1 /**************************************************************************
2 * Copyright(c) 1998-1999, ALICE Experiment at CERN, All rights reserved. *
4 * Author: The ALICE Off-line Project. *
5 * Contributors are mentioned in the code where appropriate. *
7 * Permission to use, copy, modify and distribute this software and its *
8 * documentation strictly for non-commercial purposes is hereby granted *
9 * without fee, provided that the above copyright notice appears in all *
10 * copies and that both the copyright notice and this permission notice *
11 * appear in the supporting documentation. The authors make no claims *
12 * about the suitability of this software for any purpose. It is *
13 * provided "as is" without express or implied warranty. *
14 **************************************************************************/
18 #include "AliPythia.h"
19 #include "AliPythiaRndm.h"
20 #include "AliFastGlauber.h"
21 #include "AliQuenchingWeights.h"
22 #include "AliOmegaDalitz.h"
24 #include "TLorentzVector.h"
25 #include "PyquenCommon.h"
30 # define pyclus pyclus_
31 # define pycell pycell_
32 # define pyshow pyshow_
33 # define pyrobo pyrobo_
34 # define pyquen pyquen_
35 # define pyevnw pyevnw_
36 # define pyshowq pyshowq_
37 # define qpygin0 qpygin0_
38 # define pytune pytune_
39 # define py2ent py2ent_
42 # define pyclus PYCLUS
43 # define pycell PYCELL
44 # define pyrobo PYROBO
45 # define pyquen PYQUEN
46 # define pyevnw PYEVNW
47 # define pyshowq PYSHOWQ
48 # define qpygin0 QPYGIN0
49 # define pytune PYTUNE
50 # define py2ent PY2ENT
51 # define type_of_call _stdcall
54 extern "C" void type_of_call pyclus(Int_t & );
55 extern "C" void type_of_call pycell(Int_t & );
56 extern "C" void type_of_call pyshow(Int_t &, Int_t &, Double_t &);
57 extern "C" void type_of_call pyrobo(Int_t &, Int_t &, Double_t &, Double_t &, Double_t &, Double_t &, Double_t &);
58 extern "C" void type_of_call pyquen(Double_t &, Int_t &, Double_t &);
59 extern "C" void type_of_call pyevnw(){;}
60 extern "C" void type_of_call pyshowq(Int_t &, Int_t &, Double_t &);
61 extern "C" void type_of_call pytune(Int_t &);
62 extern "C" void type_of_call py2ent(Int_t &, Int_t&, Int_t&, Double_t&);
63 extern "C" void type_of_call qpygin0();
64 //_____________________________________________________________________________
66 AliPythia* AliPythia::fgAliPythia=NULL;
68 AliPythia::AliPythia():
81 // Default Constructor
84 if (!AliPythiaRndm::GetPythiaRandom())
85 AliPythiaRndm::SetPythiaRandom(GetRandom());
87 fQuenchingWeights = 0;
89 for (i = 0; i < 501; i++) fDefMDCY[i] = 0;
90 for (i = 0; i < 2001; i++) fDefMDME[i] = 0;
91 for (i = 0; i < 4; i++) fZQuench[i] = 0;
94 AliPythia::AliPythia(const AliPythia& pythia):
105 fQuenchingWeights(0),
111 for (i = 0; i < 501; i++) fDefMDCY[i] = 0;
112 for (i = 0; i < 2001; i++) fDefMDME[i] = 0;
113 for (i = 0; i < 4; i++) fZQuench[i] = 0;
117 void AliPythia::ProcInit(Process_t process, Float_t energy, StrucFunc_t strucfunc, Int_t itune)
119 // Initialise the process to generate
120 if (!AliPythiaRndm::GetPythiaRandom())
121 AliPythiaRndm::SetPythiaRandom(GetRandom());
127 fStrucFunc = strucfunc;
128 //...Switch off decay of pi0, K0S, Lambda, Sigma+-, Xi0-, Omega-.
129 SetMDCY(Pycomp(111) ,1,0); // pi0
130 SetMDCY(Pycomp(310) ,1,0); // K0S
131 SetMDCY(Pycomp(3122),1,0); // kLambda
132 SetMDCY(Pycomp(3112),1,0); // sigma -
133 SetMDCY(Pycomp(3222),1,0); // sigma +
134 SetMDCY(Pycomp(3312),1,0); // xi -
135 SetMDCY(Pycomp(3322),1,0); // xi 0
136 SetMDCY(Pycomp(3334),1,0); // omega-
137 // Select structure function
139 SetMSTP(51, AliStructFuncType::PDFsetIndex(strucfunc));
140 // Particles produced in string fragmentation point directly to either of the two endpoints
141 // of the string (depending in the side they were generated from).
145 // Pythia initialisation for selected processes//
149 for (Int_t i=1; i<= 200; i++) {
152 // select charm production
155 case kPyOldUEQ2ordered: //Old underlying events with Q2 ordered QCD processes
156 // Multiple interactions on.
158 // Double Gaussian matter distribution.
164 // Reference energy for pT0 and energy rescaling pace.
167 // String drawing almost completely minimizes string length.
170 // ISR and FSR activity.
176 case kPyOldUEQ2ordered2:
177 // Old underlying events with Q2 ordered QCD processes
178 // Multiple interactions on.
180 // Double Gaussian matter distribution.
186 // Reference energy for pT0 and energy rescaling pace.
188 SetPARP(90,0.16); // here is the difference with kPyOldUEQ2ordered
189 // String drawing almost completely minimizes string length.
192 // ISR and FSR activity.
