sPHENIX discussion of software

[Joe Osborn <jdosbo AT umich.edu>]

Hi Jin,

One minor comment, sqs=200 GeV doesn't make sense here as these are single particle simulations and not e.g. PYTHIA simulations.

The plots have HIJING 0-4 fm and single photon on them, which I think conveys the appropriate information (and I think is what you were suggesting, unless I'm mistaken).

Another follow up on the average energy plot: for the p-CDR plots, the mean is estimated over broader energy-axis range then what is plotted. The higher energy tail, although not plotted, may shift up the mean value by a small fraction.

This is true, I have noticed this makes a slight difference; however, it is not too significant. I can replace the numbers with the means taken over the full range of the plots though, which includes the very high energy tails. 

Joe

---------------------------------------

Joe Osborn

Ph.D Candidate

University of Michigan

jdosbo AT umich.edu

(859) 433-8738

On Fri, Apr 27, 2018 at 10:40 AM, Huang, Jin <jhuang AT bnl.gov> wrote:

Hi, Joe and Craig

 

One minor comment on the embedding resolution plot: Please use label specify the HIJING configuration, e.g. “single photon embedded\n Au+Au collision \sqrt(s) = 200 GeV, b = 0-4 fm”

 

Another follow up on the average energy plot: for the p-CDR plots, the mean is estimated over broader energy-axis range then what is plotted. The higher energy tail, although not plotted, may shift up the mean value by a small fraction.

 

As in another thread in the sPHENIX general list, Gunther is wondering about photon simulation plot, which is discussed in this thread. Meanwhile, UPP of 15%/sqrt(E) does not include a constant term, which is very strong constraint on the high energy end. Since we are refining the UPP, I would suggest include a constant term, e.g. 4% and specify that is defined in p+p collisions.

 

Cheers

 

Jin

 

______________________________

 

Jin HUANG

 

Associate Physicist

Brookhaven National Laboratory

Physics Department, Bldg 510 C

Upton, NY 11973-5000

 

Office: 631-344-5898

Cell:   757-604-9946

______________________________

 

From: sPHENIX-software-l [mailto:sphenix-software-l-bounces AT lists.bnl.gov] On Behalf Of Joe Osborn
Sent: Friday, April 27, 2018 10:01 AM
To: Woody, Craig <woody AT bnl.gov>
Cc: sphenix-emcal-l AT lists.bnl.gov; sphenix-software-l AT lists.bnl.gov
Subject: Re: [Sphenix-software-l] EMCal Simulations For CDR

 

Hi Craig,

 

Okay, I will update the plots and corresponding text today. 

 

For the single particle resolutions, I guess the difference between the resolutions on slides 2 and 4 are that the plots on slide 4 use the new clusterizer. Is that correct ?

 

This is correct. However there is not significant difference between the two, which Sasha had already shown in his clusterizer studies; basically the new clusterizer doesn't change the performance significantly in single particle simulations, but the new clusterizer performs much better in HIJING background.

 

We also agreed that we should show the same plots with embedding, which don't look so good, but this is what the Monte Carlo is now telling us. We all felt that at this stage of our design, we better show that we at least looked at how the calorimeter will perform in central heavy ion collisions, or someone will surely ask why we haven't. This will require changing what we claim about our design goal for the energy resolution, but I think we need to do that anyway since we never specified a constant term, which any calorimeter person would question why not. 

 

Sounds good, I will put the corresponding plots in with some text accompanying them.

 

Joe Osborn 

 

 

 

 

---------------------------------------

 

Joe Osborn

Ph.D Candidate

University of Michigan

jdosbo AT umich.edu

(859) 433-8738

 

On Thu, Apr 26, 2018 at 10:59 AM, Craig Woody <woody AT bnl.gov> wrote:

Hi Joe,
  Sorry for not having gotten back to you sooner, but I guess we're all just getting back from China. John, Anne and I did discuss your new plots when we were in China (unfortunately Jin wasn't there when we did), but the three of us agreed that we should include your new plots in the CDR. For the single particle resolutions, I guess the difference between the resolutions on slides 2 and 4 are that the plots on slide 4 use the new clusterizer. Is that correct ? If so, we should use the plots with the new clusterizer, which look a bit better, along with the corresponding new linearity plots. I think just showing electrons and photons should be fine.
    We also agreed that we should show the same plots with embedding, which don't look so good, but this is what the Monte Carlo is now telling us. We all felt that at this stage of our design, we better show that we at least looked at how the calorimeter will perform in central heavy ion collisions, or someone will surely ask why we haven't. This will require changing what we claim about our design goal for the energy resolution, but I think we need to do that anyway since we never specified a constant term, which any calorimeter person would question why not.
   Lastly, since the average energy per tower and in a 3x3 cluster sum doesn't change very much, there is no harm in updating those, and we would then be able to claim these values are with the correct geometry. If you could update those plots along with the corresponding text in the CDR, that would be great. Jin may have some further comments or suggestions about this as well. I'll be away again starting tomorrow, but I'll be around next week and we can all discuss things more then.

Cheers,
Craig

On 4/26/2018 10:06 AM, Joe Osborn wrote:

Hi all,

 

Since I haven't heard anything about these, should I go ahead and replace the current CDR plots with these? Since the CDR is supposed to be frozen by the end of the month that gives me a few days to add them and adjust the text associated with the plots.

 

Joe

 

 

---------------------------------------

 

Joe Osborn

Ph.D Candidate

University of Michigan

jdosbo AT umich.edu

(859) 433-8738

 

On Fri, Apr 20, 2018 at 6:51 PM, Joe Osborn <jdosbo AT umich.edu> wrote:

Hi all,

 

Sorry to reply to my own email, but after discussion with Jin I implemented a tower geometry cut such that the mean energy per tower is only for the 2D projective towers; therefore, it is a direct comparison to what is currently in the CDR (and I guess is from the pCDR a while ago). The mean tower energy in 0-4 fm HIJING events is roughly 34 MeV, and the 3x3 tower sum is roughly 321 MeV. Not substantially different than what I sent previously, and also not substantially different from what is in the pCDR; however, it is a little less than these values. I can make these plots look nicer if we agree that they should replace those in the current draft of the CDR.

 

Joe

 

 

---------------------------------------

 

Joe Osborn

Ph.D Candidate

University of Michigan

jdosbo AT umich.edu

(859) 433-8738

 

On Fri, Apr 20, 2018 at 10:40 AM, Joe Osborn <jdosbo AT umich.edu> wrote:

Hi EMCalers,

 

Per the discussion from the EMCal meeting on Tuesday, I have generated some new simulation plots with the existing single particle simulations and the most recent HIJING backgrounds that the jet structure topical group is also using for their jet studies. I have attached some slides which document some results for the CDR that we can now say definitely include the tilted geometry of the EMCal. In particular, these plots can be compared to the existing figures 4.21-4.23 of the current CDR. 

 

Craig specifically requested that we try and replace figure 4.17 of the CDR with the 2016 test beam data. My next task is to try and run some test beam simulations and see if I can overlay the simulation data on top of the 2016 publication data. 

 

Let me know if you have any comments on the current work,

 

Joe Osborn

 

 

---------------------------------------

 

Joe Osborn

Ph.D Candidate

University of Michigan

jdosbo AT umich.edu

(859) 433-8738

 

 

 

_______________________________________________

sPHENIX-software-l mailing list

sPHENIX-software-l AT lists.bnl.gov

https://lists.bnl.gov/mailman/listinfo/sphenix-software-l