diff --git a/desmume/src/GPU.cpp b/desmume/src/GPU.cpp
index cde834c55..01e1ca16e 100644
--- a/desmume/src/GPU.cpp
+++ b/desmume/src/GPU.cpp
@@ -1314,10 +1314,10 @@ void GPUEngineBase::ParseReg_BGnHOFS()
 	const IOREG_BGnHOFS &BGnHOFS = this->_IORegisterMap->BGnOFS[LAYERID].BGnHOFS;
 	this->_BGLayer[LAYERID].BGnHOFS = BGnHOFS;
 	
-#ifdef LOCAL_LE
-	this->_BGLayer[LAYERID].xOffset = BGnHOFS.Offset;
-#else
+#ifdef MSB_FIRST
 	this->_BGLayer[LAYERID].xOffset = LOCAL_TO_LE_16(BGnHOFS.value) & 0x01FF;
+#else
+	this->_BGLayer[LAYERID].xOffset = BGnHOFS.Offset;
 #endif
 }
 
@@ -1327,10 +1327,10 @@ void GPUEngineBase::ParseReg_BGnVOFS()
 	const IOREG_BGnVOFS &BGnVOFS = this->_IORegisterMap->BGnOFS[LAYERID].BGnVOFS;
 	this->_BGLayer[LAYERID].BGnVOFS = BGnVOFS;
 	
-#ifdef LOCAL_LE
-	this->_BGLayer[LAYERID].yOffset = BGnVOFS.Offset;
-#else
+#ifdef MSB_FIRST
 	this->_BGLayer[LAYERID].yOffset = LOCAL_TO_LE_16(BGnVOFS.value) & 0x01FF;
+#else
+	this->_BGLayer[LAYERID].yOffset = BGnVOFS.Offset;
 #endif
 }
 
@@ -1483,9 +1483,7 @@ void GPUEngineBase::_LineColorCopy(void *__restrict dstBuffer, const void *__res
 		
 #if defined(ENABLE_SSE2)
 		MACRODO_N( GPU_FRAMEBUFFER_NATIVE_WIDTH / (sizeof(__m128i) / PIXELBYTES), _mm_stream_si128((__m128i *)dst + (X), _mm_load_si128((__m128i *)src + (X))) );
-#elif LOCAL_LE
-		memcpy(dst, src, GPU_FRAMEBUFFER_NATIVE_WIDTH * PIXELBYTES);
-#else
+#elif MSB_FIRST
 		if (NEEDENDIANSWAP)
 		{
 			for (size_t i = 0; i < GPU_FRAMEBUFFER_NATIVE_WIDTH; i++)
@@ -1504,6 +1502,8 @@ void GPUEngineBase::_LineColorCopy(void *__restrict dstBuffer, const void *__res
 		{
 			memcpy(dst, src, GPU_FRAMEBUFFER_NATIVE_WIDTH * PIXELBYTES);
 		}
+#else
+		memcpy(dst, src, GPU_FRAMEBUFFER_NATIVE_WIDTH * PIXELBYTES);
 #endif
 	}
 	else if (!NATIVEDST && !NATIVESRC)
@@ -1515,9 +1515,7 @@ void GPUEngineBase::_LineColorCopy(void *__restrict dstBuffer, const void *__res
 		void *__restrict dst = (USELINEINDEX) ? (u8 *)dstBuffer + (lineIndex * lineWidth * PIXELBYTES) : (u8 *)dstBuffer;
 		const void *__restrict src = (USELINEINDEX) ? (u8 *)srcBuffer + (lineIndex * lineWidth * PIXELBYTES) : (u8 *)srcBuffer;
 		
-#if defined(LOCAL_LE)
-		memcpy(dst, src, lineWidth * lineCount * PIXELBYTES);
-#else
+#if defined(MSB_FIRST)
 		if (NEEDENDIANSWAP)
 		{
 			for (size_t i = 0; i < lineWidth * lineCount; i++)
@@ -1536,6 +1534,8 @@ void GPUEngineBase::_LineColorCopy(void *__restrict dstBuffer, const void *__res
 		{
 			memcpy(dst, src, lineWidth * lineCount * PIXELBYTES);
 		}
+#else
+		memcpy(dst, src, lineWidth * lineCount * PIXELBYTES);
 #endif
 	}
 	else if (NATIVEDST && !NATIVESRC)
@@ -2422,7 +2422,7 @@ void GPUEngineBase::_RenderPixelIterate_Final(GPUEngineCompositorInfo &compInfo,
 	IOREG_BGnX x = param.BGnX;
 	IOREG_BGnY y = param.BGnY;
 	
-#ifdef LOCAL_BE
+#ifdef MSB_FIRST
 	// This only seems to work in the unrotated/unscaled case. I'm not too sure
 	// about how these bits should really be arranged on big-endian, but at
 	// least this arrangement fixes a bunch of games that use affine or extended
diff --git a/desmume/src/GPU.h b/desmume/src/GPU.h
index 47c1455db..db4d02acb 100644
--- a/desmume/src/GPU.h
+++ b/desmume/src/GPU.h
@@ -116,38 +116,7 @@ typedef union
 	
 	struct
 	{
-#ifdef LOCAL_LE
-		u8 BG_Mode:3;						//  0- 2: A+B;
-		u8 BG0_3D:1;						//     3: A  ; 0=2D,         1=3D
-		u8 OBJ_Tile_mapping:1;				//     4: A+B; 0=2D (32KB),  1=1D (32..256KB)
-		u8 OBJ_BMP_2D_dim:1;				//     5: A+B; 0=128x512,    1=256x256 pixels
-		u8 OBJ_BMP_mapping:1;				//     6: A+B; 0=2D (128KB), 1=1D (128..256KB)
-		u8 ForceBlank:1;					//     7: A+B;
-		
-		u8 BG0_Enable:1;					//     8: A+B; 0=Disable, 1=Enable
-		u8 BG1_Enable:1;					//     9: A+B; 0=Disable, 1=Enable
-		u8 BG2_Enable:1;					//    10: A+B; 0=Disable, 1=Enable
-		u8 BG3_Enable:1;					//    11: A+B; 0=Disable, 1=Enable
-		u8 OBJ_Enable:1;					//    12: A+B; 0=Disable, 1=Enable
-		u8 Win0_Enable:1;					//    13: A+B; 0=Disable, 1=Enable
-		u8 Win1_Enable:1;					//    14: A+B; 0=Disable, 1=Enable
-		u8 WinOBJ_Enable:1;					//    15: A+B; 0=Disable, 1=Enable
-		
-		u8 DisplayMode:2;					// 16-17: A+B; coreA(0..3) coreB(0..1) GBA(Green Swap)
-											//        0=off (white screen)
-											//        1=on (normal BG & OBJ layers)
-											//        2=VRAM display (coreA only)
-											//        3=RAM display (coreA only, DMA transfers)
-		u8 VRAM_Block:2;					// 18-19: A  ; VRAM block (0..3=A..D)
-		u8 OBJ_Tile_1D_Bound:2;				//    20: A+B;
-		u8 OBJ_BMP_1D_Bound:1;				// 21-22: A  ;
-		u8 OBJ_HBlank_process:1;			//    23: A+B; OBJ processed during HBlank (GBA bit5)
-		
-		u8 CharacBase_Block:3;				// 24-26: A  ; Character Base (64K step)
-		u8 ScreenBase_Block:3;				// 27-29: A  ; Screen Base (64K step)
-		u8 ExBGxPalette_Enable:1;			//    30: A+B; 0=Disable, 1=Enable BG extended Palette
-		u8 ExOBJPalette_Enable:1;			//    31: A+B; 0=Disable, 1=Enable OBJ extended Palette
-#else
+#ifdef MSB_FIRST
 		u8 ForceBlank:1;					//     7: A+B;
 		u8 OBJ_BMP_mapping:1;				//     6: A+B; 0=2D (128KB), 1=1D (128..256KB)
 		u8 OBJ_BMP_2D_dim:1;				//     5: A+B; 0=128x512,    1=256x256 pixels
@@ -178,6 +147,37 @@ typedef union
 		u8 ExBGxPalette_Enable:1;			//    30: A+B; 0=Disable, 1=Enable BG extended Palette
 		u8 ScreenBase_Block:3;				// 27-29: A  ; Screen Base (64K step)
 		u8 CharacBase_Block:3;				// 24-26: A  ; Character Base (64K step)
+#else
+		u8 BG_Mode:3;						//  0- 2: A+B;
+		u8 BG0_3D:1;						//     3: A  ; 0=2D,         1=3D
+		u8 OBJ_Tile_mapping:1;				//     4: A+B; 0=2D (32KB),  1=1D (32..256KB)
+		u8 OBJ_BMP_2D_dim:1;				//     5: A+B; 0=128x512,    1=256x256 pixels
+		u8 OBJ_BMP_mapping:1;				//     6: A+B; 0=2D (128KB), 1=1D (128..256KB)
+		u8 ForceBlank:1;					//     7: A+B;
+		
+		u8 BG0_Enable:1;					//     8: A+B; 0=Disable, 1=Enable
+		u8 BG1_Enable:1;					//     9: A+B; 0=Disable, 1=Enable
+		u8 BG2_Enable:1;					//    10: A+B; 0=Disable, 1=Enable
+		u8 BG3_Enable:1;					//    11: A+B; 0=Disable, 1=Enable
+		u8 OBJ_Enable:1;					//    12: A+B; 0=Disable, 1=Enable
+		u8 Win0_Enable:1;					//    13: A+B; 0=Disable, 1=Enable
+		u8 Win1_Enable:1;					//    14: A+B; 0=Disable, 1=Enable
+		u8 WinOBJ_Enable:1;					//    15: A+B; 0=Disable, 1=Enable
+		
+		u8 DisplayMode:2;					// 16-17: A+B; coreA(0..3) coreB(0..1) GBA(Green Swap)
+											//        0=off (white screen)
+											//        1=on (normal BG & OBJ layers)
+											//        2=VRAM display (coreA only)
+											//        3=RAM display (coreA only, DMA transfers)
+		u8 VRAM_Block:2;					// 18-19: A  ; VRAM block (0..3=A..D)
+		u8 OBJ_Tile_1D_Bound:2;				//    20: A+B;
+		u8 OBJ_BMP_1D_Bound:1;				// 21-22: A  ;
+		u8 OBJ_HBlank_process:1;			//    23: A+B; OBJ processed during HBlank (GBA bit5)
+		
+		u8 CharacBase_Block:3;				// 24-26: A  ; Character Base (64K step)
+		u8 ScreenBase_Block:3;				// 27-29: A  ; Screen Base (64K step)
+		u8 ExBGxPalette_Enable:1;			//    30: A+B; 0=Disable, 1=Enable BG extended Palette
+		u8 ExOBJPalette_Enable:1;			//    31: A+B; 0=Disable, 1=Enable OBJ extended Palette
 #endif
 	};
 } IOREG_DISPCNT;							// 0x400x000: Display control (Engine A+B)
@@ -222,36 +222,37 @@ typedef union
 	
