[Solid_software] GEMC benefits and larger frameworks

[Zhiwen Zhao]

hi, Rich

"Copyof" in GEMC is using one logical volume with many physical volumes, 
refer to
https://eic.jlab.org/wiki/index.php/Detector_simulation_general_note#define_type_and_dimensions
So our baffle can be modified to use it.
Long time ago I used it for baffle and don't remember it gave big 
improvement on speed and mem use, but I could be wrong.

Ole said last time artg4 as an example of simulation for art is fairly 
independent of art.
I would be nice if someone can look into how it actually works in detail 
and evaluate if any existing geant4 code like GEMC can be connected to art.

Zhiwen

On 3/4/2016 11:51 AM, Richard S. Holmes wrote:
> On Thu, Mar 3, 2016 at 5:28 PM, Maurizio Ungaro <ungaro at jlab.org
> <mailto:ungaro at jlab.org>> wrote:
>
>     - FADC output - mode 1 right now but mode 5 and 7 coming soon. This
>     is a 4ns sampled signal, same as real life.
>
>
> This isn't something I know a lot about; is it applicable to SoLID?
>
>     - Documentation is getting there ;-) gemc.jlab.org
>     <http://gemc.jlab.org>
>
>
> I wasn't aware of the new site, looks good so far.
>
>     - Coming next week: merging of background events into generator
>
>
> Potentially significant for us I think.
>
>     This is indeed a limitation. Currently in clas12 we do use the same
>     parameters as reconstruction, because they come from a common DB.
>     The calculations of parameters are done using the reconstruction
>     java library.
>     If the reconstruction is in C++ this can be achieved with the plugin
>     capability of gemc - using a shared library that is used in
>     reconstruction as well.
>
>
> This is something I'd like to know more about.
>
>     Having said that, it’s not feasible to include in reconstruction ALL
>     the passive volumes.
>
>
> I don't think there's any need to.
>
>     If the mysql DB is used in the API, the geometry is run and
>     variation indexed. So it’s uniquely associated with an output (where
>     run, variation and ALL geometry parameters are stored).
>
>
> I need a few more words about this. Given an output file, how does one
> determine the parameters and the Perl script versions used to create the
> geometry?
>
>>     Also note: GEMC does not allow (I think) taking advantage of
>>     GEANT4's ability to replicate logical volumes; e.g. each
>>     calorimeter module must be a separately-defined entry in the database.
>     Replicas are there. Copies are there.
>
>
> They are? Are we using GEMC wrong? Currently our Perl scripts generate,
> for instance, 30 copies of each baffle block, one for each sector,
> rather than one logical volume that's instantiated as 30 physical volumes.
>
>     Admittedly gemc was not designed to be part of a framework, however
>     a partial re-design is in process that could help in that regard.
>     For example, using plugins for hit processes, for generator and
>     output mechanisms.
>
>
> That sounds promising.
>
> --
> - Richard S. Holmes
>    Physics Department
>    Syracuse University
>    Syracuse, NY 13244
>    315-443-5977
>
>
> _______________________________________________
> Solid_software mailing list
> Solid_software at jlab.org
> https://mailman.jlab.org/mailman/listinfo/solid_software
>