROFTRACK dvtrack;
+    dvtrack.pdwTime = &pprof->listTimes.back();
+    dvtrack.pprof = pprof;
+    dvtrack.dwUserData = 0;
+
+    g_listCurTracking.push_back(dvtrack);
+
+    // add to all profiler list
+    g_listAllProfilers.push_back(pprof);
+}
+
+void DVProfEnd(u32 dwUserData)
+{
+    if (!g_bWriteProfile)
+        return;
+    if (g_listCurTracking.size() == 0)
+        return;
+
+    DVPROFTRACK dvtrack = g_listCurTracking.back();
+
+    _assert_( dvtrack.pdwTime != NULL && dvtrack.pprof != NULL );
+
+    dvtrack.pdwTime->dwTime = GET_PROFILE_TIME()- dvtrack.pdwTime->dwTime;
+    dvtrack.pdwTime->dwUserData= dwUserData;
+
+    g_listCurTracking.pop_back();
+}
+
+struct DVTIMEINFO
+{
+    DVTIMEINFO() : uInclusive(0), uExclusive(0) {}
+    u64 uInclusive, uExclusive;
+};
+
+std::map<std::string, DVTIMEINFO> mapAggregateTimes;
+
+u64 DVProfWriteStruct(FILE* f, const DVPROFSTRUCT* p, int ident)
+{
+    fprintf(f, "%*s%s - ", ident, "", p->pname);
+    std::list<DVPROFSTRUCT::DATA>::const_iterator ittime = p->listTimes.begin();
+    u64 utime = 0;
+    while (ittime != p->listTimes.end()) {
+        utime += ittime->dwTime;
+        if (ittime->dwUserData) 
+            fprintf(f, "time: %d, user: 0x%8.8x", (u32)ittime->dwTime, ittime->dwUserData);
+        else
+            fprintf(f, "time: %d", (u32)ittime->dwTime);
+        ++ittime;
+    }
+
+    // yes this is necessary, maps have problems with constructors on their type
+    std::map<std::string, DVTIMEINFO>::iterator ittimes = mapAggregateTimes.find(p->pname);
+    if (ittimes == mapAggregateTimes.end()) {
+        ittimes = mapAggregateTimes.insert(std::map<std::string, DVTIMEINFO>::value_type(p->pname, DVTIMEINFO())).first;
+        ittimes->second.uExclusive = 0;
+        ittimes->second.uInclusive = 0;
+    }
+
+    ittimes->second.uInclusive += utime;
+
+    fprintf(f, "\n");
+
+    std::list<DVPROFSTRUCT*>::const_iterator itprof = p->listpChild.begin();
+
+    u64 uex = utime;
+    while (itprof != p->listpChild.end()) {
+        uex -= DVProfWriteStruct(f, *itprof, ident+4);
+        ++itprof;
+    }
+
+    if (uex > utime) {
+        uex = 0;
+    }
+
+    ittimes->second.uExclusive += uex;
+    return utime;
+}
+
+void DVProfWrite(const char* pfilename, u32 frames)
+{
+    _assert_( pfilename != NULL );
+    FILE* f = fopen(pfilename, "w");
+
+    // pop back any unused
+    mapAggregateTimes.clear();
+    std::list<DVPROFSTRUCT>::iterator it = g_listProfilers.begin();
+
+    while (it != g_listProfilers.end() ) {
+        DVProfWriteStruct(f, &(*it), 0);
+        ++it;
+    }
+
+    std::map<std::string, DVTIMEINFO>::const_iterator iter;
+    fprintf(f, "\n\n-------------------------------------------------------------------\n\n");
+
+    u64 uTotal[2] = {0};
+    double fiTotalTime[2];
+
+    for (iter = mapAggregateTimes.begin(); iter != mapAggregateTimes.end(); ++iter) {
+        uTotal[0] += iter->second.uExclusive;
+        uTotal[1] += iter->second.uInclusive;
+    }
+
+    fprintf(f, "total times (%d): ex: %Lu ", frames, 1000000 * uTotal[0] / (luPerfFreq*(u64)frames));
+    fprintf(f, "inc: %Lu\n", 1000000 * uTotal[1]/(luPerfFreq*(u64)frames));
+
+    fiTotalTime[0] = 1.0 / (double)uTotal[0];
+    fiTotalTime[1] = 1.0 / (double)uTotal[1];
+
+    // output the combined times
+    for (iter = mapAggregateTimes.begin(); iter != mapAggregateTimes.end(); ++iter) {
+        fprintf(f, "%s - ex: %f inc: %f\n", iter->first.c_str(), (float)((double)iter->second.uExclusive * fiTotalTime[0]),
+            (float)((double)iter->second.uInclusive * fiTotalTime[1]));
+    }
+
+    fclose(f);
+}
+
+void DVProfClear()
+{
+    g_listCurTracking.clear();
+    g_listProfilers.clear();
+    g_listAllProfilers.clear();
+}
diff --git a/Source/Core/VideoCommon/Src/Statistics.cpp b/Source/Core/VideoCommon/Src/Statistics.cpp
index df0dbec877..b686195e39 100644
--- a/Source/Core/VideoCommon/Src/Statistics.cpp
+++ b/Source/Core/VideoCommon/Src/Statistics.cpp
@@ -1,43 +1,43 @@
-// Copyright (C) 2003-2008 Dolphin Project.

-

-// This program is free software: you can redistribute it and/or modify

-// it under the terms of the GNU General Public License as published by

-// the Free Software Foundation, version 2.0.

-

-// This program is distributed in the hope that it will be useful,

-// but WITHOUT ANY WARRANTY; without even the implied warranty of

-// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

-// GNU General Public License 2.0 for more details.

-

-// A copy of the GPL 2.0 should have been included with the program.

-// If not, see http://www.gnu.org/licenses/

-

-// Official SVN repository and contact information can be found at

-// http://code.google.com/p/dolphin-emu/

-

-#include <string.h>

-

-#include "Statistics.h"

-

-Statistics stats;

-

-template <class T>

-void Xchg(T& a, T&b)

-{

-	T c = a;

-	a = b;

-	b = c;

-}

-

-void Statistics::ResetFrame()

-{

-	memset(&thisFrame, 0, sizeof(ThisFrame));

-}

-

-void Statistics::SwapDL()

-{

-	Xchg(stats.thisFrame.numDLPrims, stats.thisFrame.numPrims);

-	Xchg(stats.thisFrame.numXFLoadsInDL, stats.thisFrame.numXFLoads);

-	Xchg(stats.thisFrame.numCPLoadsInDL, stats.thisFrame.numCPLoads);

-	Xchg(stats.thisFrame.numBPLoadsInDL, stats.thisFrame.numBPLoads);

-}

+// Copyright (C) 2003-2008 Dolphin Project.
+
+// This program is free software: you can redistribute it and/or modify
+// it under the terms of the GNU General Public License as published by
+// the Free Software Foundation, version 2.0.
+
+// This program is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU General Public License 2.0 for more details.
+
+// A copy of the GPL 2.0 should have been included with the program.
+// If not, see http://www.gnu.org/licenses/
+
+// Official SVN repository and contact information can be found at
+// http://code.google.com/p/dolphin-emu/
+
+#include <string.h>
+
+#include "Statistics.h"
+
+Statistics stats;
+
+template <class T>
+void Xchg(T& a, T&b)
+{
+	T c = a;
+	a = b;
+	b = c;
+}
+
+void Statistics::ResetFrame()
+{
+	memset(&thisFrame, 0, sizeof(ThisFrame));
+}
+
+void Statistics::SwapDL()
+{
+	Xchg(stats.thisFrame.numDLPrims, stats.thisFrame.numPrims);
+	Xchg(stats.thisFrame.numXFLoadsInDL, stats.thisFrame.numXFLoads);
+	Xchg(stats.thisFrame.numCPLoadsInDL, stats.thisFrame.numCPLoads);
+	Xchg(stats.thisFrame.numBPLoadsInDL, stats.thisFrame.numBPLoads);
+}
diff --git a/Source/Core/VideoCommon/Src/TextureDecoder.cpp b/Source/Core/VideoCommon/Src/TextureDecoder.cpp
index 9f2b7c00bd..e20b5368d6 100644
--- a/Source/Core/VideoCommon/Src/TextureDecoder.cpp
+++ b/Source/Core/VideoCommon/Src/TextureDecoder.cpp
@@ -1,1380 +1,1380 @@
-// Copyright (C) 2003-2008 Dolphin Project.

-

-// This program is free software: you can redistribute it and/or modify

-// it under the terms of the GNU General Public License as published by

-// the Free Software Foundation, version 2.0.

-

-// This program is distributed in the hope that it will be useful,

-// but WITHOUT ANY WARRANTY; without even the implied warranty of

-// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

-// GNU General Public License 2.0 for more details.

-

-// A copy of the GPL 2.0 should have been included with the program.

-// If not, see http://www.gnu.org/licenses/

-

-// Official SVN repository and contact information can be found at

-// http://code.google.com/p/dolphin-emu/

-

-#include "Common.h"

-

-#include "TextureDecoder.h"

-#include "LookUpTables.h"

-#include <emmintrin.h>

-

-//Uncomment this to enable Texture Format ID overlays

-#define OVERLAY_TEXFMT

-

-#ifdef OVERLAY_TEXFMT

-bool TexFmt_Overlay_Enable=false;

-bool TexFmt_Overlay_Center=false;

-#endif

-

-// TRAM

-// STATE_TO_SAVE

-u8 texMem[TMEM_SIZE];

-

-//////////////////////////////////////////////////////////////////////////

-// Gamecube/Wii texture decoder

-//////////////////////////////////////////////////////////////////////////

-// Decodes all known Gamecube/Wii texture formats.

-// by ector

-//////////////////////////////////////////////////////////////////////////

-int TexDecoder_GetTexelSizeInNibbles(int format)

-{

-    switch (format & 0x3f) {

-    case GX_TF_I4: return 1;   

-    case GX_TF_I8: return 2;

-    case GX_TF_IA4: return 2;

-    case GX_TF_IA8: return 4;

-    case GX_TF_RGB565: return 4;

-    case GX_TF_RGB5A3: return 4;

-    case GX_TF_RGBA8: return 8;

-    case GX_TF_C4: return 1;

-    case GX_TF_C8: return 2;

-    case GX_TF_C14X2: return 4;

-    case GX_TF_CMPR: return 1;

-    default: return 1;

-    }

-}

-

-int TexDecoder_GetTextureSizeInBytes(int width, int height, int format)

-{

-	return (width * height * TexDecoder_GetTexelSizeInNibbles(format)) / 2;

-}

-

-u32 TexDecoder_GetTlutHash(const u16 *src, int len)

-{

-	u32 hash = 0xbeefbabe;

-	for (int i = 0; i < len / 2; i += 2) {

-		hash = _rotl(hash, 17) ^ ((u32 *)src)[i];

-	}

-	return hash;

-}

-

-u32 TexDecoder_GetSafeTextureHash(const u8 *src, int width, int height, int texformat, u32 seed)

-{

-	int sz = TexDecoder_GetTextureSizeInBytes(width, height, texformat);

-	

-	u32 hash = seed ? seed : 0x1337c0de;

-

-	if (sz < 2048) {

-		for (int i = 0; i < sz / 4; i += 13) {

-			hash = _rotl(hash, 17) ^ ((u32 *)src)[i];

-		}

-		return hash;

-	} else {

-		int step = sz / 13 / 4;

-		for (int i = 0; i < sz / 4; i += step) {

-			hash = _rotl(hash, 17) ^ ((u32 *)src)[i];

-		}

-	}

-	return hash;

-}

-

-int TexDecoder_GetBlockWidthInTexels(int format)

-{

-    switch (format) {

-    case GX_TF_I4: return 8;   

-    case GX_TF_I8: return 8;

-    case GX_TF_IA4: return 8;

-    case GX_TF_IA8: return 4;

-    case GX_TF_RGB565: return 4;

-    case GX_TF_RGB5A3: return 4;

-    case GX_TF_RGBA8: return  4;

-    case GX_TF_C4: return 8;

-    case GX_TF_C8: return 8;

-    case GX_TF_C14X2: return 4;

-    case GX_TF_CMPR: return 8;

-    default: return 8;

-    }

-}

-

-//returns bytes

-int TexDecoder_GetPaletteSize(int format)

-{

-    switch (format) {

-    case GX_TF_C4: return 16*2;

-    case GX_TF_C8: return 256*2;

-    case GX_TF_C14X2: return 16384*2;

-    default:

-        return 0;

-    }

-}

-

-inline u32 decode565(u16 val)

-{

-    int r,g,b,a;

-    r=lut5to8[(val>>11) & 0x1f];

-    g=lut6to8[(val>>5 ) & 0x3f];

-    b=lut5to8[(val    ) & 0x1f];

-    a=0xFF;

-    return (a<<24) | (r<<16) | (g<<8) | b;

-}

-

-inline u32 decodeIA8(u16 val)

-{

-    int a=val>>8;

-    int i=val&0xFF;

-    return (a<<24) | (i<<16) | (i<<8) | i;

-}

-

-inline u32 decode5A3(u16 val)

-{

-    int r,g,b,a;

-    if ((val&0x8000))

-    {

-        r=lut5to8[(val>>10) & 0x1f];

-        g=lut5to8[(val>>5 ) & 0x1f];

-        b=lut5to8[(val    ) & 0x1f];

-        a=0xFF;

-    }

-    else

-    {

-        a=lut3to8[(val>>12) & 0x7];

-        r=lut4to8[(val>>8 ) & 0xf];

-        g=lut4to8[(val>>4 ) & 0xf];

-        b=lut4to8[(val    ) & 0xf];

-    }

-    return (a<<24) | (r<<16) | (g<<8) | b;

-}

-

-

-

-struct DXTBlock

-{

-    u16 color1;

-    u16 color2;

-    u8 lines[4];

-};

-

-inline int expand8888(const int j)

-{

-    int i = j | (j<<8);

-    return i | (i<<16);

-}

-

-//inline void decodebytesI4(u32 *dst, const u8 *src, int numbytes)

-inline void decodebytesI4(u32 *dst, const u8 *src)

-{

-    for (int x = 0; x < 4; x++)

-    {

-        int val = src[x];

-		*dst++ = expand8888(lut4to8[val>>4]);

-		*dst++ = expand8888(lut4to8[val&15]);

-    }

-}

-

-inline void decodebytesI8_8(u32 *dst, const u8 *src)

-{

-	for (int x = 0; x < 8; x++)

-        dst[x] = src[x] * 0x01010101; //expand8888(src[x]);  *0x... may or may not be faster. not sure. Should be faster on P4 at least.

-}

-

-//inline void decodebytesC4(u32 *dst, const u8 *src, int numbytes, int tlutaddr, int tlutfmt)

-inline void decodebytesC4(u32 *dst, const u8 *src, int tlutaddr, int tlutfmt)

-{

-    u16 *tlut = (u16*)(texMem + tlutaddr);

-    for (int x = 0; x < 4; x++)

-    {

-        int val = src[x];

-        switch (tlutfmt) {

-        case 0:

-            *dst++ = decodeIA8(Common::swap16(tlut[val >> 4]));

-            *dst++ = decodeIA8(Common::swap16(tlut[val & 15]));

-            break;

-        case 1:

-            *dst++ = decode565(Common::swap16(tlut[val >> 4]));

-            *dst++ = decode565(Common::swap16(tlut[val & 15]));

-            break;

-        case 2:

-            *dst++ = decode5A3(Common::swap16(tlut[val >> 4]));

-            *dst++ = decode5A3(Common::swap16(tlut[val & 15]));

-            break;

-        case 3: //ERROR

-            *dst++ = 0xFFFF00FF;

-            *dst++ = 0xFFFF00FF;

-            break;

-        }

-    }

-}

-

-//inline void decodebytesC8(u32 *dst, const u8 *src, int numbytes, int tlutaddr, int tlutfmt)

-inline void decodebytesC8(u32 *dst, const u8 *src, int tlutaddr, int tlutfmt)

-{ 

-    u16 *tlut = (u16*)(texMem+tlutaddr);

-    for (int x = 0; x < 8; x++)

-    {

-        int val = src[x];

-        switch (tlutfmt) {

-        case 0:

-            *dst++ = decodeIA8(Common::swap16(tlut[val]));

-            break;

-        case 1:

-            *dst++ = decode565(Common::swap16(tlut[val]));

-            break;

-        case 2:

-            *dst++ = decode5A3(Common::swap16(tlut[val]));

-            break;

-        case 3: //ERROR

-            *dst++ = 0xFFFF00FF;

-            break;

-        }

-    }

-}

-

-

-//inline void decodebytesC14X2(u32 *dst, const u16 *src, int numpixels, int tlutaddr, int tlutfmt)

-inline void decodebytesC14X2(u32 *dst, const u16 *src, int tlutaddr, int tlutfmt)

-{

-    u16 *tlut = (u16*)(texMem+tlutaddr);

-    for (int x = 0; x < 4; x++)

-    {

-        int val = Common::swap16(src[x]);

-        switch (tlutfmt) {

-        case 0:

-            *dst++ = decodeIA8(Common::swap16(tlut[(val&0x3FFF)]));

-            break;

-        case 1:

-            *dst++ = decode565(Common::swap16(tlut[(val&0x3FFF)]));

-            break;

-        case 2:

-            *dst++ = decode5A3(Common::swap16(tlut[(val&0x3FFF)]));

-            break;

-        case 3: //ERROR

-            *dst++ = 0xFFFF00FF;

-            break;

-        }

-    }

-}

-

-//inline void decodebytesRGB565(u32 *dst, const u16 *src, int numpixels)

-inline void decodebytesRGB565(u32 *dst, const u16 *src)

-{

-    for (int x = 0; x < 4; x++)

-        *dst++ = decode565(Common::swap16(src[x]));

-}

-

-//inline void decodebytesIA4(u32 *dst, const u8 *src, int numbytes)

-inline void decodebytesIA4(u32 *dst, const u8 *src)

-{

-    for (int x = 0; x < 8; x++)

-    {

-        int val = src[x];

-        int a = lut4to8[val>>4];

-        int r = lut4to8[val&15];

-        dst[x] = (a<<24) | (r<<16) | (r<<8) | r;

-    }

-}

-

-//inline void decodebytesIA8(u32 *dst, const u16 *src, int numpixels)

-inline void decodebytesIA8(u32 *dst, const u16 *src)

-{

-    for (int x = 0; x < 4; x++)

-        dst[x] = decodeIA8(Common::swap16(src[x]));

-}

-

-//inline void decodebytesRGB5A3(u32 *dst, const u16 *src, int numpixels)

-inline void decodebytesRGB5A3(u32 *dst, const u16 *src)

-{

-    for (int x = 0; x < 4; x++)

-        dst[x] = decode5A3(Common::swap16(src[x]));

-}

-

-// This one is used by many video formats. It'd therefore be good if it was fast.

