💥DO NOT MERGE💥 ❗ add new TPC alignment, hit errors and slewing corrections for 2024 #702

fisyak · 2024-08-08T14:30:20Z

new TPC Alignment 24
new TPC hit errors and slewing correction
The detailed description can be found at
https://drupal.star.bnl.gov/STAR/system/files/TPC24c.Export.Mini_.pdf

StRoot/StDbUtilities/StTpcCoordinate.cxx

plexoos · 2024-08-21T19:07:41Z

StRoot/StDbUtilities/StTpcCoordinate.cxx

@@ -21,6 +22,8 @@ ostream& operator<<(ostream& os, const StTpcCoordinate& a) {
  return os << OS;
 }
 //________________________________________________________________________________
+void StTpcCoordinate::Print(Option_t *option) const {cout << *this << endl;}


Does the use of cout and endl prompt the inclusion of Stiostream.h? If so, just add using std::cout; using std::endl; somewhere in this file

plexoos · 2024-08-22T13:45:02Z

StRoot/StBFChain/BigFullChain.h

@@ -213,9 +213,9 @@ Bfc_st BFC[] = { // standard chains
   ,                                                                                       "","","",kFALSE},
  {"RC.pp.y2005"     ,"","","pp2005a,tofdat,OSpaceZ2,OGridLeak3D"                         ,"","","",kFALSE},
  {"RC.pp.y2006"     ,"","","pp2006b,OSpaceZ2,OGridLeak3D"                                ,"","","",kFALSE},
-  {"RC.y2007"        ,"","","DbV20080418,B2007g,IAna,KeepSvtHit,hitfilt,VFMinuit3,emcDY2,ftpc,trgd,"


These changes have been merged in #709. Does this imply that StHitFilterMaker was never actually amending the output?

StRoot/Sti/StiHit.cxx

klendathu2k

Several MagFactor tables added and/or modified. It looks like most just implement the ideal magnetic field for a given year. But MagFactor.y2016.C sets a non-ideal value of -1.002. Why this value when everything else is an ideal +/-1 ?
StEvent/StTpcHit.h / cxx extends the data stored and should result in a ID passed to the ClassDef macro. Unless there is a good reason to, we should keep these IDs consecutive, so the next value is 11 not 13 as per the PR.

There is a helper function defined in the proposed StTpcHit to return a format string for building a ROOT/TGeo of one of our geometry models of the TPC. This should go into a helper function somewhere other than StEvent.

MDF4 Hit Error Model

This should be enabled / disabled by a BFC option, not a compile time flag.

This should use the standard Sti interface... you should not need to modify StiDetector to apply this. A wrapper class which inherits from StiHitErrorCalculator can forward the calculateError call to your StiHitErrorMDF4 class. (calculateError will need to be modified to accept an StiHit).

With that change, I think that the changes to StiTrackNodeHelper lines 1117 to 1136 can be moved into the wrapper class's calculateError function.

The easiest way to do this is with a wrapper function which inherits from StiHitErrorCalculator (or maybe StiTpcHitErrorCalculator). The wrapper class will forward the calculateError call to the MDF4 model.

mgr/ConsDefs.pm

StarDb/tpc/tsspars/tsspar.year_1a.C

klendathu2k · 2024-08-24T17:29:48Z

StarDb/RunLog/onl/starClockOnl.C

@@ -7,7 +9,7 @@ TDataSet *CreateTable() {
  memset(&row,0,tableSet->GetRowSize());
  row.runNumber	 =    2214033; // run number  ;
  row.time	 =  996787667; // unix time of entry  ;
-  row.frequency	 =    9383160; // frequency in Hz  ;
+  row.frequency	 =    9399962; // local clock


Confused by this change. There are starClockOnl.y2001.C ... starClockOnl.y2014.C as well, which I though take precedence over this file. The frequency is set to 9399970 there.

Or is this change in effect for all years that don't have a corresponding y20xx file? If so, why is it appropriate to change those past runs?

It would be good to understand this. @fisyak, can you comment on this question?

We start run with cosmics (local clock) and sometime data base is not populated. This allows to use proper clock for laser runs at the very begin on run.

klendathu2k · 2024-08-24T17:35:27Z

StarDb/RunLog/MagFactor.y2016.C

+St_MagFactor *tableSet = new St_MagFactor("MagFactor",1);
+//
+memset(&row,0,tableSet->GetRowSize());
+    row.ScaleFactor	 =        -1.002; // ;


Why are we changing the magnetic field in 2016?

