diff --git a/src/xenia/base/memory_generic.cc b/src/xenia/base/memory_generic.cc
index b25327770..e5c2d0957 100644
--- a/src/xenia/base/memory_generic.cc
+++ b/src/xenia/base/memory_generic.cc
@@ -44,10 +44,12 @@ void copy_and_swap_16_aligned(uint16_t* dest, const uint16_t* src,
 void copy_and_swap_16_unaligned(uint16_t* dest, const uint16_t* src,
                                 size_t count) {
   size_t i;
+  __m128i input, output;
+
   for (i = 0; i + 8 <= count; i += 8) {
-    __m128i s = _mm_loadu_si128((__m128i*)&src[i]);
-    __m128i d = _mm_or_si128(_mm_slli_epi16(s, 8), _mm_srli_epi16(s, 8));
-    _mm_storeu_si128((__m128i*)&dest[i], d);
+    input = _mm_loadu_si128((__m128i*)&src[i]);
+    output = _mm_or_si128(_mm_slli_epi16(input, 8), _mm_srli_epi16(input, 8));
+    _mm_storeu_si128((__m128i*)&dest[i], output);
   }
 
   for (; i < count; ++i) {  // handle residual elements
@@ -62,7 +64,31 @@ void copy_and_swap_32_aligned(uint32_t* dest, const uint32_t* src,
 
 void copy_and_swap_32_unaligned(uint32_t* dest, const uint32_t* src,
                                 size_t count) {
-  for (size_t i = 0; i < count; ++i) {
+  size_t i;
+  __m128i input, byte1, byte2, byte3, byte4, output;
+  __m128i byte2mask = _mm_set1_epi32(0x00FF0000);
+  __m128i byte3mask = _mm_set1_epi32(0x0000FF00);
+
+  for (i = 0; i + 4 <= count; i += 4) {
+    input = _mm_loadu_si128((__m128i*)&src[i]);
+
+    // Do the four shifts
+    byte1 = _mm_slli_epi32(input, 24);
+    byte2 = _mm_slli_epi32(input, 8);
+    byte3 = _mm_srli_epi32(input, 8);
+    byte4 = _mm_srli_epi32(input, 24);
+
+    // Or bytes together
+    output = _mm_or_si128(byte1, byte4);
+    byte2 = _mm_and_si128(byte2, byte2mask);
+    output = _mm_or_si128(output, byte2);
+    byte3 = _mm_and_si128(byte3, byte3mask);
+    output = _mm_or_si128(output, byte3);
+
+    _mm_storeu_si128((__m128i*)&dest[i], output);
+  }
+
+  for (; i < count; ++i) {  // handle residual elements
     dest[i] = byte_swap(src[i]);
   }
 }