Fortran Data Types and Variables

Analogy categoryAnalogies

Fortran provides five intrinsic data types: integer for whole numbers, real for floating-point numbers, complex for numbers with real and imaginary parts, character for text, and logical for true/false values. Each variable must be declared with one of these types before it is used, typically at the top of a program unit, in a form like integer :: count or real :: temperature. This static typing lets the compiler catch type mismatches at compile time rather than causing errors at runtime.

🏏

Cricket analogy: Like a scorecard having distinct columns for runs, overs, wickets, and extras, each tracking a specific kind of number, Fortran has distinct types, integer, real, character, logical, complex, each tracking a specific kind of data.

Declaring and Initializing Variables

Variables are declared using the syntax type, attributes :: name, optionally followed by an initial value, for example integer :: score = 0 or character(len=20) :: username. The parameter attribute creates a named constant whose value cannot change after compilation, such as real, parameter :: pi = 3.14159265. Character variables require a specified length using len=, since Fortran strings are fixed-length by default unless declared as allocatable or deferred-length with the character(:) syntax.

🏏

Cricket analogy: Like fixing a player's squad number before the tournament starts and never changing it, real, parameter :: pi = 3.14159265 fixes a value at compile time that can never change during the program.

Numeric Precision: Kind Specifiers

Plain real variables typically use single precision (about 6-7 significant decimal digits), which is often insufficient for scientific work requiring more accuracy. Modern Fortran addresses this with kind specifiers, for example real(kind=8) :: precise_value or, more portably, using the selected_real_kind function or the iso_fortran_env module's real64 constant, which guarantees double precision (about 15-16 significant digits) regardless of the compiler or platform.

🏏

Cricket analogy: Like the difference between a basic scoreboard showing only runs and a Hawk-Eye system tracking ball trajectory to the millimeter, real(kind=8) gives far more precision than plain single-precision real.

Never compare real (floating-point) values for exact equality with ==. Due to rounding, 0.1 + 0.2 may not exactly equal 0.3. Instead check that the absolute difference is smaller than a small tolerance, e.g. abs(a - b) < 1.0e-6.

fortran

program data_types_demo
    use iso_fortran_env, only: real64
    implicit none

    integer :: age = 30
    real(kind=real64) :: pi = 3.14159265358979_real64
    character(len=10) :: name = "Ada"
    logical :: is_valid = .true.
    complex :: z = (2.0, 3.0)

    print *, "Name:", name, "Age:", age
    print *, "Pi:", pi, "Valid:", is_valid
end program data_types_demo

Fortran has five intrinsic types: integer, real, complex, character, and logical.
Variables must be declared with a type before use, in the form type :: name.
The parameter attribute creates a compile-time constant that cannot be reassigned.
Character variables need an explicit length via len=, since strings are fixed-length by default.
Plain real is single precision (~6-7 digits); real(kind=8) or real64 gives double precision (~15-16 digits).
The iso_fortran_env module provides portable kind constants like real64 across compilers.
Never compare floating-point values with == due to rounding error; use a tolerance-based comparison instead.

Practice what you learned

1. Which of these is NOT one of Fortran's five intrinsic data types?

2. What does the parameter attribute do when declaring a variable?

3. Why must character variables typically specify a length with len=?

4. What module provides the portable real64 kind constant for double precision?

5. Why should you avoid comparing two real (floating-point) values with == in Fortran?

Was this page helpful?

Fortran Program Structure

Your First Fortran Program

Related Study Notes in Programming

Browse all study notes

Programming

Apache Spark Study Notes

Programming · 30 topics

Programming

Apache Flink Study Notes

Programming · 30 topics

Programming

Hadoop Study Notes

Programming · 30 topics

Programming

Snowflake Study Notes

Programming · 30 topics

Programming

Apache Airflow Study Notes

Programming · 30 topics

Programming

dbt (Data Build Tool) Study Notes

Programming · 30 topics

Topics covered

#Programming#FortranStudyNotes#FortranDataTypesAndVariables#Fortran#Data#Types#Variables#StudyNotes#SkillVeris#ExamPrep

Fortran Data Types and Variables