2010.02.08 StEemcGammaFilterMaker QA: QCD vs. gamma-jet ccept/reject rates

StEemcGammaFilterMaker QA

Pythia generated processes

Pythia gamma-jet	Pythia QCD 2->2 processes
14 f + fbar -> g + gamma	11 f + f' -> f + f' (QCD)
18 f + fbar -> gamma + gamma	12 f + fbar -> f' + fbar'
29 f + g -> f + gamma	13 f + fbar -> g + g
114 g + g -> gamma + gamma	28 f + g -> f + g
115 g + g -> g + gamma	53 g + g -> f + fbar
	68 g + g -> g + g

Number of generated events per parton pt bin

	Number of generated events (100events/job)
Parton pt range (GeV)	2-3	3-4	4-6	6-9	9-15	15-25
Pythia gamma-jet	50K	50K	50K	50K	50K	50K
Pythia QCD 2->2 processes	100K	100K	50K	50K	50K	50K

Filter configuration

Filter parameter	Value	Notes
mConeRadius	0.24
mSeedThreshold	2.5	Cluster seed energy threshold
mClusterThreshold	3.7	Cluster Et threshold
mEtaLow	0.9	EEMC acceptance
mEtaHigh	2.1	EEMC acceptance
mSmearEnergy	0	Disabled
mThrowTracks	0	Disabled
mCalDepth	279.5	ZDC SMD depth
mMinPartEnergy	1e-05	Disabled by mThrowTracks=0
mHadronScale	0.4	Downscale factor for hadron energy
mFilterMode	0	Accepting all events

stasim/Pythia options

• detp geometry y2006h
• Calorimeter cut for electromagnetic processes: emc_10keV
• call pytune(100): PYTUNE v1.015; CDF Tune A

Figures

Figure 1: Pythia Eemc gamma filter QA:
(left) False rejection, (right) fraction of accepted events

Comments

• Overall false rejection is < 0.02% (10^-4)
  for both QCD and gamma-jet simulations.
• In parton pt range 6-15 GeV the acceptance rate
  for the gamma-jet MC is ~ 50%
  (I think this is due to rapidity fiducial cut,
  otherwise it should be closer to 100%).
  For the QCD sample (in the same pt region)
  acceptance rate is 8-25%.
• Somehow for highest parton pt filter acceptance is the
  same for gamma-jet and QCD Monte-Carlo.

2010.02.09 BFC and Pythia QA: relative to trigger

QCD and gamma-jet data samples are described here

Pythia filter configuration

StEemcGammaFilter:: running the TEST mode (accepting all events). Set mFilterMode=1 to actually reject events
StEemcGammaFilter:: mConeRadius 0.22 mSeedThreshold 2.1 mClusterThreshold 3.25 mEtaLow 0.95 mEtaHigh 2.1
StEemcGammaFilter:: mCalDepth 279.5 mMinPartEnergy 1e-05 mHadronScale 0.4 mFilterMode 0 mPrintLevel 0

BFC filter configuration

StChain:INFO - Init() : Seed energy threshold = 2.8 GeV
StChain:INFO - Init() : Cluster eT threshold = 4.2 GeV
StChain:INFO - Init() : Maximum vertex = +/- 120 cm
StChain:INFO - Init() : Running the TEST mode (accepting all events). Set mFilterMode=1 to actually reject events in BFC

Accept/Reject relative to the total number of Pythia generated events

Figure 1: Fraction of accepted events

Figure 2: False rejection (Y-axis scale is 10^-3)

Accept/Reject relative to the number of triggered events

Figure 3: Fraction of accepted events (relative to triggered events)

Figure 4: False rejection relative to triggered events

2010.02.10 Money plots for W cross section

2010.02.11 BFC and Pythia QA: Gain no-gain-spread, mean=1.05

Click here for discussion and results with spread=0.05/gain=0.95

QCD and gamma-jet data samples are described here

Resultys without gain shift can be found here
(Note: ignore parton pt=25-35GeV for the gamma-jet sample since all jobs failed)

