[DD4hep] Minor size correction of 0.1 to 7 degrees in phi for tube shapes that are non-sensitive TIB pillars and ribs #34088
There are no files selected for viewing
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -11,7 +11,7 @@ | |
| <!-- support ring solids - refer to outer disk comments --> | ||
| <Algorithm name="track:DDCutTubsFromPoints"> | ||
| <rParent name="pixfwdDisks:PixelForwardDiskZminus"/> | ||
| <String name="SolidName" value="PixelForwardInnerDiskOuterRingZminus"/> | ||
|
||
| <String name="Material" value="pixfwdMaterials:C_C_InnerOuterRing"/> | ||
| <Numeric name="zPos" value="[OuterRingZ]"/> | ||
| <Numeric name="rMin" value="[pixfwdSupportRingParameters:InnerDiskOuterRingRMin]"/> | ||
|
|
@@ -22,7 +22,7 @@ | |
| </Algorithm> | ||
| <Algorithm name="track:DDCutTubsFromPoints"> | ||
| <rParent name="pixfwdDisks:PixelForwardDiskZminus"/> | ||
| <String name="SolidName" value="PixelForwardInnerDiskCFOuterRingZminus"/> | ||
| <String name="Material" value="pixfwdMaterials:FPix_CFSkin_InnerOuterRing"/> | ||
| <Numeric name="zPos" value="[OuterRingZ]"/> | ||
| <Numeric name="rMin" value="[pixfwdSupportRingParameters:InnerDiskOuterRingCFRMin]"/> | ||
|
|
@@ -33,7 +33,7 @@ | |
| </Algorithm> | ||
| <Algorithm name="track:DDCutTubsFromPoints"> | ||
| <rParent name="pixfwdDisks:PixelForwardDiskZminus"/> | ||
| <String name="SolidName" value="PixelForwardInnerDiskInnerRingZminus"/> | ||
| <String name="Material" value="pixfwdMaterials:C_C_InnerInnerRing"/> | ||
| <Numeric name="zPos" value="[InnerRingZ]"/> | ||
| <Numeric name="rMin" value="[pixfwdSupportRingParameters:InnerDiskInnerRingRMin]"/> | ||
|
|
@@ -44,7 +44,7 @@ | |
| </Algorithm> | ||
| <Algorithm name="track:DDCutTubsFromPoints"> | ||
| <rParent name="pixfwdDisks:PixelForwardDiskZminus"/> | ||
| <String name="SolidName" value="PixelForwardInnerDiskCFInnerRingZminus"/> | ||
| <String name="Material" value="pixfwdMaterials:FPix_CFSkin_InnerInnerRing"/> | ||
| <Numeric name="zPos" value="[InnerRingZ]"/> | ||
| <Numeric name="rMin" value="[pixfwdSupportRingParameters:InnerDiskInnerRingCFRMin]"/> | ||
|
|
||
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -409,6 +409,7 @@ static long algorithm(Detector& /* description */, cms::DDParsingContext& ctxt, | |
| dy = 0.5 * pitchThick; | ||
| dz = 0.5 * pitchHeight; | ||
| solid = Box(dx, dy, dz); | ||
| name += "Box"; | ||
|
||
| ns.addSolidNS(name, solid); | ||
| edm::LogVerbatim("TECGeom") << "Solid:\t" << name << " " << solid.name() << " Box made of " << pitchMat | ||
| << " of dimensions " << dx << ", " << dy << ", " << dz; | ||
|
|
@@ -419,6 +420,7 @@ static long algorithm(Detector& /* description */, cms::DDParsingContext& ctxt, | |
| bl2 = 0.5 * pitchHeight - 0.5 * dz * sin(detTilt); | ||
| thet = atan((bl1 - bl2) / (2. * dz)); | ||
| solid = Trap(dz, thet, 0, h1, bl1, bl1, 0, h1, bl2, bl2, 0); | ||
| name += "Trap"; | ||
| ns.addSolidNS(name, solid); | ||
| edm::LogVerbatim("TECGeom") << "Solid:\t" << name << " " << solid.name() << " Trap made of " << pitchMat | ||
| << " of dimensions " << dz << ", " << convertRadToDeg(thet) << ", 0, " << h1 << ", " | ||
|
|
||
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -252,7 +252,7 @@ static long algorithm(Detector& /* description */, cms::DDParsingContext& contex | |
| double dz = 0.5 * layerL - 2. * fillerDz; | ||
| double _rmi = std::min(rin + 0.5 * dd4hep::mm, rout - 0.5 * dd4hep::mm); | ||
| double _rma = std::max(rin + 0.5 * dd4hep::mm, rout - 0.5 * dd4hep::mm); | ||
| solid = ns.addSolidNS(name, Tube(_rmi, _rma, dz, -0.5 * width, 0.5 * width)); | ||
|
Contributor
There was a problem hiding this comment. @cvuosalo - it is strange that this did not come out in validation. Could you, please, share a snapshot of the geometry fragment before and after? Thanks.