 	struct
 	{
-#ifdef LOCAL_LE
-		u8 Priority:2;						//  0- 1: Rendering priority; 0...3, where 0 is highest priority and 3 is lowest priority
-		u8 CharacBase_Block:4;				//  2- 5: individual character base offset (n*16KB)
-		u8 Mosaic:1;						//     6: Mosaic render: 0=Disable, 1=Enable
+#ifdef MSB_FIRST
 		u8 PaletteMode:1;					//     7: Color/palette mode; 0=16 palettes of 16 colors each, 1=Single palette of 256 colors
+		u8 Mosaic:1;						//     6: Mosaic render: 0=Disable, 1=Enable
+		u8 CharacBase_Block:4;				//  2- 5: individual character base offset (n*16KB)
+		u8 Priority:2;						//  0- 1: Rendering priority; 0...3, where 0 is highest priority and 3 is lowest priority
 		
-		u8 ScreenBase_Block:5;				//  8-12: individual screen base offset (text n*2KB, BMP n*16KB)
-		u8 PaletteSet_Wrap:1;				//    13: BG0 extended palette set 0=set0, 1=set2
-											//        BG1 extended palette set 0=set1, 1=set3
-											//        BG2 overflow area wraparound 0=off, 1=wrap
-											//        BG3 overflow area wraparound 0=off, 1=wrap
 		u8 ScreenSize:2;					// 14-15: text    : 256x256 512x256 256x512 512x512
 											//        x/rot/s : 128x128 256x256 512x512 1024x1024
 											//        bmp     : 128x128 256x256 512x256 512x512
 											//        large   : 512x1024 1024x512 - -
+		u8 PaletteSet_Wrap:1;				//    13: BG0 extended palette set 0=set0, 1=set2
+											//        BG1 extended palette set 0=set1, 1=set3
+											//        BG2 overflow area wraparound 0=off, 1=wrap
+											//        BG3 overflow area wraparound 0=off, 1=wrap
+		u8 ScreenBase_Block:5;				//  8-12: individual screen base offset (text n*2KB, BMP n*16KB)
 #else
-		u8 PaletteMode:1;					//     7: Color/palette mode; 0=16 palettes of 16 colors each, 1=Single palette of 256 colors
-		u8 Mosaic:1;						//     6: Mosaic render: 0=Disable, 1=Enable
-		u8 CharacBase_Block:4;				//  2- 5: individual character base offset (n*16KB)
+
 		u8 Priority:2;						//  0- 1: Rendering priority; 0...3, where 0 is highest priority and 3 is lowest priority
+		u8 CharacBase_Block:4;				//  2- 5: individual character base offset (n*16KB)
+		u8 Mosaic:1;						//     6: Mosaic render: 0=Disable, 1=Enable
+		u8 PaletteMode:1;					//     7: Color/palette mode; 0=16 palettes of 16 colors each, 1=Single palette of 256 colors
 		
-		u8 ScreenSize:2;					// 14-15: text    : 256x256 512x256 256x512 512x512
-											//        x/rot/s : 128x128 256x256 512x512 1024x1024
-											//        bmp     : 128x128 256x256 512x256 512x512
-											//        large   : 512x1024 1024x512 - -
+		u8 ScreenBase_Block:5;				//  8-12: individual screen base offset (text n*2KB, BMP n*16KB)
 		u8 PaletteSet_Wrap:1;				//    13: BG0 extended palette set 0=set0, 1=set2
 											//        BG1 extended palette set 0=set1, 1=set3
 											//        BG2 overflow area wraparound 0=off, 1=wrap
 											//        BG3 overflow area wraparound 0=off, 1=wrap
-		u8 ScreenBase_Block:5;				//  8-12: individual screen base offset (text n*2KB, BMP n*16KB)
+		u8 ScreenSize:2;					// 14-15: text    : 256x256 512x256 256x512 512x512
+											//        x/rot/s : 128x128 256x256 512x512 1024x1024
+											//        bmp     : 128x128 256x256 512x256 512x512
+											//        large   : 512x1024 1024x512 - -
 #endif
 	};
 } IOREG_BGnCNT;								// 0x400x008, 0x400x00A, 0x400x00C, 0x400x00E: BGn layer control (Engine A+B)
@@ -309,14 +310,14 @@ typedef union
 	
 	struct
 	{
-#ifdef LOCAL_LE
-		u32 Fraction:8;
-		s32 Integer:20;
+#ifdef MSB_FIRST
 		s32 :4;
+		s32 Integer:20;
+		u32 Fraction:8;
 #else
-		s32 :4;
-		s32 Integer:20;
 		u32 Fraction:8;
+		s32 Integer:20;
+		s32 :4;
 #endif
 	};
 } IOREG_BGnX;								// 0x400x028, 0x400x038: BGn X-coordinate (Engine A+B)
@@ -408,22 +409,22 @@ typedef union
 	
 	struct
 	{
-#ifdef LOCAL_LE
-		u8 BG0_Enable:1;					//     0: Layer BG0 display; 0=Disable, 1=Enable
-		u8 BG1_Enable:1;					//     1: Layer BG1 display; 0=Disable, 1=Enable
-		u8 BG2_Enable:1;					//     2: Layer BG2 display; 0=Disable, 1=Enable
-		u8 BG3_Enable:1;					//     3: Layer BG3 display; 0=Disable, 1=Enable
-		u8 OBJ_Enable:1;					//     4: Layer OBJ display; 0=Disable, 1=Enable
-		u8 Effect_Enable:1;					//     5: Color special effect; 0=Disable, 1=Enable
+#ifdef MSB_FIRST
 		u8 :2;								//  6- 7: Unused bits
+		u8 Effect_Enable:1;					//     5: Color special effect; 0=Disable, 1=Enable
+		u8 OBJ_Enable:1;					//     4: Layer OBJ display; 0=Disable, 1=Enable
+		u8 BG3_Enable:1;					//     3: Layer BG3 display; 0=Disable, 1=Enable
+		u8 BG2_Enable:1;					//     2: Layer BG2 display; 0=Disable, 1=Enable
+		u8 BG1_Enable:1;					//     1: Layer BG1 display; 0=Disable, 1=Enable
+		u8 BG0_Enable:1;					//     0: Layer BG0 display; 0=Disable, 1=Enable
 #else
-		u8 :2;								//  6- 7: Unused bits
-		u8 Effect_Enable:1;					//     5: Color special effect; 0=Disable, 1=Enable
-		u8 OBJ_Enable:1;					//     4: Layer OBJ display; 0=Disable, 1=Enable
-		u8 BG3_Enable:1;					//     3: Layer BG3 display; 0=Disable, 1=Enable
-		u8 BG2_Enable:1;					//     2: Layer BG2 display; 0=Disable, 1=Enable
-		u8 BG1_Enable:1;					//     1: Layer BG1 display; 0=Disable, 1=Enable
 		u8 BG0_Enable:1;					//     0: Layer BG0 display; 0=Disable, 1=Enable
+		u8 BG1_Enable:1;					//     1: Layer BG1 display; 0=Disable, 1=Enable
+		u8 BG2_Enable:1;					//     2: Layer BG2 display; 0=Disable, 1=Enable
+		u8 BG3_Enable:1;					//     3: Layer BG3 display; 0=Disable, 1=Enable
+		u8 OBJ_Enable:1;					//     4: Layer OBJ display; 0=Disable, 1=Enable
+		u8 Effect_Enable:1;					//     5: Color special effect; 0=Disable, 1=Enable
+		u8 :2;								//  6- 7: Unused bits
 #endif
 	};
 } IOREG_WIN0IN;								// 0x400x048: Control of inside of Window 0 (highest priority)
@@ -437,18 +438,18 @@ typedef union
 	
 	struct
 	{
-#ifdef LOCAL_LE
-		u32 BG_MosaicH:4;					//  0- 3: Mosaic pixel width for BG layers; 0...15
+#ifdef MSB_FIRST
 		u32 BG_MosaicV:4;					//  4- 7: Mosaic pixel height for BG layers; 0...15
+		u32 BG_MosaicH:4;					//  0- 3: Mosaic pixel width for BG layers; 0...15
 		
-		u32 OBJ_MosaicH:4;					//  8-11: Mosaic pixel width for OBJ layer; 0...15
 		u32 OBJ_MosaicV:4;					// 12-15: Mosaic pixel height for OBJ layer; 0...15
+		u32 OBJ_MosaicH:4;					//  8-11: Mosaic pixel width for OBJ layer; 0...15
 #else
-		u32 BG_MosaicV:4;					//  4- 7: Mosaic pixel height for BG layers; 0...15
 		u32 BG_MosaicH:4;					//  0- 3: Mosaic pixel width for BG layers; 0...15
+		u32 BG_MosaicV:4;					//  4- 7: Mosaic pixel height for BG layers; 0...15
 		