-inline void decodebytesARGB8_4(u32 *dst, const u16 *src, const u16 *src2)

-{

-    for (int x = 0; x < 4; x++) {

-        dst[x] = Common::swap32((src2[x] << 16) | src[x]);

-    }

-

-	// This can probably be done in a few SSE pack/unpack instructions + pshufb

-	// some unpack instruction x2:

-	// ABABABABABABABAB 1212121212121212 ->

-	// AB12AB12AB12AB12 AB12AB12AB12AB12

-	// 2x pshufb-> 

-	// 21BA21BA21BA21BA 21BA21BA21BA21BA

-	// and we are done.

-}

-

-inline u32 makecol(int r, int g, int b, int a)

-{

-    return (a<<24)|(r<<16)|(g<<8)|b;

-}

-

-void decodeDXTBlock(u32 *dst, const DXTBlock *src, int pitch)

-{

-    u16 c1 = Common::swap16(src->color1);

-    u16 c2 = Common::swap16(src->color2);

-    int blue1 = lut5to8[c1 & 0x1F];

-    int blue2 = lut5to8[c2 & 0x1F];

-    int green1 = lut6to8[(c1>>5) & 0x3F];

-    int green2 = lut6to8[(c2>>5) & 0x3F];

-    int red1 = lut5to8[(c1>>11) & 0x1F];

-    int red2 = lut5to8[(c2>>11) & 0x1F];

-    

-    int colors[4];

-

-    if (c1 > c2)

-    {

-        colors[0] = makecol(red1, green1, blue1, 255);

-        colors[1] = makecol(red2, green2, blue2, 255);

-        colors[2] = makecol(red1+(red2-red1)/3, green1+(green2-green1)/3, blue1+(blue2-blue1)/3, 255);

-        colors[3] = makecol(red2+(red1-red2)/3, green2+(green1-green2)/3, blue2+(blue1-blue2)/3, 255);

-    }

-    else

-    {

-        colors[0] = makecol(red1, green1, blue1, 255);

-        colors[1] = makecol(red2, green2, blue2, 255);

-        colors[2] = makecol((red1+red2)/2, (green1+green2)/2, (blue1+blue2)/2, 255);

-        colors[3] = makecol(0,0,0,0); //transparent

-    }

-

-    for (int y = 0; y < 4; y++)

-    { 

-        int val = src->lines[y];

-        for (int x = 0; x < 4; x++)

-        {

-            dst[x] = colors[(val >> 6) & 3];

-            val <<= 2;

-        }

-        dst += pitch;

-    }

-}

-

-

-//switch endianness, unswizzle

-//TODO: to save memory, don't blindly convert everything to argb8888

-//also ARGB order needs to be swapped later, to accommodate modern hardware better

-//need to add DXT support too

-#ifdef OVERLAY_TEXFMT

-PC_TexFormat TexDecoder_Decode_real(u8 *dst, const u8 *src, int width, int height, int texformat, int tlutaddr, int tlutfmt)

-#else

-PC_TexFormat TexDecoder_Decode(u8 *dst, const u8 *src, int width, int height, int texformat, int tlutaddr, int tlutfmt)

-#endif

-{

-    switch (texformat)

-    {

-    case GX_TF_C4:

-        {

-            for (int y = 0; y < height; y += 8)

-                for (int x = 0; x < width; x += 8)

-                    for (int iy = 0; iy < 8; iy++, src += 4)

-                        //decodebytesC4((u32*)dst+(y+iy)*width+x, src, 4, tlutaddr, tlutfmt);

-                        decodebytesC4((u32*)dst+(y+iy)*width+x, src, tlutaddr, tlutfmt);

-        }

-        return PC_TEX_FMT_BGRA32;

-    case GX_TF_I4:

-        {

-        	// TODO: SSSE3 variant (pshufb), THP videos use this format.

-        	// SSSE3 variant could bring even more speed

-#if (defined(_WIN32) || (defined (_M_X64) && !defined(_WIN32)))

-			__m128i Lmask = _mm_set1_epi8 (0x0F);

-			__m128i Hmask = _mm_set1_epi8 (0xF0);

-			__m128i* sseSrc  = (__m128i *)src;

-			__m128i* sseDst  = (__m128i *)dst;

-            for (int y = 0; y < height; y += 8)

-	            for (int x = 0; x < width; x += 8)

-					for (int iy = 0; iy < 8; iy++, sseSrc++) {

-						// TODO (mb2): func and don't let the optimizer perform all the clean up by itself

-						__m128i s = _mm_load_si128 (sseSrc);			// ab cd ef gh ...

-						__m128i sl = _mm_and_si128 (s, Lmask);			// 0b 0d 0f 0h ...

-						__m128i sls = _mm_slli_epi16 (sl, 4);			// b0 d0 f0 h0 ...

-						__m128i sl_ = _mm_or_si128 (sl, sls);			// bb dd ff ff ...

-

-						__m128i sh = _mm_and_si128 (s, Hmask);			// a0 c0 e0 g0 ...

-						__m128i shs = _mm_srli_epi16 (sh, 4);			// 0a 0c 0e g0 ...

-						__m128i sh_ = _mm_or_si128 (sh, shs);			// aa cc ee gg ...

-						__m128i rl = _mm_unpacklo_epi8 (sh_, sl_);		// bb aa dd cc ...

-						__m128i rh = _mm_unpackhi_epi8 (sh_, sl_);		// 

-

-						__m128i ral = _mm_unpacklo_epi8 (rl, rl);		// bb bb aa aa ...

-						__m128i rah = _mm_unpackhi_epi8 (rl, rl);		// 

-						// result part a

-						__m128i rall = _mm_unpacklo_epi16 (ral, ral);	// bb bb bb bb ...	-> done

-						__m128i ralh = _mm_unpackhi_epi16 (ral, ral);	//					-> done

-						__m128i rahl = _mm_unpacklo_epi16 (rah, rah);	//					-> done

-						__m128i rahh = _mm_unpackhi_epi16 (rah, rah);	//					-> done

-

-						__m128i rbl = _mm_unpacklo_epi8 (rh, rh);		// 

-						__m128i rbh = _mm_unpackhi_epi8 (rh, rh);		// 

-						// result part b

-						__m128i rbll = _mm_unpacklo_epi16 (rbl, rbl);	//					-> done

-						__m128i rblh = _mm_unpackhi_epi16 (rbl, rbl);	//					-> done

-						__m128i rbhl = _mm_unpacklo_epi16 (rbh, rbh);	//					-> done

-						__m128i rbhh = _mm_unpackhi_epi16 (rbh, rbh);	//					-> done

-						// store

-						sseDst = (__m128i*)(dst+((y+iy)*width+x)*4);	// that sucks... too lazy 

-						_mm_store_si128 (sseDst++, rall); 

-						_mm_store_si128 (sseDst, ralh);

-						iy++;

-						sseDst = (__m128i*)(dst+((y+iy)*width+x)*4);

-						_mm_store_si128 (sseDst++, rahl);

-						_mm_store_si128 (sseDst, rahh);

-						iy++;

-						sseDst = (__m128i*)(dst+((y+iy)*width+x)*4);

-						_mm_store_si128 (sseDst++, rbll);

-						_mm_store_si128 (sseDst, rblh);

-						iy++;

-						sseDst = (__m128i*)(dst+((y+iy)*width+x)*4);

-						_mm_store_si128 (sseDst++, rbhl);

-						_mm_store_si128 (sseDst, rbhh);

-					}

-#else

-            for (int y = 0; y < height; y += 8)

-                for (int x = 0; x < width; x += 8)

-                    for (int iy = 0; iy < 8; iy++, src += 4)

-                        //decodebytesI4((u32*)dst+(y+iy)*width+x, src, 4);

-                        decodebytesI4((u32*)dst+(y+iy)*width+x, src);

-#endif

-        }

-        return PC_TEX_FMT_BGRA32;

-    case GX_TF_C8:

-        {

-            for (int y = 0; y < height; y += 4)

-                for (int x = 0; x < width; x += 8)

-                    for (int iy = 0; iy < 4; iy++, src += 8)

-                        //decodebytesC8((u32*)dst+(y+iy)*width+x, src, 8, tlutaddr, tlutfmt);

-                        decodebytesC8((u32*)dst+(y+iy)*width+x, src, tlutaddr, tlutfmt);

-        }

-        return PC_TEX_FMT_BGRA32;

-    case GX_TF_I8:  // speed critical

-		{

-#if (defined(_WIN32) || (defined (_M_X64) && !defined(_WIN32)))

-			__m128i *sseSrc  = (__m128i *)src;

-			__m128i *sseDst  = (__m128i *)dst;

-            for (int y = 0; y < height; y += 4)

-                for (int x = 0; x < width; x += 8)

-					for (int iy = 0; iy < 4; iy++, sseSrc++) {

-						// TODO (mb2): func and don't let the optimizer perform all the clean up by itself

-						__m128i s = _mm_load_si128 (sseSrc);			// ab cd ef gh ...

-						__m128i rl = _mm_unpacklo_epi8 (s, s);			// ab ab cd cd ...

-						__m128i rh = _mm_unpackhi_epi8 (s, s);			// 

-						// result

-						__m128i rll = _mm_unpacklo_epi8 (rl, rl);		// ab ab ab ab 

-						__m128i rlh = _mm_unpackhi_epi8 (rl, rl);

-						__m128i rhl = _mm_unpacklo_epi8 (rh, rh);

-						__m128i rhh = _mm_unpackhi_epi8 (rh, rh);

-						// store

-						sseDst = (__m128i*)(dst+((y+iy)*width+x)*4);

-						_mm_store_si128 (sseDst++, rll); 

-						_mm_store_si128 (sseDst, rlh);

-						iy++;

-						sseDst = (__m128i*)(dst+((y+iy)*width+x)*4);

-						_mm_store_si128 (sseDst++, rhl);

-						_mm_store_si128 (sseDst, rhh);

-					}

-#else

-            for (int y = 0; y < height; y += 4)

-                for (int x = 0; x < width; x += 8)

-                    for (int iy = 0; iy < 4; iy++, src += 8)

-                        decodebytesI8_8((u32*)dst+(y+iy)*width+x, src);

-#endif

-		}

-        return PC_TEX_FMT_BGRA32;

-    case GX_TF_IA4:

-        {

-            for (int y = 0; y < height; y += 4)

-                for (int x = 0; x < width; x += 8)

-                    for (int iy = 0; iy < 4; iy++, src += 8)

-                        //decodebytesIA4((u32*)dst+(y+iy)*width+x, src, 8);

-                        decodebytesIA4((u32*)dst+(y+iy)*width+x, src);

-        }

-        return PC_TEX_FMT_BGRA32;

-    case GX_TF_IA8:

-        {

-            for (int y = 0; y < height; y += 4)

-                for (int x = 0; x < width; x += 4)

-                    for (int iy = 0; iy < 4; iy++, src += 8)

-                        //decodebytesIA8((u32*)dst+(y+iy)*width+x, (u16*)src, 4);

-                        decodebytesIA8((u32*)dst+(y+iy)*width+x, (u16*)src);

-        }

-        return PC_TEX_FMT_BGRA32;

-    case GX_TF_C14X2: 

-        {

-            for (int y = 0; y < height; y += 4)

-                for (int x = 0; x < width; x += 4)

-                    for (int iy = 0; iy < 4; iy++, src += 8)

-                        //decodebytesC14X2((u32*)dst+(y+iy)*width+x, (u16*)src, 4, tlutaddr, tlutfmt);

-                        decodebytesC14X2((u32*)dst+(y+iy)*width+x, (u16*)src, tlutaddr, tlutfmt);

-        }

-        return PC_TEX_FMT_BGRA32;

-    case GX_TF_RGB565:

-        {

-            for (int y = 0; y < height; y += 4)

-                for (int x = 0; x < width; x += 4)

-                    for (int iy = 0; iy < 4; iy++, src += 8)

-                        //decodebytesRGB565((u32*)dst+(y+iy)*width+x, (u16*)src, 4);

-                        decodebytesRGB565((u32*)dst+(y+iy)*width+x, (u16*)src);

-        }

-        return PC_TEX_FMT_BGRA32;

-    case GX_TF_RGB5A3:

-        {

-            for (int y = 0; y < height; y += 4)

-                for (int x = 0; x < width; x += 4)

-                    for (int iy = 0; iy < 4; iy++, src += 8)

-                        //decodebytesRGB5A3((u32*)dst+(y+iy)*width+x, (u16*)src, 4);

-                        decodebytesRGB5A3((u32*)dst+(y+iy)*width+x, (u16*)src);

-        }

-        return PC_TEX_FMT_BGRA32;

-    case GX_TF_RGBA8:  // speed critical

-        {

-			for (int y = 0; y < height; y += 4) {

-                for (int x = 0; x < width; x += 4)

-                {

-					for (int iy = 0; iy < 4; iy++) {

-                        decodebytesARGB8_4((u32*)dst + (y+iy)*width + x, (u16*)src + 4 * iy, (u16*)src + 4 * iy + 16);

-					}

-					src += 64;

-                }

-			}

-        }

-        return PC_TEX_FMT_BGRA32;

-    case GX_TF_CMPR:  // speed critical

-        {

-			// TODO: Shuffle to PC S3TC (DXTC) format instead of converting

-            // 11111111 22222222 55555555 66666666

-            // 33333333 44444444 77777777 88888888

-			// The metroid games use this format almost exclusively.

-            for (int y = 0; y < height; y += 8)

-                for (int x = 0; x < width; x += 8)

-                {

-                    decodeDXTBlock((u32*)dst+y*width+x, (DXTBlock*)src, width);

-					src += sizeof(DXTBlock);

-                    decodeDXTBlock((u32*)dst+y*width+x+4, (DXTBlock*)src, width);

-					src += sizeof(DXTBlock);

-                    decodeDXTBlock((u32*)dst+(y+4)*width+x, (DXTBlock*)src, width);

-					src += sizeof(DXTBlock);

-                    decodeDXTBlock((u32*)dst+(y+4)*width+x+4, (DXTBlock*)src, width);

-					src += sizeof(DXTBlock);

-                }

-        }

-        return PC_TEX_FMT_BGRA32;

-    }

-

-	// The "copy" texture formats, too?

