STAR HardProbes PWG

[Rosi Reed <rosijreed AT lehigh.edu>]

Hi Ahmed,

I believe adjusting how many BSMD strips are saved is something that can be easily done - given the definition of the TSP we felt 3 towers was sufficient, but a different definition is totally possible.  My advice here is to make the picoDst definition match what is needed - because this is the data structure that should be used for all such analyses in the future. 

I would be very interested in documentation of the changing TSP definition.  Unfortunately with the pandemic and other issues, Annika was not able to complete the full body of work I had intended, but I wanted to look at the differences both in data and in simulation.  I wasn't aware that there were different TSP definitions - though it was clear that the BSMD gain was not as well calibrated as one might hope.  Since most analyses using this trigger were simply ratios of gamma-rich vs pi0-rich populations, this issue mostly fell out.  But for true complementarity between the very nice Atlas/CMS results and for the (hopefully) soon-to-be-produced sPHENIX results we will have to fully understand the efficiency and background. 

Cheers,

Rosi

On Sun, Apr 2, 2023 at 11:26 AM Ahmed M. Hamed <ahmed.hamed2008 AT gmail.com> wrote:

Hi Rosi;

Thanks for your email.

Regarding the trigger, my main point is not use the trigger patch as the spin folks in pp. L2gamma was very selective and suitable for the heavy ion environment.

Regarding; Annika’s work; reproducing the TSP is not sufficient that all are in place. The TSP itself changed over years with the change in the BSMD gain. Anyway; should I understand that only those strips under the 3x3 fired towers are kept, and the rest are thrown away? If yes; then I have to think before concluding the impact on the transverse shower profile Algo.

Best Regards 

On Sun, Apr 2, 2023 at 5:04 PM Rosi Reed <rosijreed AT lehigh.edu> wrote:

Hi All,

My understanding is that BHT3-L2Gamma used the same L0 BHT3 trigger plus a L2Gamma algorithm.  When I look at 2012 or earlier, it looks like there was also a barrel threshold in the L0 bits that seems to match with the 3rd level of barrel trigger that we used.  (At that time they were called NPE triggers apparently.)  This is all recorded in the trigger details in runlog, though interpretation sometimes takes a little bit.

> 3) If it is an option to run either over pico or mudst root files;
> then why do we need to request the mudst? Again; I don't know what was
> thrown away for the BSMD in the pico format. Any feedback?!

>If we plan and request to have only Mudst production, then we will not

Maybe I haven't been paying close enough attention, but I do not see any reason why only the MuDst would be requested.  The whole point of developing the picoDst structure was so that STAR can continue to function with the increased data rate.  The last few years of data are going to be huge.  It's possible that the L2gamma portion is smallish given the relative cross-sections, but my suspicion is that we will quickly need to really balance disk space.

As for the BSMD in the pico format, what Annika did was add the BSMD trip information that was within 3 towers of any tower that fired the trigger.  She was able to essentially duplicate the TSP using either the mudst or the picodst.  It would probably be worthwhile for someone who is planning on analyzing the run 23 data to confirm that everything is there.  It should be possible to reconstruct everything, though it may require grabbing the data in a way that is different than in the mudst.  (I think one can grab the track projection, for instance, through the emc pid traits class.)

In any case, my advice would be that anyone who want to do photon analyses in STAR in run 25, which will be a beautiful, high luminosity data set, should absolutely spend the time now to confirm that everything that is needed can be grabbed from the picoDst (or add it if not).

Cheers,

Rosi

On Sun, Apr 2, 2023 at 7:59 AM Nihar Sahoo via Star-hp-l <star-hp-l AT lists.bnl.gov> wrote:

Dear Ahmed,
Please find my replies inline.

On 2023-04-02 00:59, Ahmed M. Hamed wrote:
> Dear Nihar;
> I think I failed in making my questions/concerns clear, so let me try
> again:
> 1) For the old version of L2gamma trigger (07,08,10,11,12); we don't
> mix it with the trigger patch, indeed as you stated the highest tower
> was triggered at 5.76 GeV and the next to highest tower out of the 3x3
> tower patch was set to 7.44 GeV. Just the sum of two towers (highest +
> next-to-highest). The offline clustering algo. doesn't work for the
> asymmetric decay modes of pi0. This will make a significant difference
> in the environment of high background AuAu..Surely there was a reason
> to add the BHT3+L2gamma for run 14 which I don't know
Not sure why only for Run14 and 16 it was labeled as BHT3+L2gamma.
I have seen other runs it was labeled as "L2gamma" only.
But when I look at previous AuAu runs the thresholds were the same with
Run14/16.
Sorry, I don't know the reason.

