Commit 85576f56 authored by William Patrick Mccormack's avatar William Patrick Mccormack Committed by Ewelina Maria Lobodzinska
Browse files

Update of SuperChic DD jobOptions

parent 104d5596
The athena SuperChic interface described here was developed by Patrick McCormack (, who now works for CMS. For interface questions please contact Aleksandra Dimitrievska ( or Simone Pagan Griso ( For generator-specific questions, you might want to contact the SuperChic authors ([Lucian Harland-Lang]( or [Marek Tasevsky](
# Introduction
SuperChic 4 ([ref]( and [manual]( is a generator for exclusive and photon-induced physics. The current (As of May 2021) version of the athena interface of SuperChic has been strongly reworked from prior releases (those preceeding April 2021), now relying primarily on the standalone executables from LCG, rather than the Fortran code that works behind the scenes for SuperChic. The standalone executables can be found at e.g. /cvmfs/, /cvmfs/, or /cvmfs/
## Intro to standalone SuperChic
When running independently, SuperChic requires:
1) Folders named outputs/, inputs/, and evrecs/ in the directory that you want to run from
2) An input configuration file, probably named input.DAT. The format must follow that found in the directories with the standalone executables referenced above
3) The packages apfel and lhapdf must also be in the ROOT_INCLUDE_PATH and LD_LIBRARY_PATH of your environment. This is taken care of if you source a recent 21.6,AthGeneration release (21.6.70 or later)
Standalone SuperChic is run in two steps:
1) First, you would run "init < input.DAT", which generates files in the inputs/ directory. This step performs some initial integrations that get referenced in the next step.
2) Next, you would run "superchic < input.DAT", using the same input.DAT file. This step generates an LHE/HepMC/HEP-EVT in the evrecs/ folder that contains all of the generated events' information. This LHE file can then be fed into pythia for showering
# SuperChic Interface
The new athena SuperChic interface proceeds primarily as standalone SuperChic would. An example of SuperChic without showering can be found at /athena/Generators/Superchic_i/share/ It proceeds fairly linearly:
1) The necessary inputs/, outputs/, and evrecs/ folders are created if they don't already exist
2) An input.DAT file is generated using arguments that are given to "Init". The class type for Init is defined here: /athena/Generators/Superchic_i/python/ You can see all of the various input variables that can be given. The file is written with the writeInputDAT function in that python macro.
3) SuperChicRun also runs the init and superchic executables, creating an LHE file. It is important that the output format be LHE, because the next step, which creates an EVNT file is only compatible with the LHE format.
4) The EVNT file is filled using the /athena/Generators/Superchic_i/python/ macro. To use a format other than LHE, a different version of this macro would have to be written.
An example jobOptions file with pythia showering can be found at /athena/Generators/Superchic_i/share/ The key difference here is of course that Pythia8 is added to the genSeq, and that we have to be careful about what pythia command options are included. The /athena/Generators/Superchic_i/share/, /athena/Generators/Superchic_i/share/, /athena/Generators/Superchic_i/share/, and /athena/Generators/Superchic_i/share/ contain the latest recommended options for the case of photon-induced di-lepton for the cases of double-dissociated production, single dissociative production on one side and the other, and elasic production, respectively.
## Some Notes
* The S^2 model can be changed by adjusting the Init.isurv value
* The Init.diff should be either 'el', 'sda', 'sdb', or 'dd'. Be careful to make sure that the "BeamRemnants:unresolvedHadron = 0" component of the pythia commands is adjusted accordingly below (el -> 0, sda -> 2, sdb ->1, and dd -> 3)
* The list of processes for Init.proc can be found in section 7 of the [manual](
* Output kinematics can of course be changed from what is in the samples. Two body decays use "a" and "b" particles (as in pta and ptb). Four body decays use "a", "b", "c", "d", and so forth.
* Cuts that have "x" in them refer to the hard process as a whole, so e.g. the di-lepton system in a di-lepton event
* When running without showering with pythia, an important drawback of using LHE files should be noted: they do not use high enough precision. What I mean is the following. In, you can see that you enter in a four vector that contains every particle's px, py, pz, and E. The mass of the particles is then computed based on those values. Unfortunately, for highly boosted objects (i.e. those where E >> m), such as the final state protons, the calculation of four-vector mass with the precision given in the LHE files is slightly off. So in the output EVNT files, you'll see protons with masses in the range of 900 - 1000 MeV. I have set the "generated_mass" of each particle to the right mass, but if you retrieve "m_m" for a truth-level particle, it will be slightly incorrect.
# Running locally
To run locally:
1) Make a folder to hold the athena release, go inside it, and run: asetup 21.6,latest,AthGeneration (21.6.70 or any later release should do)
2) (optional) make a run folder and go inside it
3) mkdir 999999
4) put a joboptions file in 999999/. You may need to change the prefix from "jobOptions." to "mc.".
5) Run with e.g.:
``` --ecmEnergy=13000.0 --maxEvents=1000 --firstEvent=1 --randomSeed=14 --outputEVNTFile=test.pool.root --jobConfig 999999
# Releases
A functional version of the SuperChic interface was first incorporated into the nightly release 21.6,AthGeneration,r2021-04-30T0933. Any full release more recent than 21.6.70 should contain a functional interface. However, a slightly tweak was implemented in to put the initial state protons into the EVNT files in the unshowered case. This will likely be in release 21.6.72 and onward.
...@@ -68,7 +68,6 @@ genSeq.Pythia8.Commands += [ ...@@ -68,7 +68,6 @@ genSeq.Pythia8.Commands += [
"SpaceShower:dipoleRecoil = on", "SpaceShower:dipoleRecoil = on",
"SpaceShower:pTmaxMatch = 2", "SpaceShower:pTmaxMatch = 2",
"BeamRemnants:primordialKT = off", "BeamRemnants:primordialKT = off",
"SpaceShower:QEDshowerByQ = off",
"BeamRemnants:unresolvedHadron = 0", "BeamRemnants:unresolvedHadron = 0",
"SpaceShower:pTdampMatch=1" "SpaceShower:pTdampMatch=1"
] ]
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment