Very interesting, just tested on 0.5 and vectorized can still be a magnitude slower. ulia> @time vect(x, y, a, b) 0.000038 seconds (29 allocations: 37.406 KB

Very interesting, just tested on 0.5 and vectorized can still be a magnitude slower. ulia> @time vect(x, y, a, b) 0.000038 seconds (29 allocations: 37.406 KB) 1×2 Array{Float64,2}: julia> @time devect(x, y, a, b) 0.000006 seconds (10 allocations: 512 bytes) 1×2 Array{Float64,2}: julia> @time vect(x, y, a, b) 0.002324 seconds (73 allocations: 3.075 MB) 416×2 Array{Float64,2}: julia> @time devect(x, y, a, b) 0.000206 seconds (18 allocations: 23.359 KB) 416×2 Array{Float64,2}: