STAR task force for tuning PYTHIA

[Matthew Kelsey <mkelsey AT wayne.edu>]

——————————
Matthew Kelsey 
Postdoctoral Fellow
Wayne State University
<he/him/his>

Begin forwarded message:

From: Christian Bierlich <christian.bierlich AT thep.lu.se>

Subject: Re: Tuning pythia8 to STAR pp data at sqrt-s=200 GeV

Date: November 17, 2020 at 3:29:59 AM EST

To: Raghav Kunnawalkam Elayavalli <kunnawalkamraghav AT gmail.com>

Cc: Matthew Kelsey <mkelsey AT lbl.gov>

Hi Raghav,

Yes, I would definitely recommend Professor. In my experience it is very efficient, though you should not expect things to be completely automatic - some physics intuition is required. I know that Holger Schultz, who is the main person behind Professor, recently made an updated version which is allegedly much faster, but I have not used it myself yet, so I cannot testify to it. If you want to check it out, it is here: https://iamholger.gitbook.io/apprentice/

Cheers,

Christian

On 11/16/20 7:36 PM, Raghav Kunnawalkam Elayavalli wrote:

Hi Christian, 

So we have initiated the task force to tune PYTHIA 8 and it is being chaired by Matt Kelsey (cc'ed here). Based on your earlier email, we are discussing the procedure for the tuning and we wanted some input from you. 

so we have chosen the parameters we want to tune and then we want to basically use some sort of toolkit that does the comparison. some of us were thinking of using Professor and RIVET to compare with our STAR pp data for particle production and also jet spectra and substructure since we have that published now. It looks like this was also done in the recent CMS paper https://arxiv.org/abs/2011.03422 where they tuned HERWIG. I have not used Professor in the past (but I'm sure we can set it up) so is that procedure what you recommend? Any thoughts/comments?  

Thanks! 

Cheers

Raghav 

**************************************

Raghav Kunnawalkam Elayavalli

email: raghavke AT wayne.edu 

Post Doctoral Fellow 

Wayne State University

666 W Hancock St. RM 314 

Detroit MI. 48201

<he/him/his>

**************************************

On Oct 2, 2020, at 9:10 AM, Raghav Kunnawalkam Elayavalli <kunnawalkamraghav AT gmail.com> wrote:

Thanks for the quick reply Christian! 

Yes from our experience of tuning PYTHIA 6, the first parameters we will start working with is the pT0ref and the expPow to see if we can reproduce the pion spectrum and hadron multiplicities. Thanks very much for your guidance and i will contact you if we get stuck anywhere and be sure to not touch the fragmentation parameters :)

Cheers

Raghav 

**************************************

Raghav Kunnawalkam Elayavalli

email: raghavke AT wayne.edu 

Post Doctoral Fellow 

Wayne State University

666 W Hancock St. RM 314 

Detroit MI. 48201

<he/him/his>

**************************************

On Oct 2, 2020, at 3:24 AM, Christian Bierlich <christian.bierlich AT thep.lu.se> wrote:

Hi Raghav,

Sounds very good indeed!

The main parameter to tune, in this case, would be the MultipartonInteractions:pT0Ref, which should probably be lower than default values, since the extrapolation procedure for this parameter is a bit off, and the default value is tuned to sqrts = 7 TeV. If you can change this parameter to describe minimum bias data (total multiplicities and such) at 200 GeV (and then also change MultipartonInteractions:ecmRef to 200), I would suspect that you have a pretty decent baseline tune.

Another (but probably of second order importance) is the parameter controlling the proton overlaps in impact parameter space, which could also be altered, as the protons are on average smaller at lower energies. This is the parameter MultipartonInteractions:expPow.

The main pitfall people usually fall into is to start tuning fragmentation parameters. Since they are tuned to LEP data, they should really not be touched, unless you have a really good reason for doing so.

Hope it helped, happy to answer detailed questions as they come up, and also to discuss some concrete results once you have those.

Cheers,

Christian

On 10/1/20 5:14 PM, Raghav Kunnawalkam Elayavalli wrote:

Dear Christian,

Hope you are doing well. We are initiating a task-force within STAR to work on getting a PYTHIA 8 tune to match our pp data at sqrts=200 (possibly also 500) GeV. I would like to ask if you can can point us to the established procedure that if you are familiar with that. Are there any pitfalls that we need to be vary off? Anything that we should not do? Just any comments really help :)
Thanks!

Cheers
Raghav


**************************************
Raghav Kunnawalkam Elayavalli
email: raghavke AT wayne.edu

Post Doctoral Fellow
Wayne State University
666 W Hancock St. RM 314
Detroit MI. 48201
<he/him/his>
**************************************