Links

Content Skeleton

This Page

Previous topic

Photon Level Debug

Next topic

Propagated Flags Mismatch

Material Pair Debug

Thoughts on how to proceed

There is a wealth of details and curiosities (asymmetries, unexpected positions), but do they matter ?

Some will be artifacts of selection techniques, presentation techniques, truncation.

Avoid getting bogged down in these now until have Geant4 equivalents to compare against, otherwise risk spending time barking up the wrong trees. This will provide a candle against which it should be easier to see which aspects need attention.

So, in summary aim for Geant4 comparison earlier rather than later : once have interface into main features of the propagation. (currently missing surfaces).

Geant4 step level comparison

Collect G4 steps into VBO like structure ?

Machinery

  1. More intuitive to select photons that do a particular step rather than select steps. How to implement ?

DAEPhotonsAnalyzer.get_material_pairs

Both end material check

Pushing up to –max-slots 30 reduces asymmetry peculiarities a lot. Ten slots (which is effectively 8 slots) is simply not enough. Pushing to –max-slots 100 suffers from crashes when in animation modes and slow interface in other modes.

With selection based on from/to materials. Peculiarities:

Tyvek,Tyvek                     9755    # not real, just 0,0
GdDopedLS,Acrylic               6209   confetti and noodle modes show many steps starting on upper acylic of inner AD, almost none from the bottom
       # large numbers along muon path are expected, its almost inevitable that any step staring inside GDLS is going to interesect to acrylic

udp.py --material GdDopedLS,ANY
       # generalize, still almost nothing downwards


Acrylic,LiquidScintillator      5809
LiquidScintillator,Acrylic      5402
MineralOil,Acrylic              5000
Acrylic,MineralOil              3575
ESR,Air                         2079
udp.py --material ESR,Air  --style noodles,confetti   ## almost all at base, pointed down
pushing up max-slots to 30 equalizes this a lot, so a truncation artifact

StainlessSteel,IwsWater         1178
Acrylic,Air                     482
Acrylic,GdDopedLS               402     # huge asymmetry, whereas not for LS ?
MineralOil,Pyrex                306     # OK: Mostly just inside the MineralOil, nr Acrylic, and all pointing at PMTs. Top orthographic clearest
StainlessSteel,MineralOil       258
MineralOil,StainlessSteel       240
Air,ESR                         237    udp.py --material Air,ESR  --style noodles,confetti   ## also, almost all at base
Vacuum,Bialkali                 191    # efficiency ?
Pyrex,Vacuum                    165    # OK: all inside PMTs, for clear view make geometry invisible with I
Bialkali,Vacuum                 130
UnstStainlessSteel,MineralOil   82
Pyrex,MineralOil                59
IwsWater,IwsWater               16
Vacuum,OpaqueVacuum             14
StainlessSteel,Water            9
MineralOil,UnstStainlessSteel   8
LiquidScintillator,Teflon       8
LiquidScintillator,GdDopedLS    7
LiquidScintillator,MineralOil   6
GdDopedLS,LiquidScintillator    4
OwsWater,Pyrex                  3
OpaqueVacuum,Vacuum             3
IwsWater,StainlessSteel         2
Teflon,LiquidScintillator       2
Water,StainlessSteel            2
StainlessSteel,GdDopedLS        2
StainlessSteel,NitrogenGas      2
OwsWater,UnstStainlessSteel     1
Air,Acrylic                     1
IwsWater,Water                  1
g4daeview.sh  --load 1 --wipepropagate --debugkernel --debugphoton 0 --max-slots 30

