Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Fortran.io – a Fortran Web Framework

fortran.io

Post reply on HN
Print view View original

251–255 of 255 posts

Re: Fortran.io – a Fortran Web Framework

#251

Post by tobmlt »

jofer wrote: post #244
tobmlt wrote: post #219

Earlier quoted context omitted.

Sure - I am speaking broadly of the job market for computational programmers outside of academia and national labs, and I'd better say that I am in the USA. So if you did a masters or PhD in CFD, or the finite element for structures, (and I wager also, E&M/Maxwell, computational physics, and numerical simulation generally (a catch all because the boundaries are fuzzy), but I think "as it goes in CFD and FEA for struc…

I'll second all of that. It's well worth actually learning C++ (and not C masquerading as C++) if you want to go into the broader scientific computing job market. I certainly wish I had focused more on "real" C++. I have maintained some very large C++ codebases, but I still don't really know C++, just "C with classes". Thankfully, C++ is not my day-job anymore. I'd also give a quiet nod to trying to get familiar with…

Thanks, and it's nice to hear that about Python! I don't know why I am surprised, as the more forward thinking folks at (even such a place as) my work are having a Python api built for the stuff they look after presently (their bit does structural analysis). That should make design generation/variation/optimization a lot more fun.

I guess my fear was that all Python jobs (that I'll find) are going to be "machine learning" -- but really that would just mean data munging to feed Tensor Flow (or similar lib) and post process. Total snooze fest unless "post process" means something more like design generation/variation/optimization - and we are back to the api.

Re: Fortran.io – a Fortran Web Framework

#252

Post by TheRealKing »

0xff00ffee wrote: post #81

I'd like to hear some thoughts as to why one would choose FORTRAN over R, JMP, Excel, Python... other than for fun or familiarity?

As a computational scientist who knows C, Fortran, IDL, MATLAB, Mathematica, Pascal, Python, and R for about two decades, I can tell you that the unique advantage of (modern) Fortran (2008/2018) over any other languages mentioned here is in its great flexibility and high-performance for numerical computing and basically any mathematical problem. It has a coding and syntax flexibility comparable to MATLAB and Python, but the runtime speed of C, and in many cases, beyond C. It is the only high-performance programming language that has native built-in mechanism for one-sided parallel programming on shared and distributed architectures that can scale your code from your laptop to the largest supercomputers in the world (Coarray Fortran parallelism). Checkout the OpenCorrays software on GitHub and Intel's amazing Coarray-enabled Fortran compiler. Also, Fortran along with C, are the only two languages for which the MPI standard is written. OpenMP parallelism also has its roots in the Fortran language. Nvidia is also working on a new F18 compiler for Fortran which enables GPU parallel computing via the same Coarray Fortran syntax and rules. One parallel programming API for all architectures and platforms. No programming language has this capability that I am aware of.

Typically, simulations that I run in MATLAB, Python or R, take 100 to 500 times longer than the equivalents that I write in Fortran. I am sure you can achieve a similar performance with C as well. But the development cost in C is much higher for numerical computing than in Fortran, given Fortran's native array-based syntax, parallelism features, optimization hints to the compiler, and the new high-level string and memory manipulation tools that it has.

That said, every language has its usage and place in the world of programming. In my case, all of the post-processing analyses of my Fortran/C simulation results are done in either Python or MATLAB, and sometimes in R. They complement each other rather than being rivals to each other.

Re: Fortran.io – a Fortran Web Framework

#253

Post by TheRealKing »

krapp wrote: post #101
0xff00ffee wrote: post #81

I'd like to hear some thoughts as to why one would choose FORTRAN over R, JMP, Excel, Python... other than for fun or familiarity?

No one would choose it. This is a terrible idea. No one asked for this. No one wanted this. No one needed this, yet here it is. Like some many-angled Lovecraftian horror, it spites reality, sanity and common sense merely by existing.

There are people who want it. Those who know the value of each language.

Re: Fortran.io – a Fortran Web Framework

#255

Post by TheRealKing »

UncleSlacky wrote: post #59
greglindahl wrote: post #15

Earlier quoted context omitted.

Fortran 77 has a lot of luxuries compared to FORTRAN IV, I have to say.

I can write Fortran in any language!

I have a Fortran codebase which takes about a week to finish a simulation on 10 cores. Rewriting this code in Python and running would take about 50 years to finish the same simulation in serial mode, and at best 5 years if run in parallel on 10 cores.
Post reply on HN