[Rivet] Problems when simulating an analysis

[Chiara Zampolli]

Hi Andy,

I am adding Andreas in CC to this thread, since he is more expert than me in ALICE simulations.

Cheers,

Chiara


On Apr 19, 2013, at 12:20 PM, Andy Buckley wrote:

On 19/04/13 12:08, Chiara Zampolli wrote:
Hi James,


On Apr 19, 2013, at 11:54 AM, James Monk wrote:


On 19 Apr 2013, at 11:01, Chiara Zampolli wrote:

Dear James,

Thanks for your reply!

When you generate the Pythia events, make sure the particle lifetime cut is consistent in the ALICE analysis and your standalone Pythia run.  Experiments typically generate events with a ctau (lifetime) cut of 10mm, meaning that any particle whose species has an average lifetime of more than 10mm will not be decayed, but this setting may well not be present in a standalone run of Pythia unless you set it yourself.


I will check, this might be a good point in fact. How to I set this in Rivet in case?

It's a generator setting, not a Rivet one.  Assuming, as it appears, that you are using AGILe and Pythia 6 then you need

-p "MSTJ(22)=2"

will set a ctau of 10mm (the commonly used value).  If you need another ctau then add PARJ(71)=xx to set it


Ok, I thought it was a cut for the counting of the final states.

agile-runmc Pythia6:426 --beams=LHC:7000 -p "PYTUNE=100" -n 1000000 -o /tmp/zampolli/hepmc_TuneA_7TeV.fifo &

The fact that including or excluding the weight gives me differences means that the generators assigns weights to the events. Even though you write that min bias generation should not… ???
So, in any case the safest is that I simply do not include them, since as you say for MB events there should be no weight.


No - you should keep the weights in the histogram filling and find out why you are generating weighted events when you don't intend to!  When I run the command you list there I only see weights of 1, as I would expect. Have you checked the HepMC file to see what the weights are (or even filled a histogram of weight values).  The weight in the HepMC file should be the last number on the line beginning with "E."


They are all 1, in fact. I will cross check again. Maybe I messed up something..

Note also that the command you list there does not include diffractive processes, whereas the ALICE sample may or may not include those processes - you may need to be more explicit in your command to generate exactly the same setup used by ALICE.  I expect the generator experts on ALICE will be able to help you with that.


Thanks! I will then ask ALICE experts what I should add.

ALICE historically used a "PYTHIA NSD" definition, i.e. the data would
be modified (I don't want to say "corrected"!) to subtract the effect of
PYTHIA6's modelling of single diffractive events.

If that's still what's being done, then you should run with DD and ND
processes only (but be aware that the comparison to generators other
than PYTHIA6 might be a bit funny). If the only correction is for
detector effects, then run with SD, DD and ND.

AGILe already has PYTHIA6 steering card fragments for these two situtations:

* NSD: http://agile.hepforge.org/svn/trunk/data/params/fpythia-nsd.params
* ~Reality ;-) :
http://agile.hepforge.org/svn/trunk/data/params/fpythia-minbias.params

You can pull these in by just adding e.g. -P fpythia-nsd.params to the
agile-runmc command line. If everything works as intended, it should
even tab-complete :-)

Hope that helps,
Andy




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.hepforge.org/lists-archive/rivet/attachments/20130419/5dff6472/attachment.html>