-    return PC_TEX_FMT_NONE;

-}

-

-void TexDecoder_SetTexFmtOverlayOptions(bool enable, bool center)

-{

-#ifdef OVERLAY_TEXFMT

-	TexFmt_Overlay_Enable = enable;

-	TexFmt_Overlay_Center = center;

-#endif

-}

-

-#ifdef OVERLAY_TEXFMT

-extern const char* texfmt[];

-extern const unsigned char sfont_map[];

-extern const unsigned char sfont_raw[][9*10];

-

-PC_TexFormat TexDecoder_Decode(u8 *dst, const u8 *src, int width, int height, int texformat, int tlutaddr, int tlutfmt)

-{

-	PC_TexFormat  retval = TexDecoder_Decode_real(dst,src,width,height,texformat,tlutaddr,tlutfmt);

-

-	if((!TexFmt_Overlay_Enable)||(retval==PC_TEX_FMT_NONE))

-		return retval;

-

-	// assume ABGR/ARGB (32bit)

-	int  *dtp = (int*)dst;

-

-	int w = min(width,40);

-	int h = min(height,10);

-

-	int xoff = (width-w)>>1;

-	int yoff = (height-h)>>1;

-

-	if(!TexFmt_Overlay_Center)

-	{

-		xoff=0;

-		yoff=0;

-	}

-

-	const char* fmt = texfmt[texformat&15];

-	while(*fmt)

-	{

-		int xcnt = 0;

-		int nchar = sfont_map[(int)*fmt];

-		

-		const unsigned char *ptr = sfont_raw[nchar]; // each char is up to 9x10

-

-		for(int x=0;x<9;x++)

-		{

-			if(ptr[x]==0x78)

-				break;

-			xcnt++;

-		}

-		for(int y=0;y<10;y++)

-		{

-			for(int x=0;x<xcnt;x++)

-			{

-				dtp[(y+yoff)*width + x+xoff] = ptr[x]?0xFFFFFFFF:0xFF000000;

-			}

-			ptr+=9;

-		}

-		xoff+=xcnt;

-		fmt++;

-	}

-

-	return retval;

-}

-

-const char* texfmt[] = {

-	// pixel

-	"I4",		"I8",		"IA4",		"IA8",

-	"RGB565",	"RGB5A3",	"RGBA8",	"C4",

-	"C8",		"C14X2",	"0x0A",		"0x0B",

-	"0x0C",		"0x0D",		"CMPR",		"0x0F",

-	// Z-buffer

-	"0x10",		"Z8",		"0x12",		"Z16",

-	"0x14",		"0x15",		"Z24X8",	"0x17",

-	"0x18",		"0x19",		"0x1A",		"0x1B",

-	"0x1C",		"0x1D",		"0x1E",		"0x1F",

-	// pixel + copy

-	"CR4",		"0x21",		"CRA4",		"CRA8",

-	"0x24",		"0x25",		"CYUVA8",	"CA8",

-	"CR8",		"CG8",		"CB8",		"CRG8",

-	"CGB8",		"0x2D",		"0x2E",		"0x2F",

-	// Z + copy

-	"CZ4",		"0x31",		"0x32",		"0x33",

-	"0x34",		"0x35",		"0x36",		"0x37",

-	"0x38",		"CZ8M",		"CZ8L",		"0x3B",

-	"CZ16L",	"0x3D",		"0x3E",		"0x3F",

-};

-

-const unsigned char sfont_map[] = {

-	10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,

-	10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,

-	10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,

-	 0, 1, 2, 3, 4, 5, 6, 7, 8, 9,10,10,10,10,10,10,

-	10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,

-	26,27,28,29,30,31,32,33,34,35,36,10,10,10,10,10,

-	10,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,

-	52,53,54,55,56,57,58,59,60,61,62,10,10,10,10,10,

-	10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,

-	10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,

-	10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,

-	10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,

-	10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,

-	10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,

-	10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,

-	10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,

-};

-

-const unsigned char sfont_raw[][9*10] = {

-	{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78,

-	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78,

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78,

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78,

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 

-	0xff, 0x00, 0x00, 0xff, 0xff, 0xff, 0x00, 0x00, 0xff, 

-	0xff, 0x00, 0x00, 0xff, 0xff, 0xff, 0x00, 0x00, 0xff, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78,

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0x00, 0x00, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0xff, 0x00, 0xff, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0xff, 0x00, 0xff, 

-	0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 

-	0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0xff, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0xff, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78,

-	0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	},{

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 

-	},

-};

