Directory:
    ${PENCIL_HOME}/samples/most-modules
SVN Id:
    $Id$
Maintainer:
    Axel Brandenburg <brandenb/nordita[dot]org>
    Dhrubaditya Mitra <dhruba.mitra/gmail[dot]com>
    Matthias Rheinhardt <mreinhardt/nordita[dot]org>
    Wladimir Lyra <wladimir.lyra/gmail[dot]com>
    Natalia Babkovskaia <nbabkovskaia/gmail[dot]com>
    Philippe Bourdin <pc/bourdin[dot]ch>
    Boris Dintrans <boris.dintrans/gmail[dot]com>
    Anders Johansen <anders/astro[dot]lu.se>
    Nils E.L. Haugen <Nils.E.Haugen/gmail[dot]com>
    Tobias Heinemann <tobias.heinemann/gmail[dot]com>
    Julien Savre <juliensavre/gmail[dot]com>
    Chao-Chin Yang <ccyang/ua[dot]edu>
    Jörn Warnecke <joern[dot]warnecke/gmail[dot]com>
    Sven Bingert <sven[dot]bingert/gwdg[dot]de>
Added:
    13-Jan-2007
Status:
    succeeds # One of [succeeds|failing since <date>|obsolete| ...]
Recommended resolution:
    128x128x128 (nu=eta=1.3e-3)
Comments:
    Use as many physics modules as possible; together with samples/no-modules,
    this constitutes the minimal set of auto-tests.

    The ultimate auto-test: Evolves chiral passive scalars with the occasional
    supernova explosion in the planetary dust, and of course all of this with
    mean-field MHD, radiative transfer and self-gravity.
    It took some time to tune this such that it passes one Euler step at
    dt=1e-20. A time step of 1e-19 is already way too large...

    This now works, but with lots of smaller changes to the code that will get
    checked in next. Expect the next ~ 10 commits to contain questions like
    `Anders, is this OK?', `Tony, please check', etc.

    This test complements the no-modules test, but the two don't cover all
    modules we have. The next thing to do is to create a coverage report and
    check which important .f90 files are still missing. Then we can add
    `some-modules-1' and maybe `some-modules-2' to have a close-to-complete
    set of four tests.

    Fred: added non-zero entropy to start.in such that the ISM temperature is
    10^4 K instead of 10^{-1} K. The timestep therefore increases to ~1e-4 from
    ~1e-20, without crashing. Increased the number of iterations to 5 rather
    than 1 to allow sensitivity to more changes to the code.