Contributor
Author
There was a problem hiding this comment. Yes, I will reproduce the bug and post the before and after. It is a very small difference, so that is maybe why it wasn't detected earlier.
Contributor
There was a problem hiding this comment. @cvuosalo, I suspect, that old variant of the Tube was correct. The signature of the Tube:
Contributor
Author
There was a problem hiding this comment. We can compare old DD with DD4hep for these tubes. Here is old DD: Here is DD4hep WITHOUT the fix: The problem comes from the change in the Tube constructor between old DD and DD4hep. In old DD, the last parameter is and the corresponding line of code is: but with DD4hep the last parameter is so the last the last parameter with DD4hep needs to be It is unfortunate that this bug was not found until now. I think it affects about 20 solids. I would think we would want to fix it now, rather than leaving it for all of Run 3.
Contributor
There was a problem hiding this comment. @cvuosalo , thank you for the explanation but in my view what DD4Hep did is a scandal - they change signature in a place which is very difficult to debug. We may have thousands similar places.
Contributor
Author
There was a problem hiding this comment. @civanch I did a systematic comparison of all solids between old DD and DD4hep. That's how I discovered the minor discrepancies discussed in this PR. I don't think there are any more. Probably, in most cases for tubes,
Contributor
There was a problem hiding this comment. @cvuosalo - it would be good to see a snapshot before and after the fix. I'd assume there would be an overlap detected with a double delta phi. Also, it is not obvious what DD4hep is doing when constructing the shape. Sometimes the arguments may differ. |
||
| LogDebug("TIBGeom") << solid.name() << " Tubs made of " << ribMat << " from " << -0.5 * convertRadToDeg(width) | ||
| << " to " << 0.5 * convertRadToDeg(width) << " with Rin " << rin + 0.5 * dd4hep::mm << " Rout " | ||
| << rout - 0.5 * dd4hep::mm << " ZHalf " << dz; | ||
|
|
@@ -490,7 +490,7 @@ static long algorithm(Detector& /* description */, cms::DDParsingContext& contex | |
| break; | ||
| } | ||
|
|
||
| solid = ns.addSolidNS(name, Tube(pillarRin, pillarRout, pillarDz, -pillarDPhi, pillarDPhi)); | ||
| Volume Pillar = ns.addVolumeNS(Volume(name, solid, ns.material(pillarMaterial))); | ||
| LogDebug("TIBGeom") << solid.name() << " Tubs made of " << pillarMaterial << " from " << -pillarDPhi << " to " | ||
| << pillarDPhi << " with Rin " << pillarRin << " Rout " << pillarRout << " ZHalf " << pillarDz; | ||
|
|
||
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@cvuosalo - this is an unnecessary performance bomb. A new object is returned after each +. That means that a new buffer is used each time. You are doing a ton of extra + operations - it is not efficient.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I will remove these rotation names since they are not strictly necessary.