Pythia filter configuration

StEemcGammaFilter:: running the TEST mode (accepting all events). Set mFilterMode=1 to actually reject events
StEemcGammaFilter:: mConeRadius 0.22 mSeedThreshold 2.1 mClusterThreshold 3.25 mEtaLow 0.95 mEtaHigh 2.1
StEemcGammaFilter:: mCalDepth 279.5 mMinPartEnergy 1e-05 mHadronScale 0.4 mFilterMode 0 mPrintLevel 0

BFC filter configuration

StChain:INFO - Init() : Seed energy threshold = 2.8 GeV
StChain:INFO - Init() : Cluster eT threshold = 4.2 GeV
StChain:INFO - Init() : Maximum vertex = +/- 120 cm
StChain:INFO - Init() : Running the TEST mode (accepting all events). Set mFilterMode=1 to actually reject events in BFC

Accept/Reject relative to the total number of Pythia generated events

Figure 1: Fraction of accepted events

Accept rate: fract. of generated events
GammaJet
pt bin	Bfc Filter	Pythia Filter	L2gamma Trigger
pt=2-3	0.0023	0.06264	0.00148
pt=3-4	0.0242285	0.250601	0.0126854
pt=4-6	0.103111	0.427535	0.0571313
pt=6-9	0.16828	0.48368	0.13918
pt=9-15	0.1692	0.50118	0.1619
pt=15-25	0.12708	0.42904	0.11786
pt=25-35	0.05702	0.24854	0.0509
QCD
pt bin	Bfc Filter	Pythia Filter	L2gamma Trigger
pt=2-3	3.003e-05	0.00426426	2.002e-05
pt=3-4	0.0001001	0.0122923	1.001e-05
pt=4-6	0.00078	0.03166	0.00014
pt=6-9	0.00622	0.10538	0.00216
pt=9-15	0.02822	0.27666	0.01022
pt=15-25	0.07568	0.4405	0.03086
pt=25-35	0.0761	0.35556	0.04116

Figure 2: False rejection (Y-axis scale is 10^-3)

False reject: fract. of generated events
GammaJet
pt bin	Bfc Filter	Pythia Filter	L2gamma Trigger
pt=2-3	2e-05	0	0
pt=3-4	2.00401e-05	0	0
pt=4-6	8.08081e-05	0	0
pt=6-9	6e-05	4e-05	0
pt=9-15	0.0002	0.00018	0
pt=15-25	0.0001	4e-05	0
pt=25-35	0.00018	2e-05	0
QCD
pt bin	Bfc Filter	Pythia Filter	L2gamma Trigger
pt=2-3	0	0	0
pt=3-4	0	0	0
pt=4-6	0	0	0
pt=6-9	4e-05	6e-05	0
pt=9-15	4e-05	0.00026	0
pt=15-25	0.00016	0.00018	0
pt=25-35	4e-05	4e-05	0

Accept/Reject relative to the number of triggered events

Figure 3: Fraction of accepted events (relative to triggered events)

Figure 4: False rejection relative to triggered events

Accept/Reject relative to the number of Pythia filter accepted events

Figure 5: Fraction of accepted events relative to Pythia filter accepted events

Accept rate: fract. of Pythia-filtered events
GammaJet
pt bin	Bfc Filter	Pythia Filter	L2gamma Trigger
pt=2-3	0.0363985	1	0.0220307
pt=3-4	0.0966813	1	0.0501399
pt=4-6	0.241081	1	0.133488
pt=6-9	0.347461	1	0.287587
pt=9-15	0.336286	1	0.322639
pt=15-25	0.29573	1	0.274613
pt=25-35	0.228776	1	0.204635
QCD
pt bin	Bfc Filter	Pythia Filter	L2gamma Trigger
pt=2-3	0.00704225	1	0.00234742
pt=3-4	0.00488599	1	0.000814332
pt=4-6	0.0214782	1	0.00442198
pt=6-9	0.053141	1	0.0191687
pt=9-15	0.0965806	1	0.0356394
pt=15-25	0.16958	1	0.0695573
pt=25-35	0.213185	1	0.115592