-#endif

-

+// Copyright (C) 2003-2008 Dolphin Project.
+
+// This program is free software: you can redistribute it and/or modify
+// it under the terms of the GNU General Public License as published by
+// the Free Software Foundation, version 2.0.
+
+// This program is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU General Public License 2.0 for more details.
+
+// A copy of the GPL 2.0 should have been included with the program.
+// If not, see http://www.gnu.org/licenses/
+
+// Official SVN repository and contact information can be found at
+// http://code.google.com/p/dolphin-emu/
+
+#include "Common.h"
+
+#include "TextureDecoder.h"
+#include "LookUpTables.h"
+#include <emmintrin.h>
+
+//Uncomment this to enable Texture Format ID overlays
+#define OVERLAY_TEXFMT
+
+#ifdef OVERLAY_TEXFMT
+bool TexFmt_Overlay_Enable=false;
+bool TexFmt_Overlay_Center=false;
+#endif
+
+// TRAM
+// STATE_TO_SAVE
+u8 texMem[TMEM_SIZE];
+
+//////////////////////////////////////////////////////////////////////////
+// Gamecube/Wii texture decoder
+//////////////////////////////////////////////////////////////////////////
+// Decodes all known Gamecube/Wii texture formats.
+// by ector
+//////////////////////////////////////////////////////////////////////////
+int TexDecoder_GetTexelSizeInNibbles(int format)
+{
+    switch (format & 0x3f) {
+    case GX_TF_I4: return 1;   
+    case GX_TF_I8: return 2;
+    case GX_TF_IA4: return 2;
+    case GX_TF_IA8: return 4;
+    case GX_TF_RGB565: return 4;
+    case GX_TF_RGB5A3: return 4;
+    case GX_TF_RGBA8: return 8;
+    case GX_TF_C4: return 1;
+    case GX_TF_C8: return 2;
+    case GX_TF_C14X2: return 4;
+    case GX_TF_CMPR: return 1;
+    default: return 1;
+    }
+}
+
+int TexDecoder_GetTextureSizeInBytes(int width, int height, int format)
+{
+	return (width * height * TexDecoder_GetTexelSizeInNibbles(format)) / 2;
+}
+
+u32 TexDecoder_GetTlutHash(const u16 *src, int len)
+{
+	u32 hash = 0xbeefbabe;
+	for (int i = 0; i < len / 2; i += 2) {
+		hash = _rotl(hash, 17) ^ ((u32 *)src)[i];
+	}
+	return hash;
+}
+
+u32 TexDecoder_GetSafeTextureHash(const u8 *src, int width, int height, int texformat, u32 seed)
+{
+	int sz = TexDecoder_GetTextureSizeInBytes(width, height, texformat);
+	
+	u32 hash = seed ? seed : 0x1337c0de;
+
+	if (sz < 2048) {
+		for (int i = 0; i < sz / 4; i += 13) {
+			hash = _rotl(hash, 17) ^ ((u32 *)src)[i];
+		}
+		return hash;
+	} else {
+		int step = sz / 13 / 4;
+		for (int i = 0; i < sz / 4; i += step) {
+			hash = _rotl(hash, 17) ^ ((u32 *)src)[i];
+		}
+	}
+	return hash;
+}
+
+int TexDecoder_GetBlockWidthInTexels(int format)
+{
+    switch (format) {
+    case GX_TF_I4: return 8;   
+    case GX_TF_I8: return 8;
+    case GX_TF_IA4: return 8;
+    case GX_TF_IA8: return 4;
+    case GX_TF_RGB565: return 4;
+    case GX_TF_RGB5A3: return 4;
+    case GX_TF_RGBA8: return  4;
+    case GX_TF_C4: return 8;
+    case GX_TF_C8: return 8;
+    case GX_TF_C14X2: return 4;
+    case GX_TF_CMPR: return 8;
+    default: return 8;
+    }
+}
+
+//returns bytes
+int TexDecoder_GetPaletteSize(int format)
+{
+    switch (format) {
+    case GX_TF_C4: return 16*2;
+    case GX_TF_C8: return 256*2;
+    case GX_TF_C14X2: return 16384*2;
+    default:
+        return 0;
+    }
+}
+
+inline u32 decode565(u16 val)
+{
+    int r,g,b,a;
+    r=lut5to8[(val>>11) & 0x1f];
+    g=lut6to8[(val>>5 ) & 0x3f];
+    b=lut5to8[(val    ) & 0x1f];
+    a=0xFF;
+    return (a<<24) | (r<<16) | (g<<8) | b;
+}
+
+inline u32 decodeIA8(u16 val)
+{
+    int a=val>>8;
+    int i=val&0xFF;
+    return (a<<24) | (i<<16) | (i<<8) | i;
+}
+
+inline u32 decode5A3(u16 val)
+{
+    int r,g,b,a;
+    if ((val&0x8000))
+    {
+        r=lut5to8[(val>>10) & 0x1f];
+        g=lut5to8[(val>>5 ) & 0x1f];
+        b=lut5to8[(val    ) & 0x1f];
+        a=0xFF;
+    }
+    else
+    {
+        a=lut3to8[(val>>12) & 0x7];
+        r=lut4to8[(val>>8 ) & 0xf];
+        g=lut4to8[(val>>4 ) & 0xf];
+        b=lut4to8[(val    ) & 0xf];
+    }
+    return (a<<24) | (r<<16) | (g<<8) | b;
+}
+
+
+
+struct DXTBlock
+{
+    u16 color1;
+    u16 color2;
+    u8 lines[4];
+};
+
+inline int expand8888(const int j)
+{
+    int i = j | (j<<8);
+    return i | (i<<16);
+}
+
+//inline void decodebytesI4(u32 *dst, const u8 *src, int numbytes)
+inline void decodebytesI4(u32 *dst, const u8 *src)
+{
+    for (int x = 0; x < 4; x++)
+    {
+        int val = src[x];
+		*dst++ = expand8888(lut4to8[val>>4]);
+		*dst++ = expand8888(lut4to8[val&15]);
+    }
+}
+
+inline void decodebytesI8_8(u32 *dst, const u8 *src)
+{
+	for (int x = 0; x < 8; x++)
+        dst[x] = src[x] * 0x01010101; //expand8888(src[x]);  *0x... may or may not be faster. not sure. Should be faster on P4 at least.
+}
+
+//inline void decodebytesC4(u32 *dst, const u8 *src, int numbytes, int tlutaddr, int tlutfmt)
+inline void decodebytesC4(u32 *dst, const u8 *src, int tlutaddr, int tlutfmt)
+{
+    u16 *tlut = (u16*)(texMem + tlutaddr);
+    for (int x = 0; x < 4; x++)
+    {
+        int val = src[x];
+        switch (tlutfmt) {
+        case 0:
+            *dst++ = decodeIA8(Common::swap16(tlut[val >> 4]));
+            *dst++ = decodeIA8(Common::swap16(tlut[val & 15]));
+            break;
+        case 1:
+            *dst++ = decode565(Common::swap16(tlut[val >> 4]));
+            *dst++ = decode565(Common::swap16(tlut[val & 15]));
+            break;
+        case 2:
+            *dst++ = decode5A3(Common::swap16(tlut[val >> 4]));
+            *dst++ = decode5A3(Common::swap16(tlut[val & 15]));
+            break;
+        case 3: //ERROR
+            *dst++ = 0xFFFF00FF;
+            *dst++ = 0xFFFF00FF;
+            break;
+        }
+    }
+}
+
+//inline void decodebytesC8(u32 *dst, const u8 *src, int numbytes, int tlutaddr, int tlutfmt)
+inline void decodebytesC8(u32 *dst, const u8 *src, int tlutaddr, int tlutfmt)
+{ 
+    u16 *tlut = (u16*)(texMem+tlutaddr);
+    for (int x = 0; x < 8; x++)
+    {
+        int val = src[x];
+        switch (tlutfmt) {
+        case 0:
+            *dst++ = decodeIA8(Common::swap16(tlut[val]));
+            break;
+        case 1:
+            *dst++ = decode565(Common::swap16(tlut[val]));
+            break;
+        case 2:
+            *dst++ = decode5A3(Common::swap16(tlut[val]));
+            break;
+        case 3: //ERROR
+            *dst++ = 0xFFFF00FF;
+            break;
+        }
+    }
+}
+
+
+//inline void decodebytesC14X2(u32 *dst, const u16 *src, int numpixels, int tlutaddr, int tlutfmt)
+inline void decodebytesC14X2(u32 *dst, const u16 *src, int tlutaddr, int tlutfmt)
+{
+    u16 *tlut = (u16*)(texMem+tlutaddr);
+    for (int x = 0; x < 4; x++)
+    {
+        int val = Common::swap16(src[x]);
+        switch (tlutfmt) {
+        case 0:
+            *dst++ = decodeIA8(Common::swap16(tlut[(val&0x3FFF)]));
+            break;
+        case 1:
+            *dst++ = decode565(Common::swap16(tlut[(val&0x3FFF)]));
+            break;
+        case 2:
+            *dst++ = decode5A3(Common::swap16(tlut[(val&0x3FFF)]));
+            break;
+        case 3: //ERROR
+            *dst++ = 0xFFFF00FF;
+            break;
+        }
+    }
+}
+
+//inline void decodebytesRGB565(u32 *dst, const u16 *src, int numpixels)
+inline void decodebytesRGB565(u32 *dst, const u16 *src)
+{
+    for (int x = 0; x < 4; x++)
+        *dst++ = decode565(Common::swap16(src[x]));
+}
+
+//inline void decodebytesIA4(u32 *dst, const u8 *src, int numbytes)
+inline void decodebytesIA4(u32 *dst, const u8 *src)
+{
+    for (int x = 0; x < 8; x++)
+    {
+        int val = src[x];
+        int a = lut4to8[val>>4];
+        int r = lut4to8[val&15];
+        dst[x] = (a<<24) | (r<<16) | (r<<8) | r;
+    }
+}
+
+//inline void decodebytesIA8(u32 *dst, const u16 *src, int numpixels)
+inline void decodebytesIA8(u32 *dst, const u16 *src)
+{
+    for (int x = 0; x < 4; x++)
+        dst[x] = decodeIA8(Common::swap16(src[x]));
+}
+
+//inline void decodebytesRGB5A3(u32 *dst, const u16 *src, int numpixels)
+inline void decodebytesRGB5A3(u32 *dst, const u16 *src)
+{
+    for (int x = 0; x < 4; x++)
+        dst[x] = decode5A3(Common::swap16(src[x]));
+}
+
+// This one is used by many video formats. It'd therefore be good if it was fast.
+inline void decodebytesARGB8_4(u32 *dst, const u16 *src, const u16 *src2)
+{
+    for (int x = 0; x < 4; x++) {
+        dst[x] = Common::swap32((src2[x] << 16) | src[x]);
+    }
+
+	// This can probably be done in a few SSE pack/unpack instructions + pshufb
+	// some unpack instruction x2:
+	// ABABABABABABABAB 1212121212121212 ->
+	// AB12AB12AB12AB12 AB12AB12AB12AB12
+	// 2x pshufb-> 
+	// 21BA21BA21BA21BA 21BA21BA21BA21BA
+	// and we are done.
+}
+
+inline u32 makecol(int r, int g, int b, int a)
+{
+    return (a<<24)|(r<<16)|(g<<8)|b;
+}
+
+void decodeDXTBlock(u32 *dst, const DXTBlock *src, int pitch)
+{
+    u16 c1 = Common::swap16(src->color1);
+    u16 c2 = Common::swap16(src->color2);
+    int blue1 = lut5to8[c1 & 0x1F];
+    int blue2 = lut5to8[c2 & 0x1F];
+    int green1 = lut6to8[(c1>>5) & 0x3F];
+    int green2 = lut6to8[(c2>>5) & 0x3F];
+    int red1 = lut5to8[(c1>>11) & 0x1F];
+    int red2 = lut5to8[(c2>>11) & 0x1F];
+    
+    int colors[4];
+
+    if (c1 > c2)
+    {
+        colors[0] = makecol(red1, green1, blue1, 255);
+        colors[1] = makecol(red2, green2, blue2, 255);
+        colors[2] = makecol(red1+(red2-red1)/3, green1+(green2-green1)/3, blue1+(blue2-blue1)/3, 255);
+        colors[3] = makecol(red2+(red1-red2)/3, green2+(green1-green2)/3, blue2+(blue1-blue2)/3, 255);
+    }
+    else
+    {
+        colors[0] = makecol(red1, green1, blue1, 255);
+        colors[1] = makecol(red2, green2, blue2, 255);
+        colors[2] = makecol((red1+red2)/2, (green1+green2)/2, (blue1+blue2)/2, 255);
+        colors[3] = makecol(0,0,0,0); //transparent
+    }
+
+    for (int y = 0; y < 4; y++)
+    { 
+        int val = src->lines[y];
+        for (int x = 0; x < 4; x++)
+        {
+            dst[x] = colors[(val >> 6) & 3];
+            val <<= 2;
+        }
+        dst += pitch;
+    }
+}
+
+
+//switch endianness, unswizzle
+//TODO: to save memory, don't blindly convert everything to argb8888
+//also ARGB order needs to be swapped later, to accommodate modern hardware better
+//need to add DXT support too
+#ifdef OVERLAY_TEXFMT
+PC_TexFormat TexDecoder_Decode_real(u8 *dst, const u8 *src, int width, int height, int texformat, int tlutaddr, int tlutfmt)
+#else
+PC_TexFormat TexDecoder_Decode(u8 *dst, const u8 *src, int width, int height, int texformat, int tlutaddr, int tlutfmt)
+#endif
+{
+    switch (texformat)
+    {
+    case GX_TF_C4:
+        {
+            for (int y = 0; y < height; y += 8)
+                for (int x = 0; x < width; x += 8)
+                    for (int iy = 0; iy < 8; iy++, src += 4)
+                        //decodebytesC4((u32*)dst+(y+iy)*width+x, src, 4, tlutaddr, tlutfmt);
+                        decodebytesC4((u32*)dst+(y+iy)*width+x, src, tlutaddr, tlutfmt);
+        }
+        return PC_TEX_FMT_BGRA32;
+    case GX_TF_I4:
+        {
+        	// TODO: SSSE3 variant (pshufb), THP videos use this format.
+        	// SSSE3 variant could bring even more speed
+#if (defined(_WIN32) || (defined (_M_X64) && !defined(_WIN32)))
+			__m128i Lmask = _mm_set1_epi8 (0x0F);
+			__m128i Hmask = _mm_set1_epi8 (0xF0);
+			__m128i* sseSrc  = (__m128i *)src;
+			__m128i* sseDst  = (__m128i *)dst;
+            for (int y = 0; y < height; y += 8)
+	            for (int x = 0; x < width; x += 8)
+					for (int iy = 0; iy < 8; iy++, sseSrc++) {
+						// TODO (mb2): func and don't let the optimizer perform all the clean up by itself
+						__m128i s = _mm_load_si128 (sseSrc);			// ab cd ef gh ...
+						__m128i sl = _mm_and_si128 (s, Lmask);			// 0b 0d 0f 0h ...
+						__m128i sls = _mm_slli_epi16 (sl, 4);			// b0 d0 f0 h0 ...
+						__m128i sl_ = _mm_or_si128 (sl, sls);			// bb dd ff ff ...
+
+						__m128i sh = _mm_and_si128 (s, Hmask);			// a0 c0 e0 g0 ...
+						__m128i shs = _mm_srli_epi16 (sh, 4);			// 0a 0c 0e g0 ...
+						__m128i sh_ = _mm_or_si128 (sh, shs);			// aa cc ee gg ...
+						__m128i rl = _mm_unpacklo_epi8 (sh_, sl_);		// bb aa dd cc ...
+						__m128i rh = _mm_unpackhi_epi8 (sh_, sl_);		// 
+
+						__m128i ral = _mm_unpacklo_epi8 (rl, rl);		// bb bb aa aa ...
+						__m128i rah = _mm_unpackhi_epi8 (rl, rl);		// 
+						// result part a
+						__m128i rall = _mm_unpacklo_epi16 (ral, ral);	// bb bb bb bb ...	-> done
+						__m128i ralh = _mm_unpackhi_epi16 (ral, ral);	//					-> done
+						__m128i rahl = _mm_unpacklo_epi16 (rah, rah);	//					-> done
+						__m128i rahh = _mm_unpackhi_epi16 (rah, rah);	//					-> done
+
+						__m128i rbl = _mm_unpacklo_epi8 (rh, rh);		// 
+						__m128i rbh = _mm_unpackhi_epi8 (rh, rh);		// 
+						// result part b
+						__m128i rbll = _mm_unpacklo_epi16 (rbl, rbl);	//					-> done
+						__m128i rblh = _mm_unpackhi_epi16 (rbl, rbl);	//					-> done
+						__m128i rbhl = _mm_unpacklo_epi16 (rbh, rbh);	//					-> done
+						__m128i rbhh = _mm_unpackhi_epi16 (rbh, rbh);	//					-> done
+						// store
+						sseDst = (__m128i*)(dst+((y+iy)*width+x)*4);	// that sucks... too lazy 
+						_mm_store_si128 (sseDst++, rall); 
+						_mm_store_si128 (sseDst, ralh);
+						iy++;
+						sseDst = (__m128i*)(dst+((y+iy)*width+x)*4);
+						_mm_store_si128 (sseDst++, rahl);
+						_mm_store_si128 (sseDst, rahh);
+						iy++;
+						sseDst = (__m128i*)(dst+((y+iy)*width+x)*4);
+						_mm_store_si128 (sseDst++, rbll);
+						_mm_store_si128 (sseDst, rblh);
+						iy++;
+						sseDst = (__m128i*)(dst+((y+iy)*width+x)*4);
+						_mm_store_si128 (sseDst++, rbhl);
+						_mm_store_si128 (sseDst, rbhh);
+					}
+#else
+            for (int y = 0; y < height; y += 8)
+                for (int x = 0; x < width; x += 8)
+                    for (int iy = 0; iy < 8; iy++, src += 4)
+                        //decodebytesI4((u32*)dst+(y+iy)*width+x, src, 4);
+                        decodebytesI4((u32*)dst+(y+iy)*width+x, src);
+#endif
+        }
+        return PC_TEX_FMT_BGRA32;
+    case GX_TF_C8:
+        {
+            for (int y = 0; y < height; y += 4)
+                for (int x = 0; x < width; x += 8)
+                    for (int iy = 0; iy < 4; iy++, src += 8)
+                        //decodebytesC8((u32*)dst+(y+iy)*width+x, src, 8, tlutaddr, tlutfmt);
+                        decodebytesC8((u32*)dst+(y+iy)*width+x, src, tlutaddr, tlutfmt);
+        }
+        return PC_TEX_FMT_BGRA32;
+    case GX_TF_I8:  // speed critical
+		{
+#if (defined(_WIN32) || (defined (_M_X64) && !defined(_WIN32)))
+			__m128i *sseSrc  = (__m128i *)src;
+			__m128i *sseDst  = (__m128i *)dst;
+            for (int y = 0; y < height; y += 4)
+                for (int x = 0; x < width; x += 8)
+					for (int iy = 0; iy < 4; iy++, sseSrc++) {
+						// TODO (mb2): func and don't let the optimizer perform all the clean up by itself
+						__m128i s = _mm_load_si128 (sseSrc);			// ab cd ef gh ...
+						__m128i rl = _mm_unpacklo_epi8 (s, s);			// ab ab cd cd ...
+						__m128i rh = _mm_unpackhi_epi8 (s, s);			// 
+						// result
+						__m128i rll = _mm_unpacklo_epi8 (rl, rl);		// ab ab ab ab 
+						__m128i rlh = _mm_unpackhi_epi8 (rl, rl);
+						__m128i rhl = _mm_unpacklo_epi8 (rh, rh);
+						__m128i rhh = _mm_unpackhi_epi8 (rh, rh);
+						// store
+						sseDst = (__m128i*)(dst+((y+iy)*width+x)*4);
+						_mm_store_si128 (sseDst++, rll); 
+						_mm_store_si128 (sseDst, rlh);
+						iy++;
+						sseDst = (__m128i*)(dst+((y+iy)*width+x)*4);
+						_mm_store_si128 (sseDst++, rhl);
+						_mm_store_si128 (sseDst, rhh);
+					}
+#else
+            for (int y = 0; y < height; y += 4)
+                for (int x = 0; x < width; x += 8)
+                    for (int iy = 0; iy < 4; iy++, src += 8)
+                        decodebytesI8_8((u32*)dst+(y+iy)*width+x, src);
+#endif
+		}
+        return PC_TEX_FMT_BGRA32;
+    case GX_TF_IA4:
+        {
+            for (int y = 0; y < height; y += 4)
+                for (int x = 0; x < width; x += 8)
+                    for (int iy = 0; iy < 4; iy++, src += 8)
+                        //decodebytesIA4((u32*)dst+(y+iy)*width+x, src, 8);
+                        decodebytesIA4((u32*)dst+(y+iy)*width+x, src);
+        }
+        return PC_TEX_FMT_BGRA32;
+    case GX_TF_IA8:
+        {
+            for (int y = 0; y < height; y += 4)
+                for (int x = 0; x < width; x += 4)
+                    for (int iy = 0; iy < 4; iy++, src += 8)
+                        //decodebytesIA8((u32*)dst+(y+iy)*width+x, (u16*)src, 4);
+                        decodebytesIA8((u32*)dst+(y+iy)*width+x, (u16*)src);
+        }
+        return PC_TEX_FMT_BGRA32;
+    case GX_TF_C14X2: 
+        {
+            for (int y = 0; y < height; y += 4)
+                for (int x = 0; x < width; x += 4)
+                    for (int iy = 0; iy < 4; iy++, src += 8)
+                        //decodebytesC14X2((u32*)dst+(y+iy)*width+x, (u16*)src, 4, tlutaddr, tlutfmt);
+                        decodebytesC14X2((u32*)dst+(y+iy)*width+x, (u16*)src, tlutaddr, tlutfmt);
+        }
+        return PC_TEX_FMT_BGRA32;
+    case GX_TF_RGB565:
+        {
+            for (int y = 0; y < height; y += 4)
+                for (int x = 0; x < width; x += 4)
+                    for (int iy = 0; iy < 4; iy++, src += 8)
+                        //decodebytesRGB565((u32*)dst+(y+iy)*width+x, (u16*)src, 4);
+                        decodebytesRGB565((u32*)dst+(y+iy)*width+x, (u16*)src);
+        }
+        return PC_TEX_FMT_BGRA32;
+    case GX_TF_RGB5A3:
+        {
+            for (int y = 0; y < height; y += 4)
+                for (int x = 0; x < width; x += 4)
+                    for (int iy = 0; iy < 4; iy++, src += 8)
+                        //decodebytesRGB5A3((u32*)dst+(y+iy)*width+x, (u16*)src, 4);
+                        decodebytesRGB5A3((u32*)dst+(y+iy)*width+x, (u16*)src);
+        }
+        return PC_TEX_FMT_BGRA32;
+    case GX_TF_RGBA8:  // speed critical
+        {
+			for (int y = 0; y < height; y += 4) {
+                for (int x = 0; x < width; x += 4)
+                {
+					for (int iy = 0; iy < 4; iy++) {
+                        decodebytesARGB8_4((u32*)dst + (y+iy)*width + x, (u16*)src + 4 * iy, (u16*)src + 4 * iy + 16);
+					}
+					src += 64;
+                }
+			}
+        }
+        return PC_TEX_FMT_BGRA32;
+    case GX_TF_CMPR:  // speed critical
+        {
+			// TODO: Shuffle to PC S3TC (DXTC) format instead of converting
+            // 11111111 22222222 55555555 66666666
+            // 33333333 44444444 77777777 88888888
+			// The metroid games use this format almost exclusively.
+            for (int y = 0; y < height; y += 8)
+                for (int x = 0; x < width; x += 8)
+                {
+                    decodeDXTBlock((u32*)dst+y*width+x, (DXTBlock*)src, width);
+					src += sizeof(DXTBlock);
+                    decodeDXTBlock((u32*)dst+y*width+x+4, (DXTBlock*)src, width);
+					src += sizeof(DXTBlock);
+                    decodeDXTBlock((u32*)dst+(y+4)*width+x, (DXTBlock*)src, width);
+					src += sizeof(DXTBlock);
+                    decodeDXTBlock((u32*)dst+(y+4)*width+x+4, (DXTBlock*)src, width);
+					src += sizeof(DXTBlock);
+                }
+        }
+        return PC_TEX_FMT_BGRA32;
+    }
+
+	// The "copy" texture formats, too?
+    return PC_TEX_FMT_NONE;
+}
+
+void TexDecoder_SetTexFmtOverlayOptions(bool enable, bool center)
+{
+#ifdef OVERLAY_TEXFMT
+	TexFmt_Overlay_Enable = enable;
+	TexFmt_Overlay_Center = center;
+#endif
+}
+
+#ifdef OVERLAY_TEXFMT
+extern const char* texfmt[];
+extern const unsigned char sfont_map[];
+extern const unsigned char sfont_raw[][9*10];
+
+PC_TexFormat TexDecoder_Decode(u8 *dst, const u8 *src, int width, int height, int texformat, int tlutaddr, int tlutfmt)
+{
+	PC_TexFormat  retval = TexDecoder_Decode_real(dst,src,width,height,texformat,tlutaddr,tlutfmt);
+
+	if((!TexFmt_Overlay_Enable)||(retval==PC_TEX_FMT_NONE))
+		return retval;
+
+	// assume ABGR/ARGB (32bit)
+	int  *dtp = (int*)dst;
+
+	int w = min(width,40);
+	int h = min(height,10);
+
+	int xoff = (width-w)>>1;
+	int yoff = (height-h)>>1;
+
+	if(!TexFmt_Overlay_Center)
+	{
+		xoff=0;
+		yoff=0;
+	}
+
+	const char* fmt = texfmt[texformat&15];
+	while(*fmt)
+	{
+		int xcnt = 0;
+		int nchar = sfont_map[(int)*fmt];
+		
+		const unsigned char *ptr = sfont_raw[nchar]; // each char is up to 9x10
+
+		for(int x=0;x<9;x++)
+		{
+			if(ptr[x]==0x78)
+				break;
+			xcnt++;
+		}
+		for(int y=0;y<10;y++)
+		{
+			for(int x=0;x<xcnt;x++)
+			{
+				dtp[(y+yoff)*width + x+xoff] = ptr[x]?0xFFFFFFFF:0xFF000000;
+			}
+			ptr+=9;
+		}
+		xoff+=xcnt;
+		fmt++;
+	}
+
+	return retval;
+}
+
+const char* texfmt[] = {
+	// pixel
+	"I4",		"I8",		"IA4",		"IA8",
+	"RGB565",	"RGB5A3",	"RGBA8",	"C4",
+	"C8",		"C14X2",	"0x0A",		"0x0B",
+	"0x0C",		"0x0D",		"CMPR",		"0x0F",
+	// Z-buffer
+	"0x10",		"Z8",		"0x12",		"Z16",
+	"0x14",		"0x15",		"Z24X8",	"0x17",
+	"0x18",		"0x19",		"0x1A",		"0x1B",
+	"0x1C",		"0x1D",		"0x1E",		"0x1F",
+	// pixel + copy
+	"CR4",		"0x21",		"CRA4",		"CRA8",
+	"0x24",		"0x25",		"CYUVA8",	"CA8",
+	"CR8",		"CG8",		"CB8",		"CRG8",
+	"CGB8",		"0x2D",		"0x2E",		"0x2F",
+	// Z + copy
+	"CZ4",		"0x31",		"0x32",		"0x33",
+	"0x34",		"0x35",		"0x36",		"0x37",
+	"0x38",		"CZ8M",		"CZ8L",		"0x3B",
+	"CZ16L",	"0x3D",		"0x3E",		"0x3F",
+};
+
+const unsigned char sfont_map[] = {
+	10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,
+	10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,
+	10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,
+	 0, 1, 2, 3, 4, 5, 6, 7, 8, 9,10,10,10,10,10,10,
+	10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,
+	26,27,28,29,30,31,32,33,34,35,36,10,10,10,10,10,
+	10,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,
+	52,53,54,55,56,57,58,59,60,61,62,10,10,10,10,10,
+	10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,
+	10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,
+	10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,
+	10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,
+	10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,
+	10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,
+	10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,
+	10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,
+};
+
+const unsigned char sfont_raw[][9*10] = {
+	{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78,
+	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78,
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78,
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78,
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 
+	0xff, 0x00, 0x00, 0xff, 0xff, 0xff, 0x00, 0x00, 0xff, 
+	0xff, 0x00, 0x00, 0xff, 0xff, 0xff, 0x00, 0x00, 0xff, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78,
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0x00, 0x00, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0xff, 0x00, 0xff, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0xff, 0x00, 0xff, 
+	0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 
+	0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0xff, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0xff, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78,
+	0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0x00, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	},{
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0x00, 0x00, 0x00, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	0xff, 0xff, 0xff, 0xff, 0xff, 0x78, 0x78, 0x78, 0x78, 
+	},
+};
+#endif
+
diff --git a/Source/Core/VideoCommon/Src/VertexShader.cpp b/Source/Core/VideoCommon/Src/VertexShader.cpp
index 97b7f9b426..04acf1c651 100644
--- a/Source/Core/VideoCommon/Src/VertexShader.cpp
+++ b/Source/Core/VideoCommon/Src/VertexShader.cpp
@@ -1,466 +1,466 @@
-// Copyright (C) 2003-2008 Dolphin Project.

-

-// This program is free software: you can redistribute it and/or modify

-// it under the terms of the GNU General Public License as published by

-// the Free Software Foundation, version 2.0.

-

-// This program is distributed in the hope that it will be useful,

-// but WITHOUT ANY WARRANTY; without even the implied warranty of

-// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

-// GNU General Public License 2.0 for more details.

-

-// A copy of the GPL 2.0 should have been included with the program.