@fisyak, can you answer this question? The proposed changes should only affect 2019+ productions.

klendathu2k · 2024-08-25T13:56:02Z

StRoot/Sti/StiTrackNodeHelper.cxx

-     mJoinPars.hz() = mTargetHz;
-   }
-
+   mJoinPars.hz()=mTargetHz;//////////////////////???????????????????????????????????????????????????


//////////////////////???????????????????????????????????????????????????

Not very informative comment here. This change can impact past productions.

StRoot/Sti/StiHit.h

StRoot/StiTpc/StiTpcDetectorBuilder.cxx

StRoot/StTpcDb/StTpcDbMaker.cxx

klendathu2k · 2024-08-26T22:57:41Z

StRoot/Sti/StiTrackNodeHelper.cxx

@@ -65,6 +72,10 @@ void StiTrackNodeHelper::set(StiKalmanTrackNode *pNode,StiKalmanTrackNode *sNode
  mParentNode = pNode;
  mTargetNode = sNode;
  mTargetHz = mTargetNode->getHz();
+assert(mTargetHz);
+ if (! sNode->fitPars().hz()) sNode->fitPars().hz() = mTargetHz;


Wondering what condition leads to this... Does this mean that in fixed target we are seeing nodes w/out a magnetic field?

Some times node is created with zero filed. This is check to avoid assert.

StRoot/Sti/StiTrackNodeHelper.cxx

fgeurts · 2024-08-28T03:12:38Z

Hi Yuri, I see several comments were made by STAR software experts. Can you follow up on these as your earliest convenience so that we can get the merger and new library set up by early next week? Thanks!!

Merge on 08/31/2024

fgeurts · 2024-09-05T13:58:55Z

Hi Yuri, I see several comments were made by STAR software experts. Can you follow up on these as your earliest convenience so that we can get the merger and new library set up by early next week? Thanks!!

Thanks, @fisyak. Aside from some programming related comments there were a few more specific comments regarding functionality changes that may be beyond the 2019+ alignment. It would be good to understand the motivation.

mgr/ConsDefs.pm

genevb · 2024-09-11T19:30:06Z

I ran the nightly test suite on SDCC using this TFG24c.Export.Mini branch. The first finding is that all the *_embed jobs failed with the following assertion:

StChain:WARN  - StDbSql::QueryDb(table,time) line=439 tpcTimeBucketCor  has No data for query
root4star: .sl73_gcc485/obj/StRoot/StDetectorDbMaker/StDetectorDbChairs.cxx:530: static St_tpcTimeBucketCorC* St_tpcTimeBucketCorC::instance(): Assertion `table' failed.

An example such log file is here:

/star/rcf/test/gitdev/daq_sl302.ittf/Wed/year_2012/UU193_embed/st_physics_adc_13115004_raw_2510001.log

The 64-bit jobs also all failed, unable to open libCore.so.5. That may have been something I missed, so I may try it again for tomorrow if I spot my mistake.

Other jobs generally ran (a few are still running), and people can compare the output between yesterday's dev jobs at:

/star/rcf/test/dev/*/Tue/

and today's jobs with this PR's code branch:

/star/rcf/test/gitdev/*/Wed/

-Gene

fisyak · 2024-09-11T20:57:18Z

To use nightly test is very good idea. To make track by track comparison I need event.root files in both dev and girder.

fisyak · 2024-09-11T21:06:54Z

Event root files are needed with hits, i.e. without "hitfilt".

genevb · 2024-09-12T01:14:53Z

Event root files are needed with hits, i.e. without "hitfilt".

