 | The reconstructed energy resolutions with the standard and Full SC methods for the Si/Scint and the all-Scint setups in comparison with the energy resolution fits published in \protect\cite{Frank}. The fit parameters are given in the legend, where the noise term is fixed to $c$ = 0.18\,GeV, as was determined from noise measurements in dedicated runs without beam particles as well as in random trigger events in the CERN test beam setup \cite{Frank}. |
 | Distribution of the hits of 25 GeV $\pi^-$ of CERN TB in the Si-W ECAL (left) and AHCAL (right). The different colors show the different bins and the track energy depositions. |
 | The optimized bin weights of the Si-W ECAL (left) and AHCAL (right) for 4\,GeV, 40\,GeV and 80\,GeV $\pi^-$ beams. The use of a constant value rather than the individual amplitude of each hit in the first two bins yields energy dependent weights in these bins. |
 | Distribution of the hits of 25 GeV $\pi^-$ of CERN TB in the Si-W ECAL (left) and AHCAL (right). The different colors show the different bins and the track energy depositions. |
 | Distribution of the hits of 25 GeV $\pi^-$ of CERN TB in the Si-W ECAL (left) and AHCAL (right). The different colors show the different bins and the track energy depositions. |
 | The optimized bin weights of the Si-W ECAL (left) and AHCAL (right) for 4\,GeV, 40\,GeV and 80\,GeV $\pi^-$ beams. The use of a constant value rather than the individual amplitude of each hit in the first two bins yields energy dependent weights in these bins. |
 | Distribution of the hits of 25 GeV $\pi^-$ of CERN TB in the Si-W ECAL (left) and AHCAL (right). The different colors show the different bins and the track energy depositions. |
 | The optimized bin weights of the Si-W ECAL (left) and AHCAL (right) for 4\,GeV, 40\,GeV and 80\,GeV $\pi^-$ beams. The use of a constant value rather than the individual amplitude of each hit in the first two bins yields energy dependent weights in these bins. |
 | Reconstructed energy resolutions using SC weights optimized for the respective dataset (black) and SC weights trained for a different dataset (red) and the relative deviations between them. The total (statistical and systematic) uncertainties are marked with '[]'. |
 | The reconstructed energy resolutions with the standard and Full SC methods for the Si/Scint and the all-Scint setups in comparison with the energy resolution fits published in \protect\cite{Frank}. The fit parameters are given in the legend, where the noise term is fixed to $c$ = 0.18\,GeV, as was determined from noise measurements in dedicated runs without beam particles as well as in random trigger events in the CERN test beam setup \cite{Frank}. |
 | Reconstructed energy resolutions( left) and relative improvement of the resolutions (right) with standard, ECAL SC, AHCAL SC and Full SC reconstruction methods for the Si/Scint setup. The fit parameters are given in the legend. The total (statistical and systematic) uncertainties are marked with '[]'. |
 | Reconstructed energy resolutions( left) and relative improvement of the resolutions (right) with standard, ECAL SC, AHCAL SC and Full SC reconstruction methods for the Si/Scint setup. The fit parameters are given in the legend. The total (statistical and systematic) uncertainties are marked with '[]'. |
 | Reconstructed energy resolutions( left) and relative improvement of the resolutions (right) with standard, ECAL SC, AHCAL SC and Full SC reconstruction methods for the Si/Scint setup. The fit parameters are given in the legend. The total (statistical and systematic) uncertainties are marked with '[]'. |
 | The optimized bin weights of the Si-W ECAL (left) and AHCAL (right) for 4\,GeV, 40\,GeV and 80\,GeV $\pi^-$ beams. The use of a constant value rather than the individual amplitude of each hit in the first two bins yields energy dependent weights in these bins. |
 | Reconstructed energy resolutions( left) and relative improvement of the resolutions (right) with standard, ECAL SC, AHCAL SC and Full SC reconstruction methods for the Si/Scint setup. The fit parameters are given in the legend. The total (statistical and systematic) uncertainties are marked with '[]'. |
 | Reconstructed energy resolutions using SC weights optimized for the respective dataset (black) and SC weights trained for a different dataset (red) and the relative deviations between them. The total (statistical and systematic) uncertainties are marked with '[]'. |
 | Reconstructed energy resolutions using SC weights optimized for the respective dataset (black) and SC weights trained for a different dataset (red) and the relative deviations between them. The total (statistical and systematic) uncertainties are marked with '[]'. |
 | Reconstructed energy resolutions using SC weights optimized for the respective dataset (black) and SC weights trained for a different dataset (red) and the relative deviations between them. The total (statistical and systematic) uncertainties are marked with '[]'. |