Conversation
siilieva
force-pushed
the
muonDIS_with_pythia6_run_through_G4
branch
from
031eb67 to
fe64122
Compare
It adds to the Geant4 process manager a custom process that is using Pythia6 to generate muonDIS.
to reuse the function globally in sndsw.
Using standalone CMake and don't rely on the FairRoot CMake macros as we usually do.
Better control of severity level and make the log print in colour.
In rare cases, a track is moving right along the border of two volumes. This makes the geo navigation stuck. To unblock it, one now moves along the particle trajectory with a step of 1e-6cm, smaller than any position measurement precision in SND. This way, one allows the navigator to move away from any border at a reasonable computing cost.
When going from original p6 code(Fortran) to ROOT's TPyhia6 some arrays, like the XSEC one, have their indices swapped, also counting starts from 0 in TPythia6 case. With this fix one can read the P6 xsec directly from the code, no need to look into the produced txt file. Another part of the fix is to allow using P6 or P8 when creating the DIS xsec file using existing CL options.
This is essential when running muonDIS with activated emulsions.
siilieva
force-pushed
the
muonDIS_with_pythia6_run_through_G4
branch
from
9244683 to
590ce52
Compare
olantwin
left a comment
olantwin
left a comment
There was a problem hiding this comment.
A first batch of comments. Impressive work!
| return kSUCCESS; | ||
| } | ||
|
|
||
| ClassImp(MuonDISProcessInjector) |
| muMinus = G4MuonMinus::Definition(); | ||
| muPlus = G4MuonPlus::Definition(); | ||
|
|
||
| DISProcess = new G4MuonDISProcess(); |
There was a problem hiding this comment.
Does the process-manager free this? Since the process is added to two different ones, probably not?
| char *fNucleon; //! nucleon type | ||
| char *fVolumeName; //! geometry volume name to place the DIS |
There was a problem hiding this comment.
I'd prefer using std::string here (and if necessary accessing the C-string when needed).
| const G4ParticleDefinition *muPlus; | ||
| const G4ParticleDefinition *muMinus; |
There was a problem hiding this comment.
Why are these const? They are updated in Init, I'm surprised this compiles.
| parser.add_argument('--y_range', help='y position range placing muonDIS interaction as two values: start and end', type=float, nargs=2, default=[-1e10,1e10]) | ||
| parser.add_argument('--z_range', help='z position range placing muonDIS interaction as two values: start and end', type=float, nargs=2, default=[-1e10,1e10]) | ||
| parser.add_argument('--volume', help='Name of geometry volume where to simulate the DIS; e.g. volTarget, volMuUpstreamDet, volMuDownstreamDet, volVetoPlane, Wall', default="") | ||
| parser.add_argument('--dis_xsec_file', help='Name of DIS cross section file', default="/eos/experiment/sndlhc/MonteCarlo/Pythia6/MuonDIS/muDIScrossSec.root") |
There was a problem hiding this comment.
Please use the full path here. This will only work on lxplus or on systems where /eos/ is mounted via FUSE.
| if ((fZStart - pos.z()) > 1e-3) { | ||
| // do nothing | ||
| } else { | ||
| string geoVolumePath =gGeoManager->GetPath(); |
There was a problem hiding this comment.
Can we please auto-format all the new files?
| fYEnd = y_range->at(1) * cm; | ||
| fZStart = z_range->at(0) * cm; | ||
| fZEnd = z_range->at(1) * cm; | ||
| LOG(INFO) << "G4MuonDISProcess: Setting xyz ranges[cm] for muon DIS\nx: " |
There was a problem hiding this comment.
You mix iostream and FairLogger output. I'd suggest using one or the other.
The include of FairLogger is hidden via the include of MuonDISGenerator?
| //myPythia->Pylist(1); // list table of properties for all particles | ||
| myPythia->SetMSTU(11, 6); // file number to which all program output is directed | ||
|
|
||
| delete myPythia; |
There was a problem hiding this comment.
Reusing Pythia6 instances is impossible?
| } | ||
|
|
||
| std::tuple<double, double, double> | ||
| G4MuonDISProcess::rotate(double ctheta, double stheta, double cphi, double sphi, double px, double py, double pz) |
There was a problem hiding this comment.
Is this really something we need to reimplement?
| if options.debug: checking4overlaps = True | ||
| if options.debug: | ||
| checking4overlaps = True | ||
| ROOT.fair.Logger.SetConsoleSeverity("debug") |
There was a problem hiding this comment.
What's the difference between this and SetLogScreenLevel? I'd use the same method for both cases. Ideally right next to each other, so that it's easy to understand the logging behaviour. Maybe where you set the log to be colourful.
2 participants
Create a FairTask that injects a custom process into Geant4 (since we run the MC engine through a FairRunSim instance)
In this case the process is muonDIS and is being generated using Pythia6.
The muonDIS procedure is technically a C++ copy of the muon dis generation code from muonDIS.py
Also, the famous MeanMaterianBudget is made public and is widely used in the code. A tiny update was made there allowing the geo navigator to step forward shall a particle be moving along a volume border.
Tuning was done to make sure the proposed simulation chain works correctly and efficiently! See related presentations: 2July2025 and 30July2025 and 10September2025.
On top of the previously available option to set z range for the DIS vertex, the new chain allows to fully simulate events only if DIS vertex is in a specific x-y range AND/OR a volume. This saves a lot of computing resources.
Revision of indices of TPythia6 arrays was done to correct a hiccup with the cross section, but that doesn't affect previous MC chain, just made obtaining this more difficult.
Creating draft to show result on SoftPhys meeting on 27.08.2025 first, then will go public.
ps. candidate for cherry-picking in SHiP
ps. an internal/technical note is to be prepared