11/29/10

The following is output to the screen when I try to run the calibration code for sector 1 and plane U in iteration 3:

[rcas6017] ~/calibrations/> root4star -b -q macros/doSmdGains.C\(1,3,0\) > & log 

Abort

[rcas6017] ~/calibrations/> more log

  *******************************************

  *                                         *

  *        W E L C O M E  to  R O O T       *

  *                                         *

  *   Version   5.22/00  17 December 2008   *

  *                                         *

  *  You are welcome to visit our Web site  *

  *          http://root.cern.ch            *

  *                                         *

  *******************************************

 

ROOT 5.22/00 (trunk@26997, May 28 2010, 11:40:00 on linux)

 

CINT/ROOT C/C++ Interpreter version 5.16.29, Jan 08, 2008

Type ? for help. Commands must be C++ statements.

Enclose multiple statements between { }.

*** Float Point Exception is OFF ***

 *** Start at Date : Mon Nov 29 16:39:37 2010

QAInfo:You are using STAR_LEVEL : dev, ROOT_LEVEL : 5.22.00 and node : rcas6017.

rcf.bnl.gov 

 

Processing macros/doSmdGains.C(1,3,0)...

strangeArrayNames-arrayNames-__NARRAYS__ = 0

 loading of shared libraries done

Info in <TCanvas::Print>: ps file mpvN01U.ps has been created

loaded  libraries

Here are the gains:

cuts for 01U : adcMin=40 ,adcMax=140 minSum=50 maxRelEr=0.500000

doGainCorr() for mpvN01U, average over 12-strips pl=1

avr01U001 MPV=0.82 +/- 0.16 (19.0%) 

avr01U013 MPV=0.90 +/- 0.04 (3.9%) 

avr01U025 MPV=0.89 +/- 0.04 (4.1%) 

avr01U037 MPV=0.95 +/- 0.03 (3.1%) 

avr01U049 MPV=0.92 +/- 0.03 (3.1%) 

avr01U061 MPV=0.90 +/- 0.03 (3.5%) 

avr01U073 MPV=0.91 +/- 0.03 (2.8%) 

avr01U085 MPV=0.93 +/- 0.03 (3.2%) 

avr01U097 MPV=0.92 +/- 0.03 (3.3%) 

avr01U109 MPV=0.98 +/- 0.03 (3.2%) 

avr01U121 MPV=0.99 +/- 0.04 (4.4%) 

avr01U133 MPV=1.00 +/- 0.03 (2.7%) 

avr01U145 MPV=1.05 +/- 0.04 (3.3%) 

avr01U157 MPV=1.06 +/- 0.04 (3.9%) 

avr01U169 MPV=1.01 +/- 0.03 (3.0%) 

avr01U181 MPV=1.06 +/- 0.05 (4.8%) 

avr01U193 MPV=1.01 +/- 0.04 (3.5%) 

avr01U205 MPV=1.06 +/- 0.04 (4.1%) 

avr01U217 MPV=1.10 +/- 0.06 (5.5%) 

avr01U229 MPV=1.14 +/- 0.10 (8.4%) 

avr01U241 MPV=1.14 +/- 0.09 (7.5%) 

avr01U253 MPV=0.93 +/- 0.17 (18.3%) 

avr01U265 MPV=1.78 +/- 0.17 (9.5%) 

Inside plTGraph, flag 0, shpFunc=pol0

flag 1, ig=1

flag 2, pl=1

plot mpvN01U

flag 3

flag 4

flag 5

flag 6

flag 7

Error in <StoreAndDrawFitFunction>: Empty funciton list- cannot store fitted function

 

****************************************

Minimizer is Linear

Chi2                      = 84.0399

NDf                       = 22

Par_0                     = 0.967036 +/- 0.00782767

flag 8

flag 9

flag 10

root4star: .sl53_gcc432/obj/StRoot/StEEmcPool/StEzSmdCal/SmdGains.cxx:241: void 

SmdGains::plTGraph(const Char_t*, int, int): Assertion `gnCorFn' failed.

 

11/28/10

First thing, after logging into the BNL computers, I ensure that I'm using the STAR development version of the code:

    stardev

The EEMC calibration code is In my directory:

    /star/u/jschaub/calibrations

and its sub-directories.  This is the version of the code that Alice checked into the CVS repository.  (See her instructions for how to get the code from CVS and set up the run 9 calibration run list.)  She set up the code so that it builds a directory structure:

    /star/u/jschaub/calibrations/EEMC_cal/iter-#/

where # is the iteration number.

The code is set up so that iter=0 is the initial SMD iteration, in which we determine initial SMD gains by fitting dE/dx slopes.  Iteration 1 uses these gains determined in iteration 0 to generate gains for the pre- and postshower layers, then iteration 2 uses gains from iterations 0 and 1 to generate gains for the towers.  This all works fine.

Iteration 3 should use the gains from previous iterations to generate new SMD gains.  (Then iterations 4 and 5 will re-generate gains for pre-/postshowers and towers.)  I run the code from the calibrations directory I mentioned above using a shell that Alice wrote:

    macros/fitSlopes.sh

which calls doSmdGains.C (also in macros/), which calls SmdGains.cxx in

    /star/u/jschaub/calibrations/StRoot/StEEmcPool/StEzSmdCal/

I put a bunch of print statements into doSmdGains.C (which I've now deleted) and SmdGains.cxx (which are all still there) to determine that the error comes from the line:

    gr->Fit(shpFunc);

in the function (Is "Function" the right word for C++?) plTGraph in SmdGains.cxx.

I looked up the error message on google--I thought it might be distinctive because of the word "funciton"--and found the webpage I mentioned below:

    https://savannah.cern.ch/bugs/?45607

I tried to use the non-default constructor TGraphErrors(Int_t n) in the variable declaration but still received the same error message as before.

I tried to find other code that fits a TGraphErrors.  My calibrations directory contains plTwCal.C, which uses a Fit call for a TGraphErrors, and this program seems to work fine.  I didn't notice any differences in the setup of the TGraphErrors or the Fit call between plTwCal.C and SmdGains.cxx, so I'm confused why one works and the other doesn't.

I suspect the fix for this bug would probably be fast and simple for someone more familiar with STAR's code.