Best way to declare a double precision in Fortran?

Help
21

I do the following,

integer, parameter :: r8=selected_real_kind(15,9)
real(kind=r8), parameter :: pi=4.0_r8*atan(1.0_r8)
1 Like