Difference between revisions of "GlueX Offline Meeting, July 25, 2012"

From GlueXWiki
Jump to: navigation, search
(notes as minutes)
m (Text replacement - "www/halldweb1/" to "www/halldweb/")
 
(3 intermediate revisions by the same user not shown)
Line 31: Line 31:
 
==Slides==
 
==Slides==
  
−
Talks can be deposited in the directory <code>/group/halld/www/halldweb1/html/talks/2012-3Q</code> on the JLab CUE. This directory is accessible from the web at https://halldweb1.jlab.org/talks/2012-3Q/ .
+
Talks can be deposited in the directory <code>/group/halld/www/halldweb/html/talks/2012-3Q</code> on the JLab CUE. This directory is accessible from the web at https://halldweb.jlab.org/talks/2012-3Q/ .
  
 
=Minutes=
 
=Minutes=
  
−
New Note 59
+
Present:
 +
* '''CMU''': Will Levine, Curtis Meyer
 +
* '''IU''': Ryan Mitchell, Matt Shepherd
 +
* '''JLab''': Mark Ito (chair), Manuel Lara, David Lawrence, Paul Mattione, Elton Smith, Simon Taylor, Elliott Wolin, Beni Zihlmann
  
−
Offline software meeting notes, 7/25
+
==Announcements==
  
−
elton, dude, simon, mark, elliott, paul, david
+
* Data Challenge Meetings have started on alternate Mondays at 1:30 pm.
−
ryan, matt
+
* Mark urged us to participate in the [[Private:Scientific Computing Survey, 2012|Scientific Computing Survey]].
−
curtis, will
+
* There will be no more (default) software sessions as part of the bi-weekly Tracking Meeting.
 +
* David and Elliott have started work on translation tables. If you have thoughts send them along. A proposed format was circulated on the Offline email list.
  
−
announcements
+
==Review of Minutes from Previous Meetings==
  
−
david and elliott translation tables, serious thoughts send them along, proposed format circulated
+
===Offline Meeting===
  
−
how to turn on output
+
We reviewed the [[GlueX Offline Meeting, July 11, 2012#Minutes|minutes from the July 11th meeting]].
−
no more work known to be done
+
  
−
simon positive charged tracks: pi k p
+
* Paul reported that he thinks all work has been done for now on the REST format. We can start using it in production.
−
negative k- pi-
+
* Simon made the promised changes in the default track hypotheses generated:
−
by default
+
** Positive tracks: pi, K, p
 +
** Negative tracks: pi, K
  
−
calorimeters
+
===Data Challenge Meeting===
−
• david, putting in realism in bcal smearing
+
−
∘ full electronic pulse shape, with dark hits
+
−
∘ hdgeant and mcsmear, disabled by preprocessor directives. it as the default, unsmeared file much larger
+
−
∘ mcsmear 4 times slower for single photons
+
−
∘ times of execution are comparable, hgdgeant vs mcsmear, 3 times longer for hdgeant
+
−
∘ pythia events, have been tried should be measured
+
−
∘ presentation at future meeting
+
  
−
tracking, Simon
+
We looked at the [[GlueX Data Challenge Meeting, July 16, 2012#Minutes|minutes of the July 16th meeting]] without significant comment.
−
• rate of execution, one additional swim over references eliminated,
+
−
• straight line fitting for alignment of the chambers
+
−
• algorithm to adjust parameters of each wire plane
+
−
• allow for rotation and shift of entire plane
+
−
• reference article, using silicon strips
+
−
• convergence problems, put in annealing procedure, scale errors, tighten them up
+
−
• Eltons suggestion, omega events for 1200 vs 1500 A in solenoid, looks like drop in performance due to reduction in background
+
−
• started lamda, p pi- events, no conclusions yet
+
−
• 1200 - 1500 20% loss in delta p/p, missing mass only worse by a few percent
+
  
−
Paul, topics
+
==Reconstruction Sub-Group Reports==
−
• middle of writing a kinematic fitters, 4-momentum conservation, time constraints, rf timing, vertext constraint
+
−
• hit in bcal, propagate time back to vertex, uncertainty in the time, with energy loss,
+
−
• 7 variables, p3, t, x, y, z, in covariance matrix
+
−
• hope to have results in two weeks
+
  
−
Comment
+
===Calorimeters===
  
−
• ubuntu 64 reconstruction  does not work
+
David has been working on putting more realism in the BCAL smearing.
−
•
+
* Full electronic pulse shape are generated, with dark hits.
−
•
+
* Changes have been made to hdgeant and mcsmear. They are currently disabled by pre-processor directives.
 +
* The unsmeared file is much larger; mcsmear is four times slower for single photon events
 +
* hdgeant is about three times slower than mcsmear previously.
 +
* David will give a presentation on this work at a future meeting.
 +
 
