First attempt at simulation based on AgML geometries

 

NOTE: Following blog entry likely wrong.  Pilot error in loading local vs STAR library.

 

 

We are now at the point where we can export AgML back to AgSTAR, and run a starsim simulation using this path.  (With the one exception that we use the native agstar CAVE, as hits are not generated for the agml exported cave...) Spot checking hits for one event where we throw 10 electrons, we find (using sdiff):

BEMC HITS (example)


 ====> HITS  IN DETECTOR ** CSUP ** OF SET ** CALH ** <====      ====> HITS  IN DETECTOR ** CSUP ** OF SET ** CALH ** <====

  HITS TRACK CHLV CPHI CSUP      ETA       Y         BIRK         HITS TRACK CHLV CPHI CSUP      ETA       Y         BIRK  
     1     9    1   19    1   0.275     -6.50     8.277E-04          1     9    1   19    1   0.275     -6.50     8.277E-04
     2     7    2   48    1   0.575     -6.50     4.045E-03          2     7    2   48    1   0.575     -6.50     4.045E-03
     3     7    2   47    2   0.575      6.50     0.170              3     7    2   47    2   0.575      6.50     0.170    
     4     7    2   48    2   0.675     -6.50     5.282E-04          4     7    2   48    2   0.675     -6.50     5.282E-04
     5     7    2   48    2   0.625     -6.50     1.020E-02          5     7    2   48    2   0.625     -6.50     1.020E-02
     6     7    2   48    2   0.575     -6.50     0.668              6     7    2   48    2   0.575     -6.50     0.668    
     7     7    2   47    2   0.525      6.50     4.840E-04          7     7    2   47    2   0.525      6.50     4.840E-04
     8     7    2   48    2   0.525     -6.50     3.166E-04          8     7    2   48    2   0.525     -6.50     3.166E-04
     9     7    2   48    1   0.625     -6.50     2.756E-04          9     7    2   48    1   0.625     -6.50     2.756E-04
    10     7    2   47    2   0.625      6.50     4.780E-02         10     7    2   47    2   0.625      6.50     4.780E-02
    11     7    2   48    2   0.575      6.50     1.327E-03         11     7    2   48    2   0.575      6.50     1.327E-03
    12     7    2   47    2   0.675     -6.50     3.379E-05         12     7    2   47    2   0.675     -6.50     3.379E-05
    13     7    2   47    2   0.525     -6.50     9.078E-04         13     7    2   47    2   0.525     -6.50     9.078E-04
    14     5    1   35    1   0.175      6.50     1.437E-03         14     5    1   35    1   0.175      6.50     1.437E-03
    15     3    1   57    2   0.175      6.50     8.847E-04         15     3    1   57    2   0.175      6.50     8.847E-04
    16     3    2   16    2   0.225      6.50     2.024E-04         16     3    2   16    2   0.225      6.50     2.024E-04
    17     3    2   32    1   0.525      6.50     1.038E-03         17     3    2   32    1   0.525      6.50     1.038E-03
    18     3    1   57    2   0.125     -6.50     3.101E-05         18     3    1   57    2   0.125     -6.50     3.101E-05

 ====> HITS  IN DETECTOR ** CSDA ** OF SET ** CALH ** <====      ====> HITS  IN DETECTOR ** CSDA ** OF SET ** CALH ** <====

  HITS TRACK CHLV CPHI CSDA      TYPE      ETA       ETSP         HITS TRACK CHLV CPHI CSDA      TYPE      ETA       ETSP    
     1     7    2   48    2    2.00     0.550      26.5     1        1     7    2   48    2    2.00     0.550      26.5     1
     2     7    2   48    4    4.00     0.550     0.747     7        2     7    2   48    4    4.00     0.550     0.747     7
     3     7    2   48    4    4.00     0.550      2.24     1        3     7    2   48    4    4.00     0.550      2.24     1

Looking at some of the other detectors:

[1] No differences in the ECAL
[2] No dfifferences in the TPCM
[3] No differences in the BBCM
[4] No differences in the VPDD
[5] No signal in the MUTD (none expected for e-)
[6] No signal in the BTOF (none expected due to bug in y2009a geometry)
[7] No differences in the FPDM (includes both FMS and FPD hits)

 

Conclusions:

Barring some very specific mistakes in the agstar import / agml and agml to agstar export, the detector hits are properly exported from AgML back to AgSTAR.

While there are some remaining details to ensure that we get right, specifically tracking cuts in calorimeters, cherenkov settings, etc... we are ready to base simulations on the AgML geometry.