diff --git a/examples/PoolBox/application/src/Overlay/GTETest/include/GTE_Sprite.hpp b/examples/PoolBox/application/src/Overlay/GTETest/include/GTE_Sprite.hpp
index 8ff06c08..32f8ad7a 100644
--- a/examples/PoolBox/application/src/Overlay/GTETest/include/GTE_Sprite.hpp
+++ b/examples/PoolBox/application/src/Overlay/GTETest/include/GTE_Sprite.hpp
@@ -18,22 +18,19 @@ namespace GTETest {
         }
 
         void apply(const GTE::MATRIX& matrix) {
-            static const auto apply_to = [](const GTE::MATRIX& matrix, GTE::SVECTOR vector) -> GPU::Vertex {
+            static const auto apply_to = [](const GTE::MATRIX& matrix, GPU::Vertex vertex) -> GPU::Vertex {
                 GTE::MATRIX  move_back = GTE::MATRIX{
                     GTE::ROTMATRIX::identity(), GTE::TRANSFERVECTOR::translated(-matrix.transfer.x, -matrix.transfer.y, -matrix.transfer.z)
                 };
-                GTE::SVECTOR output;
-                int32_t     flag;
-
                 GTE::comp_matrix(matrix, move_back, move_back);
-                return GTE::apply_matrix(move_back, vector, output).to<GPU::Vertex>();
+                return GTE::apply_matrix(move_back, vertex, vertex);
             };
 
             const auto& area = this->area;
-            this->display.vertex0 = apply_to(matrix, GTE::SVECTOR::from(area.get_top_left()));
-            this->display.vertex1 = apply_to(matrix, GTE::SVECTOR::from(area.get_top_right()));
-            this->display.vertex2 = apply_to(matrix, GTE::SVECTOR::from(area.get_bottom_left()));
-            this->display.vertex3 = apply_to(matrix, GTE::SVECTOR::from(area.get_bottom_right()));
+            this->display.vertex0 = apply_to(matrix, area.get_top_left());
+            this->display.vertex1 = apply_to(matrix, area.get_top_right());
+            this->display.vertex2 = apply_to(matrix, area.get_bottom_left());
+            this->display.vertex3 = apply_to(matrix, area.get_bottom_right());
         }
 
         void render() {
diff --git a/include/PSX/GTE/gte.hpp b/include/PSX/GTE/gte.hpp
index c5cd2370..7c271963 100644
--- a/include/PSX/GTE/gte.hpp
+++ b/include/PSX/GTE/gte.hpp
@@ -5,6 +5,20 @@ namespace JabyEngine {
     namespace GTE {
         static constexpr auto StackSize = 16;
 
+        /*
+            matrix: first input
+
+            Sets the 3x3 constant rotation matrix and the parallel transfer vector from input 
+        */
+        void set_matrix(const MATRIX& matrix);
+
+        /*
+            returns: current matrix
+
+            Gets the current 3x3 constant rotation matrix and the parallel transfer vector
+        */
+        MATRIX get_matrix();
+
         /*
             RotTrans
 
@@ -99,8 +113,28 @@ namespace JabyEngine {
             returns: result
 
             Applies the matrix to the vector
+            The function destroys the constant rotation matrix and transfer vector
         */
-       SVECTOR& apply_matrix(const MATRIX& m0, const SVECTOR& v0, SVECTOR& v1);
+        static SVECTOR& apply_matrix(const MATRIX& m0, const SVECTOR& v0, SVECTOR& v1) {
+            set_matrix(m0);
+
+            JabyEngine::GTE::ldv0(v0);
+            JabyEngine::GTE::rt();
+            JabyEngine::GTE::stsv(v1);
+            return v1;
+        }
+
+        /*
+            Same as apply_matrix but works on Vertex
+        */
+        static GPU::Vertex& apply_matrix(const MATRIX& m0, const GPU::Vertex& v0, GPU::Vertex& v1) {
+            set_matrix(m0);
+
+            JabyEngine::GTE::ldgv0(v0);
+            JabyEngine::GTE::rt();
+            JabyEngine::GTE::stgv(v1);
+            return v1;
+        }
 
         /*
             MulMatrix0
@@ -133,20 +167,6 @@ namespace JabyEngine {
 			return result;
         } 
 
