sPHENIX discussion of software

["Huang, Jin" <jhuang AT bnl.gov>]

Hi Hugo

 

In both cases, I was rounding them to nearest 5% in quotation in the presentations, as the occupancy distribution is so wide and has a long tail. I am also expecting much has changed from the 2018 simulation setup, in particular a few TPC digitization adjustment that can change occupancy at percent level.

> Does that make sense ?

I think so. With 250 z-bins per side, <hit> = 80 translate to 32% occupancy.

> For 170kHz, your 25% does seem a bit high though. On MC collisions I would expect 6+1.7*8 = 20%. (although by comparing the plots on your two sets of slides, if 80hits on the first corresponds to 30%, then the 60 on the second would be more like 22.5, which is closer to my numbers). So maybe there is no difference.

Here, 170kHz + MB gives about <hit> = 60 that translates to 24% occupancy. That is a bit higher than 20% from your extrapolation. In my simulation, the in-time triggered collision originate from z=0, which may affect occupancy in a small level too. Let me know if there is anything I could assist your check too.

 

Cheers

 

Jin

 

 

______________________________

 

Jin HUANG

 

Physicist, Ph.D.

Brookhaven National Laboratory

Physics Department, Bldg 510 C

Upton, NY 11973-5000

 

Office: 631-344-5898

Cell:   757-604-9946

______________________________

 

From: Hugo Pereira Da Costa <hugo.pereira-da-costa AT cea.fr>
Sent: Tuesday, August 18, 2020 4:19 PM
To: Huang, Jin <jhuang AT bnl.gov>; Anthony Frawley <afrawley AT fsu.edu>; sphenix-software-l AT lists.bnl.gov; sphenix-calibration-l AT lists.bnl.gov
Subject: Re: [Sphenix-software-l] Occupancy in the TPC

 

Hi Jin,

Thanks a lot for all the info. Looking at the numbers in your slides I actually think that they agree with the one I quoted, at least for the 200kHz case. The "trick" is that you consider central collisions for the trigger event, if I read your slide right.

From my slides today, slide 4:

in layer 7 (largest occupancy):

MB collision 0 pile-up gives 6% occupancy.

To which you add 8% for 100kHz PU, which converts in 18% for 200kHz Pile-up. So for MB collisions the total is 22%

Now based on slide 6 of my ppresentation, bottom left plot, central (0-10%) correspond to 15% occupancy, instead of the 6% for MB. So the total: 15% + 16% -> 31% which matches nicely the the number you quote on slide 5 of your first presentation.

Does that make sense ?

For 170kHz, your 25% does seem a bit high though. On MC collisions I would expect 6+1.7*8 = 20%. (although by comparing the plots on your two sets of slides, if 80hits on the first corresponds to 30%, then the 60 on the second would be more like 22.5, which is closer to my numbers). So maybe there is no difference.

In any case I will dig a bit deeper during the week to make sure that things are consistent.

 

Hugo

 

 

On 8/18/20 1:19 PM, Huang, Jin wrote:

Hi Hugo

 

Follow up with the simulation meeting, and here is the occupancy study from July 2018:

1. Central (0-4.4fm) + 200kHz AuAu pile up: https://indico.bnl.gov/event/4024/ slide 5, where I get 30% occupancy at the first TPC layer
2. MB + 170kHz AuAu pile up: https://indico.bnl.gov/event/6949/attachments/25171/37559/TPC_Charge_Data_DAQ_Workfest.pdf slide 7, where I get 25% occupancy at the first TPC layer

Both occupancies are any TpcHits (PHG4Cell in 2018) including pre & post samples.

 

Digging into the records, here is a branch for MB+170 kHz pile up simulations:

https://github.com/blackcathj/macros/blob/tpc_electronics_hepmc_mb/macros/g4simulations/Fun4All_G4_sPHENIX.C

where the TPC hits are analyzed for histogram filling directly at this analysis module:

https://sphenix-collaboration.github.io/doxygen/da/dfe/TPCIntegratedCharge_8cc_source.html#l00334

For each job, sHIjing is run to regenerate the MB sample. And the output is still available at

/sphenix/user/jinhuang/TPC/Multiplicity/AuAu200MB_170kHz_Iter2/, including the result occupancy plot AuAu200MB_170kHz_Iter2_SUM.xml_TPCIntegratedCharge.rootCheck.pdf

 

Cheers

 

Jin

 

 

______________________________

 

Jin HUANG

 

Physicist, Ph.D.

