Releases

The project MCSHAPE is a work in progress. The last version ready to download is 2.60 (July 2006). This new release includes some important new features, as explained in the table below. For the comparison with the other codes developed by our group click here.

Features	Details
Physics	includes line width
	multilayer targets (up to 5 layers)
	multielement (up to 20 elements)
	maximum energy resolution (10 eV)
	up to 100 collisions
Code	language: FORTRAN 90
	The code uses the random number generators library of Numerical Recipes and of SPRNG (only for the parallel version). For more information about random number generators click here.
	parallelization: MPICH v.1.0

The history of the most important releases is as follows.

Version 2.xx -- MULTILAYER TARGETS --

• 2.60 - Several changes were operated on the code which justified a drastic step up in the version number:
• -in transmission, the uncollided flux is now accumulated on the file TR_BEAM.DAT. For the moment the file SPECTRUM.DAT does not contain the uncollided beam. Therefore, to obtain the full spectrum in transmission mode is necessary to add the two files SPECTRUM.DAT + TR_BEAM.DAT.
• -automatic saving (and recover) of the configuration of the last computation.   It is also possible to plot again the last computed results without performing   again a run (only if the folder has been not renamed).
• 2.52 - Several changes were performed:
• -it was corrected the scale of the output spectra SPECTRUM_W and ONLY_PHOTO_W which are now independent of the chosen energy resolution.
• - the accumulators in SPECTRUM_W and ONLY_PHOTO_W now contain only data greater that 1.0e-30.
• -it has been added the possibility to choose the transport model (vector or scalar) used for the simulation. The output with the scalar option is comparable to a “traditional” MC code (i.e. one considering an average unpolarised state).

2.51 - Two changes were performed:

• -it was corrected a bug which undervalued the high-order contributions including at least one photoelectric collision.
• -the plotting routine is now refreshed before the start of a new plot.

2.50 - Several changes were operated on the code which justified a drastic step up in the version number:

• - a bug which appeared for line width spectrum near the lower energy limit was corrected.
• - the accumulators now contain only data greater that 1.0e-30.
• - the fundamental data base has been updated: the cumulated functions for all the elements were corrected.
• - the executable code is user friendly: input is completely driven by dialog entry and graphic output is handled in the same interface.
• - the scale of the output spectra is now independent of the chosen energy resolution. This feature allows straightforward comparison of spectra produced under different energy resolutions.
• - maximum energy resolution of the output is now 10 eV.
• -added output files ONLY_PHOTO.DAT and ONLY_PHOTO_W.DAT. It allows the extraction of the pure XRF terms for large number of collisions simulations, i.e. without any scattering contribution in the MS chains.

2.04 - It was discovered a bias in the profile generation which is common to all MC codes and is related to the assumption of completely populated atomic orbitals contributing to the scattering. Such assumption does not agree with the equivalence condition between the integrated Compton profile (IA) and the scattering function. An unbiased MC algorithm for the simulation of the Compton scattering has been developed and is now included in this version. An article about this algorithm wil be presented during the next IRRMA-6 symposium in Hamilton (Canada), 18-24 June 2005.

2.03 - (a) SOURCE and GEOMETRY were splitted in two different files. From this version is necessary to use MCINPUT v2.10. (b) A bias in the build-up of the Compton profile was discovered and has been partially solved using a weight. The bias is more apparent at low energy excitations. It will be definitively solved in the next version.

2.02 -A bug was discovered and has been corrected. It regards a time consuming selection of the momentum for the shell electron involved in the Compton scattering. The new algorithm is more efficient.

• 2.01 - The code can compute MULTILAYER TARGETS. The maximum number of layers is 5.

- The code can be compiled to run on a Beowulf cluster of processor. The parallel version of the code was developed using MPICH v.1.0

- The interactive starting dialog was inproved. It is possible both to read the composition of the target from the text file SAMPLE.DAT, both to enter step-by-step a new target. The geometry of the scattering and the characteristics of the source can be read from a text file of arbitrary name, or can be entered during execution. Moreover, the user has to specify the azimuthal angle of scattering.

• - The output of the code was enriched. The text files written by the code are the following.

SPECTRUM.DAT file containing the full energy spectrum SPECTRUM_W.DAT file with the intensity corrected considering the width of the characteristic lines CONTINUUM.DAT file with the continuous part of the spectrum DISCRETE.DAT file with the discrete part of the spectrum PHOTO_EFFECT.DAT file containing the spectrum of the chains with at least one photoelectric collision ONLY_COMPTON.DAT file containing the intensity due to Compton scattering only ONLY_SCATTERING.DAT file with the intensity due to pure scattering, i.e. without any photoelectric collision STOKES1.DAT file with the intensity component in function of the energy STOKES2.DAT file with the second component of the Stokes vector STOKES3.DAT file with the third component of the Stokes vector STOKES4.DAT file with the fourth component of the Stokes vector ANGLE_CHI.DAT file with the angle chi as a function of the energy POL_DEGREE.DAT file with the polarization degree of the spectrum as a function of the energy mcshape.log file with some infomation about the simulation

Please note that PHOTO_EFFECT.DAT + ONLY_SCATTERING.DAT gives SPECTRUM.DAT.

• Moreover, there are some plots (only for Linux)

Intensitylin.eps intensity spectrum in linear scale Intensity.eps intensity spectrum in logaritmic scale Q.eps plot of the second component of the Stokes vector in linear scale U.eps plot of the third component of the Stokes vector in linear scale V.eps plot of the fourth component of the Stokes vector in linear scale

- For the selection of random numbers, the codes includes now some additional libraries: Numerical Recipes for Fortran 90 v.2.06 and SPRNG v.2.0 (only for the parallel version)

 Version 1.xx

1.07 - The text file mcshape.log was added to the output files. In this file there is the summary of some important information about the simulation, i.e. name and composition of the target, geometry, configuration of the source, version of the code used, etc.

1.06 - The input-output management was modified in order to rearrenge the output of the code. The input file SAMPLE.DAT has to be placed in the directory with the name of the target to compute (i.e. Water). In this directory, after the exacution of the code, you can find the folder RES with the results of the simulation. The name of the file with the full spectrum is now FULL_SPECTRUM.DAT.

1.04 - A bug has been corrected. It regards the erroneous calculation of some probabilities for the selection of the characteristic lines for the photoelectric effect.

1.01 - The energetic range was exented from 101 keV to 1001 keV

- It is possible to choose the step for the energy resolution

- Preprocessing instructions have been introduced to compile the same version of the code on different operating systems (Windows and Linux)

- Output files changed: the code creates a text file for each element of the Stokes vector I, Q, U and V and. For the Linux version, the plots of the Stokes parameters are created using Gnuplot.

 

©Copyright 2006
ALMA MATER STUDIORUM - Università  di Bologna
Via Zamboni, 33 - I-40126 Bologna