Re: [femm] FEMM "batch processing"

["sihang" <stsc_sh@xxxxxxxx>]

Dear All,

    I can see there are so many neat suggestions on the "batch processing" features about FEMM program. And I think we all want this program will provide more features in processing kinds of jobs(not only a batch processing for animation but also batch processing for line integeral, block integeral and other....) so we can get more benefit from it.

    I also did a similar thing using the fkn for my work and a university laboratory at here. The problem I have was like most of femm usrs saying. At each time running fkn, I only need get a few quantities from the post-processor and not need open a window then do manual editing. For completing this job using femm, I called the fkn.exe in my C program
(This can be easily done by using WIN-API function CreateProcess) to do mesh and solve the model. After I get the *.ans file, I had to do the post-processing job by myself.(This was not a easy job to do and its almost need rewrite the femmview's kernel routines). But thanks for Dave's help, I finished it. But I have to commit, it was at pains for me to complete this job. 

    So I think if FEMM can provide its kernel routines as liborary or DLL and ship with its origin programs that may be useful for many users in their individual works. The benefit is we can easily gain the fem kernel from it by little programming work. And if someone not using C/C++, many other programming languages(like BASIC, PASCAL) and most scripting languages(like Python, Matlab) do have the interface for calling C/C++ functions. Even some user not know and use any programming language, its a good chance for him to gain some programming experience. 

    And Just like Peter said, it will need a lot of considerations and documentations before that will finially comming down. But a simplified version of this liborary may not difficult to provide now I think. The most important thing is to provide useful functions(like mesher, solvers, and post-processing routines) in the library, and the features for easy of using is not very urgent.

         "If you make something easy enough for a fool to use only a fool will
          want to use it."    --Ancient Saying (from a Newsnet .signature file)

Well, it's a joke :)).

Si hang


       
----- Original Message ----- 
From: Petr Krc <pekr@xxxxxxxxxxx>
To: <femm@xxxxxxxxxxxxxxx>
Sent: Saturday, April 28, 2001 12:57 AM
Subject: Re: [femm] FEMM "batch processing"


> Graham Gunderson wrote:
> > I've run into the same thing, kind of. Recently, I've been using FEMM to
> > make animations. For each frame (in this example), I have to adjust the
> > angular offset of magnetization to simulate rotating magnets and moving
> > fields.
> 
> I did similar thing. To speed-up the process of selecting and rotating
> circular geometry i wrote simple small C programm which read *.FEM file,
> do some calculations on it (only rotation at this time) and write results
> into another *.FEM file.
> It can also 'rotate' angular offset of magnetization.
> 
> Program parameters:
> FEMM_ROT.EXE infile outfile angle radius [-i|o] [-m]
>  infile   - name of input file (*.FEM)
>  outfile  - name of output file (*.FEM)
>  angle    - angle of rotation (in degrees)
>  radius   - radius of boundary between static and rotational points
>  [-i|-o]  - defines which points to rotate (inside or outside of radius)
>             default is inside
>  [-m]     - rotate angles defined in materal properties also
>             default is no
> 
> 
> But it supports only FEMM v.2 data files (tested on v.2.1a).
> 
> I can send this program including source code if there is any interest.
> 
> 
> Dave Squires wrote:
> > FEMM would have to monitor FKERN to know when it is done so it can
> > launch the next run.
> 
> It should be easy implemented in BAT file using:
> START /WAIT ...
> 
> -- 
> Regards
>            Petr Krc
> 
> 
>  
> 
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/ 
> 
> 
> 
>