diff --git a/include/PSX/GTE/gte.hpp b/include/PSX/GTE/gte.hpp
index 0c9b3335..979fc6b7 100644
--- a/include/PSX/GTE/gte.hpp
+++ b/include/PSX/GTE/gte.hpp
@@ -55,6 +55,38 @@ namespace JabyEngine {
             __asm__ volatile("ctc2 $14, $7"     :: "r"(&matrix) : "$12", "$13", "$14");
         }
 
+        /*
+            MulMatrix0
+
+            m0: first input
+            m1: second input
+            result: result of multiplication
+            returns: result
+
+            Multiplies two matrices m0 and m1.
+            The function destroys the constant rotation matrix
+        */
+        static MATRIX& mult_matrix(const MATRIX& m0, const MATRIX& m1, MATRIX& result) {
+            /*
+                Jaby: Somehow this code creates stack usage.... Investigate!!
+            */
+            set_rot_matrix(m0);
+            
+            ldclmv(m1, 0);
+            rtir();
+            stclmv(result, 0);
+            
+            ldclmv(m1, 1);
+            rtir();
+            stclmv(result, 1);
+            
+            ldclmv(m1, 2);
+            rtir();
+            stclmv(result, 2);
+
+            return result;
+        }
+
         /*
             SetGeomOffset(ofx,ofy)
 
diff --git a/include/PSX/GTE/gte_instruction.hpp b/include/PSX/GTE/gte_instruction.hpp
index 5fd9e707..d21dc3df 100644
--- a/include/PSX/GTE/gte_instruction.hpp
+++ b/include/PSX/GTE/gte_instruction.hpp
@@ -21,13 +21,33 @@ namespace JabyEngine {
             __asm__ volatile("lwc2 $5, 4(%0)" :: "r"(&vector));
         }
 
+        // Load column vector of MATRIX to universal register
+        static __always_inline void ldclmv(const MATRIX& matrix, size_t col) {
+            __asm__ volatile("lhu  $12, 0(%0)"  :: "r"(reinterpret_cast<uintptr_t>(&matrix) + (col << 1)) : "$12", "$13", "$14");
+            __asm__ volatile("lhu  $13, 6(%0)"  :: "r"(reinterpret_cast<uintptr_t>(&matrix) + (col << 1)) : "$12", "$13", "$14");
+            __asm__ volatile("lhu  $14, 12(%0)" :: "r"(reinterpret_cast<uintptr_t>(&matrix) + (col << 1)) : "$12", "$13", "$14");
+            __asm__ volatile("mtc2 $12, $9"     :: "r"(reinterpret_cast<uintptr_t>(&matrix) + (col << 1)) : "$12", "$13", "$14");
+            __asm__ volatile("mtc2 $13, $10"    :: "r"(reinterpret_cast<uintptr_t>(&matrix) + (col << 1)) : "$12", "$13", "$14");
+            __asm__ volatile("mtc2 $14, $11"    :: "r"(reinterpret_cast<uintptr_t>(&matrix) + (col << 1)) : "$12", "$13", "$14");
+        }
+
         // Store flag
         static __always_inline void stflg(int32_t& flag) {
             __asm__ volatile("cfc2 $12, $31"   :: "r"(&flag) : "$12", "memory");
             __asm__ volatile("nop"             :: "r"(&flag) : "$12", "memory");
             __asm__ volatile("sw   $12, 0(%0)" :: "r"(&flag) : "$12", "memory");
         }
-        
+
+        // Store MATRIX column from 16 bit universal register
+        static __always_inline void stclmv(MATRIX& matrix, size_t col) {
+            __asm__ volatile("mfc2 $12, $9"     :: "r"(reinterpret_cast<uintptr_t>(&matrix) + (col << 1)) : "$12", "$13", "$14", "memory");
+            __asm__ volatile("mfc2 $13, $10"    :: "r"(reinterpret_cast<uintptr_t>(&matrix) + (col << 1)) : "$12", "$13", "$14", "memory");
+            __asm__ volatile("mfc2 $14, $11"    :: "r"(reinterpret_cast<uintptr_t>(&matrix) + (col << 1)) : "$12", "$13", "$14", "memory");
+            __asm__ volatile("sh   $12, 0(%0)"  :: "r"(reinterpret_cast<uintptr_t>(&matrix) + (col << 1)) : "$12", "$13", "$14", "memory");
+            __asm__ volatile("sh   $13, 6(%0)"  :: "r"(reinterpret_cast<uintptr_t>(&matrix) + (col << 1)) : "$12", "$13", "$14", "memory");
+            __asm__ volatile("sh   $14, 12(%0)" :: "r"(reinterpret_cast<uintptr_t>(&matrix) + (col << 1)) : "$12", "$13", "$14", "memory");
+        }
+
         // Store VECTOR from 32 bit universal register
         static __always_inline void stlvnl(VECTOR& out_vector) {
             __asm__ volatile("swc2 $25, 0(%0)" :: "r"(&out_vector) : "memory");
@@ -44,5 +64,15 @@ namespace JabyEngine {
             __asm__ volatile("nop");
             __asm__ volatile("cop2 0x0480012");
         }
+
+        /*
+            Variation of gte_rt
+            (Rotation Matrix)*(16 bit universal vector)
+        */
+        static __always_inline void rtir() {
+            __asm__ volatile("nop");
+            __asm__ volatile("nop");
+            __asm__ volatile("cop2 0x049E012");
+       }
     }
 }
\ No newline at end of file