-		u32 OBJ_MosaicV:4;					// 12-15: Mosaic pixel height for OBJ layer; 0...15
 		u32 OBJ_MosaicH:4;					//  8-11: Mosaic pixel width for OBJ layer; 0...15
+		u32 OBJ_MosaicV:4;					// 12-15: Mosaic pixel height for OBJ layer; 0...15
 #endif
 		
 		u32 :16;							// 16-31: Unused bits
@@ -461,46 +462,46 @@ typedef union
 	
 	struct
 	{
-#ifdef LOCAL_LE
-		u16 BG0_Target1:1;					//     0: Select layer BG0 for 1st target; 0=Disable, 1=Enable
-		u16 BG1_Target1:1;					//     1: Select layer BG1 for 1st target; 0=Disable, 1=Enable
-		u16 BG2_Target1:1;					//     2: Select layer BG2 for 1st target; 0=Disable, 1=Enable
-		u16 BG3_Target1:1;					//     3: Select layer BG3 for 1st target; 0=Disable, 1=Enable
-		u16 OBJ_Target1:1;					//     4: Select layer OBJ for 1st target; 0=Disable, 1=Enable
-		u16 Backdrop_Target1:1;				//     5: Select backdrop for 1st target; 0=Disable, 1=Enable
+#ifdef MSB_FIRST
 		u16 ColorEffect:2;					//  6- 7: Color effect mode;
 											//        0=Disable
 											//        1=Alpha blend 1st and 2nd target, interacts with BLDALPHA (0x400x052)
 											//        2=Increase brightness, interacts with BLDY (0x400x054)
 											//        3=Decrease brightness, interacts with BLDY (0x400x054)
+		u16 Backdrop_Target1:1;				//     5: Select backdrop for 1st target; 0=Disable, 1=Enable
+		u16 OBJ_Target1:1;					//     4: Select layer OBJ for 1st target; 0=Disable, 1=Enable
+		u16 BG3_Target1:1;					//     3: Select layer BG3 for 1st target; 0=Disable, 1=Enable
+		u16 BG2_Target1:1;					//     2: Select layer BG2 for 1st target; 0=Disable, 1=Enable
+		u16 BG1_Target1:1;					//     1: Select layer BG1 for 1st target; 0=Disable, 1=Enable
+		u16 BG0_Target1:1;					//     0: Select layer BG0 for 1st target; 0=Disable, 1=Enable
 		
-		u16 BG0_Target2:1;					//     8: Select layer BG0 for 2nd target; 0=Disable, 1=Enable
-		u16 BG1_Target2:1;					//     9: Select layer BG1 for 2nd target; 0=Disable, 1=Enable
-		u16 BG2_Target2:1;					//    10: Select layer BG2 for 2nd target; 0=Disable, 1=Enable
-		u16 BG3_Target2:1;					//    11: Select layer BG3 for 2nd target; 0=Disable, 1=Enable
-		u16 OBJ_Target2:1;					//    12: Select layer OBJ for 2nd target; 0=Disable, 1=Enable
-		u16 Backdrop_Target2:1;				//    13: Select backdrop for 2nd target; 0=Disable, 1=Enable
 		u16 :2;								// 14-15: Unused bits
+		u16 Backdrop_Target2:1;				//    13: Select backdrop for 2nd target; 0=Disable, 1=Enable
+		u16 OBJ_Target2:1;					//    12: Select layer OBJ for 2nd target; 0=Disable, 1=Enable
+		u16 BG3_Target2:1;					//    11: Select layer BG3 for 2nd target; 0=Disable, 1=Enable
+		u16 BG2_Target2:1;					//    10: Select layer BG2 for 2nd target; 0=Disable, 1=Enable
+		u16 BG1_Target2:1;					//     9: Select layer BG1 for 2nd target; 0=Disable, 1=Enable
+		u16 BG0_Target2:1;					//     8: Select layer BG0 for 2nd target; 0=Disable, 1=Enable
 #else
+		u16 BG0_Target1:1;					//     0: Select layer BG0 for 1st target; 0=Disable, 1=Enable
+		u16 BG1_Target1:1;					//     1: Select layer BG1 for 1st target; 0=Disable, 1=Enable
+		u16 BG2_Target1:1;					//     2: Select layer BG2 for 1st target; 0=Disable, 1=Enable
+		u16 BG3_Target1:1;					//     3: Select layer BG3 for 1st target; 0=Disable, 1=Enable
+		u16 OBJ_Target1:1;					//     4: Select layer OBJ for 1st target; 0=Disable, 1=Enable
+		u16 Backdrop_Target1:1;				//     5: Select backdrop for 1st target; 0=Disable, 1=Enable
 		u16 ColorEffect:2;					//  6- 7: Color effect mode;
 											//        0=Disable
 											//        1=Alpha blend 1st and 2nd target, interacts with BLDALPHA (0x400x052)
 											//        2=Increase brightness, interacts with BLDY (0x400x054)
 											//        3=Decrease brightness, interacts with BLDY (0x400x054)
-		u16 Backdrop_Target1:1;				//     5: Select backdrop for 1st target; 0=Disable, 1=Enable
-		u16 OBJ_Target1:1;					//     4: Select layer OBJ for 1st target; 0=Disable, 1=Enable
-		u16 BG3_Target1:1;					//     3: Select layer BG3 for 1st target; 0=Disable, 1=Enable
-		u16 BG2_Target1:1;					//     2: Select layer BG2 for 1st target; 0=Disable, 1=Enable
-		u16 BG1_Target1:1;					//     1: Select layer BG1 for 1st target; 0=Disable, 1=Enable
-		u16 BG0_Target1:1;					//     0: Select layer BG0 for 1st target; 0=Disable, 1=Enable
 		
-		u16 :2;								// 14-15: Unused bits
-		u16 Backdrop_Target2:1;				//    13: Select backdrop for 2nd target; 0=Disable, 1=Enable
-		u16 OBJ_Target2:1;					//    12: Select layer OBJ for 2nd target; 0=Disable, 1=Enable
-		u16 BG3_Target2:1;					//    11: Select layer BG3 for 2nd target; 0=Disable, 1=Enable
-		u16 BG2_Target2:1;					//    10: Select layer BG2 for 2nd target; 0=Disable, 1=Enable
-		u16 BG1_Target2:1;					//     9: Select layer BG1 for 2nd target; 0=Disable, 1=Enable
 		u16 BG0_Target2:1;					//     8: Select layer BG0 for 2nd target; 0=Disable, 1=Enable
+		u16 BG1_Target2:1;					//     9: Select layer BG1 for 2nd target; 0=Disable, 1=Enable
+		u16 BG2_Target2:1;					//    10: Select layer BG2 for 2nd target; 0=Disable, 1=Enable
+		u16 BG3_Target2:1;					//    11: Select layer BG3 for 2nd target; 0=Disable, 1=Enable
+		u16 OBJ_Target2:1;					//    12: Select layer OBJ for 2nd target; 0=Disable, 1=Enable
+		u16 Backdrop_Target2:1;				//    13: Select backdrop for 2nd target; 0=Disable, 1=Enable
+		u16 :2;								// 14-15: Unused bits
 #endif
 	};
 } IOREG_BLDCNT;								// 0x400x050: Color effects selection (Engine A+B)
@@ -511,18 +512,18 @@ typedef union
 	
 	struct
 	{
-#ifdef LOCAL_LE
-		u16 EVA:5;							//  0- 4: Blending coefficient for 1st target; 0...31 (clamped to 16)
+#ifdef MSB_FIRST
 		u16 :3;								//  5- 7: Unused bits
+		u16 EVA:5;							//  0- 4: Blending coefficient for 1st target; 0...31 (clamped to 16)
 		
-		u16 EVB:5;							//  8-12: Blending coefficient for 2nd target; 0...31 (clamped to 16)
 		u16 :3;								// 13-15: Unused bits
+		u16 EVB:5;							//  8-12: Blending coefficient for 2nd target; 0...31 (clamped to 16)
 #else
-		u16 :3;								//  5- 7: Unused bits
 		u16 EVA:5;							//  0- 4: Blending coefficient for 1st target; 0...31 (clamped to 16)
+		u16 :3;								//  5- 7: Unused bits
 		