-// If not, see http://www.gnu.org/licenses/

-

-// Official SVN repository and contact information can be found at

-// http://code.google.com/p/dolphin-emu/

-

-#include <math.h>

-

-#include "Profiler.h"

-#include "NativeVertexFormat.h"

-

-#include "BPMemory.h"

-#include "VertexShader.h"

-

-static char text[16384];

-

-#define WRITE p+=sprintf

-

-#define LIGHTS_POS ""

-

-char *GenerateLightShader(char* p, int index, const LitChannel& chan, const char* dest, int coloralpha);

-

-char *GenerateVertexShader(u32 components, bool has_zbuffer_target)

-{

-	text[sizeof(text) - 1] = 0x7C;  // canary

-    DVSTARTPROFILE();

-

-    _assert_( bpmem.genMode.numtexgens == xfregs.numTexGens);

-    _assert_( bpmem.genMode.numcolchans == xfregs.nNumChans);

-    

-    u32 lightMask = 0;

-    if (xfregs.nNumChans > 0)

-		lightMask |= xfregs.colChans[0].color.GetFullLightMask() | xfregs.colChans[0].alpha.GetFullLightMask();

-    if (xfregs.nNumChans > 1)

-        lightMask |= xfregs.colChans[1].color.GetFullLightMask() | xfregs.colChans[1].alpha.GetFullLightMask();

-

-    bool bOutputZ = bpmem.ztex2.op==ZTEXTURE_ADD || has_zbuffer_target;

-    int ztexcoord = -1;

-

-    char *p = text;

-    WRITE(p, "//Vertex Shader: comp:%x, \n", components);

-    WRITE(p, "typedef struct {\n"

-        "  float4 T0, T1, T2;\n"

-        "  float4 N0, N1, N2;\n"

-        "} s_"I_POSNORMALMATRIX";\n\n"

-        "typedef struct {\n"

-        "  float4 t;\n"

-        "} FLT4;\n"

-        "typedef struct {\n"

-        "  FLT4 T[24];\n"

-        "} s_"I_TEXMATRICES";\n\n"

-        "typedef struct {\n"

-        "  FLT4 T[64];\n"

-        "} s_"I_TRANSFORMMATRICES";\n\n"

-        "typedef struct {\n"

-        "  FLT4 T[32];\n"

-        "} s_"I_NORMALMATRICES";\n\n"

-        "typedef struct {\n"

-        "  FLT4 T[64];\n"

-        "} s_"I_POSTTRANSFORMMATRICES";\n\n"

-        "typedef struct {\n"

-        "  float4 col;\n"

-        "  float4 cosatt;\n"

-        "  float4 distatt;\n"

-        "  float4 pos;\n"

-        "  float4 dir;\n"

-        "} Light;\n\n"

-        "typedef struct {\n"

-        "  Light lights[8];\n"

-        "} s_"I_LIGHTS";\n\n"

-        "typedef struct {\n"

-        "  float4 C0, C1, C2, C3;\n"

-        "} s_"I_MATERIALS";\n\n"

-        "typedef struct {\n"

-        "  float4 T0,T1,T2,T3;\n"

-        "} s_"I_PROJECTION";\n"

-        "typedef struct {\n"

-        "  float4 params;\n" // a, b, c, b_shift

-        "} s_"I_FOGPARAMS";\n\n");

-

-    WRITE(p, "struct VS_OUTPUT {\n");

-    WRITE(p, "  float4 pos : POSITION;\n");

-    WRITE(p, "  float4 colors[2] : COLOR0;\n");

-

-    // if outputting Z, embed the Z coordinate in the w component of a texture coordinate

-    // if number of tex gens occupies all the texture coordinates, use the last tex coord

-    // otherwise use the next available tex coord

-    for (int i = 0; i < xfregs.numTexGens; ++i) {

-        WRITE(p, "  float%d tex%d : TEXCOORD%d;\n", (i==(xfregs.numTexGens-1)&&bOutputZ)?4:3, i, i);

-    }

-    if (bOutputZ && xfregs.numTexGens == 0) {

-        ztexcoord = 0;

-        WRITE(p, "  float4 tex%d : TEXCOORD%d;\n", ztexcoord, ztexcoord);

-    }

-    else if (bOutputZ)

-        ztexcoord = xfregs.numTexGens - 1;

-

-    WRITE(p, "};\n");

-    WRITE(p, "\n");

-

-    // uniforms

-    

-    // bool bTexMtx = ((components & VB_HAS_TEXMTXIDXALL)<<VB_HAS_UVTEXMTXSHIFT)!=0; unused TODO: keep?

-

-    WRITE(p, "uniform s_"I_TRANSFORMMATRICES" "I_TRANSFORMMATRICES" : register(c%d);\n", C_TRANSFORMMATRICES);

-    WRITE(p, "uniform s_"I_TEXMATRICES" "I_TEXMATRICES" : register(c%d);\n", C_TEXMATRICES); // also using tex matrices

-    WRITE(p, "uniform s_"I_NORMALMATRICES" "I_NORMALMATRICES" : register(c%d);\n", C_NORMALMATRICES);

-    WRITE(p, "uniform s_"I_POSNORMALMATRIX" "I_POSNORMALMATRIX" : register(c%d);\n", C_POSNORMALMATRIX);

-    WRITE(p, "uniform s_"I_POSTTRANSFORMMATRICES" "I_POSTTRANSFORMMATRICES" : register(c%d);\n", C_POSTTRANSFORMMATRICES);

-    WRITE(p, "uniform s_"I_LIGHTS" "I_LIGHTS" : register(c%d);\n", C_LIGHTS);

-    WRITE(p, "uniform s_"I_MATERIALS" "I_MATERIALS" : register(c%d);\n", C_MATERIALS);

-    WRITE(p, "uniform s_"I_PROJECTION" "I_PROJECTION" : register(c%d);\n", C_PROJECTION);

-    WRITE(p, "uniform s_"I_FOGPARAMS" "I_FOGPARAMS" : register(c%d);\n", C_FOGPARAMS);

-

-    WRITE(p, "VS_OUTPUT main(\n");

-    

-    // inputs

-    if (components & VB_HAS_NRM0)

-        WRITE(p, "  float3 rawnorm0 : NORMAL,\n");

-    if (components & VB_HAS_NRM1)

-        WRITE(p, "  float3 rawnorm1 : ATTR%d,\n", SHADER_NORM1_ATTRIB);

-    if (components & VB_HAS_NRM2)

-        WRITE(p, "  float3 rawnorm2 : ATTR%d,\n", SHADER_NORM2_ATTRIB);

-    if (components & VB_HAS_COL0)

-        WRITE(p, "  float4 color0 : COLOR0,\n");

-    if (components & VB_HAS_COL1)

-        WRITE(p, "  float4 color1 : COLOR1,\n");

-    for (int i = 0; i < 8; ++i) {

-        u32 hastexmtx = (components & (VB_HAS_TEXMTXIDX0<<i));

-        if ((components & (VB_HAS_UV0<<i)) || hastexmtx )

-            WRITE(p, "  float%d tex%d : TEXCOORD%d,\n", hastexmtx ? 3 : 2, i,i);

-    }

-    if (components & VB_HAS_POSMTXIDX)

-        WRITE(p, "  half posmtx : ATTR%d,\n", SHADER_POSMTX_ATTRIB);

-

-    WRITE(p, "  float4 rawpos : POSITION) {\n");

-    WRITE(p, "VS_OUTPUT o;\n");

-

-    // transforms

-    if (components & VB_HAS_POSMTXIDX) {

-        WRITE(p, "float4 pos = float4(dot("I_TRANSFORMMATRICES".T[posmtx].t, rawpos), dot("I_TRANSFORMMATRICES".T[posmtx+1].t, rawpos), dot("I_TRANSFORMMATRICES".T[posmtx+2].t, rawpos),1);\n");

-        

-        if (components & VB_HAS_NRMALL) {

-            WRITE(p, "int normidx = posmtx >= 32 ? (posmtx-32) : posmtx;\n");

-            WRITE(p, "float3 N0 = "I_NORMALMATRICES".T[normidx].t.xyz, N1 = "I_NORMALMATRICES".T[normidx+1].t.xyz, N2 = "I_NORMALMATRICES".T[normidx+2].t.xyz;\n");

-        }

-

-        if (components & VB_HAS_NRM0)

-            WRITE(p, "half3 _norm0 = half3(dot(N0, rawnorm0), dot(N1, rawnorm0), dot(N2, rawnorm0));\n"

-                     "half3 norm0 = normalize(_norm0);\n");

-        if (components & VB_HAS_NRM1)

-            WRITE(p, "half3 _norm1 = half3(dot(N0, rawnorm1), dot(N1, rawnorm1), dot(N2, rawnorm1));\n");

-                    //"half3 norm1 = normalize(_norm1);\n");

-        if (components & VB_HAS_NRM2)

-            WRITE(p, "half3 _norm2 = half3(dot(N0, rawnorm2), dot(N1, rawnorm2), dot(N2, rawnorm2));\n");

-                    //"half3 norm2 = normalize(_norm2);\n");

-    }

-    else {

-        WRITE(p, "float4 pos = float4(dot("I_POSNORMALMATRIX".T0, rawpos), dot("I_POSNORMALMATRIX".T1, rawpos), dot("I_POSNORMALMATRIX".T2, rawpos), 1);\n");

-        if (components & VB_HAS_NRM0)

-            WRITE(p, "half3 _norm0 = half3(dot("I_POSNORMALMATRIX".N0.xyz, rawnorm0), dot("I_POSNORMALMATRIX".N1.xyz, rawnorm0), dot("I_POSNORMALMATRIX".N2.xyz, rawnorm0));\n"

-                     "half3 norm0 = normalize(_norm0);\n");

-        if (components & VB_HAS_NRM1)

-            WRITE(p, "half3 _norm1 = half3(dot("I_POSNORMALMATRIX".N0.xyz, rawnorm1), dot("I_POSNORMALMATRIX".N1.xyz, rawnorm1), dot("I_POSNORMALMATRIX".N2.xyz, rawnorm1));\n");

-                    //"half3 norm1 = normalize(_norm1);\n");

-        if (components & VB_HAS_NRM2)

-            WRITE(p, "half3 _norm2 = half3(dot("I_POSNORMALMATRIX".N0.xyz, rawnorm2), dot("I_POSNORMALMATRIX".N1.xyz, rawnorm2), dot("I_POSNORMALMATRIX".N2.xyz, rawnorm2));\n");

-                    //"half3 norm2 = normalize(_norm2);\n");

-    }

-

-    if (!(components & VB_HAS_NRM0))

-        WRITE(p, "half3 _norm0 = half3(0,0,0), norm0= half3(0,0,0);\n");

-

-    WRITE(p, "o.pos = float4(dot("I_PROJECTION".T0, pos), dot("I_PROJECTION".T1, pos), dot("I_PROJECTION".T2, pos), dot("I_PROJECTION".T3, pos));\n");

-

-    WRITE(p, "half4 mat, lacc;\n"

-    "half3 ldir, h;\n"

-    "half dist, dist2, attn;\n");

-

-    // lights/colors

-    for (int j = 0; j < xfregs.nNumChans; j++) {

-

-        // bool bColorAlphaSame = xfregs.colChans[j].color.hex == xfregs.colChans[j].alpha.hex;  unused

-        const LitChannel& color = xfregs.colChans[j].color;

-        const LitChannel& alpha = xfregs.colChans[j].alpha;

-

-        WRITE(p, "{\n");

-

-        if (color.matsource) {// from vertex

-            if (components & (VB_HAS_COL0<<j) )

-                WRITE(p, "mat = color%d;\n", j);

-            else WRITE(p, "mat = half4(1,1,1,1);\n");

-        }

-        else // from color

-            WRITE(p, "mat = "I_MATERIALS".C%d;\n", j+2);

-

-        if (color.enablelighting) {

-            if (color.ambsource) {// from vertex

-                if (components & (VB_HAS_COL0<<j) )

-                    WRITE(p, "lacc = color%d;\n", j);

-                else WRITE(p, "lacc = half4(0.0f,0.0f,0.0f,0.0f);\n");

-            }

-            else // from color

-                WRITE(p, "lacc = "I_MATERIALS".C%d;\n", j);

-        }

-

-        // check if alpha is different

-        if (alpha.matsource != color.matsource) {

-            if (alpha.matsource) {// from vertex

-                if (components & (VB_HAS_COL0<<j) )

-                    WRITE(p, "mat.w = color%d.w;\n", j);

-                else WRITE(p, "mat.w = 1;\n");

-            }

-            else // from color

-                WRITE(p, "mat.w = "I_MATERIALS".C%d.w;\n", j+2);

-        }

-

-        if (alpha.enablelighting && alpha.ambsource != color.ambsource) {

-            if (alpha.ambsource) {// from vertex

-                if (components & (VB_HAS_COL0<<j) )

-                    WRITE(p, "lacc.w = color%d.w;\n", j);

-                else WRITE(p, "lacc.w = 0;\n");

-            }

-            else // from color

-                WRITE(p, "lacc.w = "I_MATERIALS".C%d.w;\n", j);

-        }

-        

-        if (color.enablelighting && alpha.enablelighting && (color.GetFullLightMask() != alpha.GetFullLightMask() || color.lightparams != alpha.lightparams)) {

-

-            // both have lighting, except not using the same lights

-            int mask = 0; // holds already computed lights

-

-            if (color.lightparams == alpha.lightparams && (color.GetFullLightMask() & alpha.GetFullLightMask())) {

-                // if lights are shared, compute those first

-                mask = color.GetFullLightMask() & alpha.GetFullLightMask();

-                for (int i = 0; i < 8; ++i) {

-                    if (mask&(1<<i))

-                        p = GenerateLightShader(p, i, color, "lacc", 3);

-                }

-            }

-

-            // no shared lights

-            for (int i = 0; i < 8; ++i) {

-                if (!(mask&(1<<i)) && (color.GetFullLightMask() & (1<<i)) )

-                    p = GenerateLightShader(p, i, color, "lacc", 1);

-                if (!(mask&(1<<i)) && (alpha.GetFullLightMask() & (1<<i)) )

-                    p = GenerateLightShader(p, i, alpha, "lacc", 2);

-            }

-        }

-        else if (color.enablelighting || alpha.enablelighting) {

-            // either one is enabled

-            int coloralpha = (int)color.enablelighting|((int)alpha.enablelighting<<1);

-            for (int i = 0; i < 8; ++i) {

-                if (color.GetFullLightMask() & (1<<i) )

-                    p = GenerateLightShader(p, i, color.enablelighting?color:alpha, "lacc", coloralpha);

-            }

-        }

-

-        if (color.enablelighting != alpha.enablelighting) {

-            if (color.enablelighting )

-                WRITE(p, "o.colors[%d].xyz = mat.xyz * clamp(lacc.xyz,float3(0.0f,0.0f,0.0f),float3(1.0f,1.0f,1.0f));\n"

-                    "o.colors[%d].w = mat.w;\n", j, j);

-            else

-                WRITE(p, "o.colors[%d].xyz = mat.xyz;\n"

-                    "o.colors[%d].w = mat.w  * clamp(lacc.w,0.0f,1.0f);\n", j, j);

-        }

-        else {

-            if (alpha.enablelighting )

-                WRITE(p, "o.colors[%d] = mat * clamp(lacc,float4(0.0f,0.0f,0.0f,0.0f), float4(1.0f,1.0f,1.0f,1.0f));\n", j);

-            else WRITE(p, "o.colors[%d] = mat;\n", j);

-        }

-        WRITE(p, "}\n");

-    }

-

-    // zero left over channels

-    for (int i = xfregs.nNumChans; i < 2; ++i)

-		WRITE(p, "o.colors[%d] = 0;\n", i);

-

-    // transform texcoords

-    for (int i = 0; i < xfregs.numTexGens; ++i) {

-        TexMtxInfo& texinfo = xfregs.texcoords[i].texmtxinfo;

-

-        WRITE(p, "{\n");

-        switch (texinfo.sourcerow) {

-        case XF_SRCGEOM_INROW:

-            _assert_( texinfo.inputform == XF_TEXINPUT_ABC1 );

-            WRITE(p, "float4 coord = rawpos;\n"); // pos.w is 1

-            break;

-        case XF_SRCNORMAL_INROW:

-            if (components & VB_HAS_NRM0) {

-                _assert_( texinfo.inputform == XF_TEXINPUT_ABC1 );

-                WRITE(p, "float4 coord = float4(rawnorm0.xyz, 1.0);\n");

-            }

-            else WRITE(p, "float4 coord = 0;\n");

-            break;

-        case XF_SRCCOLORS_INROW:

-            _assert_( texinfo.texgentype == XF_TEXGEN_COLOR_STRGBC0 || texinfo.texgentype == XF_TEXGEN_COLOR_STRGBC1 );

-            break;

-        case XF_SRCBINORMAL_T_INROW:

-            if (components & VB_HAS_NRM1) {

-                _assert_( texinfo.inputform == XF_TEXINPUT_ABC1 );

-                WRITE(p, "float4 coord = float4(rawnorm1.xyz, 1.0);\n");

-            }

-            else WRITE(p, "float4 coord = 0;\n");

-            break;

-        case XF_SRCBINORMAL_B_INROW:

-            if (components & VB_HAS_NRM2) {

-                _assert_( texinfo.inputform == XF_TEXINPUT_ABC1 );

-                WRITE(p, "float4 coord = float4(rawnorm2.xyz, 1.0);\n");

-            }

-            else WRITE(p, "float4 coord = 0;\n");

-            break;

-        default:

-            _assert_(texinfo.sourcerow <= XF_SRCTEX7_INROW);

-            if (components & (VB_HAS_UV0<<(texinfo.sourcerow - XF_SRCTEX0_INROW)) )

-                WRITE(p, "float4 coord = float4(tex%d.x, tex%d.y, 1.0f, 1.0f);\n", texinfo.sourcerow - XF_SRCTEX0_INROW, texinfo.sourcerow - XF_SRCTEX0_INROW);

-            else

-                WRITE(p, "float4 coord = float4(0.0f, 0.0f, 1.0f, 1.0f);\n");

-            break;

-        }

-

-        // firs transformation

-        switch (texinfo.texgentype) {

-            case XF_TEXGEN_REGULAR:

-                if (components & (VB_HAS_TEXMTXIDX0<<i)) {

-                    if (texinfo.projection == XF_TEXPROJ_STQ )

-                        WRITE(p, "o.tex%d.xyz = float3(dot(coord, "I_TRANSFORMMATRICES".T[tex%d.z].t), dot(coord, "I_TRANSFORMMATRICES".T[tex%d.z+1].t), dot(coord, "I_TRANSFORMMATRICES".T[tex%d.z+2].t));\n", i, i, i, i);

-                    else {

-                        WRITE(p, "o.tex%d.xyz = float3(dot(coord, "I_TRANSFORMMATRICES".T[tex%d.z].t), dot(coord, "I_TRANSFORMMATRICES".T[tex%d.z+1].t), 1);\n", i, i, i);

-                    }

-                }

-                else {

-                    if (texinfo.projection == XF_TEXPROJ_STQ )

-                        WRITE(p, "o.tex%d.xyz = float3(dot(coord, "I_TEXMATRICES".T[%d].t), dot(coord, "I_TEXMATRICES".T[%d].t), dot(coord, "I_TEXMATRICES".T[%d].t));\n", i, 3*i, 3*i+1, 3*i+2);

-                    else

-                        WRITE(p, "o.tex%d.xyz = float3(dot(coord, "I_TEXMATRICES".T[%d].t), dot(coord, "I_TEXMATRICES".T[%d].t), 1);\n", i, 3*i, 3*i+1);

-                }

-                break;

-            case XF_TEXGEN_EMBOSS_MAP: // calculate tex coords into bump map

-

-                if (components & (VB_HAS_NRM1|VB_HAS_NRM2)) {

-                    // transform the light dir into tangent space

-                    WRITE(p, "ldir = normalize("I_LIGHTS".lights[%d].pos.xyz - pos.xyz);\n", texinfo.embosslightshift); 

-                    WRITE(p, "o.tex%d.xyz = o.tex%d.xyz + float3(dot(ldir, _norm1), dot(ldir, _norm2), 0.0f);\n", i, texinfo.embosssourceshift);

-                }

-                else _assert_(0); // should have normals

-

-                break;

-            case XF_TEXGEN_COLOR_STRGBC0:

-                _assert_(texinfo.sourcerow == XF_SRCCOLORS_INROW);

-                WRITE(p, "o.tex%d.xyz = float3(o.colors[0].x, o.colors[0].y, 1);\n", i);

-                break;

-            case XF_TEXGEN_COLOR_STRGBC1:

-                _assert_(texinfo.sourcerow == XF_SRCCOLORS_INROW);

-                WRITE(p, "o.tex%d.xyz = float3(o.colors[1].x, o.colors[1].y, 1);\n", i);

-                break;

-        }

-

-        if(xfregs.bEnableDualTexTransform && texinfo.texgentype == XF_TEXGEN_REGULAR) { // only works for regular tex gen types?