Figure 6: False rejection relative to Pythia filter accepted events

False reject: fract. of Pythia-filtered events
GammaJet
pt bin	Bfc Filter	Pythia Filter	L2gamma Trigger
pt=2-3	0.000319285	0	0
pt=3-4	7.9968e-05	0	0
pt=4-6	0.000189009	0	0
pt=6-9	0.000124049	0	0
pt=9-15	0.000399058	0	0
pt=15-25	0.000233079	0	0
pt=25-35	0.000724229	0	0
QCD
pt bin	Bfc Filter	Pythia Filter	L2gamma Trigger
pt=2-3	0	0	0
pt=3-4	0	0	0
pt=4-6	0	0	0
pt=6-9	0.000379579	0	0
pt=9-15	0.000144582	0	0
pt=15-25	0.000363224	0	0
pt=25-35	0.000112499	0	0

2010.02.11 BFC and Pythia QA: Gain spread=0.05, mean=0.95

QCD and gamma-jet data samples are described here

Resultys without gain spread can be found here
(Note: ignore parton pt=25-35GeV for the gamma-jet sample since all jobs failed)

Gain spread implementation in StEEmcSlowMaker.cxx (private version):

void StEEmcSlowMaker::setTowerGainSpread(Float_t s, Float_t mTowerGainMean)
{
  LOG_INFO << "setTowerGainSpread(): gain spread: " << s << "; gain mean value: " << mTowerGainMean << endm;
  // initialize tower gain factors to 1
  for ( Int_t sec=0;sec<kEEmcNumSectors;sec++ )
    for ( Int_t sub=0;sub<kEEmcNumSubSectors;sub++ )
      for ( Int_t eta=0;eta<kEEmcNumEtas;eta++ )
    {
      //      mTowerGainFact[sec][sub][eta]=1.0;

      Float_t f = -1.0E9;
      while ( f <= -1. || f >= 1.0 )
        f = gRandom->Gaus(0., s);

      mTowerGainFact[sec][sub][eta] = mTowerGainMean + f;

    }
}

Pythia filter configuration

StEemcGammaFilter:: running the TEST mode (accepting all events). Set mFilterMode=1 to actually reject events
StEemcGammaFilter:: mConeRadius 0.22 mSeedThreshold 2.1 mClusterThreshold 3.25 mEtaLow 0.95 mEtaHigh 2.1
StEemcGammaFilter:: mCalDepth 279.5 mMinPartEnergy 1e-05 mHadronScale 0.4 mFilterMode 0 mPrintLevel 0

 

BFC filter configuration

StChain:INFO - Init() : Seed energy threshold = 2.8 GeV
StChain:INFO - Init() : Cluster eT threshold = 4.2 GeV
StChain:INFO - Init() : Maximum vertex = +/- 120 cm
StChain:INFO - Init() : Running the TEST mode (accepting all events). Set mFilterMode=1 to actually reject events in BFC

Accept/Reject relative to the total number of Pythia generated events

Figure 1: Fraction of accepted events

Figure 2: False rejection (Y-axis scale is 10^-3)

Accept/Reject relative to the number of triggered events

Figure 3: Fraction of accepted events (relative to triggered events)

Figure 4: False rejection relative to triggered events

2010.02.12 Final Pythia and BFC EEMC-gamma-filter paramter settings

Pythia generated processes

Prompt photons (gamma-jets)		2->2 QCD
id	Process	id	Process
14	f + fbar -> g + gamma	11	f + f' -> f + f' (QCD)
18	f + fbar -> gamma + gamma	12	f + fbar -> f' + fbar'
29	f + g -> f + gamma	13	f + fbar -> g + g
114	g + g -> gamma + gamma	28	f + g -> f + g
115	g + g -> g + gamma	53	g + g -> f + fbar
		68	g + g -> g + g

