Latest News
  • November 20, 2009
  • FEMhub 0.9.8 released
  • October 15, 2009
  • New home page launched
  • September 23, 2009
  • FEMhub 0.9.7 released
  • July 30, 2009
  • FEMhub 0.9.6 released
  • July 14, 2009
  • FEMhub 0.9.5 released
  • June 29, 2009
  • FEMhub 0.9.4 released
  • June 28, 2009
  • FEMhub 0.9.3 released
  • June 26, 2009
  • FEMhub 0.9.2 released
  • June 25, 2009
  • FEMhub 0.9.1 released
  • May 18, 2009
  • femhub.org page created
  • May 13, 2009
  • FEMhub distribution started
Our Mission
  • To create an open-source distribution of many scientific computing codes enhanced with a unified Python interface. To support transparency and establish quality standards in the development of open source scientific computing codes. We want FEMhub to become a free alternative to expensive commercial codes.
How To Join
Mailing List
  • Please use this mailing list for all communications related to FEMhub.

Download and Install

First check in the README.txt file for prerequisites, these may differ for different operating systems. For example for Ubuntu, those are

$ sudo apt-get install gcc g++ gfortran make m4 perl tar
Then download, unpack, build FEMhub:

$ wget http://femhub.org/pub/femhub-0.9.8.tar
$ tar -xf femhub-0.9.8.tar
$ cd femhub-0.9.8
$ make

You can take advantage of several cores on your computer by executing

$ export MAKE="make -j9"

before typing make to compile in parallel on 9 cores.

Depending on the speed of your computer, wait between 14 minutes to 1.5 hour. That's it. It should work on any Linux, Mac and Windows (in Cygwin), if it fails, please report a bug. If you want, you can also download a binary from here, however, if it doesn't work for you, compile from source, that should always work.

If you encounter problems, let us know through the FEMhub mailing list.

Run It!

$ ./femhub
------------------------------------------------------------
| FEMhub Version 0.9.7, Release Date: 2009-09-23           |
| Type notebook() for the GUI.                             |
------------------------------------------------------------
In [1]: notebook()

and a browser will start with the web notebook. Create a new worksheet and type the following into the cell:

import hermes2d
hermes2d.demo_layer()

Press Shift+Enter (or click the "evaluate" button) and you should see the following output:

iter=00, err_est=99.05%, DOFS=1
iter=01, err_est=86.54%, DOFS=3
iter=02, err_est=82.76%, DOFS=7
iter=03, err_est=76.92%, DOFS=21
iter=04, err_est=61.67%, DOFS=39
iter=05, err_est=43.00%, DOFS=74
iter=06, err_est=27.93%, DOFS=113
iter=07, err_est=20.37%, DOFS=156
iter=08, err_est=14.32%, DOFS=218
iter=09, err_est=11.06%, DOFS=298
iter=10, err_est= 8.01%, DOFS=412
iter=11, err_est= 5.68%, DOFS=555
iter=12, err_est= 3.86%, DOFS=714
iter=13, err_est= 2.53%, DOFS=861
iter=14, err_est= 1.66%, DOFS=1056
iter=15, err_est= 1.06%, DOFS=1223
iter=16, err_est= 0.68%, DOFS=1565