Star-fst-l mailing list

["Zhang, Shenghui" <shenghui AT uic.edu>]

Hi Daniel,

I'm sorry I made a mistake. The third question is "Can the new code be run and submit jobs on RCF?"

Thanks,

Shenghui

---

From: Star-fst-l <star-fst-l-bounces AT lists.bnl.gov> on behalf of Zhang, Shenghui <shenghui AT uic.edu>
Sent: Saturday, July 25, 2020 3:08 PM
To: Ye, Zhenyu <yezhenyu AT uic.edu>; Daniel Brandenburg <dbrandenburg.ufl AT gmail.com>
Cc: star-fst-l AT lists.bnl.gov <star-fst-l AT lists.bnl.gov>
Subject: Re: [Star-fst-l] Question on FST (and sTGC) fast simulators

 

Hi Daniel,

I have three questions about the updated code using FtsFastSimulatorMaker. Can you help me check it? You can find questions at the below.

1.I used the latest code on GitHub using FtsFastSimulatorMaker. I found the efficiency is same as before (i.e. using SiRasterizer), while the reconstruction pT is significantly different. You can find the plot of pT resolution at the attachment.

2.I noticed the FST geometry is different than before (the latest version from Te-Chuan). I don't know what the changes are and why these changes.

3. The new code can be run and submit jobs on RCF.

Thanks,

Shenghui

---

From: Star-fst-l <star-fst-l-bounces AT lists.bnl.gov> on behalf of Ye, Zhenyu <yezhenyu AT uic.edu>
Sent: Wednesday, July 22, 2020 12:04 PM
To: Daniel Brandenburg <dbrandenburg.ufl AT gmail.com>
Cc: star-fst-l AT lists.bnl.gov <star-fst-l AT lists.bnl.gov>
Subject: Re: [Star-fst-l] Question on FST (and sTGC) fast simulators

 

Hi Daniel,

Thanks for the update. I was able to install and run fst_study.C, which uses the Fast Simulator in my computer. Shenghui and Gavin will check if the results are consistent with the previous version of the code using the function SiRasterizer, and also develop and test the slow simulator.

Let me try again to explain my confusion about the FST and sTGC Fast simulators:

1. In your current GitHub area, you use StFtsFastSimulatorMaker, which includes both FST and sTGC in one cxx file.

2. In Te-Chuan’s latest work area (April 14, 2020) on RCF, /star/u/tchuang/FST_simulations/README, he noted:

******************************************************
    - There are two versions of fast simulator:
        - StFtsFastSimulatorMaker: Old fast simulator which has been updated for silicon part but STGC part was obsoleted.
        - StSiSimulatorMaker and StStgcSimulatorMaker: Separated fast simulator for FST and STGC done by Daniel (dbrandenburg.ufl AT gmail.com)
            - I copied these codes from Daniel's directory. I have checked StSiSimulatorMaker that is updated and I believe StStgcSimulatorMaker is also updated.
******************************************************

So there are separated makers for FST and sTGC, StSiSimulatorMaker and StStgcSimulatorMaker, taken from YOUR directory.

We talked about separating FST and sTGC fast simulators a few times, and you agreed to get them separated. I was glad to see they were indeed separated, and thus I was under the impression that we are moving away from StFtsFastSimulatorMaker (the Old fast simulator as noted by Te-chuan), and use the separated fast simulators. But now I am totally confused by what you wrote below and what you use in GitHub for forward tracking study.

I hope this helps to explain my confusion, and you can answer my questions:

1. Which Fast Simulator (StFtsFastSimulatorMaker, StSiSimulatorMaker and StStgcSimulatorMaker) is the latest version?
2. Why do you use the StFtsFastSimulatorMaker for tracking studies

I hope we should settle down on this both for now (which I thought we did already but … ) and for the future.

