precision_problems.f90

program problems
use, intrinsic:: iso_fortran_env, only: wp=>real32
implicit none (type, external)

! look at classic problems with any floating point usages.
! To save agony, it's best for EVERY float constant to have a decimal point and _wp after EACH float number in an equation.

real(wp), parameter :: me = 9.10938356e-31_wp, mebad = 9.10938356_wp*10**(-31)

print *,'electron rest mass is',me,'kg., not',mebad

end program