diff --git a/src/backend_x64/emit_x64_vector_floating_point.cpp b/src/backend_x64/emit_x64_vector_floating_point.cpp
index 348acfbb..261bd6c8 100644
--- a/src/backend_x64/emit_x64_vector_floating_point.cpp
+++ b/src/backend_x64/emit_x64_vector_floating_point.cpp
@@ -224,28 +224,6 @@ static void EmitVectorOperation64(BlockOfCode& code, EmitContext& ctx, IR::Inst*
     ctx.reg_alloc.DefineValue(inst, result);
 }
 
-void EmitX64::EmitFPVectorAbsoluteDifference32(EmitContext& ctx, IR::Inst* inst) {
-    auto args = ctx.reg_alloc.GetArgumentInfo(inst);
-    const Xbyak::Xmm a = ctx.reg_alloc.UseScratchXmm(args[0]);
-    const Xbyak::Xmm b = ctx.reg_alloc.UseXmm(args[1]);
-
-    code.subps(a, b);
-    code.andps(a, code.MConst(xword, 0x7FFFFFFF7FFFFFFF, 0x7FFFFFFF7FFFFFFF));
-
-    ctx.reg_alloc.DefineValue(inst, a);
-}
-
-void EmitX64::EmitFPVectorAbsoluteDifference64(EmitContext& ctx, IR::Inst* inst) {
-    auto args = ctx.reg_alloc.GetArgumentInfo(inst);
-    const Xbyak::Xmm a = ctx.reg_alloc.UseScratchXmm(args[0]);
-    const Xbyak::Xmm b = ctx.reg_alloc.UseXmm(args[1]);
-
-    code.subpd(a, b);
-    code.andpd(a, code.MConst(xword, 0x7FFFFFFFFFFFFFFF, 0x7FFFFFFFFFFFFFFF));
-
-    ctx.reg_alloc.DefineValue(inst, a);
-}
-
 void EmitX64::EmitFPVectorAbs16(EmitContext& ctx, IR::Inst* inst) {
     auto args = ctx.reg_alloc.GetArgumentInfo(inst);
 
diff --git a/src/frontend/A64/translate/impl/simd_scalar_three_same.cpp b/src/frontend/A64/translate/impl/simd_scalar_three_same.cpp
index c3aba19c..3a3e94a2 100644
--- a/src/frontend/A64/translate/impl/simd_scalar_three_same.cpp
+++ b/src/frontend/A64/translate/impl/simd_scalar_three_same.cpp
@@ -173,12 +173,11 @@ bool TranslatorVisitor::CMTST_1(Imm<2> size, Vec Vm, Vec Vn, Vec Vd) {
 bool TranslatorVisitor::FABD_2(bool sz, Vec Vm, Vec Vn, Vec Vd) {
     const size_t esize = sz ? 64 : 32;
 
-    const IR::U128 operand1 = V(esize, Vn);
-    const IR::U128 operand2 = V(esize, Vm);
-    const IR::U128 difference = ir.FPVectorAbsoluteDifference(esize, operand1, operand2);
-    const IR::U128 result = ir.VectorZeroUpper(difference);
+    const IR::U32U64 operand1 = V_scalar(esize, Vn);
+    const IR::U32U64 operand2 = V_scalar(esize, Vm);
+    const IR::U32U64 result = ir.FPAbs(ir.FPSub(operand1, operand2, true));
 
-    V(128, Vd, result);
+    V_scalar(esize, Vd, result);
     return true;
 }
 
diff --git a/src/frontend/A64/translate/impl/simd_three_same.cpp b/src/frontend/A64/translate/impl/simd_three_same.cpp
index 36dfbca8..0e14e3d7 100644
--- a/src/frontend/A64/translate/impl/simd_three_same.cpp
+++ b/src/frontend/A64/translate/impl/simd_three_same.cpp
@@ -384,7 +384,7 @@ bool TranslatorVisitor::FABD_4(bool Q, bool sz, Vec Vm, Vec Vn, Vec Vd) {
 
     const IR::U128 operand1 = V(datasize, Vn);
     const IR::U128 operand2 = V(datasize, Vm);
-    const IR::U128 result = ir.FPVectorAbsoluteDifference(esize, operand1, operand2);
+    const IR::U128 result = ir.FPVectorAbs(esize, ir.FPVectorSub(esize, operand1, operand2));
 
     V(datasize, Vd, result);
     return true;
diff --git a/src/frontend/ir/ir_emitter.cpp b/src/frontend/ir/ir_emitter.cpp
index 860b9f93..cc46c63b 100644
--- a/src/frontend/ir/ir_emitter.cpp
+++ b/src/frontend/ir/ir_emitter.cpp
@@ -1577,17 +1577,6 @@ U128 IREmitter::FPVectorAbs(size_t esize, const U128& a) {
     return {};
 }
 
-U128 IREmitter::FPVectorAbsoluteDifference(size_t esize, const U128& a, const U128& b) {
-    switch (esize) {
-    case 32:
-        return Inst<U128>(Opcode::FPVectorAbsoluteDifference32, a, b);
-    case 64:
-        return Inst<U128>(Opcode::FPVectorAbsoluteDifference64, a, b);
-    }
-    UNREACHABLE();
-    return {};
-}
-
 U128 IREmitter::FPVectorAdd(size_t esize, const U128& a, const U128& b) {
     switch (esize) {
     case 32:
diff --git a/src/frontend/ir/ir_emitter.h b/src/frontend/ir/ir_emitter.h
index 91d50976..6986af67 100644
--- a/src/frontend/ir/ir_emitter.h
+++ b/src/frontend/ir/ir_emitter.h
@@ -289,7 +289,6 @@ public:
     U32 FPU64ToSingle(const U64& a, bool round_to_nearest, bool fpscr_controlled);
 
     U128 FPVectorAbs(size_t esize, const U128& a);
-    U128 FPVectorAbsoluteDifference(size_t esize, const U128& a, const U128& b);
     U128 FPVectorAdd(size_t esize, const U128& a, const U128& b);
     U128 FPVectorDiv(size_t esize, const U128& a, const U128& b);
     U128 FPVectorEqual(size_t esize, const U128& a, const U128& b);
diff --git a/src/frontend/ir/opcodes.inc b/src/frontend/ir/opcodes.inc
index ede93ec8..7a3ca289 100644
--- a/src/frontend/ir/opcodes.inc
+++ b/src/frontend/ir/opcodes.inc
@@ -415,8 +415,6 @@ OPCODE(FPS64ToSingle,                       T::U32,         T::U64,         T::U
 OPCODE(FPVectorAbs16,                       T::U128,        T::U128                                         )
 OPCODE(FPVectorAbs32,                       T::U128,        T::U128                                         )
 OPCODE(FPVectorAbs64,                       T::U128,        T::U128                                         )
-OPCODE(FPVectorAbsoluteDifference32,        T::U128,        T::U128,        T::U128                         )
-OPCODE(FPVectorAbsoluteDifference64,        T::U128,        T::U128,        T::U128                         )
 OPCODE(FPVectorAdd32,                       T::U128,        T::U128,        T::U128                         )
 OPCODE(FPVectorAdd64,                       T::U128,        T::U128,        T::U128                         )
 OPCODE(FPVectorDiv32,                       T::U128,        T::U128,        T::U128                         )