199 // Old production mechanism: Old Popcorn
202 // (D=2) Like MSTJ(12)=2 but added prod ofthe 1er rank baryon
204 // (D=1)see can be used to form baryons (BARYON JUNCTION)
210 // heavy quark masses
240 case kPyCharmUnforced:
249 case kPyBeautyUnforced:
259 // Minimum Bias pp-Collisions
262 // select Pythia min. bias model
264 SetMSUB(92,1); // single diffraction AB-->XB
265 SetMSUB(93,1); // single diffraction AB-->AX
266 SetMSUB(94,1); // double diffraction
267 SetMSUB(95,1); // low pt production
272 case kPyMbAtlasTuneMC09:
273 // Minimum Bias pp-Collisions
276 // select Pythia min. bias model
278 SetMSUB(92,1); // single diffraction AB-->XB
279 SetMSUB(93,1); // single diffraction AB-->AX
280 SetMSUB(94,1); // double diffraction
281 SetMSUB(95,1); // low pt production
286 case kPyMbWithDirectPhoton:
287 // Minimum Bias pp-Collisions with direct photon processes added
290 // select Pythia min. bias model
292 SetMSUB(92,1); // single diffraction AB-->XB
293 SetMSUB(93,1); // single diffraction AB-->AX
294 SetMSUB(94,1); // double diffraction
295 SetMSUB(95,1); // low pt production
308 // Minimum Bias pp-Collisions
311 // select Pythia min. bias model
313 SetMSUB(92,1); // single diffraction AB-->XB
314 SetMSUB(93,1); // single diffraction AB-->AX
315 SetMSUB(94,1); // double diffraction
316 SetMSUB(95,1); // low pt production
319 // Les Houches Working Group 05 Minimum Bias pp-Collisions: hep-ph/0604120
320 // -> Pythia 6.3 or above is needed
323 SetMSUB(92,1); // single diffraction AB-->XB
324 SetMSUB(93,1); // single diffraction AB-->AX
325 SetMSUB(94,1); // double diffraction
326 SetMSUB(95,1); // low pt production
328 SetMSTP(51,AliStructFuncType::PDFsetIndex(kCTEQ6ll)); // CTEQ6ll pdf
332 SetMSTP(81,1); // Multiple Interactions ON
333 SetMSTP(82,4); // Double Gaussian Model
336 SetPARP(82,2.3); // [GeV] PT_min at Ref. energy
337 SetPARP(83,0.5); // Core density in proton matter distribution (def.value)
338 SetPARP(84,0.5); // Core radius
339 SetPARP(85,0.9); // Regulates gluon prod. mechanism
340 SetPARP(90,0.2); // 2*epsilon (exponent in power law)
344 // Minimum Bias pp-Collisions
347 // select Pythia min. bias model
349 SetMSUB(95,1); // low pt production
356 SetMSTP(91,1);// Width (1=gaussian) primordial kT dist. inside hadrons
357 SetPARP(91,1.); // <kT^2> = PARP(91,1.)^2
358 SetPARP(93,5.); // Upper cut-off
360 SetPMAS(4,1,1.2); // Charm quark mass
361 SetPMAS(5,1,4.78); // Beauty quark mass
362 SetPARP(71,4.); // Defaut value
371 // Pythia Tune A (CDF)
373 SetPARP(67,2.5); // Regulates Initial State Radiation (value from best fit to D0 dijet analysis)
374 SetMSTP(82,4); // Double Gaussian Model
375 SetPARP(82,2.0); // [GeV] PT_min at Ref. energy
376 SetPARP(84,0.4); // Core radius
377 SetPARP(85,0.90) ; // Regulates gluon prod. mechanism
378 SetPARP(86,0.95); // Regulates gluon prod. mechanism
379 SetPARP(89,1800.); // [GeV] Ref. energy
380 SetPARP(90,0.25); // 2*epsilon (exponent in power law)
385 case kPyCharmPbPbMNR:
387 case kPyDPlusPbPbMNR:
388 case kPyDPlusStrangePbPbMNR:
389 // Tuning of Pythia parameters aimed to get a resonable agreement
390 // between with the NLO calculation by Mangano, Nason, Ridolfi for the
391 // c-cbar single inclusive and double differential distributions.
392 // This parameter settings are meant to work with Pb-Pb collisions
393 // (AliGenPythia::SetNuclei) and with kCTEQ4L PDFs.
394 // To get a good agreement the minimum ptHard (AliGenPythia::SetPtHard)
395 // has to be set to 2.1GeV. Example in ConfigCharmPPR.C.
407 case kPyDPlusStrangepPbMNR:
408 // Tuning of Pythia parameters aimed to get a resonable agreement
409 // between with the NLO calculation by Mangano, Nason, Ridolfi for the
410 // c-cbar single inclusive and double differential distributions.
411 // This parameter settings are meant to work with p-Pb collisions
412 // (AliGenPythia::SetNuclei) and with kCTEQ4L PDFs.
413 // To get a good agreement the minimum ptHard (AliGenPythia::SetPtHard)
414 // has to be set to 2.1GeV. Example in ConfigCharmPPR.C.
427 case kPyDPlusStrangeppMNR:
428 case kPyLambdacppMNR:
429 // Tuning of Pythia parameters aimed to get a resonable agreement
430 // between with the NLO calculation by Mangano, Nason, Ridolfi for the
431 // c-cbar single inclusive and double differential distributions.
432 // This parameter settings are meant to work with pp collisions
433 // (AliGenPythia::SetNuclei) and with kCTEQ4L PDFs.
434 // To get a good agreement the minimum ptHard (AliGenPythia::SetPtHard)
435 // has to be set to 2.1GeV. Example in ConfigCharmPPR.C.
445 case kPyCharmppMNRwmi:
446 // Tuning of Pythia parameters aimed to get a resonable agreement
447 // between with the NLO calculation by Mangano, Nason, Ridolfi for the
448 // c-cbar single inclusive and double differential distributions.
449 // This parameter settings are meant to work with pp collisions
450 // and with kCTEQ5L PDFs.