-		u16 :3;								// 13-15: Unused bits
 		u16 EVB:5;							//  8-12: Blending coefficient for 2nd target; 0...31 (clamped to 16)
+		u16 :3;								// 13-15: Unused bits
 #endif
 	};
 } IOREG_BLDALPHA;							// 0x400x052: Color effects selection, interacts with BLDCNT (0x400x050) (Engine A+B)
@@ -533,12 +534,12 @@ typedef union
 	
 	struct
 	{
-#ifdef LOCAL_LE
-		u16 EVY:5;							//  0- 4: Blending coefficient for increase/decrease brightness; 0...31 (clamped to 16)
+#ifdef MSB_FIRST
 		u16 :3;								//  5- 7: Unused bits
+		u16 EVY:5;							//  0- 4: Blending coefficient for increase/decrease brightness; 0...31 (clamped to 16)
 #else
-		u16 :3;								//  5- 7: Unused bits
 		u16 EVY:5;							//  0- 4: Blending coefficient for increase/decrease brightness; 0...31 (clamped to 16)
+		u16 :3;								//  5- 7: Unused bits
 #endif
 		u16 :8;								//  8-15: Unused bits
 	};
@@ -550,42 +551,42 @@ typedef union
 	
 	struct
 	{
-#ifdef LOCAL_LE
-		u8 EnableTexMapping:1;				//     0: Apply textures; 0=Disable, 1=Enable
-		u8 PolygonShading:1;				//     1: Polygon shading mode, interacts with POLYGON_ATTR (0x40004A4); 0=Toon Shading, 1=Highlight Shading
-		u8 EnableAlphaTest:1;				//     2: Perform alpha test, interacts with ALPHA_TEST_REF (0x4000340); 0=Disable, 1=Enable
-		u8 EnableAlphaBlending:1;			//     3: Perform alpha blending, interacts with POLYGON_ATTR (0x40004A4); 0=Disable, 1=Enable
-		u8 EnableAntiAliasing:1;			//     4: Render polygon edges with antialiasing; 0=Disable, 1=Enable
-		u8 EnableEdgeMarking:1;				//     5: Perform polygon edge marking, interacts with EDGE_COLOR (0x4000330); 0=Disable, 1=Enable
-		u8 FogOnlyAlpha:1;					//     6: Apply fog to the alpha channel only, interacts with FOG_COLOR (0x4000358) / FOG_TABLE (0x4000360); 0=Color+Alpha, 1=Alpha
+#ifdef MSB_FIRST
 		u8 EnableFog:1;						//     7: Perform fog rendering, interacts with FOG_COLOR (0x4000358) / FOG_OFFSET (0x400035C) / FOG_TABLE (0x4000360);
 											//        0=Disable, 1=Enable
+		u8 FogOnlyAlpha:1;					//     6: Apply fog to the alpha channel only, interacts with FOG_COLOR (0x4000358) / FOG_TABLE (0x4000360); 0=Color+Alpha, 1=Alpha
+		u8 EnableEdgeMarking:1;				//     5: Perform polygon edge marking, interacts with EDGE_COLOR (0x4000330); 0=Disable, 1=Enable
+		u8 EnableAntiAliasing:1;			//     4: Render polygon edges with antialiasing; 0=Disable, 1=Enable
+		u8 EnableAlphaBlending:1;			//     3: Perform alpha blending, interacts with POLYGON_ATTR (0x40004A4); 0=Disable, 1=Enable
+		u8 EnableAlphaTest:1;				//     2: Perform alpha test, interacts with ALPHA_TEST_REF (0x4000340); 0=Disable, 1=Enable
+		u8 PolygonShading:1;				//     1: Polygon shading mode, interacts with POLYGON_ATTR (0x40004A4); 0=Toon Shading, 1=Highlight Shading
+		u8 EnableTexMapping:1;				//     0: Apply textures; 0=Disable, 1=Enable
 		
-		u8 FogShiftSHR:4;					//  8-11: SHR-Divider, interacts with FOG_OFFSET (0x400035C); 0...10
-		u8 AckColorBufferUnderflow:1;		//    12: Color Buffer RDLINES Underflow; 0=None, 1=Underflow/Acknowledge
-		u8 AckVertexRAMOverflow:1;			//    13: Polygon/Vertex RAM Overflow; 0=None, 1=Overflow/Acknowledge
+		u8 :1;								//    15: Unused bits
 		u8 RearPlaneMode:1;					//    14: Use clear image, interacts with CLEAR_COLOR (0x4000350) / CLEAR_DEPTH (0x4000354) / CLRIMAGE_OFFSET (0x4000356);
 											//        0=Blank, 1=Bitmap
-		u8 :1;								//    15: Unused bits
+		u8 AckVertexRAMOverflow:1;			//    13: Polygon/Vertex RAM Overflow; 0=None, 1=Overflow/Acknowledge
+		u8 AckColorBufferUnderflow:1;		//    12: Color Buffer RDLINES Underflow; 0=None, 1=Underflow/Acknowledge
+		u8 FogShiftSHR:4;					//  8-11: SHR-Divider, interacts with FOG_OFFSET (0x400035C); 0...10
 		
 		u16 :16;							// 16-31: Unused bits
 #else
+		u8 EnableTexMapping:1;				//     0: Apply textures; 0=Disable, 1=Enable
+		u8 PolygonShading:1;				//     1: Polygon shading mode, interacts with POLYGON_ATTR (0x40004A4); 0=Toon Shading, 1=Highlight Shading
+		u8 EnableAlphaTest:1;				//     2: Perform alpha test, interacts with ALPHA_TEST_REF (0x4000340); 0=Disable, 1=Enable
+		u8 EnableAlphaBlending:1;			//     3: Perform alpha blending, interacts with POLYGON_ATTR (0x40004A4); 0=Disable, 1=Enable
+		u8 EnableAntiAliasing:1;			//     4: Render polygon edges with antialiasing; 0=Disable, 1=Enable
+		u8 EnableEdgeMarking:1;				//     5: Perform polygon edge marking, interacts with EDGE_COLOR (0x4000330); 0=Disable, 1=Enable
+		u8 FogOnlyAlpha:1;					//     6: Apply fog to the alpha channel only, interacts with FOG_COLOR (0x4000358) / FOG_TABLE (0x4000360); 0=Color+Alpha, 1=Alpha
 		u8 EnableFog:1;						//     7: Perform fog rendering, interacts with FOG_COLOR (0x4000358) / FOG_OFFSET (0x400035C) / FOG_TABLE (0x4000360);
 											//        0=Disable, 1=Enable
-		u8 FogOnlyAlpha:1;					//     6: Apply fog to the alpha channel only, interacts with FOG_COLOR (0x4000358) / FOG_TABLE (0x4000360); 0=Color+Alpha, 1=Alpha
-		u8 EnableEdgeMarking:1;				//     5: Perform polygon edge marking, interacts with EDGE_COLOR (0x4000330); 0=Disable, 1=Enable
-		u8 EnableAntiAliasing:1;			//     4: Render polygon edges with antialiasing; 0=Disable, 1=Enable
-		u8 EnableAlphaBlending:1;			//     3: Perform alpha blending, interacts with POLYGON_ATTR (0x40004A4); 0=Disable, 1=Enable
-		u8 EnableAlphaTest:1;				//     2: Perform alpha test, interacts with ALPHA_TEST_REF (0x4000340); 0=Disable, 1=Enable
-		u8 PolygonShading:1;				//     1: Polygon shading mode, interacts with POLYGON_ATTR (0x40004A4); 0=Toon Shading, 1=Highlight Shading
-		u8 EnableTexMapping:1;				//     0: Apply textures; 0=Disable, 1=Enable
 		
-		u8 :1;								//    15: Unused bits
+		u8 FogShiftSHR:4;					//  8-11: SHR-Divider, interacts with FOG_OFFSET (0x400035C); 0...10
+		u8 AckColorBufferUnderflow:1;		//    12: Color Buffer RDLINES Underflow; 0=None, 1=Underflow/Acknowledge
+		u8 AckVertexRAMOverflow:1;			//    13: Polygon/Vertex RAM Overflow; 0=None, 1=Overflow/Acknowledge
 		u8 RearPlaneMode:1;					//    14: Use clear image, interacts with CLEAR_COLOR (0x4000350) / CLEAR_DEPTH (0x4000354) / CLRIMAGE_OFFSET (0x4000356);
 											//        0=Blank, 1=Bitmap
-		u8 AckVertexRAMOverflow:1;			//    13: Polygon/Vertex RAM Overflow; 0=None, 1=Overflow/Acknowledge
-		u8 AckColorBufferUnderflow:1;		//    12: Color Buffer RDLINES Underflow; 0=None, 1=Underflow/Acknowledge
-		u8 FogShiftSHR:4;					//  8-11: SHR-Divider, interacts with FOG_OFFSET (0x400035C); 0...10
+		u8 :1;								//    15: Unused bits
 		
 		u16 :16;							// 16-31: Unused bits
 #endif
@@ -598,46 +599,46 @@ typedef union
 	
 	struct
 	{
-#ifdef LOCAL_LE
-		unsigned EVA:5;						//  0- 4: Blending coefficient for SrcA; 0...31 (clamped to 16)
+#ifdef MSB_FIRST
 		unsigned :3;						//  5- 7: Unused bits
+		unsigned EVA:5;						//  0- 4: Blending coefficient for SrcA; 0...31 (clamped to 16)
 		
-		unsigned EVB:5;						//  8-12: Blending coefficient for SrcB; 0...31 (clamped to 16)
 		unsigned :3;						// 13-15: Unused bits
+		unsigned EVB:5;						//  8-12: Blending coefficient for SrcB; 0...31 (clamped to 16)
 		
-		unsigned VRAMWriteBlock:2;			// 16-17: VRAM write target block; 0=Block A, 1=Block B, 2=Block C, 3=Block D
-		unsigned VRAMWriteOffset:2;			// 18-19: VRAM write target offset; 0=0KB, 1=32KB, 2=64KB, 3=96KB
-		unsigned CaptureSize:2;				// 20-21: Display capture dimensions; 0=128x128, 1=256x64, 2=256x128, 3=256x192
 		unsigned :2;						// 22-23: Unused bits
+		unsigned CaptureSize:2;				// 20-21: Display capture dimensions; 0=128x128, 1=256x64, 2=256x128, 3=256x192
+		unsigned VRAMWriteOffset:2;			// 18-19: VRAM write target offset; 0=0KB, 1=32KB, 2=64KB, 3=96KB
+		unsigned VRAMWriteBlock:2;			// 16-17: VRAM write target block; 0=Block A, 1=Block B, 2=Block C, 3=Block D
 		
