[Rivet] Use proton-lead MC generators with Rivet

[Andy Buckley]

Great. Please let us know if you encounter any more issues, and we 
really look forward to getting preserved HI analyses from you!

Andy


On 28/02/17 15:47, Van Haevermaet Hans wrote:
> Dear Andy and Leif,
>
> Thank you very much! Indeed, I didn’t got the first mail from Leif, but it seems exactly what is needed to avoid the problem.
> We will try it out, and if it works, then yes I think it would be very nice if this can be included in one of the next releases.
> That will allow then to run Rivet analyses on much more generators with proton-lead or lead-lead beams.
>
> I’m also surprised that no one else encountered this… it seems that we are the first ones to really try it :-)
> Hopefully in the near future, we will also be able then to publish routines with pA data from CMS.
>
> Cheers,
> Hans
>
>> On 28 Feb 2017, at 16:33, Andy Buckley <andy.buckley at cern.ch> wrote:
>>
>> Aah, I understand better now. I thought it was the .info file beams requirement that was causing trouble. I had forgotten that long-ago we put in a check to insist that all events in a run have the same beams.
>>
>> Indeed it would make sense to also drop that check in "ignore beams mode". If I can work out a patch that implements that easily, would you be able to patch your current Rivet build? Obviously another Rivet release will be required to make it out-of-the-box.
>>
>> I'm very surprised that we never had this reported before, given I was in touch with the Epos authors about putting improvements in! I guess they were just using it for nucleon collisions so far.
>>
>> Andy
>>
>>
>>
>> On 28/02/17 14:47, Sercan Sen wrote:
>>> Hi Andy,
>>>
>>> We did try --ignore-beams and --nskip options before but didn't help.
>>> Now trying again with the following steps:
>>>
>>> I copy one of the standard analysis in Rivet, CMS_2015_I1356998
>>> <http://i1356998.cc/>, to my working directory and modify the info file
>>> like the following,
>>>
>>> Beams: [ANY, ANY]
>>> Energies: [13000]
>>>
>>> compile the plug-in and run over this hepmc file
>>> "http://cern.ch/ssen/Rivet/epos_pPb.
>>> <http://cern.ch/ssen/Rivet/epos_pPb.hepmc>txt" which has 20 proton-lead
>>> events generated with Epos-LHC.
>>>
>>> I use --ignore-beams option but still get the error below.
>>>
>>> /Event beams mismatch: [PROTON, NEUTRON] @ 13000 GeV vs. first beams
>>> [PROTON @ (6500; 0, 0, 6500) GeV, PROTON @ (6500; 0, 0, -6500) GeV] @
>>> 13000 GeV/
>>>
>>> The output of the line below shows that there are 7 p-p events and  13
>>> p-n events in the hepmc file.
>>> grep -e "P 2 2212" epos_pPb.txt
>>> grep -e "P 2 2112" epos_pPb.txt
>>>
>>> I use Rivet 2.5.3, and also try the same steps with my own analysis but
>>> the error is the same.
>>>
>>> Thank you for your further suggestions
>>> Sercan
>>>
>>>
>>>> On Feb 28, 2017, at 2:40 PM, Andy Buckley <a.g.buckley at gmail.com
>>>> <mailto:a.g.buckley at gmail.com>> wrote:
>>>>
>>>> Hi Hans,
>>>>
>>>> The rivet command-line tool has an --ignore-beams option for this sort
>>>> of reason. But you haven't said what analysis you are trying to run:
>>>> if it's one that you're writing yourself, just use ANY rather than
>>>> PROTON in the Beams specification in the analysis' .info file.
>>>>
>>>> We've been upgrading tools for HI and cosmic-ray analysis over the
>>>> last year, so are hoping to start building up the collection. Very
>>>> happy to help with anything that's currently a blocker!
>>>>
>>>> Andy
>>>>
>>>>
>>>> On 28/02/17 11:10, Hans Jozef H Van Haevermaet wrote:
>>>>> Dear Rivet authors,
>>>>>
>>>>> I am a member of CMS (forward and small-x community in fact), and I’m
>>>>> trying to get Rivet running with proton-lead events coming from the EPOS
>>>>> MC generator.
>>>>> It turns out however, that this is not possible (?).
>>>>> When executing Rivet, it stops because it detected that the incoming
>>>>> beams are different for each event:
>>>>> Event beams mismatch: [PROTON, PROTON] @ 1.87656 GeV vs. first beams
>>>>> [NEUTRON @ 2511.57 GeV, NEUTRON @ 2511.57 GeV] @ 1.87914 GeV
>>>>>
>>>>> This in turn, is caused by the fact that the EPOS generator writes out
>>>>> the interacting nucleon (p or n) in the HEPMC file, and not the particle
>>>>> id code of lead itself.
>>>>>
>>>>> Could you let us know if it is possible for Rivet to get compatible with
>>>>> these kind of situations?
>>>>> One direct option (for me) is to comment out this function:
>>>>> https://rivet.hepforge.org/code/dev/AnalysisHandler_8cc_source.html
>>>>> At line 95.
>>>>>
>>>>> But that can lead to unwanted behaviour in other cases, I guess. Maybe
>>>>> one can include a parameter to turn this check on/off? So that it is the
>>>>> user’s responsibility? i.e. by default the check is done, but the user
>>>>> can turn it off manually if needed.
>>>>>
>>>>> It would be very good to solve this incompatibility, since in CMS we try
>>>>> to get Rivet routines from all our analyses, and it would be nice to be
>>>>> able to extend this to proton-lead, or even lead-lead measurements. As
>>>>> far as I can see on the web, I did not find any Heavy Ion analysis
>>>>> implemented in Rivet?
>>>>>
>>>>> Thanks!
>>>>> Best Regards,
>>>>> Hans
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Rivet mailing list
>>>>> Rivet at projects.hepforge.org
>>>>> https://www.hepforge.org/lists/listinfo/rivet
>>>>>
>>>>
>>>>
>>>> --
>>>> Dr Andy Buckley, Lecturer / Royal Society University Research Fellow
>>>> Particle Physics Expt Group, University of Glasgow
>>>
>>
>>
>> --
>> Dr Andy Buckley, Lecturer / Royal Society University Research Fellow
>> Particle Physics Expt Group, University of Glasgow
>


-- 
Dr Andy Buckley, Lecturer / Royal Society University Research Fellow
Particle Physics Expt Group, University of Glasgow