Compute_g1

The example is in the directory test/g1 with the parameter file params_g1_2d.dat.

The script prop/compute_prop.py can be used to propagate in time an initial state and can compute the g1 function.

I strongly recommend to use it from a directory which is not the one of the and-python repository (e.g. obtained using git).

Suppose you have created your local git repository by cloning the main one:

git clone git@github-and-python:delande/and-python.git

or created a local copy by any method you like, in the directory ~/git/and-python. Then, you copy the parameter file in a local directory:

cd your_local_directory

cp ~/git/and-python/prop/params_g1_2d.dat .

Then, the script can be simply ran using the command:

python ~/git/and-python/prop/compute_prop.py params_g1_2d.dat

or, if you want to use MPI:

mpiexec -n NN python ~/git/and-python/prop/compute_prop.py params_g1_2d.dat

where NN is the number of cores for the parallel computation.

If you have the message:

Warning, no chebyshev C library found, this uses the slow Python version!

this means that you do not use the fast C versions for some computations. This will of course slow down the calculation. If you are on a LKB machine, you should probably try to load the intel module using:

module load intel

and rerun the script.

Example

It corresponds to the data in Fig. 2 of T. Bardon-Brun et al, Phys. Rev. Research 2, 013297 (2020).

When running on a standard desktop computer on one core, it should print something like:

Python script runs on machine : meteor
Name of python script:  /home/delande/git/and-python/prop/compute_prop.py
Name of parameter file: /home/delande/git/and-python/tests/g1/params_g1_2d.dat

Python script started on: Fri Sep 16 10:23:20 2022

Total number of disorder realizations: 8
Number of processes: 1

Python script ended on: Fri Sep 16 10:23:33 2022
Wallclock time 12.695 seconds

CHE time             = 10.994
Max nonlinear phase  = 0.183
Max order            = 6
Expect time          = 0.102
Dummy time           = 1.531
Number of ops        = 8.6016e+10
Total_CPU time       = 12.695

The system size and the final time are slightly smaller than in the original figure, and the spatial discretization a bit too large, but the results are nevertheless good. To compare with the original figure, distances have to be multiplied by 0.158 and times divided by 2. g1 is computed at t=0, 100 and 200, which corresponds to z/zNL=0, 50, 100. You may view the 2d plot of (real part of) g1 using:

view_2d.py g1_intermediate_N.dat

with N=0, 1 or 2. You may compute an angular average using

my_extract_radial.py g1_intermediate_N.dat

and looking at the 1d data in g1_intermediate_N_Re_radial.dat

In this example, there is no disorder, the randomness is only in the initial state which is of type gaussian_with_speckle.