Many such event.root files are available from the nightly test suite I ran...at least 48 of them. See the outputs of:

ls -l /star/rcf/test/gitdev/*/Wed/year*/*/*.event.root | grep "Sep 11"
ls -l /star/rcf/test/dev/*/Tue/year*/*/*.event.root | grep "Sep 1"

...and pretty much all the chains have -hitfilt, so they are generally without hitfilt.

fisyak · 2024-09-17T16:11:24Z

I have posted comparison of dev and SL24y productions to answer the question on the effect of code modification on production of old (before iTPC) data with old chains.
https://drupal.star.bnl.gov/STAR/blog/fisyak/Track-track-and-hit-hit-analyses-dev-and-SL24y-Alignment2024-AuAu2002016-sample

I don't see any show stoppers to start the calibration production with SL24l for all fixed target samples from 2019-2021 with new alignment (CorrZ).

genevb · 2024-10-04T15:35:34Z

StRoot/Sti/StiTrackNodeHelper.cxx

+    if (tpcHit) {
+      if ((tpcHit->detector() == kTpcId || tpcHit->detector() == kiTpcId)) {
+	if (tpcHit->flag() == 2) {
+	  fudgeFactor = 16.;
+	}
+      }
+    }


Yuri suggested that these lines of code in Sti/StiTrackNodeHelper.cxx, modifying some TPC hit errors, might be responsible for the differences seen in nightly tests with this PR vs. DEV for pre-iTPC datasets where the alignment should be unaffected. Processing a few events with and without this bit of code is quite easy, and comparing the log files obviates that the bulk of the difference in track counts does disappear when removing this section of code.

However....from looking at the log file, I see that the new alignment scheme appears to alter TPC distortion corrections for these old, pre-iTPC datasets without any change to the reconstruction chain. This is in addition to the modification to TPC distortion corrections ("OSectorAlign") that Yuri wants to impose with the "CorrZ" chain option he has proposed for iTPC-era datasets for which I have expressed an interest in seeing a justification (through data, not anecdotally).

< StMagUtilities::XTWIST = -0.38 mrad < StMagUtilities::YTWIST = 0.25 mrad --- > StMagUtilities::XTWIST = 0 mrad > StMagUtilities::YTWIST = 0 mrad < StMagUtilities::WestClock = -0.43 mrad --- > StMagUtilities::WestClock = 0 mrad

I do not recall this being discussed before, neither in the context of pre-iTPC datasets, nor with the new alignment. If the TPC alignment calibration was truly performed without any correction for the distortions due to the misalignment of the E and B fields (represented by the XTWIST & YTWIST numbers), I have significant concerns about this TPC alignment. Hopefully I am missing something and this is not the case.

I have placed output of the nightly test without this particular code (but including the rest of this PR) at this location:

This PR minus hit error change :
/star/rcf/test/gitdev/daq_sl302.stica/Mon/year_2016/AuAu200_production_2016_64bit/

This is for comparison with other iterations:

DEV :
/star/rcf/test/dev/daq_sl302.stica/Sat/year_2016/AuAu200_production_2016_64bit/
This PR :
/star/rcf/test/gitdev/daq_sl302.stica/Tue/year_2016/AuAu200_production_2016_64bit/
This PR before the Sept. 11th commits :
/star/rcf/test/gitdev/daq_sl302.stica/Wed/year_2016/AuAu200_production_2016_64bit/

I am making note of the last two in particular because Yuri's study involved the output from the last one (those jobs were started on September 11th before the time of Yuri's last commits), even though data with those commits was available. (Side note, please pay no heed to the "Tue" or "Mon" in the gitdev paths, as I moved files there from other days to avoid them being over-written.) I will note that I didn't see any effective impact of those commits on this particular test's track counts (i.e. the last two outputs may be identical).

https://www.star.bnl.gov/~fisyak/star/TbyT/AuAu200_2016_HitError/

Yuri has corrected me:
The "Twist" and "Clock" distortion corrections have not actually been used for the past decade, beginning with the prior "New Tpc Alignment" that was introduced around that time. StMagUtilities prints these numbers, but there has been no distortion correction applied. The alignment work introduced in this PR has resulted in the printed values being zero, but no actual effect on the applied distortion corrections.

