SSE2 has been part of the x86-64 baseline ISA since AMD's original 2003 spec, which is why codebases from FFmpeg to game engines can rely on it unconditionally without runtime CPU detection, unlike AVX2. The load/store/arithmetic intrinsic pattern you practice here, mapping directly to real XMM register instructions, is the same pattern used throughout Intel's own IPP performance library.