diff --git a/GPU/Common/IndexGenerator.cpp b/GPU/Common/IndexGenerator.cpp
index 9c1b13d42db8..06bef591df11 100644
--- a/GPU/Common/IndexGenerator.cpp
+++ b/GPU/Common/IndexGenerator.cpp
@@ -218,7 +218,7 @@ void IndexGenerator::TranslateList(int numInds, const ITypeLE *inds, int indexOf
 	indexOffset = index_ - indexOffset;
 	// We only bother doing this minor optimization in triangle list, since it's by far the most
 	// common operation that can benefit.
-	if (sizeof(ITypeLE) == sizeof(inds_[0]) && indexOffset == 0) {
+	if (sizeof(ITypeLE) == sizeof(inds_[0]) && indexOffset == 0 && clockwise) {
 		memcpy(inds_, inds, numInds * sizeof(ITypeLE));
 		inds_ += numInds;
 		count_ += numInds;
diff --git a/GPU/GPUCommon.cpp b/GPU/GPUCommon.cpp
index dc53f28cdd20..2030f23f01fa 100644
--- a/GPU/GPUCommon.cpp
+++ b/GPU/GPUCommon.cpp
@@ -1627,8 +1627,13 @@ void GPUCommon::Execute_Prim(u32 op, u32 diff) {
 			break;
 		}
 		case GE_CMD_VADDR:
+			gstate.cmdmem[data >> 24] = data;
 			gstate_c.vertexAddr = gstate_c.getRelativeAddress(data & 0x00FFFFFF);
 			break;
+		case GE_CMD_IADDR:
+			gstate.cmdmem[data >> 24] = data;
+			gstate_c.indexAddr = gstate_c.getRelativeAddress(data & 0x00FFFFFF);
+			break;
 		case GE_CMD_OFFSETADDR:
 			gstate.cmdmem[GE_CMD_OFFSETADDR] = data;
 			gstate_c.offsetAddr = data << 8;
@@ -1646,8 +1651,10 @@ void GPUCommon::Execute_Prim(u32 op, u32 diff) {
 			// flip face by indices for triangles
 			cullMode = data & 1;
 			break;
+		case GE_CMD_TEXFLUSH:
 		case GE_CMD_NOP:
 		case GE_CMD_NOP_FF:
+			gstate.cmdmem[data >> 24] = data;
 			break;
 		case GE_CMD_BONEMATRIXNUMBER:
 			gstate.cmdmem[GE_CMD_BONEMATRIXNUMBER] = data;
@@ -1690,6 +1697,12 @@ void GPUCommon::Execute_Prim(u32 op, u32 diff) {
 			break;
 		}
 
+		case GE_CMD_TEXBUFWIDTH0:
+		case GE_CMD_TEXADDR0:
+			if (data != gstate.cmdmem[data >> 24])
+				goto bail;
+			break;
+
 		default:
 			// All other commands might need a flush or something, stop this inner loop.
 			goto bail;