fix: add safeDivide to cylinderSurface pathCorrection #3254
Conversation
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #3254 +/- ##
=======================================
Coverage 47.65% 47.65%
=======================================
Files 507 507
Lines 29205 29212 +7
Branches 14010 14013 +3
=======================================
+ Hits 13917 13922 +5
- Misses 5264 5265 +1
- Partials 10024 10025 +1 ☔ View full report in Codecov by Sentry. |
📊: Physics performance monitoring for 8fa208cphysmon summary
|
|
I believe I saw similar problems in the past and they seemed rather related to navigation issues since a 90° incline should be almost impossible. But apart from potential navigation problems - which should be ruled out first - we could have a bound on the path correction either where we apply it or directly in the surface function because the surface will never be infinitely thick. |
|
I guess constraining |
|
Yes, that is what safeDivide does 😊 |
|
Indirectly, yes. But it is hiding the cut and the threshold is not tunable which is not applicable in all situations. |
|
Closed in favour of a possible better solution and tracked here: |
Have *almost* same settings for GX2F and KF makes it easier to understand in the physmon output and check if our results are good. ## Why *almost*? Had also 3 FPEs which fixed here, but did not choose this approach: - #3254 This issue only occurred for `2.5 < |eta| < 3.0`. To get better physmon until the problem is fixed (maybe in a later stage of the GX2F-development), the eta-range is reduced. The full eta-range could be used after solving: - #3267
…ct#3248) Have *almost* same settings for GX2F and KF makes it easier to understand in the physmon output and check if our results are good. ## Why *almost*? Had also 3 FPEs which fixed here, but did not choose this approach: - acts-project#3254 This issue only occurred for `2.5 < |eta| < 3.0`. To get better physmon until the problem is fixed (maybe in a later stage of the GX2F-development), the eta-range is reduced. The full eta-range could be used after solving: - acts-project#3267
I encountered some FPEs in during the navigation. This would avoid them and not crash the programme by giving a good estimate, in case we end up with 2 perpendicular vectors in
double cosAlpha = normalT.dot(direction);FPE
In gx2f physmon I got the following output:
One case had the following values before:
position:
164.63
-48.368
26.7465
normalT:
0.959448
-0.281884
0
direction:
-0.0299944
-0.102092
0.994323
cosAlpha: 0
blocked by: