diff --git a/rpcs3/Emu/RSX/rsx_methods.cpp b/rpcs3/Emu/RSX/rsx_methods.cpp
index 27a2df6725..5a8be5090a 100644
--- a/rpcs3/Emu/RSX/rsx_methods.cpp
+++ b/rpcs3/Emu/RSX/rsx_methods.cpp
@@ -51,6 +51,7 @@ namespace rsx
 	template<> struct vertex_data_type_from_element_type<f16> { static const vertex_base_type type = vertex_base_type::sf; };
 	template<> struct vertex_data_type_from_element_type<u8> { static const vertex_base_type type = vertex_base_type::ub; };
 	template<> struct vertex_data_type_from_element_type<u16> { static const vertex_base_type type = vertex_base_type::s32k; };
+	template<> struct vertex_data_type_from_element_type<s16> { static const vertex_base_type type = vertex_base_type::s1; };
 
 	namespace nv406e
 	{
@@ -149,10 +150,16 @@ namespace rsx
 			sema.semaphore[index].timestamp = rsx->timestamp();
 		}
 
-		template<u32 id, u32 index, int count, typename type>
+		/**
+		 * id = base method register
+		 * index = register index in method
+		 * count = element count per attribute
+		 * register_count = number of registers consumed per attribute. E.g 3-element methods have padding
+		 */
+		template<u32 id, u32 index, int count, int register_count, typename type>
 		void set_vertex_data_impl(thread* rsx, u32 arg)
 		{
-			static const size_t increment_per_array_index = (count * sizeof(type)) / sizeof(u32);
+			static const size_t increment_per_array_index = (register_count * sizeof(type)) / sizeof(u32);
 
 			static const size_t attribute_index = index / increment_per_array_index;
 			static const size_t vertex_subreg = index % increment_per_array_index;
@@ -176,7 +183,7 @@ namespace rsx
 		{
 			static void impl(thread* rsx, u32 _reg, u32 arg)
 			{
-				set_vertex_data_impl<NV4097_SET_VERTEX_DATA4UB_M, index, 4, u8>(rsx, arg);
+				set_vertex_data_impl<NV4097_SET_VERTEX_DATA4UB_M, index, 4, 4, u8>(rsx, arg);
 			}
 		};
 
@@ -185,7 +192,7 @@ namespace rsx
 		{
 			static void impl(thread* rsx, u32 _reg, u32 arg)
 			{
-				set_vertex_data_impl<NV4097_SET_VERTEX_DATA1F_M, index, 1, f32>(rsx, arg);
+				set_vertex_data_impl<NV4097_SET_VERTEX_DATA1F_M, index, 1, 1, f32>(rsx, arg);
 			}
 		};
 
@@ -194,7 +201,7 @@ namespace rsx
 		{
 			static void impl(thread* rsx, u32 _reg, u32 arg)
 			{
-				set_vertex_data_impl<NV4097_SET_VERTEX_DATA2F_M, index, 2, f32>(rsx, arg);
+				set_vertex_data_impl<NV4097_SET_VERTEX_DATA2F_M, index, 2, 2, f32>(rsx, arg);
 			}
 		};
 
@@ -203,8 +210,8 @@ namespace rsx
 		{
 			static void impl(thread* rsx, u32 _reg, u32 arg)
 			{
-				//NOTE: attributes are 16-byte aligned (Rachet & Clank 2)
-				set_vertex_data_impl<NV4097_SET_VERTEX_DATA3F_M, index, 4, f32>(rsx, arg);
+				//Register alignment is only 1, 2, or 4 (Rachet & Clank 2)
+				set_vertex_data_impl<NV4097_SET_VERTEX_DATA3F_M, index, 3, 4, f32>(rsx, arg);
 			}
 		};
 
@@ -213,7 +220,7 @@ namespace rsx
 		{
 			static void impl(thread* rsx, u32 _reg, u32 arg)
 			{
-				set_vertex_data_impl<NV4097_SET_VERTEX_DATA4F_M, index, 4, f32>(rsx, arg);
+				set_vertex_data_impl<NV4097_SET_VERTEX_DATA4F_M, index, 4, 4, f32>(rsx, arg);
 			}
 		};
 
@@ -222,7 +229,7 @@ namespace rsx
 		{
 			static void impl(thread* rsx, u32 _reg, u32 arg)
 			{
-				set_vertex_data_impl<NV4097_SET_VERTEX_DATA2S_M, index, 2, u16>(rsx, arg);
+				set_vertex_data_impl<NV4097_SET_VERTEX_DATA2S_M, index, 2, 2, u16>(rsx, arg);
 			}
 		};
 
@@ -231,7 +238,7 @@ namespace rsx
 		{
 			static void impl(thread* rsx, u32 _reg, u32 arg)
 			{
-				set_vertex_data_impl<NV4097_SET_VERTEX_DATA4S_M, index, 4, u16>(rsx, arg);
+				set_vertex_data_impl<NV4097_SET_VERTEX_DATA4S_M, index, 4, 4, u16>(rsx, arg);
 			}
 		};
 
@@ -240,8 +247,7 @@ namespace rsx
 		{
 			static void impl(thread* rsx, u32 _reg, u32 arg)
 			{
-				LOG_ERROR(RSX, "SCALED_4S vertex data format is not properly implemented");
-				set_vertex_data_impl<NV4097_SET_VERTEX_DATA4S_M, index, 4, u16>(rsx, arg);
+				set_vertex_data_impl<NV4097_SET_VERTEX_DATA_SCALED4S_M, index, 4, 4, s16>(rsx, arg);
 			}
 		};