Thanks,
Zhenyu

> On Jul 17, 2020, at 3:10 PM, Daniel Brandenburg <dbrandenburg.ufl AT gmail.com> wrote:
>
> Hi Zhenyu,
> I pushed the updated README to the github main branch. It includes a section about running 'fst_study.C'
>
> I used the Fast simulator that Te-Chuan told me was up-to-date. I checked with him twice to be sure I got the correct one- once before the discussion in FST meeting and then again once after. If something needs to be changed just have them send me a pull request with the changes.
>
> > Also, can you tell me if there is any difference between the combined and separated fast simulator for how the FST hits are handled (how the hits are smeared, what hit error matrix is put in etc)?
> I am not sure what you mean by "combined and separated fast simulator"
>
> If you are referring to the Si Fast simulator that I wrote then you can find it here:
> https://github.com/jdbrice/star-sw/blob/9e7d22967c643c0c3ea0e7174414537c81900730/StRoot/StgMaker/StgMaker.cxx#L67
> My version is much simpler than the FstFast simulator and simply places the hit at the center of the r-phi area for a given sensor. Maybe it is missing something that the full fast simulator does but it is easier to change the parameters.
>
> And the code to compute the covariance matrix by inverting the jacobian transform matrix is here:
> https://github.com/jdbrice/star-sw/blob/9e7d22967c643c0c3ea0e7174414537c81900730/StRoot/StgMaker/StgMaker.cxx#L325
>
> Please have them let me know if there are any problems. I ran a few tests but I have not yet run a large statistics run since making the swap to the FST Fast Simulator.
>
> Best,
> Daniel
>
>
> On Wed, Jul 15, 2020 at 7:38 PM Ye, Zhenyu <yezhenyu AT uic.edu> wrote:
> Hi Daniel,
>
> I resend you these emails in case you missed them. I appreciate timely reply from you.
>
> Thanks,
> Zhenyu
>
> > On Jul 15, 2020, at 12:26 AM, Ye, Zhenyu <yezhenyu AT uic.edu> wrote:
> >
> > Hi Daniel,
> >
> > I have a related request. You reported on Monday’s FST meeting that you updated the StgMaker to use FST Fast Simulator, and you were going to update the README and config file for the forward tracking package. Please correct me if I am wrong, but I don’t think the latter has been done yet as far as I can tell by looking at your GitHub area https://github.com/jdbrice/star-fwd-dev
> >
> > Please kindly make these updates at your earliest convenience, so that Gavin and Shenghui can start running the updated tracking code with the Fast Simulator.
> >
> > Thanks and Best,
> > Zhenyu
> >
> >> On Jul 14, 2020, at 12:39 PM, Ye, Zhenyu <yezhenyu AT uic.edu> wrote:
> >>
> >> Hi Daniel,
> >>
> >> In your GitHub area for Forward Simulation and Tracking, https://github.com/jdbrice/star-fwd-dev 
> >> I see the FST+sTGC combined fast simulator (StFtsFastSimulatorMaker) is being used.
> >>
> >> When I look into Te-Chuan’s work area on RCF /star/u/tchuang/FST_simulations/README
> >>
> >> ************************
> >>   - There are two versions of fast simulator:
> >>       - StFtsFastSimulatorMaker: Old fast simulator which has been updated for silicon part but STGC part was obsoleted.
> >>           - Use runFtsSim.C to run the fast simulator.
> >>           - I added some QA histograms about secondary particles in g2t tables. You can see the variable "isShower" declared at line 539 and histograms called "hTrutHitRShower[i]" which are the variable
> >> isShower vs. the mc truth hit position in r direction on different FST disks.
> >>       - StSiSimulatorMaker and StStgcSimulatorMaker: Separated fast simulator for FST and STGC done by Daniel (dbrandenburg.ufl AT gmail.com)
> >>           - I copied these codes from Daniel's directory. I have checked StSiSimulatorMaker that is updated and I believe StStgcSimulatorMaker is also updated.
> >>           - Use sim.C to run the fast simulator.
> >>           - Daniel didn't consider FST inefficiency in sim.C but you can see how to implement this in runFtsSim.C.
> >> ************************
> >>
> >> So I am wondering which version is the latest version. I tend to believe the separated ones are, because we have discussed about this in some meetings a while back (last year), and I believe we have agreed to split the two. But then I don’t understand why the combined version is in GitHub.
> >>
> >> Also, can you tell me if there is any difference between the combined and separated fast simulator for how the FST hits are handled (how the hits are smeared, what hit error matrix is put in etc)?
> >>
> >> Thanks,
> >> Zhenyu
> >
>

_______________________________________________
Star-fst-l mailing list
Star-fst-l AT lists.bnl.gov
https://lists.bnl.gov/mailman/listinfo/star-fst-l