[Rivet] MC plots - Xi+/- and Omega+/- pt spectra

[Peter Skands]

Hi Boris,

Wonderful. The first people to contact with a new Rivet implementation for LHC 
data are the Rivet authors, that I am also cc'ing here. They periodically 
release new Rivet versions with new analyses that we can then pick up and 
display on mcplots. There is a small amount of work involved also in the latter 
(basically we need, in addition, some HTML and LaTeX names for the observables 
and cuts, that we use to populate the menus and plots on the site, but this is 
really minimal and we can return to that once a Rivet analysis is there).

@ Rivet authors: the context is ALICE and CMS identified-particle measurements 
of Xi and Omega baryons, that Boris and Antonin wanted to see how they could get 
on mcplots, so I pointed them to what would be needed, the first being a 
HepMC-based Rivet analysis (the data is already in HepData).

Cheers,
Peter

On 3/28/11 6:38 PM, Boris Hippolyte wrote:
> Hi Peter,
> the points at 900 GeV were already sent to hepdata;
> we'll work on providing the function and try to run some
> generators with Rivet locally then send you (or someone
> else ?) the function.
> Best regards,
> --B.
>
> On Mar 28, 2011, at 6:29 PM, Peter Skands wrote:
>
>> Hi Antonin,
>>
>> That sounds encouraging. Rivet is an extremely useful tool for generator validation and tuning. Only two things are needed:
>>
>>   1) providing the (particle-level-corrected) data to the HepData repository, which Rivet uses for the experimental data points. (This is basically already default at the LHC, as far as I know.)
>>
>>   2) providing a function or routine that encodes the corresponding analysis as it would be run on a HepMC output file from a Monte Carlo. Since all Monte Carlos basically are able to provide HepMC output files, Rivet is then able to run any Monte Carlo and compare it directly to the experimental data.
>>
>> All we really do on mcplots is give a graphical front end to this, showing as many Rivet plots as we can, for as many generators as we can manage to implement, but as you see, basically anything that is in HepData and any generator that can provide HepMC output can be in the game. The only step inbetween is a function that represents the experimental analysis as it would be run on a HepMC file. Also, the Rivet authors are quite responsive if we need help, especially in the mcplots context, as we have already provided some amount of validation and bug fixes for Rivet in the process of setting up the site.
>>
>> Cheers,
>> Peter
>>
>> On 3/28/11 6:11 PM, Antonin Maire (IPHC) wrote:
>>> Hi Peter,
>>>
>>> thanks for this really prompt answer.
>>> At the moment, I would guess the data I point to are not Rivet based.
>>> But before answering further, I would need to understand what is hiden behind
>>> Rivet.
>>> I have found 2 links browsing from the MC plots page :
>>> - http://projects.hepforge.org/rivet/ - source code
>>> - http://arxiv.org/abs/1003.0694 - Rivet manual
>>>
>>> I will discuss this with Boris.
>>>
>>> Cheers,
>>> Antonin
>>>
>>> Peter Skands wrote:
>>>> Hi,
>>>>
>>>> Thanks for pointing this out - mcplots is definitely intended as a resource
>>>> for the LHC experiments, and the identified-particle spectra are, as you
>>>> clearly already know, quite important to constrain certain aspects of the
>>>> models. That said, yes, it does take time and resources to add distributions,
>>>> but to help with this the LPCC at CERN is actually offering some limited
>>>> support for people, e.g. from within the experiments, to spend time with us to
>>>> add specific things. Since we are entirely based on the Rivet tool, the first
>>>> order of business is that the analyses need to be implemented in Rivet. From
>>>> there, the step to have them appear on mcplots is very short. So first I
>>>> should ask you whether the analyses you point to are available in Rivet? If
>>>> not, perhaps someone from ALICE and CMS could be identified who could help,
>>>> keeping the possibility of LPCC support in mind, to implement them, and in the
>>>> process add them to mcplots?
>>>>
>>>> Cheers,
>>>> Peter
>>>>
>>>> On 3/28/11 5:41 PM, Antonin Maire (IPHC) wrote:
>>>>> Good afternoon,
>>>>>
>>>>> I have a question about the site MC plots.
>>>>> I would like to know if I could ask for other species to be computed
>>>>> regarding the identified pt spectra.
>>>>> I have in mind the Xi- and Xi+, Omega- and Omega+
>>>>> for pp 7 TeV (and potentially 900 GeV)
>>>>> in |y|<  0.5
>>>>> to be compared with :
>>>>> - CMS results (900 GeV + 7 TeV) : http://arxiv.org/abs/1102.4282
>>>>> - ALICE results (900 GeV) : http://arxiv.org/abs/1012.3257
>>>>> (The data points do not seem available on HEPdata, yet.)
>>>>>
>>>>> Note I do not know how much time/work it would cost to produce such
>>>>> plots, I would perfectly understand this is not your priority.
>>>>> I am just asking in case it is trivial - and of some interest - for you
>>>>> to produce them.
>>>>>
>>>>> Cheers,
>>>>> Antonin
>>>>>
>>>>> Ps
>>>>> On the Xi- and Xi+
>>>>> (http://mcplots.cern.ch/?query=plots,pp%2Fppbar,uemb-soft,Xim_pt,,)
>>>>> the STAR reference seems to be mixed with the pion, kaon proton reference :
>>>>> the reference http://arxiv.org/pdf/nucl-ex/0601033v2 is quoted while I
>>>>> think it may rather be http://arxiv.org/pdf/nucl-ex/0607033
>>>>> = http://inspirebeta.net/record/722757?ln=en
>>>>>
>
> -----------------------------------------------------------------------------------
> HIPPOLYTE Boris       B214               hippolyt at in2p3.fr
> Département de Recherches Subatomiques
> Institut Pluridisciplinaire Hubert Curien
> 23, rue du Loess Bat20 BP28           phone: +33(0)388106646
> F-67037 Strasbourg cedex2             fax:        +33(0)388106614
> -----------------------------------------------------------------------------------
>
>