Robert presented detailed systematic uncertainties study for Upsilon analysis in U+U collisions.

1) Addressed question to the pT shift for the integrated mb yields based on a simple exponential function. Result looks reasonable, but need to check if the equation for calculating correction factor is correct.

2) The  last bin for the pT spectrum is not significant, suggest use upper limit.

3) Updated R_AA with including systematic uncerainties, questions raised for the detailed systematic uncerainties study.

4) Question about the acceptance correction uncertainty. Anthony confirms that the acceptance correction uncertainties matter, but do not expect change by species.

5) The details for factorization statistical fluctuation and shape uncertainties are required. Robert agreed to provide some slides for the details. Suggest use the shape extracted from mb 0-60% for other centralities.

6) The signal was fitted starting from 8 GeV, double-exponetinal function for the background starts from 6.5 GeV. Need update the plot on the poster and check the uncertainties of the background estimation by varying the fitting mass range.

7) The origin of the broad Gaussian distribution under the dE/dx distributions is not clear. Need to see in log scale and see its dependence on pT. For the given 4.5-5.5 GeV/c bin, it is not likely from the proton contribution. The uncertainties are from the deviations between the linear functions fitting to the points of 3-Gauss and 4-Gauss. Zhenyu asked why not use the difference between the data points themselves.

8) Need to confirm if the tower matching distribution unit is in delta-phi and delta-eta, or cm. The correct uncertainty should be defined as the following (for example):
difference between a) and b)
a) data with cut 0.035 / efficiency from embedding with cut 0.035
b) data with cut 0.045 / efficiency from embedding with cut 0.045

9) Detailed comparison between embedding and data for the ADC, triggered ADC distributions is needed.

10) nSigmaE uncertainty should not be cancelled when calculating R_AA, since different calibrations were performed in different year runs.

11) The Upsilon statistics are not good to determine which effect, mass smearing or shift, is dominant. Need detailed study on J/psi signal and use the conclusion/result to constrain Upsilon signal.