NewStats: 3,263,467 , 8,180,217 topics. Date: Friday, 06 June 2025 at 04:42 AM 6148b

Fortran: Anyone who thinks FORTRAN is some ancient archaic language is speaking out of nothing but sheer ignorance. Lauded by many as the numero uno of Programming Languages in Scientific Computation combined with its amazing speed (the fastest high level language), It seats on my number one spot in computational science. It's easy to learn, use and its imperative structure and enables fast prototyping coupled with for arrays, modular programming and generic programming, object-oriented programming and concurrent programming.

It is used in computationally intensive areas such as numerical weather prediction, finite element analysis, computational fluid dynamics, computational physics and computational chemistry. (wikipedia)

Cons: Poor File I/O .

Python: With the numpy library, ipython interactive shell and SciPy Library, Python is preferred by many in area of scientific computation. Its rise in fame over the past decade coupled with its legacy to put code readability over every other thing, it has been tipped by many to lead the trail in scientific computation in a decade from now.

Cons: Slow for time critical tasks, Most people use it for prototyping and implement code in Fortran or C.

C: Has numerous libraries for Scientific Computation. Its speed and power has made it the preferred choice in actual code implementation.

Cons: It's a low level language, makes fast prototyping difficult.

MATLAB: Preferred by scientist who don't have a solid programming background because of its interactive environment and visualization and easy to learn nature. Developed by a Numerical analyst and its fame in academia makes it a much preferred choice to use.

Cons: Slow, most of the critical parts are eventually written partly or entirely in C and Fortran.

The order is totally my opinion, feel free to disagree or/and add yours.

http://www.solomonubani.com/2014/09/favorite-languages-for-numerical.html

How will this languages put food on the table

This your numerical this arithmetical that can't take Nigerian coders forward.

Javanian: Fortran: Anyone who thinks FORTRAN is some ancient archaic language is speaking out of nothing but sheer ignorance. Lauded by many as the King of Programming Languages in Scientific Computation combined with its amazing speed (the fastest high level language), It seats on my number one spot in computational science. It's easy to learn, use and its imperative structure and enables fast prototyping coupled with for arrays, modular programming and generic programming, object-oriented programming and concurrent programming.

It is used in computationally intensive areas such as numerical weather prediction, finite element analysis, computational fluid dynamics, computational physics and computational chemistry. (wikipedia)

Cons: Poor File I/O .

Python: With the numpy library, ipython interactive shell and SciPy Library, Python is preferred by many in area of scientific computation. It's rise in fame over the past decade coupled with its legacy to put code readability over every other thing, it has been tipped by many to lead the trail in scientific computation in a decade from now.

Cons: Slow for time critical tasks, Most people use it for prototyping and implement code in Fortran or C.

C: Has numerous libraries for Scientific Computation. It's speed and power has made it the preferred choice in actual code implementation.

Cons: It's a low level language, makes fast prototyping difficult.

MATLAB: Preferred by scientist who don't have a solid programming background because of it's interactive environment and visualization and easy to learn nature. Developed by a Numerical analyst and it's fame in academia makes it a much preferred choice to use.

Cons: Slow, most of the critical parts are eventually written partly or entirely in C and Fortran.

The order is totally my opinion, feel free to disagree or/and add yours.

I agree with you.

You can also check out R. From their site (http://www.r-project.org/): "R is a free software environment for statistical computing and graphics".

sarutobi:

I agree with you.

You can also check out R. From their site (http://www.r-project.org/): "R is a free software environment for statistical computing and graphics".

Read some bad reviews about it, heard the language is poorly designed although it is excellent as a Computation Science language.

GodMode: How will this languages put food on the table

This your numerical this arithmetical that can't take Nigerian coders forward.

Thread is exclusive to scientists.

GodMode: How will this languages put food on the table

This your numerical this arithmetical that can't take Nigerian coders forward.

Bless Lawd!

There's always one

C++ my love , a very good companion of mine as an engineer

Javanian: Fortran: Anyone who thinks FORTRAN is some ancient archaic language is speaking out of nothing but sheer ignorance. Lauded by many as the numero uno of Programming Languages in Scientific Computation combined with its amazing speed (the fastest high level language), It seats on my number one spot in computational science. It's easy to learn
....
.......