-		unsigned SrcA:1;					//    24: SrcA target; 0=Current framebuffer, 1=3D render buffer
+		unsigned CaptureEnable:1;			//    31: Display capture status; 0=Disable/Ready 1=Enable/Busy
+		unsigned CaptureSrc:2;				// 29-30: Select capture target; 0=SrcA, 1=SrcB, 2=SrcA+SrcB blend, 3=SrcA+SrcB blend
+		unsigned :1;						//    28: Unused bit
+		unsigned VRAMReadOffset:2;			// 26-27: VRAM read target offset; 0=0KB, 1=32KB, 2=64KB, 3=96KB
 		unsigned SrcB:1;					//    25: SrcB target;
 											//        0=VRAM block, interacts with DISPCNT (0x4000000)
 											//        1=Main memory FIFO, interacts with DISP_MMEM_FIFO (0x4000068)
-		unsigned VRAMReadOffset:2;			// 26-27: VRAM read target offset; 0=0KB, 1=32KB, 2=64KB, 3=96KB
-		unsigned :1;						//    28: Unused bit
-		unsigned CaptureSrc:2;				// 29-30: Select capture target; 0=SrcA, 1=SrcB, 2=SrcA+SrcB blend, 3=SrcA+SrcB blend
-		unsigned CaptureEnable:1;			//    31: Display capture status; 0=Disable/Ready 1=Enable/Busy
+		unsigned SrcA:1;					//    24: SrcA target; 0=Current framebuffer, 1=3D render buffer
 #else
-		unsigned :3;						//  5- 7: Unused bits
 		unsigned EVA:5;						//  0- 4: Blending coefficient for SrcA; 0...31 (clamped to 16)
+		unsigned :3;						//  5- 7: Unused bits
 		
-		unsigned :3;						// 13-15: Unused bits
 		unsigned EVB:5;						//  8-12: Blending coefficient for SrcB; 0...31 (clamped to 16)
+		unsigned :3;						// 13-15: Unused bits
 		
-		unsigned :2;						// 22-23: Unused bits
-		unsigned CaptureSize:2;				// 20-21: Display capture dimensions; 0=128x128, 1=256x64, 2=256x128, 3=256x192
-		unsigned VRAMWriteOffset:2;			// 18-19: VRAM write target offset; 0=0KB, 1=32KB, 2=64KB, 3=96KB
 		unsigned VRAMWriteBlock:2;			// 16-17: VRAM write target block; 0=Block A, 1=Block B, 2=Block C, 3=Block D
+		unsigned VRAMWriteOffset:2;			// 18-19: VRAM write target offset; 0=0KB, 1=32KB, 2=64KB, 3=96KB
+		unsigned CaptureSize:2;				// 20-21: Display capture dimensions; 0=128x128, 1=256x64, 2=256x128, 3=256x192
+		unsigned :2;						// 22-23: Unused bits
 		
-		unsigned CaptureEnable:1;			//    31: Display capture status; 0=Disable/Ready 1=Enable/Busy
-		unsigned CaptureSrc:2;				// 29-30: Select capture target; 0=SrcA, 1=SrcB, 2=SrcA+SrcB blend, 3=SrcA+SrcB blend
-		unsigned :1;						//    28: Unused bit
-		unsigned VRAMReadOffset:2;			// 26-27: VRAM read target offset; 0=0KB, 1=32KB, 2=64KB, 3=96KB
+		unsigned SrcA:1;					//    24: SrcA target; 0=Current framebuffer, 1=3D render buffer
 		unsigned SrcB:1;					//    25: SrcB target;
 											//        0=VRAM block, interacts with DISPCNT (0x4000000)
 											//        1=Main memory FIFO, interacts with DISP_MMEM_FIFO (0x4000068)
-		unsigned SrcA:1;					//    24: SrcA target; 0=Current framebuffer, 1=3D render buffer
+		unsigned VRAMReadOffset:2;			// 26-27: VRAM read target offset; 0=0KB, 1=32KB, 2=64KB, 3=96KB
+		unsigned :1;						//    28: Unused bit
+		unsigned CaptureSrc:2;				// 29-30: Select capture target; 0=SrcA, 1=SrcB, 2=SrcA+SrcB blend, 3=SrcA+SrcB blend
+		unsigned CaptureEnable:1;			//    31: Display capture status; 0=Disable/Ready 1=Enable/Busy
 #endif
 	};
 	
@@ -651,20 +652,20 @@ typedef union
 	
 	struct
 	{
-#ifdef LOCAL_LE
-		u32 Intensity:5;					//  0- 4: Brightness coefficient for increase/decrease brightness; 0...31 (clamped to 16)
+#ifdef MSB_FIRST
 		u32 :3;								//  5- 7: Unused bits
+		u32 Intensity:5;					//  0- 4: Brightness coefficient for increase/decrease brightness; 0...31 (clamped to 16)
 		
-		u32 :6;								//  8-13: Unused bits
 		u32 Mode:2;							// 14-15: Brightness mode; 0=Disable, 1=Increase, 2=Decrease, 3=Reserved
+		u32 :6;								//  8-13: Unused bits
 		
 		u32 :16;							// 16-31: Unused bits
 #else
-		u32 :3;								//  5- 7: Unused bits
 		u32 Intensity:5;					//  0- 4: Brightness coefficient for increase/decrease brightness; 0...31 (clamped to 16)
+		u32 :3;								//  5- 7: Unused bits
 		
-		u32 Mode:2;							// 14-15: Brightness mode; 0=Disable, 1=Increase, 2=Decrease, 3=Reserved
 		u32 :6;								//  8-13: Unused bits
+		u32 Mode:2;							// 14-15: Brightness mode; 0=Disable, 1=Increase, 2=Decrease, 3=Reserved
 		
 		u32 :16;							// 16-31: Unused bits
 #endif
@@ -799,7 +800,7 @@ enum GPUEngineID
 #define ADDRESS_STEP_512KB	   0x80000
 #define ADDRESS_MASK_256KB	   (ADDRESS_STEP_256KB-1)
 
-#ifdef LOCAL_BE
+#ifdef MSB_FIRST
 struct _TILEENTRY
 {
 /*14*/	unsigned Palette:4;
@@ -830,7 +831,7 @@ typedef union
 */
 
 struct _COLOR { // abgr x555
-#ifdef LOCAL_BE
+#ifdef MSB_FIRST
 	unsigned alpha:1;    // sometimes it is unused (pad)
 	unsigned blue:5;
 	unsigned green:5;
@@ -867,47 +868,7 @@ typedef union
 	
 	struct
 	{
-#ifdef LOCAL_LE
-		union
-		{
-			u16 attr0;
-			
-			struct
-			{
-				u16 Y:8;					//  0- 7: Sprite Y-coordinate; 0...255
-				u16 RotScale:1;				//     8: Perform rotation/scaling; 0=Disable, 1=Enable
-				u16 Disable:1;				//     9: OBJ disable flag, only if Bit8 is cleared; 0=Perform render, 1=Do not perform render
-				u16 Mode:2;					// 10-11: OBJ mode; 0=Normal, 1=Transparent, 2=Window, 3=Bitmap
-				u16 Mosaic:1;				//    12: Mosaic render: 0=Disable, 1=Enable
-				u16 PaletteMode:1;			//    13: Color/palette select; 0=16 palettes of 16 colors each, 1=Single palette of 256 colors
-				u16 Shape:2;				// 14-15: OBJ shape; 0=Square, 1=Horizontal, 2=Vertical, 3=Prohibited
-			};
-			
-			struct
-			{
-				u16 :8;
-				u16 :1;
-				u16 DoubleSize:1;			//     9: Perform double-size render, only if Bit8 is set; 0=Disable, 1=Enable
-				u16 :6;
-			};
-		};
-		
-		s16 X:9;							// 16-24: Sprite X-coordinate; 0...511
-		u16 RotScaleIndex:3;				// 25-27: Rotation/scaling parameter selection; 0...31
-		u16 HFlip:1;						//    28: Flip sprite horizontally; 0=Normal, 1=Flip
-		u16 VFlip:1;						//    29: Flip sprite vertically; 0=Normal, 1=Flip
-		u16 Size:2;							// 30-31: OBJ size, interacts with Bit 14-15
-											//
-											//        Size| Square | Horizontal | Vertical
-											//           0:   8x8       16x8        8x16
-											//           1:  16x16      32x8        8x32
-											//           2:  32x32      32x16      16x32
-											//           3:  64x64      64x32      32x64
-		u16 TileIndex:10;					// 32-41: Tile index; 0...1023
-		
-		u16 Priority:2;						// 42-43: Rendering priority; 0...3, where 0 is highest priority and 3 is lowest priority
-		u16 PaletteIndex:4;					// 44-47: Palette index; 0...15
-#else
+#ifdef MSB_FIRST
 		union
 		{
 			u16 attr0;
@@ -949,6 +910,46 @@ typedef union
 		u16 PaletteIndex:4;					// 44-47: Palette index; 0...15
 		u16 Priority:2;						// 42-43: Rendering priority; 0...3, where 0 is highest priority and 3 is lowest priority
 		u16 TileIndex:10;					// 32-41: Tile index; 0...1023
+#else
+		union
+		{
+			u16 attr0;
+			
+			struct
+			{
+				u16 Y:8;					//  0- 7: Sprite Y-coordinate; 0...255
+				u16 RotScale:1;				//     8: Perform rotation/scaling; 0=Disable, 1=Enable
+				u16 Disable:1;				//     9: OBJ disable flag, only if Bit8 is cleared; 0=Perform render, 1=Do not perform render
+				u16 Mode:2;					// 10-11: OBJ mode; 0=Normal, 1=Transparent, 2=Window, 3=Bitmap
+				u16 Mosaic:1;				//    12: Mosaic render: 0=Disable, 1=Enable
+				u16 PaletteMode:1;			//    13: Color/palette select; 0=16 palettes of 16 colors each, 1=Single palette of 256 colors
+				u16 Shape:2;				// 14-15: OBJ shape; 0=Square, 1=Horizontal, 2=Vertical, 3=Prohibited
+			};
+			
+			struct
+			{
+				u16 :8;
+				u16 :1;
+				u16 DoubleSize:1;			//     9: Perform double-size render, only if Bit8 is set; 0=Disable, 1=Enable
+				u16 :6;
+			};
+		};
+		
+		s16 X:9;							// 16-24: Sprite X-coordinate; 0...511
+		u16 RotScaleIndex:3;				// 25-27: Rotation/scaling parameter selection; 0...31
+		u16 HFlip:1;						//    28: Flip sprite horizontally; 0=Normal, 1=Flip
+		u16 VFlip:1;						//    29: Flip sprite vertically; 0=Normal, 1=Flip
+		u16 Size:2;							// 30-31: OBJ size, interacts with Bit 14-15
+											//
+											//        Size| Square | Horizontal | Vertical
+											//           0:   8x8       16x8        8x16
+											//           1:  16x16      32x8        8x32
+											//           2:  32x32      32x16      16x32
+											//           3:  64x64      64x32      32x64
+		u16 TileIndex:10;					// 32-41: Tile index; 0...1023
+		
+		u16 Priority:2;						// 42-43: Rendering priority; 0...3, where 0 is highest priority and 3 is lowest priority
+		u16 PaletteIndex:4;					// 44-47: Palette index; 0...15
 #endif
 		