-            if (xfregs.texcoords[i].postmtxinfo.normalize)

-                WRITE(p, "o.tex%d.xyz = normalize(o.tex%d.xyz);\n", i, i);

-

-            //multiply by postmatrix

-            int postidx = xfregs.texcoords[i].postmtxinfo.index;

-            WRITE(p, "float4 P0 = "I_POSTTRANSFORMMATRICES".T[%d].t;\n"

-                "float4 P1 = "I_POSTTRANSFORMMATRICES".T[%d].t;\n"

-                "float4 P2 = "I_POSTTRANSFORMMATRICES".T[%d].t;\n",

-                postidx&0x3f, (postidx+1)&0x3f, (postidx+2)&0x3f);

-            WRITE(p, "o.tex%d.xyz = float3(dot(P0.xyz, o.tex%d.xyz) + P0.w, dot(P1.xyz, o.tex%d.xyz) + P1.w, dot(P2.xyz, o.tex%d.xyz) + P2.w);\n", i, i, i, i);

-        }

-

-        WRITE(p, "}\n");

-    }

-

-    if (ztexcoord >= 0 )

-        WRITE(p, "o.tex%d.w = o.pos.z/o.pos.w;\n", ztexcoord);

-

-//    if (bpmem.fog.c_proj_fsel.fsel != 0) {

-//        switch (bpmem.fog.c_proj_fsel.fsel) {

-//            case 1: // linear

-//                break;

-//            case 4: // exp

-//                break;

-//            case 5: // exp2

-//                break;

-//            case 6: // backward exp

-//                break;

-//            case 7: // backward exp2

-//                break;

-//        }

-//

-//        WRITE(p, "o.fog = o.pos.z/o.pos.w;\n");

-//    }

-

-    WRITE(p, "return o;\n}\n");

-

-	if (text[sizeof(text) - 1] != 0x7C)

-		PanicAlert("VertexShader generator - buffer too small, canary has been eaten!");

-    return text;

-}

-

-// coloralpha - 1 if color, 2 if alpha

-char* GenerateLightShader(char* p, int index, const LitChannel& chan, const char* dest, int coloralpha)

-{

-    const char* swizzle = "xyzw";

-    if (coloralpha == 1 ) swizzle = "xyz";

-    else if (coloralpha == 2 ) swizzle = "w";

-

-    if (!(chan.attnfunc & 1)) {

-        // atten disabled

-        switch (chan.diffusefunc) {

-            case LIGHTDIF_NONE:

-                WRITE(p, "%s.%s += "I_LIGHTS".lights[%d].col.%s;\n", dest, swizzle, index, swizzle);

-                break;

-            case LIGHTDIF_SIGN:

-            case LIGHTDIF_CLAMP:

-                WRITE(p, "ldir = normalize("I_LIGHTS".lights[%d].pos.xyz - pos.xyz);\n", index);

-                WRITE(p, "%s.%s += %sdot(ldir, norm0)) * "I_LIGHTS".lights[%d].col.%s;\n",

-                    dest, swizzle, chan.diffusefunc != LIGHTDIF_SIGN ? "max(0.0f," :"(", index, swizzle);

-                break;

-            default: _assert_(0);

-        }

-    }

-    else { // spec and spot

-        WRITE(p, "ldir = "I_LIGHTS".lights[%d].pos.xyz - pos.xyz;\n", index);

-

-        if (chan.attnfunc == 3) { // spot

-            WRITE(p, "dist2 = dot(ldir, ldir);\n"

-                "dist = sqrt(dist2);\n"

-                "ldir = ldir / dist;\n"

-                "attn = max(0.0f, dot(ldir, "I_LIGHTS".lights[%d].dir.xyz));\n",index);

-            WRITE(p, "attn = max(0.0f, dot("I_LIGHTS".lights[%d].cosatt.xyz, half3(1, attn, attn*attn))) / dot("I_LIGHTS".lights[%d].distatt.xyz, half3(1,dist,dist2));\n", index, index);

-        }

-        else if (chan.attnfunc == 1) { // specular

-            WRITE(p, "attn = dot(norm0, "I_LIGHTS".lights[%d].pos.xyz) > 0 ? max(0.0f, dot(norm0, "I_LIGHTS".lights[%d].dir.xyz)) : 0;\n", index, index);

-            WRITE(p, "ldir = half3(1,attn,attn*attn);\n");

-            WRITE(p, "attn = max(0.0f, dot("I_LIGHTS".lights[%d].cosatt.xyz, ldir)) / dot("I_LIGHTS".lights[%d].distatt.xyz, ldir);\n", index, index);

-        }

-

-        switch (chan.diffusefunc) {

-            case LIGHTDIF_NONE:

-                WRITE(p, "%s.%s += attn * "I_LIGHTS".lights[%d].col.%s;\n", dest, swizzle, index, swizzle);

-                break;

-            case LIGHTDIF_SIGN:

-            case LIGHTDIF_CLAMP:

-                WRITE(p, "%s.%s += attn * %sdot(ldir, norm0)) * "I_LIGHTS".lights[%d].col.%s;\n",

-                    dest, swizzle, chan.diffusefunc != LIGHTDIF_SIGN ? "max(0.0f," :"(", index, swizzle);

-                break;

-            default: _assert_(0);

-        }

-    }

-    WRITE(p, "\n");

-

-    return p;

