AVX and AVX2 implementation of SIMD2_64x needs improvement
Issue #29
new
Implementation of SIMD2 types for 64b scalars is not using intrinsic types. This should be fixed.