The order is totally my opinion, feel free to disagree or/and add yours.

... Fixed ...... :-P

Nice post.... FORTRAN's grammar allows more aggressive optimizations and vectorization than any other language I know of... For some domain, it boils down to optimizers....
. . .
Python's scipy is what is universally preferred for Academic projects......

C++ is the apparent choice of computationally intensive apps like Maya, Cenima4D, Adobe AE, PS... etc.. Autodesk guys... etc..

WhiZTiM:
... Fixed ...... :-P

FORTRAN is the fastest high level language. I just hope you are not considering C++ a high level language because it isn't. I stand to be corrected though.

unfortunately MATLAB

Javanian:




FORTRAN is the fastest high level language. I just hope you are not considering C++ a high level language because it isn't. I stand to be corrected though.

That's from a biased angle...
.
. See.. That's the more reason to love C++.. ... Some think its a low level, some, middle, others high... At least it models an abstract machine just enough to accommodate almost every program's needs.

Scientific computing is multidisciplinary... And the Us aren't getting faster by clock rates... Rather, by Logical cores and multithreading capacity....

Hence, its now about parallelism...
And believe me, not many algorithms and computationally intensive routines can be paralleled.... That's where the next breakthrough is, concurrency!

So, whether C++ or FORTRAN, .... Concurrency counts. Ofcause the OS task scheduler and language runtimes matter... C++ would suit large scientific applications without really sacrificing speed in my opinion.....

WhiZTiM:

That's from a biased angle...
.
. See.. That's the more reason to love C++.. ... Some think its a low level, some, middle, others high... At least it models an abstract machine just enough to accommodate almost every program's needs.

Scientific computing is multidisciplinary... And the Us aren't getting faster by clock rates... Rather, by Logical cores and multithreading capacity....

Hence, its now about parallelism...
And believe me, not many algorithms and computationally intensive routines can be paralleled.... That's where the next breakthrough is, concurrency!

So, whether C++ or FORTRAN, .... Concurrency counts. Ofcause the OS task scheduler and language runtimes matter... C++ would suit large scientific applications without really sacrificing speed in my opinion.....

Biased opinion? I wasn't saying fortran is faster than C++, i only said C++ won't be considered a High level language for obvious reasons hence Fortran stands as the fastest high level language. I concur with every other thing you said. I won't even consider fortran when it comes to mission critical and resource intensive applications. In that area C++ is certainly the way to go.

Ever heard of Julia ?

Javanian:
Fortran: Anyone who thinks FORTRAN is some ancient archaic language is speaking out of nothing but sheer ignorance. Lauded by many as the numero uno of Programming Languages in Scientific Computation combined with its amazing speed (the fastest high level language), It seats on my number one spot in computational science. It's easy to learn, use and its imperative structure and enables fast prototyping coupled with for arrays, modular programming and generic programming, object-oriented programming and concurrent programming.

It is used in computationally intensive areas such as numerical weather prediction, finite element analysis, computational fluid dynamics, computational physics and computational chemistry. (wikipedia)

Cons: Poor File I/O .

Python: With the numpy library, ipython interactive shell and SciPy Library, Python is preferred by many in area of scientific computation. Its rise in fame over the past decade coupled with its legacy to put code readability over every other thing, it has been tipped by many to lead the trail in scientific computation in a decade from now.

Cons: Slow for time critical tasks, Most people use it for prototyping and implement code in Fortran or C.

C: Has numerous libraries for Scientific Computation. Its speed and power has made it the preferred choice in actual code implementation.

Cons: It's a low level language, makes fast prototyping difficult.

MATLAB: Preferred by scientist who don't have a solid programming background because of its interactive environment and visualization and easy to learn nature. Developed by a Numerical analyst and its fame in academia makes it a much preferred choice to use.

Cons: Slow, most of the critical parts are eventually written partly or entirely in C and Fortran.

The order is totally my opinion, feel free to disagree or/and add yours.

http://www.solomonubani.com/2014/09/favorite-languages-for-numerical.html

You also have Haskell, Scala, and Julia.