-}

+// Copyright (C) 2003-2008 Dolphin Project.
+
+// This program is free software: you can redistribute it and/or modify
+// it under the terms of the GNU General Public License as published by
+// the Free Software Foundation, version 2.0.
+
+// This program is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU General Public License 2.0 for more details.
+
+// A copy of the GPL 2.0 should have been included with the program.
+// If not, see http://www.gnu.org/licenses/
+
+// Official SVN repository and contact information can be found at
+// http://code.google.com/p/dolphin-emu/
+
+#include <math.h>
+
+#include "Profiler.h"
+#include "NativeVertexFormat.h"
+
+#include "BPMemory.h"
+#include "VertexShader.h"
+
+static char text[16384];
+
+#define WRITE p+=sprintf
+
+#define LIGHTS_POS ""
+
+char *GenerateLightShader(char* p, int index, const LitChannel& chan, const char* dest, int coloralpha);
+
+char *GenerateVertexShader(u32 components, bool has_zbuffer_target)
+{
+	text[sizeof(text) - 1] = 0x7C;  // canary
+    DVSTARTPROFILE();
+
+    _assert_( bpmem.genMode.numtexgens == xfregs.numTexGens);
+    _assert_( bpmem.genMode.numcolchans == xfregs.nNumChans);
+    
+    u32 lightMask = 0;
+    if (xfregs.nNumChans > 0)
+		lightMask |= xfregs.colChans[0].color.GetFullLightMask() | xfregs.colChans[0].alpha.GetFullLightMask();
+    if (xfregs.nNumChans > 1)
+        lightMask |= xfregs.colChans[1].color.GetFullLightMask() | xfregs.colChans[1].alpha.GetFullLightMask();
+
+    bool bOutputZ = bpmem.ztex2.op==ZTEXTURE_ADD || has_zbuffer_target;
+    int ztexcoord = -1;
+
+    char *p = text;
+    WRITE(p, "//Vertex Shader: comp:%x, \n", components);
+    WRITE(p, "typedef struct {\n"
+        "  float4 T0, T1, T2;\n"
+        "  float4 N0, N1, N2;\n"
+        "} s_"I_POSNORMALMATRIX";\n\n"
+        "typedef struct {\n"
+        "  float4 t;\n"
+        "} FLT4;\n"
+        "typedef struct {\n"
+        "  FLT4 T[24];\n"
+        "} s_"I_TEXMATRICES";\n\n"
+        "typedef struct {\n"
+        "  FLT4 T[64];\n"
+        "} s_"I_TRANSFORMMATRICES";\n\n"
+        "typedef struct {\n"
+        "  FLT4 T[32];\n"
+        "} s_"I_NORMALMATRICES";\n\n"
+        "typedef struct {\n"
+        "  FLT4 T[64];\n"
+        "} s_"I_POSTTRANSFORMMATRICES";\n\n"
+        "typedef struct {\n"
+        "  float4 col;\n"
+        "  float4 cosatt;\n"
+        "  float4 distatt;\n"
+        "  float4 pos;\n"
+        "  float4 dir;\n"
+        "} Light;\n\n"
+        "typedef struct {\n"
+        "  Light lights[8];\n"
+        "} s_"I_LIGHTS";\n\n"
+        "typedef struct {\n"
+        "  float4 C0, C1, C2, C3;\n"
+        "} s_"I_MATERIALS";\n\n"
+        "typedef struct {\n"
+        "  float4 T0,T1,T2,T3;\n"
+        "} s_"I_PROJECTION";\n"
+        "typedef struct {\n"
+        "  float4 params;\n" // a, b, c, b_shift
+        "} s_"I_FOGPARAMS";\n\n");
+
+    WRITE(p, "struct VS_OUTPUT {\n");
+    WRITE(p, "  float4 pos : POSITION;\n");
+    WRITE(p, "  float4 colors[2] : COLOR0;\n");
+
+    // if outputting Z, embed the Z coordinate in the w component of a texture coordinate
+    // if number of tex gens occupies all the texture coordinates, use the last tex coord
+    // otherwise use the next available tex coord
+    for (int i = 0; i < xfregs.numTexGens; ++i) {
+        WRITE(p, "  float%d tex%d : TEXCOORD%d;\n", (i==(xfregs.numTexGens-1)&&bOutputZ)?4:3, i, i);
+    }
+    if (bOutputZ && xfregs.numTexGens == 0) {
+        ztexcoord = 0;
+        WRITE(p, "  float4 tex%d : TEXCOORD%d;\n", ztexcoord, ztexcoord);
+    }
+    else if (bOutputZ)
+        ztexcoord = xfregs.numTexGens - 1;
+
+    WRITE(p, "};\n");
+    WRITE(p, "\n");
+
+    // uniforms
+    
+    // bool bTexMtx = ((components & VB_HAS_TEXMTXIDXALL)<<VB_HAS_UVTEXMTXSHIFT)!=0; unused TODO: keep?
+
+    WRITE(p, "uniform s_"I_TRANSFORMMATRICES" "I_TRANSFORMMATRICES" : register(c%d);\n", C_TRANSFORMMATRICES);
+    WRITE(p, "uniform s_"I_TEXMATRICES" "I_TEXMATRICES" : register(c%d);\n", C_TEXMATRICES); // also using tex matrices
+    WRITE(p, "uniform s_"I_NORMALMATRICES" "I_NORMALMATRICES" : register(c%d);\n", C_NORMALMATRICES);
+    WRITE(p, "uniform s_"I_POSNORMALMATRIX" "I_POSNORMALMATRIX" : register(c%d);\n", C_POSNORMALMATRIX);
+    WRITE(p, "uniform s_"I_POSTTRANSFORMMATRICES" "I_POSTTRANSFORMMATRICES" : register(c%d);\n", C_POSTTRANSFORMMATRICES);
+    WRITE(p, "uniform s_"I_LIGHTS" "I_LIGHTS" : register(c%d);\n", C_LIGHTS);
+    WRITE(p, "uniform s_"I_MATERIALS" "I_MATERIALS" : register(c%d);\n", C_MATERIALS);
+    WRITE(p, "uniform s_"I_PROJECTION" "I_PROJECTION" : register(c%d);\n", C_PROJECTION);
+    WRITE(p, "uniform s_"I_FOGPARAMS" "I_FOGPARAMS" : register(c%d);\n", C_FOGPARAMS);
+
+    WRITE(p, "VS_OUTPUT main(\n");
+    
+    // inputs
+    if (components & VB_HAS_NRM0)
+        WRITE(p, "  float3 rawnorm0 : NORMAL,\n");
+    if (components & VB_HAS_NRM1)
+        WRITE(p, "  float3 rawnorm1 : ATTR%d,\n", SHADER_NORM1_ATTRIB);
+    if (components & VB_HAS_NRM2)
+        WRITE(p, "  float3 rawnorm2 : ATTR%d,\n", SHADER_NORM2_ATTRIB);
+    if (components & VB_HAS_COL0)
+        WRITE(p, "  float4 color0 : COLOR0,\n");
+    if (components & VB_HAS_COL1)
+        WRITE(p, "  float4 color1 : COLOR1,\n");
+    for (int i = 0; i < 8; ++i) {
+        u32 hastexmtx = (components & (VB_HAS_TEXMTXIDX0<<i));
+        if ((components & (VB_HAS_UV0<<i)) || hastexmtx )
+            WRITE(p, "  float%d tex%d : TEXCOORD%d,\n", hastexmtx ? 3 : 2, i,i);
+    }
+    if (components & VB_HAS_POSMTXIDX)
+        WRITE(p, "  half posmtx : ATTR%d,\n", SHADER_POSMTX_ATTRIB);
+
+    WRITE(p, "  float4 rawpos : POSITION) {\n");
+    WRITE(p, "VS_OUTPUT o;\n");
+
+    // transforms
+    if (components & VB_HAS_POSMTXIDX) {
+        WRITE(p, "float4 pos = float4(dot("I_TRANSFORMMATRICES".T[posmtx].t, rawpos), dot("I_TRANSFORMMATRICES".T[posmtx+1].t, rawpos), dot("I_TRANSFORMMATRICES".T[posmtx+2].t, rawpos),1);\n");
+        
+        if (components & VB_HAS_NRMALL) {
+            WRITE(p, "int normidx = posmtx >= 32 ? (posmtx-32) : posmtx;\n");
+            WRITE(p, "float3 N0 = "I_NORMALMATRICES".T[normidx].t.xyz, N1 = "I_NORMALMATRICES".T[normidx+1].t.xyz, N2 = "I_NORMALMATRICES".T[normidx+2].t.xyz;\n");
+        }
+
+        if (components & VB_HAS_NRM0)
+            WRITE(p, "half3 _norm0 = half3(dot(N0, rawnorm0), dot(N1, rawnorm0), dot(N2, rawnorm0));\n"
+                     "half3 norm0 = normalize(_norm0);\n");
+        if (components & VB_HAS_NRM1)
+            WRITE(p, "half3 _norm1 = half3(dot(N0, rawnorm1), dot(N1, rawnorm1), dot(N2, rawnorm1));\n");
+                    //"half3 norm1 = normalize(_norm1);\n");
+        if (components & VB_HAS_NRM2)
+            WRITE(p, "half3 _norm2 = half3(dot(N0, rawnorm2), dot(N1, rawnorm2), dot(N2, rawnorm2));\n");
+                    //"half3 norm2 = normalize(_norm2);\n");
+    }
+    else {
+        WRITE(p, "float4 pos = float4(dot("I_POSNORMALMATRIX".T0, rawpos), dot("I_POSNORMALMATRIX".T1, rawpos), dot("I_POSNORMALMATRIX".T2, rawpos), 1);\n");
+        if (components & VB_HAS_NRM0)
+            WRITE(p, "half3 _norm0 = half3(dot("I_POSNORMALMATRIX".N0.xyz, rawnorm0), dot("I_POSNORMALMATRIX".N1.xyz, rawnorm0), dot("I_POSNORMALMATRIX".N2.xyz, rawnorm0));\n"
+                     "half3 norm0 = normalize(_norm0);\n");
+        if (components & VB_HAS_NRM1)
+            WRITE(p, "half3 _norm1 = half3(dot("I_POSNORMALMATRIX".N0.xyz, rawnorm1), dot("I_POSNORMALMATRIX".N1.xyz, rawnorm1), dot("I_POSNORMALMATRIX".N2.xyz, rawnorm1));\n");
+                    //"half3 norm1 = normalize(_norm1);\n");
+        if (components & VB_HAS_NRM2)
+            WRITE(p, "half3 _norm2 = half3(dot("I_POSNORMALMATRIX".N0.xyz, rawnorm2), dot("I_POSNORMALMATRIX".N1.xyz, rawnorm2), dot("I_POSNORMALMATRIX".N2.xyz, rawnorm2));\n");
+                    //"half3 norm2 = normalize(_norm2);\n");
+    }
+
+    if (!(components & VB_HAS_NRM0))
+        WRITE(p, "half3 _norm0 = half3(0,0,0), norm0= half3(0,0,0);\n");
+
+    WRITE(p, "o.pos = float4(dot("I_PROJECTION".T0, pos), dot("I_PROJECTION".T1, pos), dot("I_PROJECTION".T2, pos), dot("I_PROJECTION".T3, pos));\n");
+
+    WRITE(p, "half4 mat, lacc;\n"
+    "half3 ldir, h;\n"
+    "half dist, dist2, attn;\n");
+
+    // lights/colors
+    for (int j = 0; j < xfregs.nNumChans; j++) {
+
+        // bool bColorAlphaSame = xfregs.colChans[j].color.hex == xfregs.colChans[j].alpha.hex;  unused
+        const LitChannel& color = xfregs.colChans[j].color;
+        const LitChannel& alpha = xfregs.colChans[j].alpha;
+
+        WRITE(p, "{\n");
+
+        if (color.matsource) {// from vertex
+            if (components & (VB_HAS_COL0<<j) )
+                WRITE(p, "mat = color%d;\n", j);
+            else WRITE(p, "mat = half4(1,1,1,1);\n");
+        }
+        else // from color
+            WRITE(p, "mat = "I_MATERIALS".C%d;\n", j+2);
+
+        if (color.enablelighting) {
+            if (color.ambsource) {// from vertex
+                if (components & (VB_HAS_COL0<<j) )
+                    WRITE(p, "lacc = color%d;\n", j);
+                else WRITE(p, "lacc = half4(0.0f,0.0f,0.0f,0.0f);\n");
+            }
+            else // from color
+                WRITE(p, "lacc = "I_MATERIALS".C%d;\n", j);
+        }
+
+        // check if alpha is different
+        if (alpha.matsource != color.matsource) {
+            if (alpha.matsource) {// from vertex
+                if (components & (VB_HAS_COL0<<j) )
+                    WRITE(p, "mat.w = color%d.w;\n", j);
+                else WRITE(p, "mat.w = 1;\n");
+            }
+            else // from color
+                WRITE(p, "mat.w = "I_MATERIALS".C%d.w;\n", j+2);
+        }
+
+        if (alpha.enablelighting && alpha.ambsource != color.ambsource) {
+            if (alpha.ambsource) {// from vertex
+                if (components & (VB_HAS_COL0<<j) )
+                    WRITE(p, "lacc.w = color%d.w;\n", j);
+                else WRITE(p, "lacc.w = 0;\n");
+            }
+            else // from color
+                WRITE(p, "lacc.w = "I_MATERIALS".C%d.w;\n", j);
+        }
+        
+        if (color.enablelighting && alpha.enablelighting && (color.GetFullLightMask() != alpha.GetFullLightMask() || color.lightparams != alpha.lightparams)) {
+
+            // both have lighting, except not using the same lights
+            int mask = 0; // holds already computed lights
+
+            if (color.lightparams == alpha.lightparams && (color.GetFullLightMask() & alpha.GetFullLightMask())) {
+                // if lights are shared, compute those first
+                mask = color.GetFullLightMask() & alpha.GetFullLightMask();
+                for (int i = 0; i < 8; ++i) {
+                    if (mask&(1<<i))
+                        p = GenerateLightShader(p, i, color, "lacc", 3);
+                }
+            }
+
+            // no shared lights
+            for (int i = 0; i < 8; ++i) {
+                if (!(mask&(1<<i)) && (color.GetFullLightMask() & (1<<i)) )
+                    p = GenerateLightShader(p, i, color, "lacc", 1);
+                if (!(mask&(1<<i)) && (alpha.GetFullLightMask() & (1<<i)) )
+                    p = GenerateLightShader(p, i, alpha, "lacc", 2);
+            }
+        }
+        else if (color.enablelighting || alpha.enablelighting) {
+            // either one is enabled
+            int coloralpha = (int)color.enablelighting|((int)alpha.enablelighting<<1);
+            for (int i = 0; i < 8; ++i) {
+                if (color.GetFullLightMask() & (1<<i) )
+                    p = GenerateLightShader(p, i, color.enablelighting?color:alpha, "lacc", coloralpha);
+            }
+        }
+
+        if (color.enablelighting != alpha.enablelighting) {
+            if (color.enablelighting )
+                WRITE(p, "o.colors[%d].xyz = mat.xyz * clamp(lacc.xyz,float3(0.0f,0.0f,0.0f),float3(1.0f,1.0f,1.0f));\n"
+                    "o.colors[%d].w = mat.w;\n", j, j);
+            else
+                WRITE(p, "o.colors[%d].xyz = mat.xyz;\n"
+                    "o.colors[%d].w = mat.w  * clamp(lacc.w,0.0f,1.0f);\n", j, j);
+        }
+        else {
+            if (alpha.enablelighting )
+                WRITE(p, "o.colors[%d] = mat * clamp(lacc,float4(0.0f,0.0f,0.0f,0.0f), float4(1.0f,1.0f,1.0f,1.0f));\n", j);
+            else WRITE(p, "o.colors[%d] = mat;\n", j);
+        }
+        WRITE(p, "}\n");
+    }
+
+    // zero left over channels
+    for (int i = xfregs.nNumChans; i < 2; ++i)
+		WRITE(p, "o.colors[%d] = 0;\n", i);
+
+    // transform texcoords
+    for (int i = 0; i < xfregs.numTexGens; ++i) {
+        TexMtxInfo& texinfo = xfregs.texcoords[i].texmtxinfo;
+
+        WRITE(p, "{\n");
+        switch (texinfo.sourcerow) {
+        case XF_SRCGEOM_INROW:
+            _assert_( texinfo.inputform == XF_TEXINPUT_ABC1 );
+            WRITE(p, "float4 coord = rawpos;\n"); // pos.w is 1
+            break;
+        case XF_SRCNORMAL_INROW:
+            if (components & VB_HAS_NRM0) {
+                _assert_( texinfo.inputform == XF_TEXINPUT_ABC1 );
+                WRITE(p, "float4 coord = float4(rawnorm0.xyz, 1.0);\n");
+            }
+            else WRITE(p, "float4 coord = 0;\n");
+            break;
+        case XF_SRCCOLORS_INROW:
+            _assert_( texinfo.texgentype == XF_TEXGEN_COLOR_STRGBC0 || texinfo.texgentype == XF_TEXGEN_COLOR_STRGBC1 );
+            break;
+        case XF_SRCBINORMAL_T_INROW:
+            if (components & VB_HAS_NRM1) {
+                _assert_( texinfo.inputform == XF_TEXINPUT_ABC1 );
+                WRITE(p, "float4 coord = float4(rawnorm1.xyz, 1.0);\n");
+            }
+            else WRITE(p, "float4 coord = 0;\n");
+            break;
+        case XF_SRCBINORMAL_B_INROW:
+            if (components & VB_HAS_NRM2) {
+                _assert_( texinfo.inputform == XF_TEXINPUT_ABC1 );
+                WRITE(p, "float4 coord = float4(rawnorm2.xyz, 1.0);\n");
+            }
+            else WRITE(p, "float4 coord = 0;\n");
+            break;
+        default:
+            _assert_(texinfo.sourcerow <= XF_SRCTEX7_INROW);
+            if (components & (VB_HAS_UV0<<(texinfo.sourcerow - XF_SRCTEX0_INROW)) )
+                WRITE(p, "float4 coord = float4(tex%d.x, tex%d.y, 1.0f, 1.0f);\n", texinfo.sourcerow - XF_SRCTEX0_INROW, texinfo.sourcerow - XF_SRCTEX0_INROW);
+            else
+                WRITE(p, "float4 coord = float4(0.0f, 0.0f, 1.0f, 1.0f);\n");
+            break;
+        }
+
+        // firs transformation
+        switch (texinfo.texgentype) {
+            case XF_TEXGEN_REGULAR:
+                if (components & (VB_HAS_TEXMTXIDX0<<i)) {
+                    if (texinfo.projection == XF_TEXPROJ_STQ )
+                        WRITE(p, "o.tex%d.xyz = float3(dot(coord, "I_TRANSFORMMATRICES".T[tex%d.z].t), dot(coord, "I_TRANSFORMMATRICES".T[tex%d.z+1].t), dot(coord, "I_TRANSFORMMATRICES".T[tex%d.z+2].t));\n", i, i, i, i);
+                    else {
+                        WRITE(p, "o.tex%d.xyz = float3(dot(coord, "I_TRANSFORMMATRICES".T[tex%d.z].t), dot(coord, "I_TRANSFORMMATRICES".T[tex%d.z+1].t), 1);\n", i, i, i);
+                    }
+                }
+                else {
+                    if (texinfo.projection == XF_TEXPROJ_STQ )
+                        WRITE(p, "o.tex%d.xyz = float3(dot(coord, "I_TEXMATRICES".T[%d].t), dot(coord, "I_TEXMATRICES".T[%d].t), dot(coord, "I_TEXMATRICES".T[%d].t));\n", i, 3*i, 3*i+1, 3*i+2);
+                    else
+                        WRITE(p, "o.tex%d.xyz = float3(dot(coord, "I_TEXMATRICES".T[%d].t), dot(coord, "I_TEXMATRICES".T[%d].t), 1);\n", i, 3*i, 3*i+1);
+                }
+                break;
+            case XF_TEXGEN_EMBOSS_MAP: // calculate tex coords into bump map
+
+                if (components & (VB_HAS_NRM1|VB_HAS_NRM2)) {
+                    // transform the light dir into tangent space
+                    WRITE(p, "ldir = normalize("I_LIGHTS".lights[%d].pos.xyz - pos.xyz);\n", texinfo.embosslightshift); 
+                    WRITE(p, "o.tex%d.xyz = o.tex%d.xyz + float3(dot(ldir, _norm1), dot(ldir, _norm2), 0.0f);\n", i, texinfo.embosssourceshift);
+                }
+                else _assert_(0); // should have normals
+
+                break;
+            case XF_TEXGEN_COLOR_STRGBC0:
+                _assert_(texinfo.sourcerow == XF_SRCCOLORS_INROW);
+                WRITE(p, "o.tex%d.xyz = float3(o.colors[0].x, o.colors[0].y, 1);\n", i);
+                break;
+            case XF_TEXGEN_COLOR_STRGBC1:
+                _assert_(texinfo.sourcerow == XF_SRCCOLORS_INROW);
+                WRITE(p, "o.tex%d.xyz = float3(o.colors[1].x, o.colors[1].y, 1);\n", i);
+                break;
+        }
+
+        if(xfregs.bEnableDualTexTransform && texinfo.texgentype == XF_TEXGEN_REGULAR) { // only works for regular tex gen types?
+            if (xfregs.texcoords[i].postmtxinfo.normalize)
+                WRITE(p, "o.tex%d.xyz = normalize(o.tex%d.xyz);\n", i, i);
+
+            //multiply by postmatrix
+            int postidx = xfregs.texcoords[i].postmtxinfo.index;
+            WRITE(p, "float4 P0 = "I_POSTTRANSFORMMATRICES".T[%d].t;\n"
+                "float4 P1 = "I_POSTTRANSFORMMATRICES".T[%d].t;\n"
+                "float4 P2 = "I_POSTTRANSFORMMATRICES".T[%d].t;\n",
+                postidx&0x3f, (postidx+1)&0x3f, (postidx+2)&0x3f);
+            WRITE(p, "o.tex%d.xyz = float3(dot(P0.xyz, o.tex%d.xyz) + P0.w, dot(P1.xyz, o.tex%d.xyz) + P1.w, dot(P2.xyz, o.tex%d.xyz) + P2.w);\n", i, i, i, i);
+        }
+
+        WRITE(p, "}\n");
+    }
+
+    if (ztexcoord >= 0 )
+        WRITE(p, "o.tex%d.w = o.pos.z/o.pos.w;\n", ztexcoord);
+
+//    if (bpmem.fog.c_proj_fsel.fsel != 0) {
+//        switch (bpmem.fog.c_proj_fsel.fsel) {
+//            case 1: // linear
+//                break;
+//            case 4: // exp
+//                break;
+//            case 5: // exp2
+//                break;
+//            case 6: // backward exp
+//                break;
+//            case 7: // backward exp2
+//                break;
+//        }
+//
+//        WRITE(p, "o.fog = o.pos.z/o.pos.w;\n");
+//    }
+
+    WRITE(p, "return o;\n}\n");
+
+	if (text[sizeof(text) - 1] != 0x7C)
+		PanicAlert("VertexShader generator - buffer too small, canary has been eaten!");
+    return text;
+}
+
+// coloralpha - 1 if color, 2 if alpha
+char* GenerateLightShader(char* p, int index, const LitChannel& chan, const char* dest, int coloralpha)
+{
+    const char* swizzle = "xyzw";
+    if (coloralpha == 1 ) swizzle = "xyz";
+    else if (coloralpha == 2 ) swizzle = "w";
+
+    if (!(chan.attnfunc & 1)) {
+        // atten disabled
+        switch (chan.diffusefunc) {
+            case LIGHTDIF_NONE:
+                WRITE(p, "%s.%s += "I_LIGHTS".lights[%d].col.%s;\n", dest, swizzle, index, swizzle);
+                break;
+            case LIGHTDIF_SIGN:
+            case LIGHTDIF_CLAMP:
+                WRITE(p, "ldir = normalize("I_LIGHTS".lights[%d].pos.xyz - pos.xyz);\n", index);
+                WRITE(p, "%s.%s += %sdot(ldir, norm0)) * "I_LIGHTS".lights[%d].col.%s;\n",
+                    dest, swizzle, chan.diffusefunc != LIGHTDIF_SIGN ? "max(0.0f," :"(", index, swizzle);
+                break;
+            default: _assert_(0);
+        }
+    }
+    else { // spec and spot
+        WRITE(p, "ldir = "I_LIGHTS".lights[%d].pos.xyz - pos.xyz;\n", index);
+
+        if (chan.attnfunc == 3) { // spot
+            WRITE(p, "dist2 = dot(ldir, ldir);\n"
+                "dist = sqrt(dist2);\n"
+                "ldir = ldir / dist;\n"
+                "attn = max(0.0f, dot(ldir, "I_LIGHTS".lights[%d].dir.xyz));\n",index);
+            WRITE(p, "attn = max(0.0f, dot("I_LIGHTS".lights[%d].cosatt.xyz, half3(1, attn, attn*attn))) / dot("I_LIGHTS".lights[%d].distatt.xyz, half3(1,dist,dist2));\n", index, index);
+        }
+        else if (chan.attnfunc == 1) { // specular
+            WRITE(p, "attn = dot(norm0, "I_LIGHTS".lights[%d].pos.xyz) > 0 ? max(0.0f, dot(norm0, "I_LIGHTS".lights[%d].dir.xyz)) : 0;\n", index, index);
+            WRITE(p, "ldir = half3(1,attn,attn*attn);\n");
+            WRITE(p, "attn = max(0.0f, dot("I_LIGHTS".lights[%d].cosatt.xyz, ldir)) / dot("I_LIGHTS".lights[%d].distatt.xyz, ldir);\n", index, index);
+        }
+
+        switch (chan.diffusefunc) {
+            case LIGHTDIF_NONE:
+                WRITE(p, "%s.%s += attn * "I_LIGHTS".lights[%d].col.%s;\n", dest, swizzle, index, swizzle);
+                break;
+            case LIGHTDIF_SIGN:
+            case LIGHTDIF_CLAMP:
+                WRITE(p, "%s.%s += attn * %sdot(ldir, norm0)) * "I_LIGHTS".lights[%d].col.%s;\n",
+                    dest, swizzle, chan.diffusefunc != LIGHTDIF_SIGN ? "max(0.0f," :"(", index, swizzle);
+                break;
+            default: _assert_(0);
+        }
+    }
+    WRITE(p, "\n");
+
+    return p;
+}
diff --git a/Source/Core/VideoCommon/Src/VideoState.cpp b/Source/Core/VideoCommon/Src/VideoState.cpp
index f1149be7f3..66f3611b9c 100644
--- a/Source/Core/VideoCommon/Src/VideoState.cpp
+++ b/Source/Core/VideoCommon/Src/VideoState.cpp
@@ -1,51 +1,51 @@
-// Copyright (C) 2003-2008 Dolphin Project.

-

-// This program is free software: you can redistribute it and/or modify

-// it under the terms of the GNU General Public License as published by

-// the Free Software Foundation, version 2.0.

-

-// This program is distributed in the hope that it will be useful,

-// but WITHOUT ANY WARRANTY; without even the implied warranty of

-// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

-// GNU General Public License 2.0 for more details.

-

-// A copy of the GPL 2.0 should have been included with the program.