451 // Added multiple interactions according to ATLAS tune settings.
452 // To get a "reasonable" agreement with MNR results, events have to be
453 // generated with the minimum ptHard (AliGenPythia::SetPtHard)
455 // To get a "perfect" agreement with MNR results, events have to be
456 // generated in four ptHard bins with the following relative
472 case kPyBeautyPbPbMNR:
473 // Tuning of Pythia parameters aimed to get a resonable agreement
474 // between with the NLO calculation by Mangano, Nason, Ridolfi for the
475 // b-bbar single inclusive and double differential distributions.
476 // This parameter settings are meant to work with Pb-Pb collisions
477 // (AliGenPythia::SetNuclei) and with kCTEQ4L PDFs.
478 // To get a good agreement the minimum ptHard (AliGenPythia::SetPtHard)
479 // has to be set to 2.75GeV. Example in ConfigBeautyPPR.C.
491 case kPyBeautypPbMNR:
492 // Tuning of Pythia parameters aimed to get a resonable agreement
493 // between with the NLO calculation by Mangano, Nason, Ridolfi for the
494 // b-bbar single inclusive and double differential distributions.
495 // This parameter settings are meant to work with p-Pb collisions
496 // (AliGenPythia::SetNuclei) and with kCTEQ4L PDFs.
497 // To get a good agreement the minimum ptHard (AliGenPythia::SetPtHard)
498 // has to be set to 2.75GeV. Example in ConfigBeautyPPR.C.
511 // Tuning of Pythia parameters aimed to get a resonable agreement
512 // between with the NLO calculation by Mangano, Nason, Ridolfi for the
513 // b-bbar single inclusive and double differential distributions.
514 // This parameter settings are meant to work with pp collisions
515 // (AliGenPythia::SetNuclei) and with kCTEQ4L PDFs.
516 // To get a good agreement the minimum ptHard (AliGenPythia::SetPtHard)
517 // has to be set to 2.75GeV. Example in ConfigBeautyPPR.C.
532 case kPyBeautyppMNRwmi:
533 // Tuning of Pythia parameters aimed to get a resonable agreement
534 // between with the NLO calculation by Mangano, Nason, Ridolfi for the
535 // b-bbar single inclusive and double differential distributions.
536 // This parameter settings are meant to work with pp collisions
537 // and with kCTEQ5L PDFs.
538 // Added multiple interactions according to ATLAS tune settings.
539 // To get a "reasonable" agreement with MNR results, events have to be
540 // generated with the minimum ptHard (AliGenPythia::SetPtHard)
542 // To get a "perfect" agreement with MNR results, events have to be
543 // generated in four ptHard bins with the following relative
566 //Inclusive production of W+/-
572 // //f fbar -> gamma W+
579 // Initial/final parton shower on (Pythia default)
580 // With parton showers on we are generating "W inclusive process"
581 SetMSTP(61,1); //Initial QCD & QED showers on
582 SetMSTP(71,1); //Final QCD & QED showers on
588 //Inclusive production of Z
593 // // f fbar -> g Z/gamma
595 // // f fbar -> gamma Z/gamma
597 // // f g -> f Z/gamma
599 // // f gamma -> f Z/gamma
602 //only Z included, not gamma
605 // Initial/final parton shower on (Pythia default)
606 // With parton showers on we are generating "Z inclusive process"
607 SetMSTP(61,1); //Initial QCD & QED showers on
608 SetMSTP(71,1); //Final QCD & QED showers on
617 if (itune > -1) Pytune(itune);
620 SetMSTP(41,1); // all resonance decays switched on
621 Initialize("CMS","p","p",fEcms);
625 Int_t AliPythia::CheckedLuComp(Int_t kf)
627 // Check Lund particle code (for debugging)
629 printf("\n Lucomp kf,kc %d %d",kf,kc);
633 void AliPythia::SetNuclei(Int_t a1, Int_t a2, Int_t pdf)
635 // Treat protons as inside nuclei with mass numbers a1 and a2
636 // The MSTP array in the PYPARS common block is used to enable and
637 // select the nuclear structure functions.
638 // MSTP(52) : (D=1) choice of proton and nuclear structure-function library
639 // =1: internal PYTHIA acording to MSTP(51)
640 // =2: PDFLIB proton s.f., with MSTP(51) = 1000xNGROUP+NSET
641 // If the following mass number both not equal zero, nuclear corrections of the stf are used.
642 // MSTP(192) : Mass number of nucleus side 1
643 // MSTP(193) : Mass number of nucleus side 2
644 // MSTP(194) : Nuclear structure function: 0: EKS98 1:EPS08
652 AliPythia* AliPythia::Instance()
654 // Set random number generator
658 fgAliPythia = new AliPythia();
663 void AliPythia::PrintParticles()
665 // Print list of particl properties
667 char* name = new char[16];
668 for (Int_t kf=0; kf<1000000; kf++) {
669 for (Int_t c = 1; c > -2; c-=2) {
670 Int_t kc = Pycomp(c*kf);
672 Float_t mass = GetPMAS(kc,1);
673 Float_t width = GetPMAS(kc,2);
674 Float_t tau = GetPMAS(kc,4);
680 printf("\n mass, width, tau: %6d %s %10.3f %10.3e %10.3e",
681 c*kf, name, mass, width, tau);
685 printf("\n Number of particles %d \n \n", np);
688 void AliPythia::ResetDecayTable()
690 // Set default values for pythia decay switches
692 for (i = 1; i < 501; i++) SetMDCY(i,1,fDefMDCY[i]);
693 for (i = 1; i < 2001; i++) SetMDME(i,1,fDefMDME[i]);
696 void AliPythia::SetDecayTable()
698 // Set default values for pythia decay switches
701 for (i = 1; i < 501; i++) fDefMDCY[i] = GetMDCY(i,1);
702 for (i = 1; i < 2001; i++) fDefMDME[i] = GetMDME(i,1);
705 void AliPythia::Pyclus(Int_t& njet)
707 // Call Pythia clustering algorithm
712 void AliPythia::Pycell(Int_t& njet)
714 // Call Pythia jet reconstruction algorithm
719 void AliPythia::Pyshow(Int_t ip1, Int_t ip2, Double_t qmax)
721 // Call Pythia jet reconstruction algorithm
723 pyshow(ip1, ip2, qmax);
726 void AliPythia::Pyrobo(Int_t imi, Int_t ima, Double_t the, Double_t phi, Double_t bex, Double_t bey, Double_t bez)
728 pyrobo(imi, ima, the, phi, bex, bey, bez);
731 void AliPythia::Pytune(Int_t itune)