54068 Tyvek,Tyvek                        # dummy

12936 LiquidScintillator,Acrylic
12365 Acrylic,LiquidScintillator
12324 MineralOil,Acrylic

9128 Acrylic,MineralOil
8646 GdDopedLS,Acrylic
2761 Acrylic,GdDopedLS

2678 ESR,Air
2169 Acrylic,Air
1425 StainlessSteel,IwsWater
1139 MineralOil,StainlessSteel

759  Air,ESR
697  Bialkali,Vacuum
655  MineralOil,Pyrex
638  Vacuum,Bialkali
636  Pyrex,Vacuum
525  Pyrex,MineralOil

295  StainlessSteel,MineralOil
255  Vacuum,Pyrex
240  Vacuum,OpaqueVacuum
226  OpaqueVacuum,Vacuum
105  UnstStainlessSteel,MineralOil
62   MineralOil,UnstStainlessSteel
50   UNKNOWN,UNKNOWN
30   LiquidScintillator,GdDopedLS
23   Teflon,LiquidScintillator
21   LiquidScintillator,MineralOil
19   GdDopedLS,LiquidScintillator
17   LiquidScintillator,Teflon
16   MineralOil,LiquidScintillator
10   StainlessSteel,Water
5    IwsWater,IwsWater
4    Air,Acrylic
3    StainlessSteel,GdDopedLS
3    GdDopedLS,Teflon
3    Water,StainlessSteel
3    OwsWater,Pyrex
3    Teflon,GdDopedLS
3    StainlessSteel,NitrogenGas
2    IwsWater,StainlessSteel
2    IwsWater,Water
1    OwsWater,UnstStainlessSteel
g4daeview.sh  --load 1 --wipepropagate --debugkernel --debugphoton 0 --max-slots 100

304179 LiquidScintillator,LiquidScintillator    # dummy zeros

20825 MineralOil,Acrylic
20400 LiquidScintillator,Acrylic
19592 Acrylic,LiquidScintillator

15015 Acrylic,MineralOil
11249 GdDopedLS,Acrylic
5097 Acrylic,GdDopedLS

3423 Acrylic,Air
2886 ESR,Air
2318 MineralOil,StainlessSteel
1680 StainlessSteel,IwsWater              # all absorption steps presumably

1561 Bialkali,Vacuum
1128 Vacuum,Bialkali
1126 Pyrex,Vacuum
1094 MineralOil,Pyrex
1006 Pyrex,MineralOil
963  Air,ESR

526  Vacuum,Pyrex
466  Vacuum,OpaqueVacuum
454  OpaqueVacuum,Vacuum
371  StainlessSteel,MineralOil
189  LiquidScintillator,GdDopedLS
187  Teflon,LiquidScintillator
155  LiquidScintillator,MineralOil
123  UnstStainlessSteel,MineralOil
110  MineralOil,UnstStainlessSteel
95   GdDopedLS,LiquidScintillator
83   MineralOil,LiquidScintillator
65   LiquidScintillator,Teflon
51   UNKNOWN,UNKNOWN
15   IwsWater,StainlessSteel
14   Water,StainlessSteel
12   StainlessSteel,Water
6    IwsWater,IwsWater
6    IwsWater,Water
5    Air,Acrylic
3    GdDopedLS,Teflon
3    Teflon,GdDopedLS
3    StainlessSteel,GdDopedLS
3    OwsWater,Pyrex
3    StainlessSteel,NitrogenGas
2    Nitrogen,Water
1    Water,IwsWater
1    Vacuum,Acrylic
1    OwsWater,UnstStainlessSteel
1    Nitrogen,MineralOil
1    MineralOil,Nitrogen
1    MineralOil,Vacuum
1    Acrylic,Vacuum
1    Vacuum,MineralOil

Implement step selection based in from and to materials

Need to expand VBO to hold material info.

