CK: Slide 7: How PE is generated in MC? GY: MeV to PE conversion from SFGD data for which is 70 PE/MeV per fiber (before attenuation and MPPC correction). CK: How did they get that? DD: The value are extracted from 5x5 cm early prototype. GY: We will look at CERN proton data to further tune this conversion. 

CM: Slide 8: simulation with full detector? DD: Yes, full detector.

CK: Slide 12: MC and data are different so you should use a different PE cut on data.

CM: Slide 14: You have a neutron interactions and you have a single protons tracks, correct? GY: This is correct. Caveat: The spectrum of the neutron is flat. GM: How many events? EC: 17k neutron events. CM: last bullet: 

CM: What is the chi2 you are using? GY: It is defined as the minimal distance between the line and the voxel. CR: This is not a chi2, but the minimum of the minimized function (FCN in Minuit language).

CM: This chi2 seems a different definition of linearity. 

CK: Suggestion: Showing efficiency and purity is good, but you should look first at the chi2 distribution first and the linearity distribution.

DL: Slide 18: That the energy drop at 725 MeV looks odd. Make sure that the energy computation is correct, in particular the time window. 

CK: Slide 18: Why some bins are empty? DL: Those bins are finer than the resolution. GY: This is well know.

CK: Slide 20: How can linearity be 0.9922 for this track? Linearity is not working, then. 

We will finish this discussion next week. 

CK: Slide 7: why two blue points at 0? GY: The one at (0,0) is a reference.

CK: Why there are three blue and no black? Slide 11: They overlap. 

CK: Beamline scientist should have information about the shape of the beam spot. GY: For 90 m location and the collimator we used they do not have this information.

CR: Slide 6: Can you add the RMS for each plot? It will help in visualizing the spread.