735 C ITUNE NAME (detailed descriptions below)
736 C 0 Default : No settings changed => linked Pythia version's defaults.
737 C ====== Old UE, Q2-ordered showers ==========================================
738 C 100 A : Rick Field's CDF Tune A
739 C 101 AW : Rick Field's CDF Tune AW
740 C 102 BW : Rick Field's CDF Tune BW
741 C 103 DW : Rick Field's CDF Tune DW
742 C 104 DWT : Rick Field's CDF Tune DW with slower UE energy scaling
743 C 105 QW : Rick Field's CDF Tune QW (NB: needs CTEQ6.1M pdfs externally)
744 C 106 ATLAS-DC2: Arthur Moraes' (old) ATLAS tune (ATLAS DC2 / Rome)
745 C 107 ACR : Tune A modified with annealing CR
746 C 108 D6 : Rick Field's CDF Tune D6 (NB: needs CTEQ6L pdfs externally)
747 C 109 D6T : Rick Field's CDF Tune D6T (NB: needs CTEQ6L pdfs externally)
748 C ====== Intermediate Models =================================================
749 C 200 IM 1 : Intermediate model: new UE, Q2-ordered showers, annealing CR
750 C 201 APT : Tune A modified to use pT-ordered final-state showers
751 C ====== New UE, interleaved pT-ordered showers, annealing CR ================
752 C 300 S0 : Sandhoff-Skands Tune 0
753 C 301 S1 : Sandhoff-Skands Tune 1
754 C 302 S2 : Sandhoff-Skands Tune 2
755 C 303 S0A : S0 with "Tune A" UE energy scaling
756 C 304 NOCR : New UE "best try" without colour reconnections
757 C 305 Old : New UE, original (primitive) colour reconnections
758 C 306 ATLAS-CSC: Arthur Moraes' (new) ATLAS tune (needs CTEQ6L externally)
759 C ======= The Uppsala models =================================================
760 C ( NB! must be run with special modified Pythia 6.215 version )
761 C ( available from http://www.isv.uu.se/thep/MC/scigal/ )
762 C 400 GAL 0 : Generalized area-law model. Old parameters
763 C 401 SCI 0 : Soft-Colour-Interaction model. Old parameters
764 C 402 GAL 1 : Generalized area-law model. Tevatron MB retuned (Skands)
769 void AliPythia::Py2ent(Int_t idx, Int_t pdg1, Int_t pdg2, Double_t p){
770 // Inset 2-parton system at line idx
771 py2ent(idx, pdg1, pdg2, p);
775 void AliPythia::InitQuenching(Float_t cMin, Float_t cMax, Float_t k, Int_t iECMethod, Float_t zmax, Int_t ngmax)
778 // (1) The quenching model using quenching weights according to C. Salgado and U. Wiedemann
779 // (2) The nuclear geometry using the Glauber Model
782 fGlauber = AliFastGlauber::Instance();
784 fGlauber->SetCentralityClass(cMin, cMax);
786 fQuenchingWeights = new AliQuenchingWeights();
787 fQuenchingWeights->InitMult();
788 fQuenchingWeights->SetK(k);
789 fQuenchingWeights->SetECMethod(AliQuenchingWeights::kECMethod(iECMethod));
796 void AliPythia::Quench()
800 // Simple Jet Quenching routine:
801 // =============================
802 // The jet formed by all final state partons radiated by the parton created
803 // in the hard collisions is quenched by a factor (1-z) using light cone variables in
804 // the initial parton reference frame:
805 // (E + p_z)new = (1-z) (E + p_z)old
810 // The lost momentum is first balanced by one gluon with virtuality > 0.
811 // Subsequently the gluon splits to yield two gluons with E = p.
815 static Float_t eMean = 0.;
816 static Int_t icall = 0;
821 Int_t klast[4] = {-1, -1, -1, -1};
823 Int_t numpart = fPyjets->N;
824 Double_t px = 0., py = 0., pz = 0., e = 0., m = 0., p = 0., pt = 0., theta = 0., phi = 0.;
825 Double_t pxq[4], pyq[4], pzq[4], eq[4], yq[4], mq[4], pq[4], phiq[4], thetaq[4], ptq[4];
827 Double_t wjtKick[4] = {0., 0., 0., 0.};
833 // Sore information about Primary partons
836 // 0, 1 partons from hard scattering
837 // 2, 3 partons from initial state radiation
839 for (Int_t i = 2; i <= 7; i++) {
841 // Skip gluons that participate in hard scattering
842 if (i == 4 || i == 5) continue;
843 // Gluons from hard Scattering
844 if (i == 6 || i == 7) {
846 pxq[j] = fPyjets->P[0][i];
847 pyq[j] = fPyjets->P[1][i];
848 pzq[j] = fPyjets->P[2][i];
849 eq[j] = fPyjets->P[3][i];
850 mq[j] = fPyjets->P[4][i];
852 // Gluons from initial state radiation
854 // Obtain 4-momentum vector from difference between original parton and parton after gluon
855 // radiation. Energy is calculated independently because initial state radition does not
856 // conserve strictly momentum and energy for each partonic system independently.