Broken chain

p_flags[1077]                  p_lht[1077]
         [[ 10.017   10.017 ]  [[  1255      3      5      0] GdDopedLS  Acrylic
          [ 10.017   19.2213]   [   991      5     11      1] Acrylic    LiquidScin
          [ 10.017   19.3094]   [   635     11      5      2] LiquidScin Acrylic
          [ 10.017   21.8971]   [   371      5      6      3] Acrylic    MineralOil
          [ 10.017   22.0028]   [632049      5      1      4] Acrylic    Air
          [ 10.017   22.0614]   [632341      1      4      5] Air        ESR
          [ 10.017   22.0616]   [632354      4      1      6] ESR        Air
B_ABSORB  [ 10.017   22.0616]   [    -1      4      1      7] ESR        Air
B_ABSORB  [ 10.017   22.0616]   [    -1      4      1      7] ESR        Air
B_ABSORB  [ 10.017   22.0616]]  [    -1      4      1      7]] ESR        Air
p_post[1077]                                                     p_dirw[1077]                                p_polw[1077]                        p_ccol[1077]
[[ -18007.123  -799808.25     -7062.793       10.017 ] -1        [[  -0.2251    0.4768   -0.8497  487.7488]  [[ 0.7983 -0.4097 -0.4414  1.    ]  [[ 1.  1.  1.  1.]
 [ -18423.6504 -798926.       -8635.          19.2213] 1850.32    [  -0.224     0.4745   -0.8513  487.7488]   [-0.3634  0.7698  0.5247  1.    ]   [ 1.  1.  1.  1.]
 [ -18427.5977 -798917.625    -8650.          19.3094] 17.6273    [  -0.2251    0.4768   -0.8497  487.7488]   [-0.3628  0.7684  0.5273  1.    ]   [ 1.  1.  1.  1.]
 [ -18544.6973 -798669.625    -9092.          21.8971] 520.173    [  -0.224     0.4745   -0.8513  487.7488]   [-0.3634  0.7698  0.5247  1.    ]   [ 1.  1.  1.  1.]
 [ -18549.4336 -798659.5625   -9110.          22.0028] 21.1586    [  -0.2287    0.4843   -0.8445  487.7488]   [-0.3605  0.7637  0.5356  1.    ]   [ 1.  1.  1.  1.]
 [ -18552.1133 -798653.875    -9119.9004      22.0614] 11.728     [  -0.3425    0.7255   -0.5969  487.7488]   [-0.2548  0.5397  0.8023  1.    ]   [ 1.  1.  1.  1.]
 [ -18552.1426 -798653.8125   -9119.9502      22.0616] 0.0851179  [  -0.3426    0.7257   -0.5966  487.7488]   [-0.2547  0.5395  0.8026  1.    ]   [ 1.  1.  1.  1.]
 [ -18552.1426 -798653.8125   -9119.9502      22.0616] 0.0        [  -0.3426    0.7257   -0.5966  487.7488]   [-0.2547  0.5395  0.8026  1.    ]   [ 1.  0.  0.  1.]
 [ -18552.1426 -798653.8125   -9119.9502      22.0616] 0.0        [  -0.3426    0.7257   -0.5966  487.7488]   [-0.2547  0.5395  0.8026  1.    ]   [ 1.  0.  0.  1.]
 [ -18334.7812 -799114.25     -8299.5605      17.2575]] 965.551   [  -0.2251    0.4768   -0.8497  487.7488]]  [ 0.7983 -0.4097 -0.4414  1.    ]]  [ 1.  1.  1.  1.]]
t_post[1077]                                          t_dirw[1077]                              t_polw[1077]                      t_ccol[1077]

Examine the big hitters

Despite the crash when attempt to animate, can use other modes with –max-slots=100 which eliminates trunctions.:

g4daeview.sh  --load 1 --wipepropagate --debugkernel --debugphoton 0 --max-slots 100

Interface is slow however. The photon disappearance technique does not prevent the processing of the invisibles.

Observations

  1. big hitters are mostly chroma level truncated
::
h = plt.hist(z.slots[z.slots>50], bins=np.linspace(50,100,51)) udp.py –sid 2190 –style spagetti,confetti,noodles

Long lived

daephotonsanalyzer.sh --load 1

In [2]: z.tl.min(), z.tl.max()
Out[2]: (0.0, 254.15379)

In [4]: np.where(z.tl > 200)
Out[4]:
(array([ 181,  415,  809, 1127, 1230, 1244, 1305, 1784, 1798, 1852, 1856,
       2365, 2439, 2547, 2593, 2799, 2926, 2938, 3024, 3077, 3158, 3655,
       3659, 3831]),)
udp.py --sid 181 --style spagetti,confetti,noodles