Tunes and CPM
Comparison between D6T, Z1, Z2, and Z2star
This is a comparison between the Pythia tunes
D6T, Z1, Z2, and Z2star in describing the charged particle multiplicity from
this CMS article.
To make the cfg.py files for Z1 and D6T, they must first be placed in the CMSSW_4_1_7/src/Configuration/GenProduction/python by running the commands:
cvs co Configuration/GenProduction/python/MinBias_TuneZ1_7TeV_pythia6_cff.py
cvs co Configuration/GenProduction/python/MinBias_TuneD6T_7TeV_pythia6_cff.py
in the CMSSW_4_1_7/src directory, and then do the
scram b
. After that the cfg.py files can be made using cmsDriver and this code from the
Rivet Twiki:
cmsDriver.py Configuration/GenProduction/python/MinBias_TuneZ2_7TeV_pythia6_cff.py -s GEN --datatier=GEN-SIM-RAW --conditions auto:mc --eventcontent RAWSIM --no_exec -n 10000 --python_filename=rivet_cfg.py --customise=Configuration/GenProduction/rivet_customize.py
where the MinBias_TuneZ2_7TeV_pythia6_cff.py should be changed to the correct filename.
Z2star can be made from a modification of Z2 without further problems by setting the parameters parp(82) = 1.93 and parp(90)=0.23 in the .py file.
The plots plotted in
ROOT are presented in
CPM_tunes.pdf.
Tuning of reconnection parameter
In order to improve the agreement between MC and data the parameters of Pythia can be tuned. Here the color reconnection is tuned from the MinBias Z2 tune.
The parameters that can be tuned to change the color reconnection in Pythia 6 are:
'MSTP(95)=6 ! CR (color reconnection parameters)'
'PARP(77)=1.016 ! CR'
'PARP(78)=0.538 ! CR'
The first paramter decides if reconnection is allowed at all and if it is, what model is used for it. The value 6 indicates that reconnection is only on for hadron-hadron collisions simulated with PYEVNW, and some more restrictions apply.
The second parameter decreases the probability for fast moving sting pieces to reconnect if larger than zero.
The third parameter controls the general probability for keeping the color history and not reconnecting:
P
Keep = (1 - PARP(78) )
n,
where n is the number of interactions that occurred in the event. (see also the
Pythia manual, and in the
update notes)
The results are shown in
CPM_RP_tunes.pdf, and since the relative error is hard to see for low multiplicities, the results can also be seen in the plots in
Parp77Tune and
Parp78Tune where it is plotted using
rivet-mkhtml.
--
RasmusSlothHansen - 05-Jul-2011