3.4.1 Application of Weights

Explanation of the cluster containers:

suffix	Content
_em	EM scale topo cluster
_ooc	after out of cluster, before dead material corrections (to come later)
_w	weighted clusters
_EM	egamma cluster, with seed and noise conditions of (6,3,3)

the weighting procedure is as follows

1. In the weighting procedure is first a noise cut applied (ratio energy/noise must be > 2, noise is the noise level of the individual cells)
2. There is a cut on the weights applied (0.5 < w < 3 / 5 / 10 for EM, FCAL / HEC / tile)
3. in the last step the DM corrections are applied (see !CaloClusterCorrection/share/CaloTopoLocalCalibDM_jobOptions.py )
   
   

Now the results look more reasonable and are also comparable with what is achieve when weighting with the official weight tables

3.4.1.1. Linearity comparison in eta and phi

There are still crucial differences between the weighting methods, whcih presumably come from differences in the noise cuts, when creating the weights (needs to be checked). Here, the cone weighting at the moment is done such, that there is NO CUT on the ratio energy/noise and NO CUT on the weights themselves, there is only a cut on the energy density in the cells. It needs to be checked, how the results change with the ratio cut and possibly the weight cuts.

The DM corrections are not yet applied.

Cone-Method

Cluster-Method

!

EM scale (only phi)

Main.kristin - 18 Apr 2007