GSI Forum
GSI Helmholtzzentrum für Schwerionenforschung

Home » R3BRoot » Simulation Issues » [NeuLAND] New Neuland Digitizer
[NeuLAND] New Neuland Digitizer [message #18412] Thu, 13 August 2015 13:48 Go to next message
Jan Mayer is currently offline  Jan Mayer
Messages: 20
Registered: August 2015
occasional visitor
From: *ikp.uni-koeln.de
Hello everyone,

I have extracted the logic of the digitization process from land/R3BLandDigitizer into a separate "digitizing engine", used by neuland/R3BNeulandDigitizer.
https://subversion.gsi.de/trac/fairroot/browser/r3broot/trunk/neuland/di gitizing/DigitizingEngine.h

The previous code was quite convoluted, but it turned out that it actually did not do "that much" with many LoC. I hope the smaller, more object oriented pieces are easier to understand.
The output should be about the same, however some of the "constants" that were previously set depending on the physical setup or the beam energy might need some reevaluation, e.g.: https://subversion.gsi.de/trac/fairroot/browser/r3broot/trunk/neuland/di gitizing/DigitizingEngine.cxx#L22
In addition, the code remaining under Exec() in R3BNeulandDigitizer (xyz positions) needs some additional love & care.

I would like to invite everyone to have a look at the code, especially the few parts where actual physics is involved (e.g. parts with '@mheil:').

Any questions or comments is appreciated.

Jan
Re: [NeuLAND] New Neuland Digitizer [message #18415 is a reply to message #18412] Mon, 17 August 2015 08:06 Go to previous messageGo to next message
Michael Heil is currently offline  Michael Heil
Messages: 4
Registered: July 2015
occasional visitor
From: *gsi.de
Dear Jan,
thank you for cleaning up the code and digging into that mess. However, I am pretty sure that the functionality of the old code was not so bad since we have tested it against LAND data. Instead of going line by line through the new code, I suggest that you take one of the old simulation examples and compare the output of the new and old code (with same Parameter Settings). In case you find difference we should trace them back and see whether it was an error in the old code or due to different coding in the new one.
Re: [NeuLAND] New Neuland Digitizer [message #18420 is a reply to message #18412] Mon, 17 August 2015 14:30 Go to previous messageGo to next message
Dmytro Kresan is currently offline  Dmytro Kresan
Messages: 166
Registered: June 2004
first-grade participant
From: *gsi.de
Hello Jan,
Current version is not compiling on OS X. The reason: Macintosh file system is not case-sensitive, so rootcint can not distinguish between file R3BNeuLandDigitizer.h from land directory and your new R3BNeulandDigitizer.h from neuland directory (you have them both in INCLUDE_DIRECTORIES in CMakeLists.txt).
Solution: please remove land/R3BNeuLandDigitizer.* from the repository. The code was not touched for 5 years and implementation shows that it actually does nothing.
Best regards,
Dima
Re: [NeuLAND] New Neuland Digitizer [message #18421 is a reply to message #18420] Mon, 17 August 2015 15:10 Go to previous messageGo to next message
Jan Mayer is currently offline  Jan Mayer
Messages: 20
Registered: August 2015
occasional visitor
From: *ikp.uni-koeln.de
I have removed land/R3BNeuLandDigitizer.* as requested. It only contained some histogram filling anyway.
There is a lot more stuff with questionable purpose in the land/ directory, but I hesitate cleaning it up more without knowing if it is still required.

Jan


PS: There is some diff noise in the commit due to whitespace. Maybe trailing whitespace removal could be made mandatory?
Re: [NeuLAND] New Neuland Digitizer [message #18425 is a reply to message #18421] Tue, 18 August 2015 10:07 Go to previous messageGo to next message
Dmytro Kresan is currently offline  Dmytro Kresan
Messages: 166
Registered: June 2004
first-grade participant
From: *gsi.de
We will clean the land directory step-by-step.

Together with migration to git, we will also introduce a coding style, with possibility of automated formatting.

Please also consider following warning in R3BNeulandDigitizer.cxx:

https://cdash.gsi.de/viewBuildError.php?type=1&buildid=60535

Best regards,
Dima
Re: [NeuLAND] New Neuland Digitizer [message #18431 is a reply to message #18425] Tue, 25 August 2015 10:48 Go to previous message
Jan Mayer is currently offline  Jan Mayer
Messages: 20
Registered: August 2015
occasional visitor
From: *ikp.uni-koeln.de
I fixed the warning. Note that on my system, it only appeared after manually adding -Wformat-security. If I traced it back correctly, the Fatal method is actually the one from ROOT, not the one from FairLogger as I first thought. The fix is easy and revolves around the ancient code in TError.cxx.

We might see this warning again in other classes, depending on compiler warning flags.

[Updated on: Tue, 25 August 2015 10:49]

Report message to a moderator

Previous Topic: CALIFA Event Display
Next Topic: Some of LandFirstHits are empty - Neutrons simulation.
Goto Forum:
  


Current Time: Fri Dec 06 18:55:27 CET 2024

Total time taken to generate the page: 0.00679 seconds