COMP 210 · Scientific Computing with Fortran

Led by John Backus Simulacrum

Scientific computing with Fortran — arrays, numerical methods, LAPACK, modern language features and the Fortran-Python bridge. For scientists and engineers who need fast numerical code.

1. Module 1

   Formula Translation Fundamentals

   Led by John Backus Simulacrum, with Kahanian Numerical Precision Simulacrum (guest for Unit 2)

   The question

   Fortran means Formula Translation. If your code does not look like the mathematics it implements, something has gone wrong. How do you write code that a mathematician can read — with correct precision and proper use of array operations?

   Outcome

   The student writes Fortran that reads like mathematics, uses KIND parameters, and prefers intrinsic array operations. (Foundational)

   Sub-units

   1. ○ 1.1 IMPLICIT NONE and KIND Parameters Open this unit → ✓ ↻
   2. ○ 1.2 Arrays and Intrinsic Functions Open this unit → ✓ ↻
   Begin the course → Print certificate of completion
2. Module 2

   Array Programming

   Led by John Backus Simulacrum

   The question

   The transformation from scalar loops to array operations is the central skill. A DO loop processes one element at a time. An array operation describes the result. How do you think in arrays?

   Outcome

   The student can transform scalar code into array operations and understands column-major layout and its cache implications. (Practical)

   Sub-units

   1. ○ 2.1 Array Operations and Sections Open this unit → ✓ ↻
   2. ○ 2.2 WHERE, FORALL and Memory Layout Open this unit → ✓ ↻
   Open this module → Print certificate of completion
3. Module 3

   Program Structure

   Led by John Backus Simulacrum, with Metcalfian Modern Fortran Simulacrum (guest)

   The question

   COMMON blocks are dead. Modules, derived types and explicit interfaces are how modern Fortran programs are built. What do they give you that the old style did not?

   Outcome

   The student can structure programs with modules, explicit INTENT, derived types and generic interfaces. (Practical)

   Sub-units

   1. ○ 3.1 Modules and Procedures Open this unit → ✓ ↻
   2. ○ 3.2 Derived Types and Interfaces Open this unit → ✓ ↻
   Open this module → Print certificate of completion
4. Module 4

   Modern Fortran

   Led by Metcalfian Modern Fortran Simulacrum

   The question

   DO CONCURRENT, allocatable arrays, pure procedures, C interoperability — each exists because a real programmer needed it. What do they enable?

   Outcome

   The student can use DO CONCURRENT, manage dynamic memory safely, and interface Fortran with C. (Advanced)

   Sub-units

   1. ○ 4.1 DO CONCURRENT and Allocatable Arrays Open this unit → ✓ ↻
   2. ○ 4.2 Pure Procedures and C Interoperability Open this unit → ✓ ↻
   Open this module → Print certificate of completion
5. Module 5

   Scientific Applications

   Led by John Backus Simulacrum, with Molerian Matrix Computation Simulacrum (guest)

   The question

   LAPACK solves linear systems. f2py bridges Fortran and Python. Together they connect your Fortran kernels to the modern scientific stack. How do you use them?

   Outcome

   The student can call LAPACK, implement numerical methods, and expose Fortran code to Python with measured performance. (Project)

   Sub-units

   1. ○ 5.1 LAPACK and Numerical Methods Open this unit → ✓ ↻
   2. ○ 5.2 Fortran-Python Bridge Open this unit → ✓ ↻
   Open this module → Print certificate of completion