> 2) Are those percentages 0.006 and 0.002 the ones of gamma-rich which
> passed all offline cuts per minbias?
These are not from offline cuts, these numbers are from online L2gamma
trigger events over total collected events (in a particular week and
week-to-week variation is negligible during Run14).
> If so; then I think the more
> meaningful ones would be the percentage per triggered event in each
> luminosity; as those are the ones needed for the projected reduction
> in the statistical errs.The trigger rates can be controlled easily
> during the run; tuning the threshold.

> 3) If it is an option to run either over pico or mudst root files;
> then why do we need to request the mudst? Again; I don't know what was
> thrown away for the BSMD in the pico format. Any feedback?!
If we plan and request to have only Mudst production, then we will not
request picodst for L2gamma trigger sample. But if data sample size will
be huge (which may not be the case) then we may ask for picodst. I have
not gone through the picodst files yet to tell you in detail which info
are not present to do some additional study on TSP. AFAS I know Saskia's
one student has been working on it and he produces almost the same TSP
profile from picodst as our mudst but there are some additional
information missing in picodst (for example either some strips info or
projected track info , I have forgotten) to do further study on TSP. On
this topic we can discuss offline.

Thank you
Nihar

> Best Regards
>
> On Sat, Apr 1, 2023 at 2:47 PM Nihar Sahoo <nihar AT rcf.rhic.bnl.gov>
> wrote:
>
>> Dear James and Ahmend,
>>
>> Thank you for your input and comments.
>> Please find my replies inline.
>>
>> On 2023-04-01 05:05, James Dunlop wrote:
>>> Ahmed, Nihar,
>>> The real question is what threshold you need.
>>> BHT1, 2, 3 is just 3 different thresholds in Et for a single High
>>> Tower,
>>> which you can choose/negotiate with other PWG's.  e.g. Upsilon
>>> tends to want a specific Et cut to avoid cutting too hard into the
>>> invariant
>>> mass distribution beyond, say, 9 GeV in M (and so 4.5 GeV in Et,
>> try
>>> to keep it more like 4 so
>>> you can understand the combinatoric background),
>>> and UPC J/Psi wants their Et cut very low.
>>>
>> Thank you for pointing this. I was thinking on this direction either
>> we
>> should increase or lower (BHT3 is 5.76 and 3x3 patch 7.44
>> thresholds)
>> for L2gmma trigger thresholds.
>> As we use 6 GeV cutoff to high tower trigger energy for our offline
>> anlysis, probably it might be Ok to keep the above thresholds for
>> Run23.
>> and  could that be decided during run time if it is needed?
>> But I welcome your suggestion and we can further discuss at trigger
>> board meeting.
>>
>>> L2 is somewhat useful to moderate a strict high tower
>>> into something that incorporates information from other
>>> towers, but it can be problematic in the high Underlying Event
>>> condition of Au+Au collisions.
>>>
>>> --J
>>>
>>>> On Mar 31, 2023, at 6:45 PM, Ahmed M. Hamed via Star-hp-l
>>>> <star-hp-l AT lists.bnl.gov> wrote:
>>>>
>>>> Dear Nihar;
>>>> Thanks a lot for the slides.
>>>> Two comments and one question:
>>>> 1) I don't understand why not L2gamma and not BHT3+L2gamma; I
>> really
>>>> don't know what BHT3 adds to the trigger components?
>>
>> As James mentioned BHT1,2,3 are different threshold used during
>> Run14
>> and 16.
>> "BHT3+L2gamma" label used for above last runs with  high tower
>> threshold
>> is 5.76 GeV and 3x3 patch threshold is 7.44 GeV. Again as James
>> pointed
>> out We may need to think on these thresholds for Run23 (and also
>> Run25
>> very high lumi) which depend upon the low and high lumi.
>> I will discuss this topic at trigger borad meeting to come up with
>> these
>> thresholds (either we can set these numbers now or during the run
>> time,
>> I need suggestion in this direction)
>>
>>>> 2) I think what you need to quote here (slide # 6) are the
>> numbers
>>>> of events as a percentage from the number of triggered events per
>>>> luminosity, and not all events.
>> I quickly calculated the following numbers from Run14 runlog the
