PYTHIA muon software for PHENIX

LAST UPDATES (dec 2004)

january 8th, 2003 - created
january 14th, 2003 - add PYTHIA 6.205
december 21th, 2004 - add PYTHIA 6.205 for SL3 (Rob Hobbs)

This tutorial aims to help beginners to run through the PYTHIA event generation process for PHENIX:

PYTHIA documentation : provides links to PYTHIA homepage, PYTHIA manuals for version 5.7 (current PHENIX version) and version 6.2, and Pythia Particle/parton data tables for both versions 5.7 and 6.2 (be cautious, they are different !).
PYTHIA muon code : provides fortran code and makefiles to produce events with PYTHIA. Note that upgrade effort is made on PYTHIA 6.205 only.
PYTHIA muon cards : provides information about datacards used to drive the event generation and selection. Provides, in addition, some control parameter files to use as starting examples.
PYTHuple : gives information about PYTHIA output rootuple. Also provides a short example to analyze this rootuple.

PYTHIA documentation

Web pages
- PYTHIA (and JETSET) homepage
- PYTHIA 7 (proof-of-concept C++ version)
Paper docs
- PYTHIA 6.200 manual
- PYTHIA 5.7 manual (current PHENIX version)
Pythia Particle/parton data table
- PYTHIA 5.7
- PYTHIA 6.205
Notes:
- Some KF particle codes have been changed between PYTHIA 5.7 and PYTHIA 6.205. As an example, psi' is quoted as KF=30443 in PYTHIA 5.7 and as KF=100443 in PYTHIA 6.205. These changes shouldn't affect the KF particle code to GEANT3 (PISA) particle code translation (see SUBROUTINE GETPAR in gen_pythia_muon.f).
- In addition IDC decay codes and KC codes have been changed between the two PYTHIA versions. These changes should be taken into account when one is requesting some specific decay channels in the .par files.

PYTHIA muon code

how to install and run the code

download pythia57.tar (version 5.7) or pythia6205.tar (version 6.205) or pythia6205_SL3.tar (version 6.205 for SL3)
> tar xvf pythiaXX.tar
> cd srcpythia
at CCF: > cp CCF_makefile makefile at RCF: > cp RCF_makefile makefile
> make
> cd ../bin
> ln -sf ../parpythia/Jpsi.par pythia.par
> gen_pythia_muon

PYTHIA muon cards

The PYTHIA muon production is controlled by a set of datacards put together in one control parameters (.par) file. Datacards are gathered in five different sets controlling five different steps in the data production:

control cards for process description : contains pythia datacards to define the process.
control datacards for event generation : number of generated events and center of mass energy.
control datacards for geometrical selection : definition of North and South acceptance region.
control datacards for signal description : select particles to be written in the output.
control datacards for output selection : select information to be written in the output.

Some control parameter files can be find here.

control cards for process description

General settings
- MSTP(51) and/or MSTP(52) = choice of proton parton-distribution set and library.
- MSEL = a switch to select between full user control and some preprogrammed alternatives.
  - MSEL = 0 : full user control. Desired subprocesses have to be switched on in MSUB.
  - MSEL = 1 : QCD high-pT processes.
Process selection
- MSUB = array to be set when MSEL=0 to choose which subset of subprocesses to include in the generation. Note that ISUB is the process code number (as an example, ISUB=86 is gg->J/psi+g).
  - MSUB(ISUB) = 0 : the subprocess ISUB is excluded.
  - MSUB(ISUB) = 1 : the subprocess ISUB is included.
- KPFR(ISUB,J) = give the KF flavor codes of the products produced in subprocess ISUB.
  - KPFR(86,1) = 553 : produce upsilon (1s) instead of J/psi for process 86.
Particle decays
- MDME(IDC,1) = on/off switch for individual decay channel IDC.
  - MDME(IDC,1) = 0 : channel is switched off.
  - MDME(IDC,1) = 1 : channel is switched on.

control datacards for event generation

tnumevt = number of generated events
sqrts = center of mass energy

control datacards for geometrical selection

With these datacards, one defines the acceptance regions. In fact, during the generation process, Pythia will stop generating events when tnumevt signal events (defined in control datacards for signal description) are generated within these acceptance regions.

thmi_no = north theta min : north object if theta.GE.thmi_no (deg)
thma_no = north theta max : north object if theta.LE.thma_no (deg)
no_on = if .true., switch north detector on
thmi_su = south theta min : south object if theta.GE.thmi_su (deg)
thma_su = south theta max : south object if theta.LE.thma_su (deg)
su_on = if .true., switch south detector on

control datacards for signal description

With these datacards, one defines the signal events. In fact, during the generation process, Pythia will stop generating events when tnumevt signal events are generated within the acceptance regions (defined in control datacards for geometrical description).

