diff --git a/plugins/GSdx/GSLocalMemory.cpp b/plugins/GSdx/GSLocalMemory.cpp
index e475697929..d62a37b114 100644
--- a/plugins/GSdx/GSLocalMemory.cpp
+++ b/plugins/GSdx/GSLocalMemory.cpp
@@ -465,20 +465,9 @@ GSLocalMemory::~GSLocalMemory()
 {
 	VirtualFree(m_vm8, 0, MEM_RELEASE);
 
-	for(hash_map<uint32, BlockOffset*>::iterator i = m_bomap.begin(); i != m_bomap.end(); i++)
-	{
-		_aligned_free(i->second);
-	}
-
-	for(hash_map<uint32, PixelOffset*>::iterator i = m_pomap.begin(); i != m_pomap.end(); i++)
-	{
-		_aligned_free(i->second);
-	}
-
-	for(hash_map<uint32, PixelOffset4*>::iterator i = m_po4map.begin(); i != m_po4map.end(); i++)
-	{
-		_aligned_free(i->second);
-	}
+	for_each(m_bomap.begin(), m_bomap.end(), aligned_free_second());
+	for_each(m_pomap.begin(), m_pomap.end(), aligned_free_second());
+	for_each(m_po4map.begin(), m_po4map.end(), aligned_free_second());
 }
 
 GSLocalMemory::BlockOffset* GSLocalMemory::GetBlockOffset(uint32 bp, uint32 bw, uint32 psm)
diff --git a/plugins/GSdx/GSState.cpp b/plugins/GSdx/GSState.cpp
index dbb7dd9f24..a64b465165 100644
--- a/plugins/GSdx/GSState.cpp
+++ b/plugins/GSdx/GSState.cpp
@@ -1163,57 +1163,88 @@ void GSState::Move()
 	}
 	else if(m_env.BITBLTBUF.SPSM == PSM_PSMT8 && m_env.BITBLTBUF.DPSM == PSM_PSMT8)
 	{
-		for(int y = 0; y < h; y++, sy += yinc, dy += yinc)
+		if(xinc > 0)
 		{
-			uint8* RESTRICT s = &m_mem.m_vm8[spo->row[sy]];
-			uint8* RESTRICT d = &m_mem.m_vm8[dpo->row[dy]];
-
-			int* RESTRICT scol = &spo->col[sy & 7][sx];
-			int* RESTRICT dcol = &dpo->col[dy & 7][dx];
-
-			for(int x = 0; x < w; x++, scol += xinc, dcol += xinc)
+			for(int y = 0; y < h; y++, sy += yinc, dy += yinc)
 			{
-				d[*dcol] = s[*scol];
+				uint8* RESTRICT s = &m_mem.m_vm8[spo->row[sy]];
+				uint8* RESTRICT d = &m_mem.m_vm8[dpo->row[dy]];
+
+				int* RESTRICT scol = &spo->col[sy & 7][sx];
+				int* RESTRICT dcol = &dpo->col[dy & 7][dx];
+
+				for(int x = 0; x < w; x++) d[dcol[x]] = s[scol[x]];
+			}
+		}
+		else
+		{
+			for(int y = 0; y < h; y++, sy += yinc, dy += yinc)
+			{
+				uint8* RESTRICT s = &m_mem.m_vm8[spo->row[sy]];
+				uint8* RESTRICT d = &m_mem.m_vm8[dpo->row[dy]];
+
+				int* RESTRICT scol = &spo->col[sy & 7][sx];
+				int* RESTRICT dcol = &dpo->col[dy & 7][dx];
+
+				for(int x = 0; x > -w; x--) d[dcol[x]] = s[scol[x]];
 			}
 		}
 	}
 	else if(m_env.BITBLTBUF.SPSM == PSM_PSMT4 && m_env.BITBLTBUF.DPSM == PSM_PSMT4)
 	{
-		for(int y = 0; y < h; y++, sy += yinc, dy += yinc, sx -= xinc * w, dx -= xinc * w)
+		if(xinc > 0)
 		{
-			uint8* RESTRICT s = &m_mem.m_vm8[spo->row[sy] >> 1];
-			uint8* RESTRICT d = &m_mem.m_vm8[dpo->row[dy] >> 1];
-
-			int* RESTRICT scol = &spo->col[sy & 7][sx];
-			int* RESTRICT dcol = &dpo->col[dy & 7][dx];
-			
-			for(int x = 0; x < w; x += 2)
+			for(int y = 0; y < h; y++, sy += yinc, dy += yinc)
 			{
-				d[*dcol >> 1] = (d[*dcol >> 1] & 0xf0) | (s[*scol >> 1] & 0x0f);
+				uint32 sbase = spo->row[sy];
+				uint32 dbase = dpo->row[dy];
 