Brookhaven National Laboratory

Physics Department, Bldg 510 C

Upton, NY 11973-5000

 

Office: 631-344-5898

Cell:   757-604-9946

______________________________

 

From: sPHENIX-software-l <sphenix-software-l-bounces AT lists.bnl.gov> On Behalf Of Anthony Frawley
Sent: Wednesday, August 12, 2020 8:06 PM
To: Hugo Pereira Da Costa <hugo.pereira-da-costa AT cea.fr>; sphenix-software-l AT lists.bnl.gov; sphenix-calibration-l AT lists.bnl.gov
Subject: Re: [Sphenix-software-l] Occupancy in the TPC

 

Hi Hugo,

 

Thanks, that is helpful!

 

Tony

---

From: Hugo Pereira Da Costa <hugo.pereira-da-costa AT cea.fr>
Sent: Wednesday, August 12, 2020 5:05 PM
To: Anthony Frawley <afrawley AT fsu.edu>; sphenix-software-l AT lists.bnl.gov <sphenix-software-l AT lists.bnl.gov>; sphenix-calibration-l AT lists.bnl.gov <sphenix-calibration-l AT lists.bnl.gov>
Subject: Re: [Sphenix-software-l] Occupancy in the TPC

 

Hi Tony, others

 

I have updated the slides to add the occupancy in the TPC using 'all hits' rather then 'hits above threshold'.

Rather than sending them by email, I have put them here: https://cernbox.cern.ch/index.php/s/bx0YJ5LRCgV2sif

I have not yet updated the occupancy with "all" hits for Christof files, because the jobs are still running.

Comments welcome

 

Hugo

 

 

On 8/11/20 1:32 PM, Anthony Frawley wrote:

No, you did not miss anything. That all sounds good. 

 

Out of curiosity, is it easy to say what the occupancy would be without the threshold adc cut?

 

Thanks

Tony

 

 

---

From: Hugo Pereira Da Costa <hugo.pereira-da-costa AT cea.fr>
Sent: Tuesday, August 11, 2020 3:24 PM
To: Anthony Frawley <afrawley AT fsu.edu>; sphenix-software-l AT lists.bnl.gov <sphenix-software-l AT lists.bnl.gov>; sphenix-calibration-l AT lists.bnl.gov <sphenix-calibration-l AT lists.bnl.gov>
Subject: Re: [Sphenix-software-l] Occupancy in the TPC

 

Hi Tony,

 

Hits are TrkrHit(v1) from "trackbase". Only counted are those that have a charge (get_adc()) above threshold

My understanding (correct me if I am wrong) is that the max number of such hits per layer is NPHiBins x NZBins in this layer. Which would correspond to 100% occupancy. Did I miss something ?

 

Hugo

 

 

On 8/11/20 12:36 PM, Anthony Frawley wrote:

Hi Hugo,

 

Question: what is the definition of a "hit"?

 

Thanks

Tony

---

From: sPHENIX-software-l <sphenix-software-l-bounces AT lists.bnl.gov> on behalf of Hugo Pereira Da Costa <hugo.pereira-da-costa AT cea.fr>
Sent: Tuesday, August 11, 2020 2:15 PM
To: sphenix-software-l AT lists.bnl.gov <sphenix-software-l AT lists.bnl.gov>; sphenix-calibration-l AT lists.bnl.gov <sphenix-calibration-l AT lists.bnl.gov>
Subject: [Sphenix-software-l] Occupancy in the TPC

 

Hi all,

Following up on a question from Christof last week, I looked into the
TPC occupancy in the newly 100kHz Pile-up Hijing files derived from the
ones produces by Chris.

The results are summarized in the file attached. And could be discussed
e.g in the next simulation and software meeting.

I wanted to post them early because as you will see the occupancies I
get with these new files are significantly smaller than those obtained
with older files from Christof, and I have not been able to find the
reason for this yet.

Comments and suggestions are welcome.

(in particular: Christof, could you point me to the macros used to
generate/reconstruct the files at
"/sphenix/user/bogui/MacrosModular/macros/macros/g4simulations/clus_only/SvtxCluHijMBPu100_Mar20_1_*"
? that would help me debug further)

Hugo