-// If not, see http://www.gnu.org/licenses/

-

-// Official SVN repository and contact information can be found at

-// http://code.google.com/p/dolphin-emu/

-

-#include "VideoState.h"

-

-#include "BPMemory.h"

-#include "CPMemory.h"

-#include "XFMemory.h"

-#include "TextureDecoder.h"

-#include "Fifo.h"

-

-static void DoState(PointerWrap &p)

-{

-    // BP Memory

-    p.Do(bpmem);

-    // CP Memory

-    p.DoArray(arraybases, 16);

-    p.DoArray(arraystrides, 16);

-    p.Do(MatrixIndexA);

-    p.Do(MatrixIndexB);

-    p.Do(g_VtxDesc.Hex);

-	p.DoArray(g_VtxAttr, 8);

-

-    // XF Memory

-    p.Do(xfregs);

-    p.DoArray(xfmem, XFMEM_SIZE);

-    // Texture decoder

-    p.DoArray(texMem, TMEM_SIZE);

- 

-    // FIFO

-    Fifo_DoState(p);

-}

-

-void VideoCommon_DoState(PointerWrap &p) {

-    DoState(p);

-	//TODO: search for more data that should be saved and add it here

-}

+// Copyright (C) 2003-2008 Dolphin Project.
+
+// This program is free software: you can redistribute it and/or modify
+// it under the terms of the GNU General Public License as published by
+// the Free Software Foundation, version 2.0.
+
+// This program is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU General Public License 2.0 for more details.
+
+// A copy of the GPL 2.0 should have been included with the program.
+// If not, see http://www.gnu.org/licenses/
+
+// Official SVN repository and contact information can be found at
+// http://code.google.com/p/dolphin-emu/
+
+#include "VideoState.h"
+
+#include "BPMemory.h"
+#include "CPMemory.h"
+#include "XFMemory.h"
+#include "TextureDecoder.h"
+#include "Fifo.h"
+
+static void DoState(PointerWrap &p)
+{
+    // BP Memory
+    p.Do(bpmem);
+    // CP Memory
+    p.DoArray(arraybases, 16);
+    p.DoArray(arraystrides, 16);
+    p.Do(MatrixIndexA);
+    p.Do(MatrixIndexB);
+    p.Do(g_VtxDesc.Hex);
+	p.DoArray(g_VtxAttr, 8);
+
+    // XF Memory
+    p.Do(xfregs);
+    p.DoArray(xfmem, XFMEM_SIZE);
+    // Texture decoder
+    p.DoArray(texMem, TMEM_SIZE);
+ 
+    // FIFO
+    Fifo_DoState(p);
+}
+
+void VideoCommon_DoState(PointerWrap &p) {
+    DoState(p);
+	//TODO: search for more data that should be saved and add it here
+}
diff --git a/Source/Core/VideoCommon/Src/XFBConvert.cpp b/Source/Core/VideoCommon/Src/XFBConvert.cpp
index 3f20ab421d..2aff13c15e 100644
--- a/Source/Core/VideoCommon/Src/XFBConvert.cpp
+++ b/Source/Core/VideoCommon/Src/XFBConvert.cpp
@@ -1,131 +1,131 @@
-// Copyright (C) 2003-2008 Dolphin Project.

-

-// This program is free software: you can redistribute it and/or modify

-// it under the terms of the GNU General Public License as published by

-// the Free Software Foundation, version 2.0.

-

-// This program is distributed in the hope that it will be useful,

-// but WITHOUT ANY WARRANTY; without even the implied warranty of

-// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

-// GNU General Public License 2.0 for more details.

-

-// A copy of the GPL 2.0 should have been included with the program.

-// If not, see http://www.gnu.org/licenses/

-

-// Official SVN repository and contact information can be found at

-// http://code.google.com/p/dolphin-emu/

-

-#if _WIN32

-#include <intrin.h>

-#endif

-

-#include <xmmintrin.h>

-

-#include "XFBConvert.h"

-#include "Common.h"

-

-namespace {

-

-const __m128i _bias1 = _mm_set_epi32(128/2 << 16,        0, 128/2 << 16, 16 << 16);

-const __m128i _bias2 = _mm_set_epi32(128/2 << 16, 16 << 16, 128/2 << 16,        0);

-

-__m128i _y[256];

-__m128i _u[256];

-__m128i _v[256];

-

-__m128i _r1[256];

-__m128i _r2[256];

-__m128i _g1[256];

-__m128i _g2[256];

-__m128i _b1[256];

-__m128i _b2[256];

-

-}  // namespace

-

-void InitXFBConvTables() 

-{

-	for (int i = 0; i < 256; i++)

-	{

-		_y[i] = _mm_set_epi32(0xFFFFFFF,     76283*(i - 16),      76283*(i - 16),     76283*(i - 16));

-		_u[i] = _mm_set_epi32(        0,                  0,  -25624 * (i - 128), 132252 * (i - 128));

-		_v[i] = _mm_set_epi32(        0, 104595 * (i - 128),  -53281 * (i - 128),                  0);

-

-		_r1[i] = _mm_add_epi32(_mm_set_epi32( 28770 * i / 2,          0,  -9699 * i / 2, 16843 * i),

-							   _bias1);

-		_g1[i] = _mm_set_epi32(-24117 * i / 2,          0, -19071 * i / 2, 33030 * i);

-		_b1[i] = _mm_set_epi32( -4653 * i / 2,          0,  28770 * i / 2,  6423 * i);

-										   									

-		_r2[i] = _mm_add_epi32(_mm_set_epi32( 28770 * i / 2,  16843 * i,  -9699 * i / 2,         0),

-							   _bias2);

-		_g2[i] = _mm_set_epi32(-24117 * i / 2,  33030 * i, -19071 * i / 2,         0);

-		_b2[i] = _mm_set_epi32( -4653 * i / 2,   6423 * i,  28770 * i / 2,         0);

-	}

-}

-

-void ConvertFromXFB(u32 *dst, const u8* _pXFB, int width, int height)

-{

-	if (((size_t)dst & 0xF) != 0) {

-		PanicAlert("ConvertFromXFB - unaligned destination");

-	}

-	const unsigned char *src = _pXFB;

-	u32 numBlocks = ((width * height) / 2) / 2;

-	for (u32 i = 0; i < numBlocks; i++)

-	{

-		__m128i y1 = _y[src[0]];

-		__m128i u  = _u[src[1]];

-		__m128i y2 = _y[src[2]];

-		__m128i v  = _v[src[3]];

-		__m128i y1_2 = _y[src[4+0]];

-		__m128i u_2  = _u[src[4+1]];

-		__m128i y2_2 = _y[src[4+2]];

-		__m128i v_2  = _v[src[4+3]];

-

-		__m128i c1 = _mm_srai_epi32(_mm_add_epi32(y1, _mm_add_epi32(u, v)), 16);

-		__m128i c2 = _mm_srai_epi32(_mm_add_epi32(y2, _mm_add_epi32(u, v)), 16);

-		__m128i c3 = _mm_srai_epi32(_mm_add_epi32(y1_2, _mm_add_epi32(u_2, v_2)), 16);

-		__m128i c4 = _mm_srai_epi32(_mm_add_epi32(y2_2, _mm_add_epi32(u_2, v_2)), 16);

-

-		__m128i four_dest = _mm_packus_epi16(_mm_packs_epi32(c1, c2), _mm_packs_epi32(c3, c4));

-		_mm_store_si128((__m128i *)dst, four_dest);

-		dst += 4;

-		src += 8;

-	}

-}

-

-void ConvertToXFB(u32 *dst, const u8* _pEFB, int width, int height)

-{

-	const unsigned char *src = _pEFB;

-

-	u32 numBlocks = ((width * height) / 2) / 4;

-	if (((size_t)dst & 0xF) != 0) {

-		PanicAlert("ConvertToXFB - unaligned XFB");

-	}

-

-	for (u32 i = 0; i < numBlocks; i++)

-	{

-		__m128i yuyv0 = _mm_srai_epi32(

-			_mm_add_epi32(

-				_mm_add_epi32(_r1[src[0]], _mm_add_epi32(_g1[src[1]], _b1[src[2]])),

-				_mm_add_epi32(_r2[src[4]], _mm_add_epi32(_g2[src[5]], _b2[src[6]]))), 16);

-		src += 8;

-		__m128i yuyv1 = _mm_srai_epi32(

-			_mm_add_epi32(

-				_mm_add_epi32(_r1[src[0]], _mm_add_epi32(_g1[src[1]], _b1[src[2]])),

-				_mm_add_epi32(_r2[src[4]], _mm_add_epi32(_g2[src[5]], _b2[src[6]]))), 16);

-		src += 8;

-		__m128i yuyv2 = _mm_srai_epi32(

-			_mm_add_epi32(

-				_mm_add_epi32(_r1[src[0]], _mm_add_epi32(_g1[src[1]], _b1[src[2]])),

-				_mm_add_epi32(_r2[src[4]], _mm_add_epi32(_g2[src[5]], _b2[src[6]]))), 16);

-		src += 8;

-		__m128i yuyv3 = _mm_srai_epi32(

-			_mm_add_epi32(

-				_mm_add_epi32(_r1[src[0]], _mm_add_epi32(_g1[src[1]], _b1[src[2]])),

-				_mm_add_epi32(_r2[src[4]], _mm_add_epi32(_g2[src[5]], _b2[src[6]]))), 16);

-		src += 8;

-		__m128i four_dest = _mm_packus_epi16(_mm_packs_epi32(yuyv0, yuyv1), _mm_packs_epi32(yuyv2, yuyv3));

-		_mm_store_si128((__m128i *)dst, four_dest);

-		dst += 4;

-	}

-}

-

+// Copyright (C) 2003-2008 Dolphin Project.
+
+// This program is free software: you can redistribute it and/or modify
+// it under the terms of the GNU General Public License as published by
+// the Free Software Foundation, version 2.0.
+
+// This program is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU General Public License 2.0 for more details.
+
+// A copy of the GPL 2.0 should have been included with the program.
+// If not, see http://www.gnu.org/licenses/
+
+// Official SVN repository and contact information can be found at
+// http://code.google.com/p/dolphin-emu/
+
+#if _WIN32
+#include <intrin.h>
+#endif
+
+#include <xmmintrin.h>
+
+#include "XFBConvert.h"
+#include "Common.h"
+
+namespace {
+
+const __m128i _bias1 = _mm_set_epi32(128/2 << 16,        0, 128/2 << 16, 16 << 16);
+const __m128i _bias2 = _mm_set_epi32(128/2 << 16, 16 << 16, 128/2 << 16,        0);
+
+__m128i _y[256];
+__m128i _u[256];
+__m128i _v[256];
+
+__m128i _r1[256];
+__m128i _r2[256];
+__m128i _g1[256];
+__m128i _g2[256];
+__m128i _b1[256];
+__m128i _b2[256];
+
+}  // namespace
+
+void InitXFBConvTables() 
+{
+	for (int i = 0; i < 256; i++)
+	{
+		_y[i] = _mm_set_epi32(0xFFFFFFF,     76283*(i - 16),      76283*(i - 16),     76283*(i - 16));
+		_u[i] = _mm_set_epi32(        0,                  0,  -25624 * (i - 128), 132252 * (i - 128));
+		_v[i] = _mm_set_epi32(        0, 104595 * (i - 128),  -53281 * (i - 128),                  0);
+
+		_r1[i] = _mm_add_epi32(_mm_set_epi32( 28770 * i / 2,          0,  -9699 * i / 2, 16843 * i),
+							   _bias1);
+		_g1[i] = _mm_set_epi32(-24117 * i / 2,          0, -19071 * i / 2, 33030 * i);
+		_b1[i] = _mm_set_epi32( -4653 * i / 2,          0,  28770 * i / 2,  6423 * i);
+										   									
+		_r2[i] = _mm_add_epi32(_mm_set_epi32( 28770 * i / 2,  16843 * i,  -9699 * i / 2,         0),
+							   _bias2);
+		_g2[i] = _mm_set_epi32(-24117 * i / 2,  33030 * i, -19071 * i / 2,         0);
+		_b2[i] = _mm_set_epi32( -4653 * i / 2,   6423 * i,  28770 * i / 2,         0);
+	}
+}
+
+void ConvertFromXFB(u32 *dst, const u8* _pXFB, int width, int height)
+{
+	if (((size_t)dst & 0xF) != 0) {
+		PanicAlert("ConvertFromXFB - unaligned destination");
+	}
+	const unsigned char *src = _pXFB;
+	u32 numBlocks = ((width * height) / 2) / 2;
+	for (u32 i = 0; i < numBlocks; i++)
+	{
+		__m128i y1 = _y[src[0]];
+		__m128i u  = _u[src[1]];
+		__m128i y2 = _y[src[2]];
+		__m128i v  = _v[src[3]];
+		__m128i y1_2 = _y[src[4+0]];
+		__m128i u_2  = _u[src[4+1]];
+		__m128i y2_2 = _y[src[4+2]];
+		__m128i v_2  = _v[src[4+3]];
+
+		__m128i c1 = _mm_srai_epi32(_mm_add_epi32(y1, _mm_add_epi32(u, v)), 16);
+		__m128i c2 = _mm_srai_epi32(_mm_add_epi32(y2, _mm_add_epi32(u, v)), 16);
+		__m128i c3 = _mm_srai_epi32(_mm_add_epi32(y1_2, _mm_add_epi32(u_2, v_2)), 16);
+		__m128i c4 = _mm_srai_epi32(_mm_add_epi32(y2_2, _mm_add_epi32(u_2, v_2)), 16);
+
+		__m128i four_dest = _mm_packus_epi16(_mm_packs_epi32(c1, c2), _mm_packs_epi32(c3, c4));
+		_mm_store_si128((__m128i *)dst, four_dest);
+		dst += 4;
+		src += 8;
+	}
+}
+
+void ConvertToXFB(u32 *dst, const u8* _pEFB, int width, int height)
+{
+	const unsigned char *src = _pEFB;
+
+	u32 numBlocks = ((width * height) / 2) / 4;
+	if (((size_t)dst & 0xF) != 0) {
+		PanicAlert("ConvertToXFB - unaligned XFB");
+	}
+
+	for (u32 i = 0; i < numBlocks; i++)
+	{
+		__m128i yuyv0 = _mm_srai_epi32(
+			_mm_add_epi32(
+				_mm_add_epi32(_r1[src[0]], _mm_add_epi32(_g1[src[1]], _b1[src[2]])),
+				_mm_add_epi32(_r2[src[4]], _mm_add_epi32(_g2[src[5]], _b2[src[6]]))), 16);
+		src += 8;
+		__m128i yuyv1 = _mm_srai_epi32(
+			_mm_add_epi32(
+				_mm_add_epi32(_r1[src[0]], _mm_add_epi32(_g1[src[1]], _b1[src[2]])),
+				_mm_add_epi32(_r2[src[4]], _mm_add_epi32(_g2[src[5]], _b2[src[6]]))), 16);
+		src += 8;
+		__m128i yuyv2 = _mm_srai_epi32(
+			_mm_add_epi32(
+				_mm_add_epi32(_r1[src[0]], _mm_add_epi32(_g1[src[1]], _b1[src[2]])),
+				_mm_add_epi32(_r2[src[4]], _mm_add_epi32(_g2[src[5]], _b2[src[6]]))), 16);
+		src += 8;
+		__m128i yuyv3 = _mm_srai_epi32(
+			_mm_add_epi32(
+				_mm_add_epi32(_r1[src[0]], _mm_add_epi32(_g1[src[1]], _b1[src[2]])),
+				_mm_add_epi32(_r2[src[4]], _mm_add_epi32(_g2[src[5]], _b2[src[6]]))), 16);
+		src += 8;
+		__m128i four_dest = _mm_packus_epi16(_mm_packs_epi32(yuyv0, yuyv1), _mm_packs_epi32(yuyv2, yuyv3));
+		_mm_store_si128((__m128i *)dst, four_dest);
+		dst += 4;
+	}
+}
+
diff --git a/Source/Core/VideoCommon/Src/XFMemory.cpp b/Source/Core/VideoCommon/Src/XFMemory.cpp
index 44000613c6..60c9c8811f 100644
--- a/Source/Core/VideoCommon/Src/XFMemory.cpp
+++ b/Source/Core/VideoCommon/Src/XFMemory.cpp
@@ -1,22 +1,22 @@
-// Copyright (C) 2003-2008 Dolphin Project.

-

-// This program is free software: you can redistribute it and/or modify

-// it under the terms of the GNU General Public License as published by

-// the Free Software Foundation, version 2.0.

-

-// This program is distributed in the hope that it will be useful,

-// but WITHOUT ANY WARRANTY; without even the implied warranty of

-// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the

-// GNU General Public License 2.0 for more details.

-

-// A copy of the GPL 2.0 should have been included with the program.

-// If not, see http://www.gnu.org/licenses/

-

-// Official SVN repository and contact information can be found at

-// http://code.google.com/p/dolphin-emu/

-

-#include "XFMemory.h"

-

-// STATE_TO_SAVE

-XFRegisters xfregs;

-u32 xfmem[XFMEM_SIZE];

+// Copyright (C) 2003-2008 Dolphin Project.
+
+// This program is free software: you can redistribute it and/or modify
+// it under the terms of the GNU General Public License as published by
+// the Free Software Foundation, version 2.0.
+
+// This program is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU General Public License 2.0 for more details.
+
+// A copy of the GPL 2.0 should have been included with the program.
+// If not, see http://www.gnu.org/licenses/
+
+// Official SVN repository and contact information can be found at
+// http://code.google.com/p/dolphin-emu/
+
+#include "XFMemory.h"
+
+// STATE_TO_SAVE
+XFRegisters xfregs;
+u32 xfmem[XFMEM_SIZE];