 		u16 attr3:16;						// 48-63: Whenever this is used, you will need to explicitly convert endianness.
@@ -1752,20 +1753,20 @@ extern CACHE_ALIGN u32 color_555_to_888[32768];
 #define COLOR555TO6665_OPAQUE_SWAP_RB(col) (color_555_to_6665_opaque_swap_rb[(col)])	// Convert a 15-bit color to an opaque sparsely packed 32-bit color containing an RGBA6665 color with R and B components swapped
 #define COLOR555TO666(col) (color_555_to_666[(col)])									// Convert a 15-bit color to a fully transparent sparsely packed 32-bit color containing an RGBA6665 color
 
-#ifdef LOCAL_LE
-	#define COLOR555TO6665(col,alpha5) (((alpha5)<<24) | color_555_to_666[(col)])		// Convert a 15-bit color to a sparsely packed 32-bit color containing an RGBA6665 color with user-defined alpha, little-endian
-#else
+#ifdef MSB_FIRST
 	#define COLOR555TO6665(col,alpha5) ((alpha5) | color_555_to_666[(col)])				// Convert a 15-bit color to a sparsely packed 32-bit color containing an RGBA6665 color with user-defined alpha, big-endian
+#else
+	#define COLOR555TO6665(col,alpha5) (((alpha5)<<24) | color_555_to_666[(col)])		// Convert a 15-bit color to a sparsely packed 32-bit color containing an RGBA6665 color with user-defined alpha, little-endian
 #endif
 
 #define COLOR555TO8888_OPAQUE(col) (color_555_to_8888_opaque[(col)])					// Convert a 15-bit color to an opaque 32-bit color
 #define COLOR555TO8888_OPAQUE_SWAP_RB(col) (color_555_to_8888_opaque_swap_rb[(col)])	// Convert a 15-bit color to an opaque 32-bit color with R and B components swapped
 #define COLOR555TO888(col) (color_555_to_888[(col)])									// Convert a 15-bit color to an opaque 24-bit color or a fully transparent 32-bit color
 
-#ifdef LOCAL_LE
-	#define COLOR555TO8888(col,alpha8) (((alpha8)<<24) | color_555_to_888[(col)])		// Convert a 15-bit color to a 32-bit color with user-defined alpha, little-endian
-#else
+#ifdef MSB_FIRST
 	#define COLOR555TO8888(col,alpha8) ((alpha8) | color_555_to_888[(col)])				// Convert a 15-bit color to a 32-bit color with user-defined alpha, big-endian
+#else
+	#define COLOR555TO8888(col,alpha8) (((alpha8)<<24) | color_555_to_888[(col)])		// Convert a 15-bit color to a 32-bit color with user-defined alpha, little-endian
 #endif
 
 //produce a 15bpp color from individual 5bit components
diff --git a/desmume/src/MMU.h b/desmume/src/MMU.h
index 97c22d4d3..a7c08e191 100644
--- a/desmume/src/MMU.h
+++ b/desmume/src/MMU.h
@@ -316,39 +316,39 @@ typedef union
 {
 	u8 value;
 	
-#ifdef LOCAL_LE
+#ifdef MSB_FIRST
 	struct
 	{
-		unsigned MST:3;
-		unsigned OFS:2;
-		unsigned :2;
 		unsigned Enable:1;
+		unsigned :2;
+		unsigned OFS:2;
+		unsigned MST:3;
 	};
 	
 	struct
 	{
-		unsigned MST_ABHI:2;
-		unsigned :1;
-		unsigned OFS_ABHI:2;
-		unsigned :2;
 		unsigned Enable_ABHI:1;
+		unsigned :2;
+		unsigned OFS_ABHI:2;
+		unsigned :1;
+		unsigned MST_ABHI:2;
 	};
 #else
 	struct
 	{
-		unsigned Enable:1;
-		unsigned :2;
-		unsigned OFS:2;
 		unsigned MST:3;
+		unsigned OFS:2;
+		unsigned :2;
+		unsigned Enable:1;
 	};
 	
 	struct
 	{
-		unsigned Enable_ABHI:1;
-		unsigned :2;
-		unsigned OFS_ABHI:2;
-		unsigned :1;
 		unsigned MST_ABHI:2;
+		unsigned :1;
+		unsigned OFS_ABHI:2;
+		unsigned :2;
+		unsigned Enable_ABHI:1;
 	};
 #endif
 } VRAMCNT;
diff --git a/desmume/src/NDSSystem.cpp b/desmume/src/NDSSystem.cpp
index a79ad0801..018ccda2d 100644
--- a/desmume/src/NDSSystem.cpp
+++ b/desmume/src/NDSSystem.cpp
@@ -436,7 +436,7 @@ bool GameInfo::loadROM(std::string fname, u32 type)
 
 	if (res)
 	{
-#ifndef LOCAL_LE
+#ifdef MSB_FIRST
 		//endian swap necessary fields. It would be better if we made accessors for these. I wonder if you could make a macro for a field accessor that would take the bitsize and do the swap on the fly
 		struct FieldSwap {
 			const size_t offset;
@@ -2760,7 +2760,7 @@ void NDS_setTouchPos(u16 x, u16 y)
 	if(movieMode != MOVIEMODE_INACTIVE && movieMode != MOVIEMODE_FINISHED)
 	{
 		// just in case, since the movie only stores 8 bits per touch coord
-#ifdef WORDS_BIGENDIAN
+#ifdef MSB_FIRST
 		rawUserInput.touch.touchX &= 0xF00F;
 		rawUserInput.touch.touchY &= 0xF00F;
 #else
diff --git a/desmume/src/addons/slot1comp_protocol.cpp b/desmume/src/addons/slot1comp_protocol.cpp
index def9f15f4..2698e01a0 100644
--- a/desmume/src/addons/slot1comp_protocol.cpp
+++ b/desmume/src/addons/slot1comp_protocol.cpp
@@ -99,10 +99,10 @@ void Slot1Comp_Protocol::write_command_KEY1(GC_Command command)
 			
 			//TODO - more endian-safe way of doing this (theres examples in R4)
 			{
-#ifdef LOCAL_LE
-				u64 cmd64 = bswap64(*(u64*)command.bytes);
-#else
+#ifdef MSB_FIRST
 				u64 cmd64 = *(u64*)command.bytes;
+#else
+				u64 cmd64 = bswap64(*(u64*)command.bytes);
 #endif
 				//todo - parse into blocknumber
 				u32 blocknumber = (cmd64>>44)&0xFFFF;
@@ -142,10 +142,10 @@ void Slot1Comp_Protocol::write_command_NORMAL(GC_Command command)
 			operation = eSlot1Operation_B7_Read;
 
 			//TODO - more endian-safe way of doing this (theres examples in R4)
-#ifdef LOCAL_LE
-			u64 cmd64 = bswap64(*(u64*)command.bytes);
-#else
+#ifdef MSB_FIRST
 			u64 cmd64 = *(u64*)command.bytes;
+#else
+			u64 cmd64 = bswap64(*(u64*)command.bytes);
 #endif
 			address = (u32)((cmd64 >> 24));
 			length = 0x200;
diff --git a/desmume/src/armcpu.h b/desmume/src/armcpu.h
index cb9781103..9e7fcddd7 100644
--- a/desmume/src/armcpu.h
+++ b/desmume/src/armcpu.h
@@ -179,7 +179,7 @@ enum Mode
 	SYS = 0x1F
 };
 
-#ifdef WORDS_BIGENDIAN
+#ifdef MSB_FIRST
 typedef union
 {
 	struct
diff --git a/desmume/src/cocoa/cocoa_output.mm b/desmume/src/cocoa/cocoa_output.mm
index 5d1ef23bf..cf75e67bd 100644
--- a/desmume/src/cocoa/cocoa_output.mm
+++ b/desmume/src/cocoa/cocoa_output.mm
@@ -763,7 +763,7 @@
 	
 	pthread_rwlock_unlock(self.rwlockProducer);
 	
-#ifdef __BIG_ENDIAN__
+#ifdef MSB_FIRST
 	uint32_t *bitmapDataEnd = bitmapData + (w * h);
 	while (bitmapData < bitmapDataEnd)
 	{
diff --git a/desmume/src/cocoa/cocoa_rom.mm b/desmume/src/cocoa/cocoa_rom.mm
index d671da76c..ff21c1fa4 100644
--- a/desmume/src/cocoa/cocoa_rom.mm
+++ b/desmume/src/cocoa/cocoa_rom.mm
@@ -435,7 +435,7 @@ static NSMutableDictionary *saveTypeValues = nil;
 	uint32_t *bitmapData = (uint32_t *)[imageRep bitmapData];
 	RomIconToRGBA8888(bitmapData);
 	
-#ifdef __BIG_ENDIAN__
+#ifdef MSB_FIRST
 	uint32_t *bitmapDataEnd = bitmapData + (w * h);
 	while (bitmapData < bitmapDataEnd)
 	{
@@ -718,7 +718,7 @@ void RomIconToRGBA8888(uint32_t *bitmapData)
 				
 				// Set the RGBA8888 bitmap pixels using our CLUT from earlier.
 				
