|
1
|
|
|
2
|
---
|
|
3
|
2023 Sep 13 (Notes from Mini KaonLT meeting on HeeP offsets, physics_iterate.f and weight iteration)
|
|
4
|
|
|
5
|
Notes by Richard
|
|
6
|
|
|
7
|
Present:
|
|
8
|
Ali
|
|
9
|
Garth
|
|
10
|
Richard
|
|
11
|
Vijay
|
|
12
|
---
|
|
13
|
|
|
14
|
Meeting to discuss HeeP offsets for Garth's analysis (mainly the discrepancy between Richard and Vijay's low energy HeeP results)
|
|
15
|
Also discussing the physics_iterate.f script for cross section functional form iterations
|
|
16
|
|
|
17
|
- HeeP Offsets
|
|
18
|
- Ali will check with his replays/simc
|
|
19
|
- If Ali and Richard match...
|
|
20
|
- We will need to compare input files for simc and hopefully it is easy to find differences with Vijay
|
|
21
|
- If that doesn't work we will then need to match our simc versions to make sure they match and address any differences
|
|
22
|
- Once SIMC is matching, we should also compare replay param files, etc. to see where these differences arise
|
|
23
|
- Ali and Richard: Check arc measurement is what is written in standard.kinematics for lower energies
|
|
24
|
- Richard: shorten range of fits so they are just the peak
|
|
25
|
- Richard: Increase sig figs to 5 for plots to better distinguish between similar mean values
|
|
26
|
- Richard: Rerun SIMC without recon_hcana script to see if they match Vijays
|
|
27
|
- The recon_hcana script changes calculation of emiss, W, and Pmx,y,z so we should see a change
|
|
28
|
- Richard: Double check that there are no offsets of SIMC input files
|
|
29
|
- Richard: Write-up email that compares recon_hcana and hcana calculations for emiss, pmx,y,z, W
|
|
30
|
- lt_analysis
|
|
31
|
- Avg Q2, W, t per t-bin in data and SIMC. Compare to make sure they look similar, but only data goes through average_kinematic.f
|
|
32
|
- The ratio of data/simc is used with the data averages per bin for xsects not SIMC. SIMC is only needed for the yields of the ratio
|
|
33
|
- physics_iterate.f
|
|
34
|
- No Fermi momentum! Remove all references to this
|
|
35
|
- Virtual photon flux factor, should be consistent with elastic (based off data)
|
|
36
|
- Code lines: 362-97, these read in the initial set of parameters. This assures you never need to run simc again. Separate code to redo this with new parameters (see recalculate weights below).
|
|
37
|
- This code does not need to be re-run!
|
|
38
|
- For kaon, change routine at top to peek and replace physics_kaon with this code.
|
|
39
|
- For kaon, replace all mass of pion with kaon.
|
|
40
|
- For kaon, in Makefile, replace physics_kaon with physics_iterate
|
|
41
|
- For kaon, no need to remove physics_kaon in event.f
|
|
42
|
- Create new branch for simc with physics_iterate instead of physics_kaon
|
|
43
|
- The returned variable of physics_iterate needs to be called in event.f
|
|
44
|
- This return value is the cross section that is the basis of the weight
|
|
45
|
- event.f
|
|
46
|
- main%sigcc = peek(vertex, main) is the read in for the physics_iterate return
|
|
47
|
- Line 1915-16: This is where the weight is calculated
|
|
48
|
- Recalculate weights
|
|
49
|
- Bill has C++ code based off FORTRAN code
|
|
50
|
- Vijay has the code (knows the location), which he can share
|
|
51
|
- Iterate each Q2 separately, with their own parameters
|
|
52
|
- Each iteration, simc needs new weight applied, then rebinned and then recalculate the yields.
|
|
53
|
- The new parameters for the weights are called in xsects via the xmodel
|
|
54
|
- xmodel should match the functional form of physics_iterate, initially (may change for kaon later if things dont converge). xmodel does not need to be adjusted each iteration, only the parameters that are read in.
|
|
55
|
- For kaon, add mass of lambda in xmodel. In general, check all masses of FORTRAN scripts
|