Tests of Photos operability for very heavy intermediate states or at very high energies

note that versions of PHOTOS (until may 2004) did not have worked well for such heavy, TeV range, objects


GENERATOR Branching ratio Maximum SDP T1 T2 Booklet
test1 test2 test1 test2 test1 test2 test1 test2 test1 test2
(n photons) -> 0 1 0 1 2 0 1 0 1 2
Comparison no 1 - older:     Z -> mumu at 2 TeV
E_test=10.0, older
KK 71.79 23.97 4.24
PHOTOS O(4) 72.31 22.41 5.27 0 0.0012 0.0375 3.113 0.202
Comparison no 1-NEW:     Z -> mumu , highly virtual Z/gamma* state at 2 TeV , 4M events
E_test=10.0
KK 71.80 28.20 71.80 23.97 4.23
PHOTOS O(2) 72.32 27.68 72.32 22.42 5.26 0 0.0052 0 0.004 0.032 1.039 3.094 0.146 0.240
PHOTOS O(4) 72.29 27.71 72.29 22.43 5.28 0 0.0055 0 0.0007 0.0292 0.986 3.07 0.154 0.150
PHOTOS EXP(7) 71.71 28.29 71.71 24.03 4.26 0 0 0 0 0 0.174 0.166 0 0
E_test=20.0
KK 76.17 23.82 76.16 20.92 2.92
PHOTOS O(2) 76.46 23.54 76.46 20.04 3.50 0 0.0023 0 0.0019 0.0198 0.567 1.752 0.052 0.091
PHOTOS O(4) 76.42 23.58 76.42 20.07 3.50 0 0.002 0 0.0005 0.015 0.498 1.689 0.047 0.053
PHOTOS EXP(7) 76.10 23.90 76.10 20.96 2.94 0 0 0 0 0 0.138 0.119 0 0
Comparison no 2:     Z -> e+ e-, highly virtual Z/gamma* state at 2 TeV
In construction of the plots we have used "muons of electron mass"
(therefore description of plots and tables may be wrong, even though the results are OK.
Plots will be replaced)
E_test=10.0
KK 59.19 40.81 59.19 31.39 9.42
PHOTOS EXP 59.23 40.77 59.23 31.32 9.45 0 0.002 0 0.002 0.0003 0.079 0.129 0.080 0.080
Comparison no 3:      Z -> mumu at LHC energies (pp at 14TeV), Z is close to its resonance peak
E_test=10.0
PYTHIA w/FSR 87.98 12.02 87.98 11.21 0.81
PHOTOS EXP 89.21 10.78 89.21 10.12 0.67 0 0.0240 0 0.025 0.003 2.468 2.453 0.278 0.263
Comparison no 4:      W+ -> mu+ nu_mu at LHC energies (pp at 14TeV), W is close to its resonance peak
E_test=10.0
WINHAC (EXP) 95.449 4.551 95.449 4.441 0.110
WINHAC+PHOTOS (EXP) 95.424 4.576 95.424 4.454 0.121 0 0.0003 0 0.0003 0.0018 0.046 0.035 0.0013 0.0013
PYTHIA (FSR ON) 94.707 5.293 94.707 5.147 0.146
PYHIA+PHOTOS (EXP) 95.416 4.584 95.416 4.463 0.121 0 0.0562 0 0.055 0.024 1.416 1.391 0.269 0.263

The purpose of these tests is to present the performance of PHOTOS at different sophistication levels, and how it compares with the matix element generator: KKMC, widely used PYTHIA and WINHAC.
As one can see, adding third and fourth order in PHOTOS improves agreement with the baseline, i.e. KKMC. However, only PHOTOS EXP has a good agreement with KKMC. In case of W decays, we perform comparison with PHOTOS EXP only. Agreement is not as perfect as in case of KKMC and Z decays, this is of no surprise, the quality of the single-photon emission kernel is not as good; see reference: Acta Phys.Polon.B34:4561-4570,2003. It will be relatively easy to improve, if necessary.

One can be surprised why we call these comparisons at very high energies, as the scale of most of the plots do not exceed 150 GeV. This is because W or Z, which decays were studies, are ultra-relativistic (as in LHC). From physics point of view, this does not provide much of importance, but in this way we test the algorithm's numerical stability in real working conditions.

The tests are being defined in the laboratory frame (the 4-momenta of leptons and photons are defined in this frame), that is why Etest=10/20 GeV should be compared to average W/Z energy and not its mass. The tests on momenta from the rest-frame of decaying particle are collected in the antenna algorithm studies page.

Note: In practical applications, problems with numerical stability may occur. These problems result in PHOTOS "STOP" messages, e.g. "Too much bremsstrahlung, PRSOFT=...", at least if PHOTOS runs with default settings.
For explanation of built-in solutions (to be activated by the user) and comments on their applicability, please refer to our stability studies page.

Last modification: Piotr Golonka, 24 August 2004, 17:35