Number of generated events per parton pt bin

	Number of generated events (100events/job)
Parton pt range (GeV)	2-3	3-4	4-6	6-9	9-15	15-25	25-35
gamma-jets	50K	50K	50K	50K	50K	50K	50K
2->2 QCD processes	100K	100K	50K	50K	50K	50K	50K

Pythia Filter configuration

StRoot/StMCFilter/StEemcGammaFilter.cxx
StRoot/StMCFilter/StEemcGammaFilter.h

Filter parameter	Value	Notes
mConeRadius	0.22
mSeedThreshold	2.6	Cluster seed energy threshold (GeV)
mClusterThreshold	3.6	Cluster Et threshold (GeV)
mEtaLow	0.95	EEMC acceptance
mEtaHigh	2.1	EEMC acceptance
mMaxVertex	120.0	Vertex z cut (cm)
mCalDepth	279.5	EEMC SMD depth (cm)
mMinPartEnergy	1e-05	Ignore track with minim energy (GeV)
mHadronScale	0.4	Downscale factor for hadron energy
mFilterMode	0 / 1	0=Accept all events; 1=reject events

BFC Filter configuration

StRoot/StFilterMaker/StEemcGammaFilterMaker.cxx
StRoot/StFilterMaker/StEemcGammaFilterMaker.h

Filter parameter	Value	Notes
mSeedEnergyThreshold	3.4	Cluter seed energy threshold (GeV)
mClusterEtThreshold	4.5	Cluster Et threshold (GeV)
mEemcSamplingFraction	0.05	Assume 5% sampling fraction for EEMC
mMaxVertex	120.0	Vertex z cut (cm)
mFilterMode	0 / 1	0=Accept all events; 1=reject events

GammaMaker configuration

Filter parameter	Value	Notes
ConeRadius	0.7
ClusterEtThreshold	5.5	(GeV)
SeedEnergyThreshold	4.2	(GeV)
ClusterEnergyThreshold	5.5	(GeV)

EEMC SlowSimulator configuration

(for a moment private version) of StEEmcSlowMaker.cxx
with modified setTowerGainSpread(Float_t s, Float_t mTowerGainMean)

Filter parameter	Value	Notes
mTowerGainMean	1.05	Overall +5% gain shift
GainSpread	0	No gain spread

GSTAR/Pythia options

• detp geometry y2006h
• Calorimeter cut for electromagnetic processes: emc_10keV
• call pytune(100): PYTUNE v1.015; CDF Tune A
  or
  call pytune(320): PYTUNE Perugia; Perugia 0 tune

2010.02.16 Pythia and BFC filter QA vs. gamma candidate pt and eta

QCD and gamma-jet data samples and filter configurtions are given here

Note: for this study I have used ideal EEMC gains (no gain shift/spread)

Note on trigger effect intepretation:
There is no requirement for the Pythia gamma-jet sample to have direct gamma
headed to the EEMC, only requirement is to have a gamma candidate in the EEMC,
so the away side jet may also contribute.

Figure 1: pt distribution of the gamma candidates
for Pythia/Bfc level filter and triggered events
Event cuts: at least one gamma candidate, |v_z| <120
Left: Pythia gamma-jet MC; (right) 2->2 Pythia QCD
Lower plots: fraction of accepted gamma candidates by filter/trigger
No parton pt weights (= ignore bumps in pt distribution for gamma-jet sample)

Figure 2: Rapidity distribution of the gamma candidates (Same conditions as in Fig. 1)

Figure 3: pt distribution of false rejection for Pythia/Bfc filters
Candidate cuts: at least one gamma candidate, l2gamm-trigger=fired, |v_z| <120
Most of false rejection (~ 1-3% for QCD) is for gamma candidates with pt < 8GeV