GSIM Project Descriptions

Introduction

GSIMKO - The GSIM Post Processor: GSIM Knock Out.

The GSIMKO code is a stand-alone post processor for GSIM output files. The code is intended to perform all those tasks that are run to run dependend and can be performed outside of the main GEANT code. The reason for this is that it would speed up the simulation effort; instead of having to run and re-run the entire simulation to figure out the effect of a dead wire on the acceptance, only GSIMKO needs to be re-run to knock this wire out of the output file. This code will eventually take care of a whole lot more....

TSIM - Trigger Simulator

Drift Chambers Geometry.

 

Robert Feuerbach

10/17/97

Status: Started.

Update the geometry of the chambers to have a seperate geometry for each sector. The geometry definition should come from the Map and BOS geometry definition files that recsis uses. Ideally the identical (new) routines for defining the DC geometry in recsis are used here to define the GSIM geometry.

DC subproject: Implement DOCA bank.

Dennis Weygand
This bank avoids the use of converting distance to time by allowing gsim to write out the distance along with the time. The gsimko program can then use this distance and use a distance to time conversion which includes effects such as resolution smearing etc.

DC subproject: Implement Distance to Time conversion.

 

Jerry Gilfoyle

Status:Near Completion?

Implement some code that will convert the distance of a track to a wire. This time should then go into the appropriate bos bank. This routine is the counter part of the recsis time-to-distance routine.

Existing routines (in recsis/dc):

dc_fill_t_to_d_table which calls dc_xvst_get_doca which calls dc_xvst_curve and also dc_fill_d_to_t_table which calls dc_xvst_get_tdrift which call dc_xvst_curve. We are still working on the underlying function, dc_xvst_curve (in fact, it may be replaced b y dc_tvsx_curve) but you might see whether this structure is compatible with GSIM.
(Mac Mestayer)

Cherencov Counters Geometry:

 

In need of a Programmer.

Update the Cherencov Counters geometry to be seperate for each sector and use more realistic (actual) numbers. The geometry should come from the MAP, and reflect the position of the forward carriage as well as the position of the individual cherencov counters,

Scintillation Counters Geometry

 

Maurik Holtrop

-

Same as Cherencov Counters for the SC.

EGN Calorimeter Geometry:

 

In need of a programmer.

Same as CC and SC for EGN.

EGN subproject:Implement EGN Timing:

 

Will Brooks

?

Status:needs porting.

Implement the code that will fill the correct bos banks with the Calorimeter timing information, so that recsis can analyze it.

Add FAST card and Fix FFREAD problems:

 

Maurik Holtrop

4/24/98

Status: Finished

A new card was added for FFREAD: FAST  this option will switch you from the "full geometry with all the bells and whistles" to a "speedier but probably less accurate" code, for those subsystems where this is implemented. Currently only SC, but that will change rapidly. See NEW FEATURES
The FFREAD input file parser was fixed (now properly processes the all the entries on a card).

Large Angle Calorimeter Geometry:

 

In need of a programmer.

Update the geometry of the LAC to reflect actual geometry. This should come from the MAP and reflect the position of the clam shell.

Add Tagger:

 

Carlos Salgado.

Write the code needed to simulate the Tagger, and then maintain this code.

Speed Improvements.

This is a "catch all" section for speed improvements. Speed improvements are much needed for the GSIM code. Currently the program can take more than to one minute to trace a single electron through the detector. The goal should be at least a factor of 100 faster than that ! The following list is a number of places where I beleive we can speed up the code significantly. Note that both "Speed Gain" and "Difficulty" are just wild quessed by me, and are much open to discussion (please do email me at maurik at improv.unh.edu if you have an oppinion !):
 
 

Description	Speed Gain	Difficulty	Programmer
Shower Library for EC	Huge (~1000)	quite	David Tedeschi
General Shower Library	Large (100?)	quite	?
Variable Step Sizes through media.	Fair	medium	?
Eliminate Particles Outside Acceptance	Fair	fair	?
Cleanup Code and Geometries.	Medium	lots to do	?
Eliminate the 'MANY' volumes by making them 'ONLY'	Fair	medium	?
More detailed control over physics processes.	not known	fair	?

 

Shower Counter Libraries / Parametrization

 

Dave Tedeschi

Status: Started

Implement code to speed up the calculation of showers in the Calorimeters. Two different methods will be investigated: 1) Fast parametrization  2) shower library, ie lookp table.
For more information see Dave's page.

Passive Elements Geometry Update: Target Cells.

Currently no material is present in the target region. We need several different target cell geometries (normal target, gas target, solid targets) which includes target cell walls and target material. The type of target should be controlled with an FFREAD switch. This affects the code for the polarized target also. The polarized target is implemented, but the target material itself is not yet.
 

Magnetic Field Maps

 

Stenpan Stepanian

Completed Jan '98

Status: Finished

Update the code that reads the magnetic field maps (Main Torus and Mini)  so that it reads the same map as is used in RECSIS. Also implement code that uses the runnumber and the MAP to set the correct field strenght for a specific run.

Add Start Counter:

 

Burin Completed.

Write the code needed to simulate the Start Counter, and then maintain this code.

Create Better Makefile

 

Maurik Holtrop

Completed 6/18/97

Status:Finished

Create a Makefile for gsim_int and gsim_bat that allow the user to only modify a few files (such as guser_kine.F) and compile and link only these modified files into an executable use the standard library for the unmodified routines. The current Makefile scheme requires one to remake the entire library. See the New Features Page.

Add lookup Table for the Magnetic Field of Polarized Targets.

 

Angela Biselli and Vladimir Sapunenko Finished.

Look-up table for the magnetic field of the PTG (99% complete); Additional shielding to suppress low energy (moliere's) electrons (in progress).

Update Beamline Geometry.

 

Finished

Update the code that currently describes the beam line, and the target region. This would include some code that can simulate various different beamline configurations.

Update Support Structure Geometry:

 

Finished

Update the code that describes the support structure of the CLAS. This would involve adding the large ring at large angles, and the small ring at forward angles and improving the accuracy of other support structures as needed. Also a great improvement to the code would be to incorporate an offset for the Forward Carriage, in case it is not perfectly in.

Update Main Toroid Geometry.

 

Maurik Holtrop

Updated: 4/25/97.

Finished

Update the code that describes the main CLAS Toroid. The current shape of the toroid is not quite the same as currently incorporated in the gsim code.
Progress: Finished. See the New Features Page.

Update Mini Toroid Geometry.

 

Dale Conwell

Updated: 4/28/97

In Final Testing Phase.

Update the code that describes the mini toroid. The mini toroid is currently the wrong shape, and it interferes with the working of the region 1 drift chambers. As a result region 1 does not work when the mini toroid is turned on.
Progress: Almost Complete, Testing phase. See the New Features Page.

Background event simulation:

 

Maurik Holtrop

Last Checked in 4/24/97

Completed

This is a module that was revived from a precursor to gsim. It simulates background electrons, photons and positrons by looking up the cross-sections and particle distributions from a table (stored in a 2d histogram) and then distributing such events randomly, scaled by the luminosity. See: Improvements and New Features.

CELEG event generator:

John Price

This is actually a completely seperate program, and it is now found in the CVS repository. The CELEG code generates particles (four vectors at the target) that can then be traced through gsim. Please look at the CELEG MANUAL PAGE for more info.