-				scol += xinc;
-				dcol += xinc;
+				int* RESTRICT scol = &spo->col[sy & 7][sx];
+				int* RESTRICT dcol = &dpo->col[dy & 7][dx];
+				
+				for(int x = 0; x < w; x++) m_mem.WritePixel4(dbase + dcol[x], m_mem.ReadPixel4(sbase + scol[x]));
+			}
+		}
+		else
+		{
+			for(int y = 0; y < h; y++, sy += yinc, dy += yinc)
+			{
+				uint32 sbase = spo->row[sy];
+				uint32 dbase = dpo->row[dy];
 
-				d[*dcol >> 1] = (d[*dcol >> 1] & 0x0f) | (s[*scol >> 1] & 0xf0);
-
-				scol += xinc;
-				dcol += xinc;
+				int* RESTRICT scol = &spo->col[sy & 7][sx];
+				int* RESTRICT dcol = &dpo->col[dy & 7][dx];
+				
+				for(int x = 0; x > -w; x--) m_mem.WritePixel4(dbase + dcol[x], m_mem.ReadPixel4(sbase + scol[x]));
 			}
 		}
 	}
 	else
 	{
-		for(int y = 0; y < h; y++, sy += yinc, dy += yinc, sx -= xinc * w, dx -= xinc * w)
+		if(xinc > 0)
 		{
-			uint32 sbase = spo->row[sy];
-			uint32 dbase = dpo->row[dy];
-
-			int* RESTRICT scol = spo->col[sy & 7];
-			int* RESTRICT dcol = dpo->col[dy & 7];
-			
-			for(int x = 0; x < w; x++, sx += xinc, dx += xinc)
+			for(int y = 0; y < h; y++, sy += yinc, dy += yinc)
 			{
-				(m_mem.*dpsm.wpa)(dbase + dcol[dx], (m_mem.*spsm.rpa)(sbase + scol[sx]));
+				uint32 sbase = spo->row[sy];
+				uint32 dbase = dpo->row[dy];
+
+				int* RESTRICT scol = &spo->col[sy & 7][sx];
+				int* RESTRICT dcol = &dpo->col[dy & 7][dx];
+				
+				for(int x = 0; x < w; x++) (m_mem.*dpsm.wpa)(dbase + dcol[x], (m_mem.*spsm.rpa)(sbase + scol[x]));
+			}
+		}
+		else
+		{
+			for(int y = 0; y < h; y++, sy += yinc, dy += yinc)
+			{
+				uint32 sbase = spo->row[sy];
+				uint32 dbase = dpo->row[dy];
+
+				int* RESTRICT scol = &spo->col[sy & 7][sx];
+				int* RESTRICT dcol = &dpo->col[dy & 7][dx];
+				
+				for(int x = 0; x > -w; x--) (m_mem.*dpsm.wpa)(dbase + dcol[x], (m_mem.*spsm.rpa)(sbase + scol[x]));
 			}
 		}
 	}
diff --git a/plugins/GSdx/stdafx.h b/plugins/GSdx/stdafx.h
index fd3a16579b..510d2c3231 100644
--- a/plugins/GSdx/stdafx.h
+++ b/plugins/GSdx/stdafx.h
@@ -63,6 +63,9 @@ extern string format(const char* fmt, ...);
 struct delete_object {template<class T> void operator()(T& p) {delete p;}};
 struct delete_first {template<class T> void operator()(T& p) {delete p.first;}};
 struct delete_second {template<class T> void operator()(T& p) {delete p.second;}};
+struct aligned_free_object {template<class T> void operator()(T& p) {_aligned_free(p);}};
+struct aligned_free_first {template<class T> void operator()(T& p) {_aligned_free(p.first);}};
+struct aligned_free_second {template<class T> void operator()(T& p) {_aligned_free(p.second);}};
 
 // syntactic sugar