-#ifdef __BIG_ENDIAN__
+#ifdef MSB_FIRST
 				*bitmapPixPtr = LOCAL_TO_LE_32(clut[(pixRowColors & 0x0F000000) >> 24]);
 				
 				bitmapPixPtr++;
diff --git a/desmume/src/cocoa/cocoa_videofilter.mm b/desmume/src/cocoa/cocoa_videofilter.mm
index 8b251f5ca..cced55964 100644
--- a/desmume/src/cocoa/cocoa_videofilter.mm
+++ b/desmume/src/cocoa/cocoa_videofilter.mm
@@ -141,7 +141,7 @@
 	uint32_t *bitmapData = (uint32_t *)[imageRep bitmapData];
 	RGBA8888ForceOpaqueBuffer((const uint32_t *)[self runFilter], bitmapData, (w * h));
 	
-#ifdef __BIG_ENDIAN__
+#ifdef MSB_FIRST
 	uint32_t *bitmapDataEnd = bitmapData + (w * h);
 	while (bitmapData < bitmapDataEnd)
 	{
diff --git a/desmume/src/ctrlssdl.cpp b/desmume/src/ctrlssdl.cpp
index 03c269148..4a7fe677e 100644
--- a/desmume/src/ctrlssdl.cpp
+++ b/desmume/src/ctrlssdl.cpp
@@ -342,7 +342,7 @@ u16 get_keypad( void)
   u16 keypad;
   keypad = ~MMU.ARM7_REG[0x136];
   keypad = (keypad & 0x3) << 10;
-#ifdef WORDS_BIGENDIAN
+#ifdef MSB_FIRST
   keypad |= ~(MMU.ARM9_REG[0x130] | (MMU.ARM9_REG[0x131] << 8)) & 0x3FF;
 #else
   keypad |= ~((u16 *)MMU.ARM9_REG)[0x130>>1] & 0x3FF;
diff --git a/desmume/src/filter/2xsai.cpp b/desmume/src/filter/2xsai.cpp
index 5cd35a734..2ff65f672 100644
--- a/desmume/src/filter/2xsai.cpp
+++ b/desmume/src/filter/2xsai.cpp
@@ -247,7 +247,7 @@ void Super2xSaI (u8 *srcPtr, u32 srcPitch,
             else
               product1a = color5;
 
-#ifdef WORDS_BIGENDIAN
+#ifdef MSB_FIRST
           product1a = (product1a << 16) | product1b;
           product2a = (product2a << 16) | product2b;
 #else
@@ -505,7 +505,7 @@ void SuperEagle (u8 *srcPtr, u32 srcPitch, u8 * /* deltaPtr */,
           //                    product2a = color2;
           //                    product2b = color3;
         }
-#ifdef WORDS_BIGENDIAN
+#ifdef MSB_FIRST
         product1a = (product1a << 16) | product1b;
         product2a = (product2a << 16) | product2b;
 #else
@@ -790,7 +790,7 @@ void _2xSaI (u8 *srcPtr, u32 srcPitch, u8 * /* deltaPtr */,
           }
         }
 
-#ifdef WORDS_BIGENDIAN
+#ifdef MSB_FIRST
         product = (colorA << 16) | product ;
         product1 = (product1 << 16) | product2 ;
 #else
diff --git a/desmume/src/filter/scanline.cpp b/desmume/src/filter/scanline.cpp
index 479cb4603..52d2e6843 100644
--- a/desmume/src/filter/scanline.cpp
+++ b/desmume/src/filter/scanline.cpp
@@ -29,22 +29,7 @@ FORCEINLINE void ScanLine32( uint32 *lpDst, uint32 *lpSrc, unsigned int Width, i
 {
 	while(Width--)
 	{
-#ifdef LOCAL_LE
-		u8* u8dst = (u8*)lpDst;
-		u8* u8src = (u8*)lpSrc;
-		*u8dst++ = *u8src++ * fac_left / 16;
-		*u8dst++ = *u8src++ * fac_left / 16;
-		*u8dst++ = *u8src++ * fac_left / 16;
-		 u8dst++;
-		
-		u8src = (u8*)lpSrc;
-		*u8dst++ = *u8src++ * fac_right / 16;
-		*u8dst++ = *u8src++ * fac_right / 16;
-		*u8dst++ = *u8src++ * fac_right / 16;
-		 u8dst++;   u8src++;
-		lpDst+=2; 
-		lpSrc++;
-#else
+#ifdef MSB_FIRST
 		u8* u8dst = (u8*)lpDst;
 		u8* u8src = (u8*)lpSrc;
 		 u8dst++;   u8src++;
@@ -59,6 +44,21 @@ FORCEINLINE void ScanLine32( uint32 *lpDst, uint32 *lpSrc, unsigned int Width, i
 		*u8dst++ = *u8src++ * fac_right / 16;
 		lpDst+=2;
 		lpSrc++;
+#else
+		u8* u8dst = (u8*)lpDst;
+		u8* u8src = (u8*)lpSrc;
+		*u8dst++ = *u8src++ * fac_left / 16;
+		*u8dst++ = *u8src++ * fac_left / 16;
+		*u8dst++ = *u8src++ * fac_left / 16;
+		 u8dst++;
+		
+		u8src = (u8*)lpSrc;
+		*u8dst++ = *u8src++ * fac_right / 16;
+		*u8dst++ = *u8src++ * fac_right / 16;
+		*u8dst++ = *u8src++ * fac_right / 16;
+		 u8dst++;   u8src++;
+		lpDst+=2; 
+		lpSrc++;
 #endif
 	}
 }
diff --git a/desmume/src/mem.h b/desmume/src/mem.h
index 3c1f2b8bc..ffe0c68a3 100644
--- a/desmume/src/mem.h
+++ b/desmume/src/mem.h
@@ -42,7 +42,7 @@ static INLINE u8 T1ReadByte(u8* const mem, const u32 addr)
 static INLINE u16 T1ReadWord_guaranteedAligned(void* const mem, const u32 addr)
 {
 	assert((addr&1)==0);
-#ifdef WORDS_BIGENDIAN
+#ifdef MSB_FIRST
    return (((u8*)mem)[addr + 1] << 8) | ((u8*)mem)[addr];
 #else
    return *(u16*)((u8*)mem + addr);
@@ -51,7 +51,7 @@ static INLINE u16 T1ReadWord_guaranteedAligned(void* const mem, const u32 addr)
 
 static INLINE u16 T1ReadWord(void* const mem, const u32 addr)
 {
-#ifdef WORDS_BIGENDIAN
+#ifdef MSB_FIRST
    return (((u8*)mem)[addr + 1] << 8) | ((u8*)mem)[addr];
 #else
    return *((u16 *) ((u8*)mem + addr));
@@ -61,7 +61,7 @@ static INLINE u16 T1ReadWord(void* const mem, const u32 addr)
 static INLINE u32 T1ReadLong_guaranteedAligned(u8* const  mem, const u32 addr)
 {
 	assert((addr&3)==0);
-#ifdef WORDS_BIGENDIAN
+#ifdef MSB_FIRST
    return (mem[addr + 3] << 24 | mem[addr + 2] << 16 |
            mem[addr + 1] << 8 | mem[addr]);
 #else
@@ -73,7 +73,7 @@ static INLINE u32 T1ReadLong_guaranteedAligned(u8* const  mem, const u32 addr)
 static INLINE u32 T1ReadLong(u8* const  mem, u32 addr)
 {
    addr &= ~3;
-#ifdef WORDS_BIGENDIAN
+#ifdef MSB_FIRST
    return (mem[addr + 3] << 24 | mem[addr + 2] << 16 |
            mem[addr + 1] << 8 | mem[addr]);
 #else
@@ -83,7 +83,7 @@ static INLINE u32 T1ReadLong(u8* const  mem, u32 addr)
 
 static INLINE u64 T1ReadQuad(u8* const mem, const u32 addr)
 {
-#ifdef WORDS_BIGENDIAN
+#ifdef MSB_FIRST
    return (u64(mem[addr + 7]) << 56 | u64(mem[addr + 6]) << 48 |
            u64(mem[addr + 5]) << 40 | u64(mem[addr + 4]) << 32 |
            u64(mem[addr + 3]) << 24 | u64(mem[addr + 2]) << 16 |
@@ -100,7 +100,7 @@ static INLINE void T1WriteByte(u8* const mem, const u32 addr, const u8 val)
 
 static INLINE void T1WriteWord(u8* const mem, const u32 addr, const u16 val)
 {
-#ifdef WORDS_BIGENDIAN
+#ifdef MSB_FIRST
    mem[addr + 1] = val >> 8;
    mem[addr] = val & 0xFF;
 #else
@@ -110,7 +110,7 @@ static INLINE void T1WriteWord(u8* const mem, const u32 addr, const u16 val)
 
 static INLINE void T1WriteLong(u8* const mem, const u32 addr, const u32 val)
 {
-#ifdef WORDS_BIGENDIAN
+#ifdef MSB_FIRST
    mem[addr + 3] = val >> 24;
    mem[addr + 2] = (val >> 16) & 0xFF;
    mem[addr + 1] = (val >> 8) & 0xFF;
@@ -122,7 +122,7 @@ static INLINE void T1WriteLong(u8* const mem, const u32 addr, const u32 val)
 
 static INLINE void T1WriteQuad(u8* const mem, const u32 addr, const u64 val)
 {
-#ifdef WORDS_BIGENDIAN
+#ifdef MSB_FIRST
 	mem[addr + 7] = (val >> 56);
 	mem[addr + 6] = (val >> 48) & 0xFF;
 	mem[addr + 5] = (val >> 40) & 0xFF;
@@ -138,7 +138,7 @@ static INLINE void T1WriteQuad(u8* const mem, const u32 addr, const u64 val)
 