 +
===Tracking===
 +
 
 +
Simon gave a report:
 +
* One final, additional swim over reference trajectory eliminated.
 +
* The straight line fitting for alignment of the chambers is working.
 +
** Working with an algorithm to adjust parameters of each wire plane.
 +
** Allows for rotation and shift of entire plane.
 +
** Some convergence difficulties being worked on presently.
 +
* Looking at omega events to compare perfomance at 1200 A vs. 1500 A in the solenoid.
 +
** Looks like there is a drop in performance due to reduction in background.
 +
** Started work on lambda to p pi- events, no conclusions yet.
 +
** In comparison of 1200/1500 A, see a 20% loss in &Delta;p/p, but missing mass only worse by a few percent.
 +
 
 +
===PID===
 +
 
 +
Paul gave a report:
 +
* He is in the middle of writing a kinematic fitter. Will incorporate 4-momentum conservation, time constraints, RF timing, vertex constraints
 +
* Using 7 variables in the covariance matrix: px, py, pz, t, x, y, and z.
 +
* Hopes to have results in two weeks.
 +
 
 +
==Ubuntu==
 +
 
 +
Beni reports that he is having problems building a working copy of the GlueX software on an Ubuntu 64-bit system.
 +
 
 +
==Review of Action Items and Repository Activity==
 +
 
 +
* Dmitry Romanov is scheduled to arrive back at JLab in early August.
 +
* We noted that Richard checked in a solution the the problem initially reported by David where hddm-xml was running out of buffer space.

Latest revision as of 11:11, 31 March 2015

GlueX Offline Software Meeting
Wednesday, July 25, 2012
1:30 pm EDT
JLab: CEBAF Center, F326/327

Agenda

  1. Announcements
    1. Data Challenge Meetings
    2. Scientific Computing Survey
    3. No more software sessions in Tracking Meetings (by default)
  2. Review of minutes from the last meeting: all
  3. Reconstruction sub-group reports
    1. Calorimeters
    2. Tracking
    3. PID
  4. Action Item Review
  5. Review of recent repository activity: all

Communication Information

Video Conferencing

Slides

Talks can be deposited in the directory /group/halld/www/halldweb/html/talks/2012-3Q on the JLab CUE. This directory is accessible from the web at https://halldweb.jlab.org/talks/2012-3Q/ .

Minutes

Present:

  • CMU: Will Levine, Curtis Meyer
  • IU: Ryan Mitchell, Matt Shepherd
  • JLab: Mark Ito (chair), Manuel Lara, David Lawrence, Paul Mattione, Elton Smith, Simon Taylor, Elliott Wolin, Beni Zihlmann

Announcements

  • Data Challenge Meetings have started on alternate Mondays at 1:30 pm.
  • Mark urged us to participate in the Scientific Computing Survey.
  • There will be no more (default) software sessions as part of the bi-weekly Tracking Meeting.
  • David and Elliott have started work on translation tables. If you have thoughts send them along. A proposed format was circulated on the Offline email list.

Review of Minutes from Previous Meetings

Offline Meeting

We reviewed the minutes from the July 11th meeting.

  • Paul reported that he thinks all work has been done for now on the REST format. We can start using it in production.
  • Simon made the promised changes in the default track hypotheses generated:
    • Positive tracks: pi, K, p
    • Negative tracks: pi, K

Data Challenge Meeting

We looked at the minutes of the July 16th meeting without significant comment.

Reconstruction Sub-Group Reports

Calorimeters

David has been working on putting more realism in the BCAL smearing.

  • Full electronic pulse shape are generated, with dark hits.
  • Changes have been made to hdgeant and mcsmear. They are currently disabled by pre-processor directives.
  • The unsmeared file is much larger; mcsmear is four times slower for single photon events
  • hdgeant is about three times slower than mcsmear previously.
  • David will give a presentation on this work at a future meeting.

Tracking

Simon gave a report:

  • One final, additional swim over reference trajectory eliminated.
  • The straight line fitting for alignment of the chambers is working.
    • Working with an algorithm to adjust parameters of each wire plane.
    • Allows for rotation and shift of entire plane.
    • Some convergence difficulties being worked on presently.
  • Looking at omega events to compare perfomance at 1200 A vs. 1500 A in the solenoid.
    • Looks like there is a drop in performance due to reduction in background.
    • Started work on lambda to p pi- events, no conclusions yet.
    • In comparison of 1200/1500 A, see a 20% loss in Δp/p, but missing mass only worse by a few percent.

PID

Paul gave a report:

  • He is in the middle of writing a kinematic fitter. Will incorporate 4-momentum conservation, time constraints, RF timing, vertex constraints
  • Using 7 variables in the covariance matrix: px, py, pz, t, x, y, and z.
  • Hopes to have results in two weeks.

Ubuntu

Beni reports that he is having problems building a working copy of the GlueX software on an Ubuntu 64-bit system.

Review of Action Items and Repository Activity

  • Dmitry Romanov is scheduled to arrive back at JLab in early August.
  • We noted that Richard checked in a solution the the problem initially reported by David where hddm-xml was running out of buffer space.
Retrieved from "https://halldweb.jlab.org/wiki/index.php?title=GlueX_Offline_Meeting,_July_25,_2012&oldid=64701"