858 // Not very clean. Should be improved !
862 pxq[j] = fPyjets->P[0][i] - fPyjets->P[0][i+2];
863 pyq[j] = fPyjets->P[1][i] - fPyjets->P[1][i+2];
864 pzq[j] = fPyjets->P[2][i] - fPyjets->P[2][i+2];
865 mq[j] = fPyjets->P[4][i];
866 eq[j] = TMath::Sqrt(pxq[j] * pxq[j] + pyq[j] * pyq[j] + pzq[j] * pzq[j] + mq[j] * mq[j]);
869 // Calculate some kinematic variables
871 yq[j] = 0.5 * TMath::Log((eq[j] + pzq[j] + 1.e-14) / (eq[j] - pzq[j] + 1.e-14));
872 pq[j] = TMath::Sqrt(pxq[j] * pxq[j] + pyq[j] * pyq[j] + pzq[j] * pzq[j]);
873 phiq[j] = TMath::Pi()+TMath::ATan2(-pyq[j], -pxq[j]);
874 ptq[j] = TMath::Sqrt(pxq[j] * pxq[j] + pyq[j] * pyq[j]);
875 thetaq[j] = TMath::ATan2(ptq[j], pzq[j]);
876 qPdg[j] = fPyjets->K[1][i];
882 fGlauber->GetI0I1ForPythiaAndXY(4, phiq, int0, int1, fXJet, fYJet, 15.);
884 for (Int_t j = 0; j < 4; j++) {
886 // Quench only central jets and with E > 10.
890 Int_t itype = (qPdg[j] == 21) ? 2 : 1;
891 Double_t eloss = fQuenchingWeights->GetELossRandomKFast(itype, int0[j], int1[j], eq[j]);
893 if (TMath::Abs(yq[j]) > 2.5 || eq[j] < 10.) {
896 if (eq[j] > 40. && TMath::Abs(yq[j]) < 0.5) {
902 Double_t l = fQuenchingWeights->CalcLk(int0[j], int1[j]);
903 wjtKick[j] = TMath::Sqrt(l * fQuenchingWeights->CalcQk(int0[j], int1[j]));
905 // Fractional energy loss
906 fZQuench[j] = eloss / eq[j];
908 // Avoid complete loss
910 if (fZQuench[j] > fZmax) fZQuench[j] = fZmax;
912 // Some debug printing
915 // printf("Initial parton # %3d, Type %3d Energy %10.3f Phi %10.3f Length %10.3f Loss %10.3f Kick %10.3f Mean: %10.3f %10.3f\n",
916 // j, itype, eq[j], phiq[j], l, eloss, wjtKick[j], eMean / Float_t(icall+1), yq[j]);
918 // fZQuench[j] = 0.8;
919 // while (fZQuench[j] >= 0.95) fZQuench[j] = gRandom->Exp(0.2);
922 quenched[j] = (fZQuench[j] > 0.01);
927 Double_t pNew[1000][4];
934 for (Int_t isys = 0; isys < 4; isys++) {
935 // Skip to next system if not quenched.
936 if (!quenched[isys]) continue;
938 nGluon[isys] = 1 + Int_t(fZQuench[isys] / (1. - fZQuench[isys]));
939 if (nGluon[isys] > fNGmax) nGluon[isys] = fNGmax;
940 zquench[isys] = 1. - TMath::Power(1. - fZQuench[isys], 1./Double_t(nGluon[isys]));
941 wjtKick[isys] = wjtKick[isys] / TMath::Sqrt(Double_t(nGluon[isys]));
947 Double_t pg[4] = {0., 0., 0., 0.};
950 // Loop on radiation events
952 for (Int_t iglu = 0; iglu < nGluon[isys]; iglu++) {
955 for (Int_t k = 0; k < 4; k++)
962 for (Int_t i = 0; i < numpart; i++)
964 imo = fPyjets->K[2][i];
965 kst = fPyjets->K[0][i];
966 pdg = fPyjets->K[1][i];
970 // Quarks and gluons only
971 if (pdg != 21 && TMath::Abs(pdg) > 6) continue;
972 // Particles from hard scattering only
974 if (imo > 8 && imo < 1000) imo = fPyjets->K[2][imo - 1];
975 Int_t imom = imo % 1000;
976 if ((isys == 0 || isys == 1) && ((imom != (isys + 7)))) continue;
977 if ((isys == 2 || isys == 3) && ((imom != (isys + 1)))) continue;
980 // Skip comment lines
981 if (kst != 1 && kst != 2) continue;
984 px = fPyjets->P[0][i];
985 py = fPyjets->P[1][i];
986 pz = fPyjets->P[2][i];
987 e = fPyjets->P[3][i];
988 m = fPyjets->P[4][i];
989 pt = TMath::Sqrt(px * px + py * py);
990 p = TMath::Sqrt(px * px + py * py + pz * pz);
991 phi = TMath::Pi() + TMath::ATan2(-py, -px);
992 theta = TMath::ATan2(pt, pz);
995 // Save 4-momentum sum for balancing
1006 // Fractional energy loss
1007 Double_t z = zquench[index];
1010 // Don't fully quench radiated gluons
1013 // This small factor makes sure that the gluons are not too close in phase space to avoid recombination
1018 // printf("z: %d %f\n", imo, z);
1025 // Transform into frame in which initial parton is along z-axis
1027 TVector3 v(px, py, pz);
1028 v.RotateZ(-phiq[index]); v.RotateY(-thetaq[index]);
1029 Double_t pxs = v.X(); Double_t pys = v.Y(); Double_t pl = v.Z();
1031 Double_t jt = TMath::Sqrt(pxs * pxs + pys * pys);
1032 Double_t mt2 = jt * jt + m * m;
1035 // Kinematic limit on z
1037 if (m > 0.) zmax = 1. - m / TMath::Sqrt(m * m + jt * jt);
1039 // Change light-cone kinematics rel. to initial parton
1041 Double_t eppzOld = e + pl;
1042 Double_t empzOld = e - pl;
1044 Double_t eppzNew = (1. - z) * eppzOld;
1045 Double_t empzNew = empzOld - mt2 * z / eppzOld;
1046 Double_t eNew = 0.5 * (eppzNew + empzNew);
1047 Double_t plNew = 0.5 * (eppzNew - empzNew);
1051 // if mt very small (or sometimes even < 0 for numerical reasons) set it to 0
1052 Double_t mt2New = eppzNew * empzNew;
1053 if (mt2New < 1.e-8) mt2New = 0.;
1055 if (m * m > mt2New) {
1057 // This should not happen
1059 Fatal("Quench()", "This should never happen %e %e %e!", m, eppzNew, empzNew);
1062 jtNew = TMath::Sqrt(mt2New - m * m);
1065 // If pT is to small (probably a leading massive particle) we scale only the energy
1066 // This can cause negative masses of the radiated gluon
1067 // Let's hope for the best ...