isig1 = particle parent's ID (443=J/psi)
iprod1 = particle ID (13=mu-)
nprod1 = number [min,max] of iprod1 (coming from isig1) particles
p1mima = momentum [min,max] of iprod1 particles
isig2 = particle parent's ID (443=J/psi)
iprod2 = particle ID (-13=mu+)
nprod2 = number [min,max] of iprod2 (coming from isig2) particles
p2mima = momentum [min,max] of iprod2 particles
Mi = invariant mass [min,max] of the iprod1, iprod2 system (nprod1 & nprod2 must be=1)

As an example, if isig1=443 / iprod1=13 / nprod1=1,2 / p1mima=0,1000, an event will be defined as a signal event only if it contains at least 1 and at most 2 mu- with momentum values between 0 and 1000 GeV; all of them coming from a J/psi decay.

Notes:

if isig1 (and/or isig2) = 0: iprod1 (and/or iprod2) particle's parent is not checked.
if iprod1 (and/or iprod2) = 0: all particles coming from isig1 (and/or isig2) are checked.

Resonance Mode: (pythia6205 only) isig1=isig2!=0 and iprod1=-iprod2

if nprod1=1,1 and nprod2=1,1: keep only events with 1 iprod1 type particle and 1 iprod2 type particle within detector acceptance (where p1mima and p2mima are included in the acceptance definition). In addition, both particles come from the same parent particle.
If nprod1=1,"large number" and nprod2=1,"large number": same as above, except that more than 1 iprod1 particle and iprod2 particle can be found within detector acceptance (no p1mima and p2mima selection for additionnal particles).

control datacards for output selection

With these datacards, one controls output files. When generating events, two files are created: pythia.dat which contains the particle information used by PISA as an input, and pythia.hist which contains one or two ntuples (depending on do1 and do2 switches) for analysis at generation level; ntuple h100 contains signal events within user defined acceptance while ntuple h1000 contains all generated signal events.

verb = verbose mode: print particle/parton data table and list the first 10 events details
wriflag = output file option: if .true. write (in pythia.dat) particle from signal selection only. If .false. write all particles from selected events.
h_file = ntuple file option: ntuple filename
do1 = ntuple file option: if .TRUE. produce ntuple before selection
do2 = ntuple file option: if .TRUE. produce ntuple after selection

Notes:

Note that for J/psi->mu+mu-, if wriflag = true only two particles, the two muons, are written in pythia.dat. Consequently, at reconstruction stage, no vertex position will be given by the vertex detectors.

some control parameter files

The following files provide, for tutorial purpose, datacard sets for some useful processes. Some processes are pretty well defined in pythia; they are indicated by a green ball. Other processes need user investigation; they are indicated by a red ball.

= 100% confidence

= 50% confidence

= 10% confidence

Pythia 5.7	Pythia 6.205
Drell-Yan	Drell-Yan
DDbar	DDbar
phi	phi
J/psi	J/psi
	BBbar->J/psi
psi'	psi'
upsilon 1S	upsilon 1S
upsilon 2S	upsilon 2S

PYTHuple

The pythia output rootuple

First of all, since PYTHIA is using HBOOK, one gets a HBOOK file as an output. In order to produce a ROOT file, type:

> h2root pythia.hist

Then, depending of what selection you make in the datacards (see do1 and do2 ), you'll get one or two TTree.

h100 is the TTree containing all the generated events.
h1000 is the TTree containing all the accepted events (the ones which are written in pythia.dat).

CAUTION: The maximum memory size allowed by HBOOK is ~130 MB, meaning that you can't produce a HBOOK file bigger than that size. Depending of the cuts you use, h100 can be huge and then reach the 130MB limit. If you want to produce more then 10000 events, you should first check what the output size should be.

The pythia rootuple variables

h100 and h1000 TTree have the same variable contain. Here are the variables description:

Event = event number
Vx = event vertex x position
Vy = event vertex y position
Vz = event vertex z position
ID = particle ID
Px = particle Px
Py = particle Py
Pz = particle Pz
E = particle E
pID = particle's parent ID
pPx = particle's parent Px
pPy = particle's parent Py
pPz = particle's parent Pz
pE = particle's parent E
gpID = particle's grand-parent ID
gpPx = particle's grand-parent Px
gpPy = particle's grand-parent Py
gpPz = particle's grand-parent Pz
gpE = particle's grand-parent E

A simple root macro to analyze pythia output

In "your pythia"/bin directory:

download h1000.C
> h2root pythia.hist
> root
root [0] .x h1000.C

Page maintained by Frédéric