 //static INLINE u8 T2ReadByte(u8* const  mem, const u32 addr)
 //{
-//#ifdef WORDS_BIGENDIAN
+//#ifdef MSB_FIRST
 //   return mem[addr ^ 1];
 //#else
 //   return mem[addr];
@@ -154,7 +154,7 @@ static INLINE u16 HostReadWord(u8* const mem, const u32 addr)
 //
 //static INLINE u32 T2ReadLong(u8* const mem, const u32 addr)
 //{
-//#ifdef WORDS_BIGENDIAN
+//#ifdef MSB_FIRST
 //   return *((u16 *) (mem + addr + 2)) << 16 | *((u16 *) (mem + addr));
 //#else
 //   return *((u32 *) (mem + addr));
@@ -163,7 +163,7 @@ static INLINE u16 HostReadWord(u8* const mem, const u32 addr)
 //
 //static INLINE void T2WriteByte(u8* const mem, const u32 addr, const u8 val)
 //{
-//#ifdef WORDS_BIGENDIAN
+//#ifdef MSB_FIRST
 //   mem[addr ^ 1] = val;
 //#else
 //   mem[addr] = val;
@@ -183,7 +183,7 @@ static INLINE void HostWriteLong(u8* const mem, const u32 addr, const u32 val)
 
 static INLINE void HostWriteTwoWords(u8* const mem, const u32 addr, const u32 val)
 {
-#ifdef WORDS_BIGENDIAN
+#ifdef MSB_FIRST
    *((u16 *) (mem + addr + 2)) = val >> 16;
    *((u16 *) (mem + addr)) = val & 0xFFFF;
 #else
diff --git a/desmume/src/rasterize.cpp b/desmume/src/rasterize.cpp
index 62a384e40..76cab1e8e 100644
--- a/desmume/src/rasterize.cpp
+++ b/desmume/src/rasterize.cpp
@@ -84,7 +84,7 @@ static u8 decal_table[32][64][64];
 //	const double _double2fixmagic = 68719476736.0*1.5;     //2^36 * 1.5,  (52-_shiftamt=36) uses limited precisicion to floor
 //	const int _shiftamt        = 16;                    //16.16 fixed point representation,
 //
-//	#ifdef WORDS_BIGENDIAN
+//	#ifdef MSB_FIRST
 //		#define iman_				1
 //	#else
 //		#define iman_				0
diff --git a/desmume/src/saves.cpp b/desmume/src/saves.cpp
index 8114c2f15..558b6f191 100644
--- a/desmume/src/saves.cpp
+++ b/desmume/src/saves.cpp
@@ -777,10 +777,7 @@ static bool ReadStateChunk(EMUFILE* is, const SFORMAT *sf, int size)
 
 		if((tmp=CheckS(guessSF,sf,sz,count,toa)))
 		{
-		#ifdef LOCAL_LE
-			// no need to ever loop one at a time if not flipping byte order
-			is->fread((char *)tmp->v,sz*count);
-		#else
+		#ifdef MSB_FIRST
 			if(sz == 1) {
 				//special case: read a huge byte array
 				is->fread((char *)tmp->v,count);
@@ -791,6 +788,9 @@ static bool ReadStateChunk(EMUFILE* is, const SFORMAT *sf, int size)
                     FlipByteOrder((u8*)tmp->v + i*sz,sz);
 				}
 			}
+		#else
+			// no need to ever loop one at a time if not flipping byte order
+			is->fread((char *)tmp->v,sz*count);
 		#endif
 			guessSF = tmp + 1;
 		}
@@ -863,10 +863,7 @@ static int SubWrite(EMUFILE* os, const SFORMAT *sf)
 			#endif
 
 
-		#ifdef LOCAL_LE
-			// no need to ever loop one at a time if not flipping byte order
-			os->fwrite((char *)sf->v,size*count);
-		#else
+		#ifdef MSB_FIRST
 			if(size == 1) {
 				//special case: write a huge byte array
 				os->fwrite((char *)sf->v,count);
@@ -878,6 +875,9 @@ static int SubWrite(EMUFILE* os, const SFORMAT *sf)
 					FlipByteOrder((u8*)sf->v + i*size, size);
 				}
 			}
+		#else
+			// no need to ever loop one at a time if not flipping byte order
+			os->fwrite((char *)sf->v,size*count);
 		#endif
 		}
 		sf++;
diff --git a/desmume/src/texcache.cpp b/desmume/src/texcache.cpp
index a936a5c1e..f5ec7dcde 100644
--- a/desmume/src/texcache.cpp
+++ b/desmume/src/texcache.cpp
@@ -296,7 +296,7 @@ public:
 		//dump the palette to a temp buffer, so that we don't have to worry about memory mapping.
 		//this isnt such a problem with texture memory, because we read sequentially from it.
 		//however, we read randomly from palette memory, so the mapping is more costly.
-		#ifdef WORDS_BIGENDIAN
+		#ifdef MSB_FIRST
 			mspal.dump16(pal);
 		#else
 			mspal.dump(pal);
@@ -764,7 +764,7 @@ public:
 								break;
 								
 							case 1:
-#ifdef LOCAL_BE
+#ifdef MSB_FIRST
 								tmp_col[2]	= ( (((tmp_col[0] & 0xFF000000) >> 1)+((tmp_col[1] & 0xFF000000)  >> 1)) & 0xFF000000 ) |
 											  ( (((tmp_col[0] & 0x00FF0000)      + (tmp_col[1] & 0x00FF0000)) >> 1)  & 0x00FF0000 ) |
 											  ( (((tmp_col[0] & 0x0000FF00)      + (tmp_col[1] & 0x0000FF00)) >> 1)  & 0x0000FF00 ) |
@@ -785,7 +785,7 @@ public:
 								
 							case 3:
 							{
-#ifdef LOCAL_BE
+#ifdef MSB_FIRST
 								const u32 r0	= (tmp_col[0]>>24) & 0x000000FF;
 								const u32 r1	= (tmp_col[1]>>24) & 0x000000FF;
 								const u32 g0	= (tmp_col[0]>>16) & 0x000000FF;
@@ -818,7 +818,7 @@ public:
 						{
 							for (size_t i = 0; i < 4; i++)
 							{
-#ifdef LOCAL_BE
+#ifdef MSB_FIRST
 								const u32 a = (tmp_col[i] >> 3) & 0x0000001F;
 								tmp_col[i] >>= 2;
 								tmp_col[i] &= 0x3F3F3F00;
diff --git a/desmume/src/types.h b/desmume/src/types.h
index 7dbb5478b..7d46faa0c 100644
--- a/desmume/src/types.h
+++ b/desmume/src/types.h
@@ -265,20 +265,8 @@ typedef int desmume_BOOL;
 #define FALSE 0
 #endif
 
-#ifdef __BIG_ENDIAN__
-#ifndef WORDS_BIGENDIAN
-#define WORDS_BIGENDIAN
-#endif
-#endif
-
-#ifdef WORDS_BIGENDIAN
-# define LOCAL_BE 1
-#else
-# define LOCAL_LE 1
-#endif
-
 /* little endian (ds' endianess) to local endianess convert macros */
-#ifdef LOCAL_BE	/* local arch is big endian */
+#ifdef MSB_FIRST	/* local arch is big endian */
 # define LE_TO_LOCAL_16(x) ((((x)&0xff)<<8)|(((x)>>8)&0xff))
 # define LE_TO_LOCAL_32(x) ((((x)&0xff)<<24)|(((x)&0xff00)<<8)|(((x)>>8)&0xff00)|(((x)>>24)&0xff))
 # define LE_TO_LOCAL_64(x) ((((x)&0xff)<<56)|(((x)&0xff00)<<40)|(((x)&0xff0000)<<24)|(((x)&0xff000000)<<8)|(((x)>>8)&0xff000000)|(((x)>>24)&0xff0000)|(((x)>>40)&0xff00)|(((x)>>56)&0xff))
diff --git a/desmume/src/windows/File_Extractor/7z_C/CpuArch.h b/desmume/src/windows/File_Extractor/7z_C/CpuArch.h
index 0a709bb26..449dbe815 100644
--- a/desmume/src/windows/File_Extractor/7z_C/CpuArch.h
+++ b/desmume/src/windows/File_Extractor/7z_C/CpuArch.h
@@ -52,7 +52,7 @@ If MY_CPU_LE_UNALIGN is not defined, we don't know about these properties of pla
 #define MY_CPU_LE
 #endif
 
-#if defined(__BIG_ENDIAN__)
+#if defined(MSB_FIRST)
 #define MY_CPU_BE
 #endif
 
diff --git a/desmume/src/windows/File_Extractor/fex/blargg_endian.h b/desmume/src/windows/File_Extractor/fex/blargg_endian.h
index d012871f2..413a9e625 100644
--- a/desmume/src/windows/File_Extractor/fex/blargg_endian.h
+++ b/desmume/src/windows/File_Extractor/fex/blargg_endian.h
@@ -36,7 +36,7 @@
 	#define BLARGG_LITTLE_ENDIAN 1
 #endif
 
-#if defined (MSB_FIRST)     || defined (__BIG_ENDIAN__) || defined (WORDS_BIGENDIAN) || \
+#if defined (MSB_FIRST)     || defined (__BIG_ENDIAN__) || \
 	defined (__sparc__)     ||  BLARGG_CPU_POWERPC || \
 	(defined (BIG_ENDIAN) && BIG_ENDIAN+0 != 4321)
 	#define BLARGG_BIG_ENDIAN 1