>> triggered event fractions (L2gamma trigger events/MB events)
>> ~0.006 for high
>> ~0.002 for low
>>>> Question: Migrating from the mudst to picodst what information
>> has
>>>> been thrown away? e.g. Whether running the cluster algorithm
>> (BSMD)
>>>> for the transverse shower profile over the picodst is still an
>>>> option; or one must run over the mudst files in order to find all
>>>> information required by the cluster algo.; i.e. the energy and
>> adc
>>>> per strip.
>> As L2gamma trigger events will be very small is size even for
>> upcoming
>> runs. we may ask for Mudst production only. But picodst production
>> is
>> also possible and both energy and adc information are there inside
>> pico.
>>
>> Thank you
>> Nihar
>>
>>>> Best Regards
>>>>
>>>> On Fri, Mar 31, 2023 at 2:23 PM Nihar Sahoo via Star-hp-l
>>>> <star-hp-l AT lists.bnl.gov> wrote:
>>>>
>>>>> Hello All,
>>>>>
>>>>> The proposed triggers for gamma+jet and HF physics have been
>>>>> updated as
>>>>> Rongrong suggested.
>>>>> (Slide#4)
>>>>>
>>>>>
>>>>
>> https://drupal.star.bnl.gov/STAR/blog/nihar/HP-pwg-Run23-discussion
>>>>>
>>>>> Thank you
>>>>> Nihar
>>>>>
>>>>> On 2023-03-31 12:17, Ma, Rongrong wrote:
>>>>>> Hello Nihar
>>>>>>
>>>>>> Thanks for the slides. I think it will be nice if we can list
>>>>>> explicitly the triggers we want.
>>>>>>
>>>>>> In terms of quarkonium physics, here are the triggers we
>>>>> propose:
>>>>>>
>>>>>> * dimuon+zdc_coin+vpd100
>>>>>> * BHT2+zdc_coin+vpd100
>>>>>> * BHT1+zdc_coin+vpd10
>>>>>>
>>>>>> * monitor triggers: vpd-zdc-novtx, vpd-zdc-10
>>>>>>
>>>>>> Best
>>>>>> Rongrong
>>>>>>
>>>>>> -------------------------
>>>>>>
>>>>>> From: Star-hp-l <star-hp-l-bounces AT lists.bnl.gov> on behalf of
>>>>> Nihar
>>>>>> Sahoo via Star-hp-l <star-hp-l AT lists.bnl.gov>
>>>>>> Sent: Thursday, March 30, 2023 12:45 AM
>>>>>> To: Star-hp L <Star-hp-l AT lists.bnl.gov>
>>>>>> Subject: [Star-hp-l] Run23 data taking plan from HP-pwg
>>>>>>
>>>>>> Hello HP-pwg members,
>>>>>>
>>>>>> We were asked to provide inputs, particularly on trigger setup,
>>>>> vertex
>>>>>>
>>>>>> selection, and BSMD usage, for Run23 data taking. You must have
>>>>> seen
>>>>>> Rongrong's emails in this direction.
>>>>>>
>>>>>> A few slides that we have prepared our response (questions can
>>>>> be
>>>>>> found
>>>>>> inside).
>>>>>>
>>>>>
>>>>
>> https://drupal.star.bnl.gov/STAR/blog/nihar/HP-pwg-Run23-discussion
>>>>>>
>>>>>> Mainly we have included STAR gamma+jet and also dimuon
>>>>> measurements
>>>>>> requirement in these slides if you want to propose/suggest any
>>>>>> specific
>>>>>> trigger/detector usage for RUn23 relevant to your analysis,
>>>>> please let
>>>>>>
>>>>>> us know asap.
>>>>>>
>>>>>> Coming Monday we plan to present at the trigger board meeting.
>>>>>>
>>>>>> Thank you
>>>>>> Nihar for HP-conveners
>>>>>> _______________________________________________
>>>>>> Star-hp-l mailing list
>>>>>> Star-hp-l AT lists.bnl.gov
>>>>>> https://lists.bnl.gov/mailman/listinfo/star-hp-l
>>>>> _______________________________________________
>>>>> Star-hp-l mailing list
>>>>> Star-hp-l AT lists.bnl.gov
>>>>> https://lists.bnl.gov/mailman/listinfo/star-hp-l
>>>> _______________________________________________
>>>> Star-hp-l mailing list
>>>> Star-hp-l AT lists.bnl.gov
>>>> https://lists.bnl.gov/mailman/listinfo/star-hp-l
>>>
>>> --
>>> He/Him/His
>>> Please do not feel obligated to respond to this message outside of
>>> your work hours.
>>> James C Dunlop Ph.: (631) 344-7781
>>> Building 510A         Cell: (631)316-8153
>>> P.O. Box 5000
>>> Brookhaven National Laboratory
>>> Upton, NY 11973
>>> dunlop AT bnl.gov
_______________________________________________
Star-hp-l mailing list
Star-hp-l AT lists.bnl.gov
https://lists.bnl.gov/mailman/listinfo/star-hp-l

--

Rosi Reed
RHIC/AGS UEC member
Associate Professor, Physics Department
Lehigh University
(610)758-3907
16 Memorial Drive East Office 406
Bethlehem, PA 18015
she/her/hers

--

Rosi Reed
RHIC/AGS UEC member
Associate Professor, Physics Department
Lehigh University
(610)758-3907
16 Memorial Drive East Office 406
Bethlehem, PA 18015
she/her/hers