Fortran 90 - Science topic

Robert Schaefer

Thanks a lot. I'll try it and see if it works.

Umat to script load and interaction film conditions

Dear Yunus Rezvanifar ,

Field variables in abaqus are quantities that are used to perform calculations when running user subroutines. They do not describe a specific

physical quantity but are often used to describe solution dependent material properties.

Integration points are referred as material points to distinguish between the

Langragian and Eulerian approach.

F90 seems to require a link library to read the HDF or NetCDF

Dear Fen Fam,

Rename the mkl_avx2.dll in (C: \SI MU L I A\Abaqus\6.14 – 5 (your Abaqus version may be different\code\bin) to mkl_avx2.dll.11.0.0.1.

If the issue is not resolved contact Abaqus support.

Best regards,

Hamed

Hi Karan,

Aside from the suggested answers, you can also use AceGen in Mathematica.

http://symech.fgg.uni-lj.si/User/AceGenPrint.pdf

Best,

Romik

If I remember correctly, the textbook "Concepts and applications of finite element analysis" by Cook et al. contains a concise section on cyclic symmetries that may help you.

It is worth checking out the language Julia. It is as easy as python, and so far at least for my atomistic Monte Carlo and Molecular Dynamics codes, is as fast as Fortran. When it is slower than Fortran the difference is measurable in percentage, never order of magnitudes. Also, I find it to be significantly faster than python + numba. However, a skilled user in python can make it much faster. The thing I like about Julia is it starts out fast, and a skilled user can make it even faster :)

As for ANN, this is out of my field but there seems to be more and more options for it in Julia.

[deep learning](http://doktormike.github.io/blog/Deep-learning-in-julia/)

to link a couple of possibly useful sources

Just my 2 cents

I think it is a no-brainer, just moving to the text editor and choosing the g++ or whatever compiler and following the syntax-defining variables and following implicit rules, makefiles aren't long enough to confuse. The syntax lines are usually very easy to follow.

Assuming:

Idontknowhowtomakefiles: Idontknowhowtomakefiles.o g++ -o Idontknowhowtomakefiles Idontknowhowtomakefiles.o Idontknowhowtomakefiles.o: Idontknowhowtomakefiles.cpp g++ -c Idontknowhowtomakefiles.cpp

Substitute Idontknowhowtomakefiles with x and life should get better.

Compile:

Run: make

Emilius Vanagas

Hello Dr. Vanagas,

I found an error in the NEA DB loadsrcnea.bat file... And after I fixed it, I compiled the code in Windows 10 x64 using Lahey/Fujitsu Fortran compiler trial version.

The error is on line 359. One should replace cd /prog with cd prog

I was looking for something less generic, to avoid lot of work i customization and to have good performances. It was a long time ago since I used Numerical Recipes, I'll refresh my memory and take a look at it anyhow.

Dear Marcus,

thank for the answer. I had gone through those link. I would like to if ask is there way to convert fortran binary file into text file without any alteration ?

@ Ishara

If you want to make flow of conditions time dependent, you need to use the subroutine as follows, for example to pause the execution of commands for 1 minute. I am copy-pasting the code and its output too. Hope that is what looking for.

program rg

implicit none

real(kind=16) :: time_t1, time_t2

real(kind=16), parameter :: time_off = 60.d0 ! pause for a minute.

call cpu_time(time_t1)

call cpu_time(time_t2)

print*, 'time begins = ', time_t2

do while ( time_t2 - time_t1 < time_off )

call cpu_time(time_t2)

end do

print*, 'time ends = ', time_t2

print*, 'time elapsed= ', time_t2 - time_t1

end program rg

_________________________

vikash@atman ~ $ cd Desktop/

vikash@atman ~/Desktop $ gfortran test.f95 -o vikash

vikash@atman ~/Desktop $ ./vikash

time begins = 1.84099999999999999999999999999999991E-0003

time ends = 60.0018250000000000000000000000000010

time elapsed= 60.0000000000000000000000000000000000

vikash@atman ~/Desktop $

Hi 

did you know how to implement it using Fortran?

did you know any good practical refrences about grid generation in Fortran?

thank you for your Answer

Really I just use gfortran under Mountain Lion with the command line from a simple xterminal, never tried Xcode.

Are you referring to the signum function as described in the link below or the sign transfer function, which is more common in FORTRAN? The former has a derivative (Dirac delta) and the latter does not.

As far as I know there is no function directly available in Fortran. The easiest way to do this is to write a function in C which uses according functions from the operating system and computes the information that you need. Then, you call this C function from within Fortran. For this you need an interface written in Fortran which will fit the interface of the C function (the Fortran keyword BIND(C) is very helpful for this).

What Hamidreza mentioned about unintentional copies being made there is a way to figure out where this occurs. The compiler will not tell you directly. But, when you compile with ifort using the flag '-check all' you will get runtime warnings when copies occur. You should be aware, though, that this compiler flag turns off all optimizations. Because of this your software will run up to 10 or 20 times slower.

Hi, do you mean to modify the shape.dat file in linux? There is a tough solution that consists of editing the file by yourself (in an usual program like qtiplot): Just change the dimension of the periodicity (for instance, Y turns into Y+delta) export the data as ascii and add them to the end of the shape.dat file (you may change also the total number of dipoles and the dipole índex -first column-). Other solution may be to run a periodic calculation (DDSCAT 7.2 and 7.3 can do it). Finally, If you are self-confident, you can modify the fortran file. If so, you may look for the section ellipsoid in the code and add as many ellipsoids as you want. Good luck!

Hi,

   Thanks Brian. The problem is due to,  The maximum number of times in one file is currently set to 10000 in WRF. Since the restart files are written one file per time, this is not an issue, but by trying to put all of the output time steps into one file, you are hitting this limit.

Another possible way for getting WRF wind at specific heights is also using NCL. This allows for using directly the WRF output files (netCDF), instead of converting them using either UPP or ARWpost.

The built-in NCL function for doing this is wrf_user_intrp_3d (https://www.ncl.ucar.edu/Document/Functions/WRF_arw/wrf_user_intrp3d.shtml)

hi

you can go for NVIDIA -CUDA GPU. but then for your software you need to get an additional license for gpu as well.

as per as operating system, these days companies are going for windows based platforms, but you can use Linux.

Thank you very much Dr. Lopez for this information

we are are trying to find the variation of resistance in terms of magnetic field, so if you encountered this problem could you give us an idea what to use, "how far is it " or even other possible software other than Oommf that can do the job.

Regards.

One basic approach to writing fast MATLAB (or R) code is to understand vectorized operations. Take a look at this PDF, which addresses this particular problem and I guess will be incredibly helpful:

Best,

José.