1069 eNew = TMath::Sqrt(plNew * plNew + mt2);
1073 // Calculate new px, py
1079 pxNew = jtNew / jt * pxs;
1080 pyNew = jtNew / jt * pys;
1082 // Double_t dpx = pxs - pxNew;
1083 // Double_t dpy = pys - pyNew;
1084 // Double_t dpz = pl - plNew;
1085 // Double_t de = e - eNew;
1086 // Double_t dmass2 = de * de - dpx * dpx - dpy * dpy - dpz * dpz;
1087 // printf("New mass (1) %e %e %e %e %e %e %e \n", dmass2, jt, jtNew, pl, plNew, e, eNew);
1088 // printf("New mass (2) %e %e \n", pxNew, pyNew);
1092 TVector3 w(pxNew, pyNew, plNew);
1093 w.RotateY(thetaq[index]); w.RotateZ(phiq[index]);
1094 pxNew = w.X(); pyNew = w.Y(); plNew = w.Z();
1096 p1[index][0] += pxNew;
1097 p1[index][1] += pyNew;
1098 p1[index][2] += plNew;
1099 p1[index][3] += eNew;
1101 // Updated 4-momentum vectors
1103 pNew[icount][0] = pxNew;
1104 pNew[icount][1] = pyNew;
1105 pNew[icount][2] = plNew;
1106 pNew[icount][3] = eNew;
1111 // Check if there was phase-space for quenching
1114 if (icount == 0) quenched[isys] = kFALSE;
1115 if (!quenched[isys]) break;
1117 for (Int_t j = 0; j < 4; j++)
1119 p2[isys][j] = p0[isys][j] - p1[isys][j];
1121 p2[isys][4] = p2[isys][3] * p2[isys][3] - p2[isys][0] * p2[isys][0] - p2[isys][1] * p2[isys][1] - p2[isys][2] * p2[isys][2];
1122 if (p2[isys][4] > 0.) {
1123 p2[isys][4] = TMath::Sqrt(p2[isys][4]);
1126 printf("Warning negative mass squared in system %d %f ! \n", isys, zquench[isys]);
1127 printf("4-Momentum: %10.3e %10.3e %10.3e %10.3e %10.3e \n", p2[isys][0], p2[isys][1], p2[isys][2], p2[isys][3], p2[isys][4]);
1128 if (p2[isys][4] < -0.01) {
1129 printf("Negative mass squared !\n");
1130 // Here we have to put the gluon back to mass shell
1131 // This will lead to a small energy imbalance
1133 p2[isys][3] = TMath::Sqrt(p2[isys][0] * p2[isys][0] + p2[isys][1] * p2[isys][1] + p2[isys][2] * p2[isys][2]);
1142 printf("zHeavy lowered to %f\n", zHeavy);
1143 if (zHeavy < 0.01) {
1144 printf("No success ! \n");
1146 quenched[isys] = kFALSE;
1150 } // iteration on z (while)
1152 // Update event record
1153 for (Int_t k = 0; k < icount; k++) {
1154 // printf("%6d %6d %10.3e %10.3e %10.3e %10.3e\n", k, kNew[k], pNew[k][0],pNew[k][1], pNew[k][2], pNew[k][3] );
1155 fPyjets->P[0][kNew[k]] = pNew[k][0];
1156 fPyjets->P[1][kNew[k]] = pNew[k][1];
1157 fPyjets->P[2][kNew[k]] = pNew[k][2];
1158 fPyjets->P[3][kNew[k]] = pNew[k][3];
1165 if (!quenched[isys]) continue;
1167 // Last parton from shower i
1168 Int_t in = klast[isys];
1170 // Continue if no parton in shower i selected
1171 if (in == -1) continue;
1173 // If this is the second initial parton and it is behind the first move pointer by previous ish
1174 if (isys == 1 && klast[1] > klast[0]) in += ish;
1179 // How many additional gluons will be generated
1181 if (p2[isys][4] > 0.05) ish = 2;
1183 // Position of gluons
1185 if (iglu == 0) igMin = iGlu;
1188 (fPyjets->N) += ish;
1191 fPyjets->P[0][iGlu] = p2[isys][0];
1192 fPyjets->P[1][iGlu] = p2[isys][1];
1193 fPyjets->P[2][iGlu] = p2[isys][2];
1194 fPyjets->P[3][iGlu] = p2[isys][3];
1195 fPyjets->P[4][iGlu] = p2[isys][4];
1197 fPyjets->K[0][iGlu] = 1;
1198 if (iglu == nGluon[isys] - 1) fPyjets->K[0][iGlu] = 1;
1199 fPyjets->K[1][iGlu] = 21;
1200 fPyjets->K[2][iGlu] = fPyjets->K[2][in] + 1000;
1201 fPyjets->K[3][iGlu] = -1;
1202 fPyjets->K[4][iGlu] = -1;
1204 pg[0] += p2[isys][0];
1205 pg[1] += p2[isys][1];
1206 pg[2] += p2[isys][2];
1207 pg[3] += p2[isys][3];
1210 // Split gluon in rest frame.
