[Clas12_high_luminosity] Simulation Update

[Whitney R. Armstrong]

Dear Latifa,

Sorry for the delayed reply. I have simulated ~300M beam-on-target events and 
am waiting for the scripts to produce the final plots. 

>1.   your results show that the occupancies in RII are higher than RI.
>   We need to understand that also.

I am still working on my simulation to make sure I haven't missed anything.  
The first thing I can think of is the fact that I don't have any detectors 
between the target and DC. But I don't think this minimal material can explain 
the two orders of magnitude difference.

>2. Can you read the  Geometry data base the same as Mauri, if you could
>   that will be the best. This way we can take out this input from the
>   equation and we can focus on the simulation part.

I am not sure what you mean by this: do you mean the ccdb, reconstruction code, 
or the perl scripts? I can read the GDML he sent.  As far as insensitive 
geometries, GDML is the best way to go.  The difficulties arise when we want to 
do sensitive geometries, which is why I re-did the drift chamber from scratch.
  
>3. The graphs that Volker asked for are produced based on which one of
>   your methods?

Those are for the beam-on-target methods.

>4. In producing these graphs could you also have them by particle type?

Yes. I am making the high statistics plots now but the low statistics is found 
at
https://clasweb.jlab.org/wiki/index.php/CLAS12_Drift_Chamber_Occupancy_Simulations#DC_Hit_Origins
I will update this page when the plots are done.

I look forward to your comments and questions.

Cheers,
Whit


On Fri, Feb 12, 2016 at 10:07:42AM -0500, Latifa Elouadrhiri wrote:
>Dear Whitney,
>
>Yes it will be good to see your results with high statistics, to look 
>at the details for the comparison between the 2 methods.
>
>I have 2 questions:
>
>1.   your results show that the occupancies in RII are higher than RI.
>   We need to understand that also.
>2. Can you read the  Geometry data base the same as Mauri, if you could
>   that will be the best. This way we can take out this input from the
>   equation and we can focus on the simulation part.
>3. The graphs that Volker asked for are produced based on which one of
>   your methods?
>4. In producing these graphs could you also have them by particle type?
>
>
>Best regards,
>
>Latifa
>
>On 2/12/16 8:14 AM, burkert wrote:
>>Whitney,
>>
>>Thanks for your response. The graphs are of interest, however, I 
>>believe they are not the same as the ones Mauri is showing. Mauri 
>>can correct me, but I think he is plotting the z position along the 
>>beamline where any of the hits in the chambers originated from. This 
>>allows one to determine the material around the beamline that caused 
>>the hits even if it goes through secondary and tertiary stages. You 
>>expect then a peak at the target for direct hits, and hits coming. 
>>e.g. from Moeller electrons produced in the target but hitting a 
>>beamline element further downstream that causes a shower and 
>>subsequently hits in the chambers.  Such a 1-D plot will allow us to 
>>compare your simulations with Mauri's and identify structures along 
>>the beam that may present obstacles to the Moeller electrons.
>>
>>Volker
>>
>>
>>
>>
>>On 2/11/16 8:42 PM, Whitney R. Armstrong wrote:
>>>Hi Volker,
>>>
>>>I have produced two graphs, however, with limited statistics 
>>>(~1.5M). Many more jobs are running on the farm right now.
>>>
>>>The first shows (in cylindrical coordinates) the radial vs z 
>>>position of the starting point for the trajectory that ultimately 
>>>produced at hit in the DC.
>>>
>>>The second shows the wire number of the DC hit vs the same z as above.
>>>I will post these results on the wiki when I have gathered more 
>>>statistics.
>>>Your comments and questions are appreciated.
>>>
>>>Cheers,
>>>Whit
>>>
>>>
>>>On Thu, Feb 11, 2016 at 08:37:45AM -0500, burkert wrote:
>>>>Hi Whitney,
>>>>
>>>>Could you produce a graph similar to the one from Maurizio 
>>>>(attached) showing where along the beamline (z-coordinate) the 
>>>>chamber hits originate from.
>>>>
>>>>Volker
>>>>
>>>>
>>>>On 2/11/16 1:41 AM, Whitney R. Armstrong wrote:
>>>>>Dear Collaborators,
>>>>>
>>>>>First, I have posted some follow-up slides to our discussion 
>>>>>on Monday. The new plots have a hit definition that is 
>>>>>consistent with Maurizio's.
>>>>>You can find my slides and the follow slides at https://clasweb.jlab.org/wiki/index.php/CLAS12_High_Luminosity_-_February_8,_2016
>>>>>
>>>>>Secondly, as suggested by Latifa, I have performed a self 
>>>>>consistency check using the beam on target method previously 
>>>>>used. Furthermore, I am using the same beamline geometry as 
>>>>>Maurizio, who has graciously provided a GDML version of the 
>>>>>geometry.
>>>>>
>>>>>Please see the following page for the latest results and 
>>>>>discussion of the simulations. https://clasweb.jlab.org/wiki/index.php/CLAS12_Drift_Chamber_Occupancy_Simulations
>>>>>
>>>>>So far the main conclusions are that my simulation is self 
>>>>>consistent, and perhaps more importantly, I am getting 
>>>>>significantly larger DC occupancies.  The cause of this 
>>>>>discrepancy remains unclear.
>>>>>
>>>>>Any comments or questions are greatly appreciated.
>>>>>
>>>>>Cheers,
>>>>>Whit
>>>>>
>>>>>
>>>>
>>>
>>>
>>>>_______________________________________________
>>>>Clas12_high_luminosity mailing list
>>>>Clas12_high_luminosity at jlab.org
>>>>https://mailman.jlab.org/mailman/listinfo/clas12_high_luminosity
>>>
>>>
>>
>>_______________________________________________
>>Clas12_high_luminosity mailing list
>>Clas12_high_luminosity at jlab.org
>>https://mailman.jlab.org/mailman/listinfo/clas12_high_luminosity
>

-- 
Whitney R. Armstrong