Elsevier Science Home
Computer Physics Communications Program Library
Full text online from Science Direct
Programs in Physics & Physical Chemistry
CPC Home

[Licence| Download | New Version Template] aedq_v2_0.tar.gz(25140 Kbytes)
Manuscript Title: carlomat, version 2 of the program for automatic computation of lowest order cross sections
Authors: Karol Kolodziej
Program title: carlomat, version 2.0
Catalogue identifier: AEDQ_v2_0
Distribution format: tar.gz
Journal reference: Comput. Phys. Commun. 185(2014)323
Programming language: Fortran 90/95.
Computer: All.
Operating system: Linux.
Keywords: Automatic calculation of cross sections, Monte Carlo, Extensions of Standard Model.
PACS: 12.15-y, 12.20-m, 12.60.Cn, 12.60.Fr.
Classification: 4.4, 11.2.

Does the new version supersede the previous version?: Yes

Nature of problem:
Leading order predictions for reactions of two particle scattering into a final state with up to 10 particles within the Standard Model and some effective models.

Solution method:
As in version 1 of the program, the matrix element in the helicity basis and multichannel Monte Carlo phase space integration routine are generated automatically for a user specified process. The color matrix is divided into smaller routines and written down as a stand alone program that is calculated prior to compilation and execution of the Monte Carlo program for computation of the cross section. The phase space integration routine is substantially shortened in order to speed up its compilation. The code generation part of the program is modified to incorporate the scalar electrodynamics and effective Lagrangians of the top quark interactions with the W and higgs bosons. Routines necessary for computing the helicity amplitudes of new couplings are added.

Reasons for new version:
The main reasons for the revision are:
  1. To adjust the program for the description of hadron collisions.
  2. To facilitate computation of the color matrix that is usually much more involved for processes of the hadron-hadron collision than for processes of electron-positron annihilation.
  3. To shorten compilation time of the generated kinematical routines.
  4. To implement some extensions of the standard model in the program.

Summary of revisions:
A few substantial modifications are introduced with respect to version 1.0 of the program. First, a single phase space parametrization is generated for the Feynman diagrams of the same topology taking into account possible differences in mappings of peaks in the individual diagrams, which speeds up a compilation time of the Monte Carlo program for multiparticle reactions by a factor 4-5 with respect to the previous version. Second, an interface to parton density functions is added that allows predictions to be made for hadron collisions. Third, calculation of the color matrix is facilitated. Fourth, the Cabibbo-Kobayashi-Maskawa mixing in the quark sector is implemented. Fifth, the effective models including scalar electrodynamics, the Wtb interaction with operators of dimension up to 5 and a general top-higgs coupling are implemented. Moreover, some minor modifications have been made and several bugs in the program have been corrected.

Restrictions:
Although the compilation time has been shortened in the current version, it still may be quite long for processes with 8 or more final state particles. Another limitation is the size of the color matrix that, if too big, may prevent compilation or result in a very long execution time of the color compilation program. This actually may happen already for some QCD processes with 7 partons such as gg → 5g, the commutation time of the color matrix of which, is about 200 hours.

Running time:
Depends strongly on the selected process and, to a lesser extent, on the Fortran compiler used. The following amounts of time are needed at different computation stages of the top quark pair production parton level process ggbudbar bbarμ-νbarμ, to produce the appended test output files on a PC with the Pentium 4 3.0 GHz processor with Absoft (GNU, Intel) Fortran compilers: code generation takes 3.7 s (3.7 s, 2.4 s), compilation, computation and simplification of the color matrix takes about 1 s (1 s, 1 s), compilation of all the generated routines takes just a few seconds and execution of the Monte Carlo program takes about 44 s (41 s, 23 s).