1212 Double_t bx = p2[isys][0] / p2[isys][3];
1213 Double_t by = p2[isys][1] / p2[isys][3];
1214 Double_t bz = p2[isys][2] / p2[isys][3];
1215 Double_t pst = p2[isys][4] / 2.;
1217 // Isotropic decay ????
1218 Double_t cost = 2. * gRandom->Rndm() - 1.;
1219 Double_t sint = TMath::Sqrt((1.-cost)*(1.+cost));
1220 Double_t phis = 2. * TMath::Pi() * gRandom->Rndm();
1222 Double_t pz1 = pst * cost;
1223 Double_t pz2 = -pst * cost;
1224 Double_t pt1 = pst * sint;
1225 Double_t pt2 = -pst * sint;
1226 Double_t px1 = pt1 * TMath::Cos(phis);
1227 Double_t py1 = pt1 * TMath::Sin(phis);
1228 Double_t px2 = pt2 * TMath::Cos(phis);
1229 Double_t py2 = pt2 * TMath::Sin(phis);
1231 fPyjets->P[0][iGlu] = px1;
1232 fPyjets->P[1][iGlu] = py1;
1233 fPyjets->P[2][iGlu] = pz1;
1234 fPyjets->P[3][iGlu] = pst;
1235 fPyjets->P[4][iGlu] = 0.;
1237 fPyjets->K[0][iGlu] = 1 ;
1238 fPyjets->K[1][iGlu] = 21;
1239 fPyjets->K[2][iGlu] = fPyjets->K[2][in] + 1000;
1240 fPyjets->K[3][iGlu] = -1;
1241 fPyjets->K[4][iGlu] = -1;
1243 fPyjets->P[0][iGlu+1] = px2;
1244 fPyjets->P[1][iGlu+1] = py2;
1245 fPyjets->P[2][iGlu+1] = pz2;
1246 fPyjets->P[3][iGlu+1] = pst;
1247 fPyjets->P[4][iGlu+1] = 0.;
1249 fPyjets->K[0][iGlu+1] = 1;
1250 if (iglu == nGluon[isys] - 1) fPyjets->K[0][iGlu+1] = 1;
1251 fPyjets->K[1][iGlu+1] = 21;
1252 fPyjets->K[2][iGlu+1] = fPyjets->K[2][in] + 1000;
1253 fPyjets->K[3][iGlu+1] = -1;
1254 fPyjets->K[4][iGlu+1] = -1;
1260 Pyrobo(iGlu + 1, iGlu + 2, 0., 0., bx, by, bz);
1263 for (Int_t ig = iGlu; ig < iGlu+ish; ig++) {
1264 Double_t px, py, pz;
1265 px = fPyjets->P[0][ig];
1266 py = fPyjets->P[1][ig];
1267 pz = fPyjets->P[2][ig];
1268 TVector3 v(px, py, pz);
1269 v.RotateZ(-phiq[isys]);
1270 v.RotateY(-thetaq[isys]);
1271 Double_t pxs = v.X(); Double_t pys = v.Y(); Double_t pzs = v.Z();
1272 Double_t r = AliPythiaRndm::GetPythiaRandom()->Rndm();
1273 Double_t jtKick = 0.3 * TMath::Sqrt(-TMath::Log(r));
1274 if (ish == 2) jtKick = wjtKick[i] * TMath::Sqrt(-TMath::Log(r)) / TMath::Sqrt(2.);
1275 Double_t phiKick = 2. * TMath::Pi() * AliPythiaRndm::GetPythiaRandom()->Rndm();
1276 pxs += jtKick * TMath::Cos(phiKick);
1277 pys += jtKick * TMath::Sin(phiKick);
1278 TVector3 w(pxs, pys, pzs);
1279 w.RotateY(thetaq[isys]);
1280 w.RotateZ(phiq[isys]);
1281 fPyjets->P[0][ig] = w.X();
1282 fPyjets->P[1][ig] = w.Y();
1283 fPyjets->P[2][ig] = w.Z();
1284 fPyjets->P[2][ig] = w.Mag();
1290 // Check energy conservation
1294 Double_t es = 14000.;
1296 for (Int_t i = 0; i < numpart; i++)
1298 kst = fPyjets->K[0][i];
1299 if (kst != 1 && kst != 2) continue;
1300 pxs += fPyjets->P[0][i];
1301 pys += fPyjets->P[1][i];
1302 pzs += fPyjets->P[2][i];
1303 es -= fPyjets->P[3][i];
1305 if (TMath::Abs(pxs) > 1.e-2 ||
1306 TMath::Abs(pys) > 1.e-2 ||
1307 TMath::Abs(pzs) > 1.e-1) {
1308 printf("%e %e %e %e\n", pxs, pys, pzs, es);
1309 // Fatal("Quench()", "4-Momentum non-conservation");
1312 } // end quenching loop (systems)
1314 for (Int_t i = 0; i < numpart; i++)
1316 imo = fPyjets->K[2][i];
1318 fPyjets->K[2][i] = fPyjets->K[2][i] % 1000;
1325 void AliPythia::Pyquen(Double_t a, Int_t ibf, Double_t b)
1327 // Igor Lokthine's quenching routine
1328 // http://lokhtin.web.cern.ch/lokhtin/pyquen/pyquen.txt
1333 void AliPythia::SetPyquenParameters(Double_t t0, Double_t tau0, Int_t nf, Int_t iengl, Int_t iangl)
1335 // Set the parameters for the PYQUEN package.