-        /*
-            matrix: first input
-
-            Sets the 3x3 constant rotation matrix and the parallel transfer vector from input 
-        */
-        void set_matrix(const MATRIX& matrix);
-
-        /*
-            returns: current matrix
-
-            Gets the current 3x3 constant rotation matrix and the parallel transfer vector
-        */
-        MATRIX get_matrix();
-
         /*
             matrix: optional input
 
diff --git a/include/PSX/GTE/gte_instruction.hpp b/include/PSX/GTE/gte_instruction.hpp
index 09735981..167fcbc6 100644
--- a/include/PSX/GTE/gte_instruction.hpp
+++ b/include/PSX/GTE/gte_instruction.hpp
@@ -29,7 +29,12 @@ namespace JabyEngine {
             __asm__ volatile("or   $12, $12, $13" :: "r"(&vector) : "$12", "$13");
             __asm__ volatile("mtc2 $12, $0"       :: "r"(&vector) : "$12", "$13");
             __asm__ volatile("lwc2 $1, 8(%0)"     :: "r"(&vector) : "$12", "$13");
-        
+        }
+
+        // Loads a GPU VERTEX type
+        static __always_inline void ldgv0(const GPU::Vertex& vertex) {
+            __asm__ volatile("lwc2 $0, 0(%0)" :: "r"(&vertex));  
+            __asm__ volatile("lwc2 $1, 0"     :: "r"(&vertex));      
         }
 
         // Load column vector of MATRIX to universal register
@@ -77,6 +82,14 @@ namespace JabyEngine {
             __asm__ volatile("sh   $14, 4(%0)" :: "r"(&out_vector) : "$12", "$13", "$14", "memory");
         }
 
+        // Stores result into a GPU Vertex type
+        static __always_inline void stgv(GPU::Vertex& out_vertex) {
+            __asm__ volatile("mfc2 $12, $9"    :: "r"(&out_vertex) : "$12", "$13", "$14", "memory");
+            __asm__ volatile("mfc2 $13, $10"   :: "r"(&out_vertex) : "$12", "$13", "$14", "memory");
+            __asm__ volatile("sh   $12, 0(%0)" :: "r"(&out_vertex) : "$12", "$13", "$14", "memory");
+            __asm__ volatile("sh   $13, 2(%0)" :: "r"(&out_vertex) : "$12", "$13", "$14", "memory");
+        }
+
         /*
             Kernel of RotTrans
             (Transfer vector)+(Rotation Matrix)*(vertex register 0)
diff --git a/include/PSX/GTE/gte_types.hpp b/include/PSX/GTE/gte_types.hpp
index f75fccba..603ecd96 100644
--- a/include/PSX/GTE/gte_types.hpp
+++ b/include/PSX/GTE/gte_types.hpp
@@ -10,13 +10,14 @@ namespace JabyEngine {
                 T x;
                 T y;
                 T z;
+                T pad;
 
                 static constexpr VECTOR create() {
                     return VECTOR::create(0, 0, 0);
                 }
 
                 static constexpr VECTOR create(T x, T y, T z) {
-                    return VECTOR{.x = x, .y = y, .z = z};
+                    return VECTOR{.x = x, .y = y, .z = z, .pad = 0};
                 }
 
                 template<typename S>
diff --git a/src/Library/src/BootLoader/start_boot.cpp b/src/Library/src/BootLoader/start_boot.cpp
index a6e11586..749c20d3 100644
--- a/src/Library/src/BootLoader/start_boot.cpp
+++ b/src/Library/src/BootLoader/start_boot.cpp
@@ -2,10 +2,7 @@
 #include "../../internal-include/GPU/gpu_internal.hpp"
 #include <stdio.h>
 
-#include <PSX/GTE/gte.hpp>
 #include <PSX/System/syscalls.hpp>
-#include "../../reference/inline_n.h"
-#include <math.h>
 
 extern "C" uint32_t __heap_start;
 extern "C" uint32_t __bss_start;
diff --git a/src/Library/src/GTE/gte.cpp b/src/Library/src/GTE/gte.cpp
index 4a7c3127..77566591 100644
--- a/src/Library/src/GTE/gte.cpp
+++ b/src/Library/src/GTE/gte.cpp
@@ -43,16 +43,6 @@ namespace JabyEngine {
         static MATRIX  Stack[StackSize];
         static MATRIX* FreeStackEntry = Stack;
 
-        SVECTOR& apply_matrix(const MATRIX& m0, const SVECTOR& v0, SVECTOR& v1) {
-            // TODO: Do we want to push here?
-            set_matrix(m0);
-
-            JabyEngine::GTE::ldv0(v0);
-            JabyEngine::GTE::rt();
-            JabyEngine::GTE::stsv(v1);
-            return v1;
-        }
-
         ROTMATRIX& multiply_matrix(const ROTMATRIX& m0, const ROTMATRIX& m1, ROTMATRIX& result) {
             set_rot_matrix(m0);