fisyak added 21 commits July 29, 2024 10:19

Add defalt values

f24b000

Clean up

b9352bd

Add switch off color for batch mode

9aae93c

Alignment2024

d152e4f

TpcRS hit errors base on MDF4 parameterization

24316ac

new Tpc hit errors

921b24e

make them visible in standard library

1fe7ee9

back to bfore HFT era

282c28f

Clean up

5b2ecc5

set ideal

2be2cf8

Warn off

d04a560

new TPc hit error parameterization

701026a

new errors and shifts from Alignment2024, supress asserts

ae31ad3

new Tpc Hit errors, supress asserts

550ae69

new errors

5c66e08

add color

10bda9e

Add Alignment2024 and TFG flavor for it

ea24401

Add Alignment2024, remove obsolete hitfit from chain options

8828f60

Add default handler for logger

b6046fe

Add ideal parameters

12fa9ff

reduce no. of committed files

8496cf4

fisyak requested review from genevb, klendathu2k, R-Witt, iraklic, plexoos, dmarkh, perevbnlgov, starsdong and jdbrice as code owners August 8, 2024 14:30

plexoos reviewed Aug 22, 2024

View reviewed changes

StRoot/Sti/StiHit.cxx Outdated Show resolved Hide resolved

StRoot/Sti/StiHit.cxx Show resolved Hide resolved

fisyak added 2 commits August 23, 2024 15:00

comment out ibreak count to catch assert case

15106c7

Merge branch 'main' into TFG24c.Export.Mini

1d9e062

klendathu2k requested changes Aug 27, 2024

View reviewed changes

fisyak added 6 commits August 31, 2024 15:51

Merge branch 'main' into TFG24c.Export.Mini

2f52e6c

Merge on 08/31/2024

fix bug with padpitch, add const

cac6321

add const

69abeeb

add correction to fixedTarget runs, update calculation of cms energy

46576ea

remove ifdef StiTpcHitErrorMDF4_h

bbaaf85

Fix typo

66b02fe

genevb reviewed Sep 9, 2024

View reviewed changes

mgr/ConsDefs.pm Show resolved Hide resolved

Add space before -DEVTGEN_EXTERNAL=1

709d840

fisyak added 2 commits September 11, 2024 16:39

use St_tpcTimeBucketCor only for new schema

987442b

Switch: ! NoTFGLDbTag => TFGDbTag

75eaec9

Merge branch 'main' into TFG24c.Export.Mini

36ca1de

genevb reviewed Oct 4, 2024

View reviewed changes

fisyak and others added 6 commits October 5, 2024 08:39

Merge branch 'main' into TFG24c.Export.Mini

d919c1d

Merge branch 'main' into TFG24c.Export.Mini

77be6db

Merge branch 'main' into TFG24c.Export.Mini

8acdf8e

Merge branch 'main' into TFG24c.Export.Mini

dc447f4

Merge branch 'main' into TFG24c.Export.Mini

8bc3333

Merge branch 'main' into TFG24c.Export.Mini

8a19542

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

💥DO NOT MERGE💥 ❗ add new TPC alignment, hit errors and slewing corrections for 2024 #702

💥DO NOT MERGE💥 ❗ add new TPC alignment, hit errors and slewing corrections for 2024 #702

fisyak commented Aug 8, 2024

plexoos Aug 21, 2024

fisyak Sep 5, 2024

plexoos Aug 22, 2024

klendathu2k left a comment

klendathu2k Aug 24, 2024

fgeurts Sep 5, 2024 •

edited

Loading

fisyak Sep 5, 2024

klendathu2k Aug 24, 2024

fgeurts Sep 5, 2024

klendathu2k Aug 25, 2024

fisyak Sep 5, 2024

klendathu2k Aug 26, 2024

fisyak Sep 5, 2024

fgeurts commented Aug 28, 2024

fgeurts commented Sep 5, 2024

genevb commented Sep 11, 2024

fisyak commented Sep 11, 2024

fisyak commented Sep 11, 2024

genevb commented Sep 12, 2024

fisyak commented Sep 17, 2024

genevb Oct 4, 2024

fisyak Oct 6, 2024

genevb Oct 9, 2024

💥DO NOT MERGE💥 ❗ add new TPC alignment, hit errors and slewing corrections for 2024 #702

Are you sure you want to change the base?

💥DO NOT MERGE💥 ❗ add new TPC alignment, hit errors and slewing corrections for 2024 #702

Conversation

fisyak commented Aug 8, 2024

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

klendathu2k left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

fgeurts Sep 5, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

fgeurts commented Aug 28, 2024

fgeurts commented Sep 5, 2024

genevb commented Sep 11, 2024

fisyak commented Sep 11, 2024

fisyak commented Sep 11, 2024

genevb commented Sep 12, 2024

fisyak commented Sep 17, 2024

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

fgeurts Sep 5, 2024 •

edited

Loading