1336 // See comments in PyquenCommon.h
1342 PYQPAR.iengl = iengl;
1343 PYQPAR.iangl = iangl;
1347 void AliPythia::Pyevnw()
1349 // New multiple interaction scenario
1353 void AliPythia::Pyshowq(Int_t ip1, Int_t ip2, Double_t qmax)
1355 // Call medium-modified Pythia jet reconstruction algorithm
1357 pyshowq(ip1, ip2, qmax);
1359 void AliPythia::Qpygin0()
1361 // New multiple interaction scenario
1365 void AliPythia::GetQuenchingParameters(Double_t& xp, Double_t& yp, Double_t z[4])
1367 // Return event specific quenching parameters
1370 for (Int_t i = 0; i < 4; i++) z[i] = fZQuench[i];
1374 void AliPythia::ConfigHeavyFlavor()
1377 // Default configuration for Heavy Flavor production
1379 // All QCD processes
1385 // No multiple interactions
1390 // Initial/final parton shower on (Pythia default)
1394 // 2nd order alpha_s
1402 void AliPythia::AtlasTuning()
1405 // Configuration for the ATLAS tuning
1406 if (fItune > -1) return;
1407 printf("ATLAS TUNE \n");
1409 SetMSTP(51, AliStructFuncType::PDFsetIndex(kCTEQ5L)); // CTEQ5L pdf
1410 SetMSTP(81,1); // Multiple Interactions ON
1411 SetMSTP(82,4); // Double Gaussian Model
1412 SetPARP(81,1.9); // Min. pt for multiple interactions (default in 6.2-14)
1413 SetPARP(82,1.8); // [GeV] PT_min at Ref. energy
1414 SetPARP(89,1000.); // [GeV] Ref. energy
1415 SetPARP(90,0.16); // 2*epsilon (exponent in power law)
1416 SetPARP(83,0.5); // Core density in proton matter distribution (def.value)
1417 SetPARP(84,0.5); // Core radius
1418 SetPARP(85,0.33); // Regulates gluon prod. mechanism
1419 SetPARP(86,0.66); // Regulates gluon prod. mechanism
1420 SetPARP(67,1); // Regulates Initial State Radiation
1423 void AliPythia::AtlasTuning_MC09()
1426 // Configuration for the ATLAS tuning
1427 if (fItune > -1) return;
1428 printf("ATLAS New TUNE MC09\n");
1429 SetMSTP(81,21); // treatment for MI, ISR, FSR and beam remnants: MI on, new model
1430 SetMSTP(82, 4); // Double Gaussian Model
1431 SetMSTP(52, 2); // External PDF
1432 SetMSTP(51, 20650); // MRST LO*
1435 SetMSTP(70, 0); // (was 2: def manual 1, def code 0) virtuality scale for ISR
1436 SetMSTP(72, 1); // (was 0: def 1) maximum scale for FSR
1437 SetMSTP(88, 1); // (was 0: def 1) strategy for qq junction to di-quark or baryon in beam remnant
1438 SetMSTP(90, 0); // (was 1: def 0) strategy of compensate the primordial kT
1440 SetPARP(78, 0.3); // the amount of color reconnection in the final state
1441 SetPARP(80, 0.1); // probability of color partons kicked out from beam remnant
1442 SetPARP(82, 2.3); // [GeV] PT_min at Ref. energy
1443 SetPARP(83, 0.8); // Core density in proton matter distribution (def.value)
1444 SetPARP(84, 0.7); // Core radius
1445 SetPARP(90, 0.25); // 2*epsilon (exponent in power law)
1446 SetPARJ(81, 0.29); // (was 0.14: def 0.29) Labmda value in running alpha_s for parton showers
1449 SetPARJ(41, 0.3); // a and b parameters of the symmm. Lund FF
1451 SetPARJ(46, 0.75); // mod. of the Lund FF for heavy end-point quarks
1452 SetPARP(89,1800.); // [GeV] Ref. energy
1455 AliPythia& AliPythia::operator=(const AliPythia& rhs)
1457 // Assignment operator
1462 void AliPythia::Copy(TObject&) const
1467 Fatal("Copy","Not implemented!\n");
1470 void AliPythia::DalitzDecays()
1474 // Replace all omega dalitz decays with the correct matrix element decays
1476 Int_t nt = fPyjets->N;
1477 for (Int_t i = 0; i < nt; i++) {
1478 if (fPyjets->K[1][i] != 223) continue;
1479 Int_t fd = fPyjets->K[3][i] - 1;
1480 Int_t ld = fPyjets->K[4][i] - 1;
1481 if (fd < 0) continue;
1482 if ((ld - fd) != 2) continue;
1483 if ((fPyjets->K[1][fd] != 111) ||
1484 ((TMath::Abs(fPyjets->K[1][fd+1]) != 11) && (TMath::Abs(fPyjets->K[1][fd+1]) != 13)))
1486 TLorentzVector omega(fPyjets->P[0][i], fPyjets->P[1][i], fPyjets->P[2][i], fPyjets->P[3][i]);
1487 Int_t pdg = TMath::Abs(fPyjets->K[1][fd+1]);
1488 fOmegaDalitz.Decay(pdg, &omega);
1489 for (Int_t j = 0; j < 3; j++) {
1490 for (Int_t k = 0; k < 4; k++) {
1491 TLorentzVector vec = (fOmegaDalitz.Products())[2-j];
1492 fPyjets->P[k][fd+j] = vec[k];