[Halld-offline] geant particle ids

[Beni Zihlmann]

Hi Kei,

the problem with geant is that every decay outside of the
standard particles have to be programmed in by the user.
As you can see in gupart.F I take as an example the omega:

       CALL GSPART(33,'OMEGA(783)',3,0.782, 0.,7.836E-23,0.,0)

this call defines the omega with id 33 together with its name,
how many decay-channels,  its mass and lifetime.  This has to
be done by the user because it is not in geant by default.
then you have to define the decay channels in this case 3 as
follows:
       IPA=33
       BRATIO(1)=89.
       BRATIO(2)=8.5
       BRATIO(3)=2.5
       MODE(1)=70809
       MODE(2)=107
       MODE(3)=908
       CALL GSDK(IPA,BRATIO,MODE)
where you have to define the decay channel ratios as %
in BRATIO and in what particles it decays with MODE.
So you see here that I put only the 3 most important
channels in and ignore the rest. And I made sure that
the sum is 100%.  Also there is a maximum of 6 decay
channels possible.

pythia is based on pdg publications(decay data) and the string
model. As such it is supposed to do a much better job than geant
could ever do.

The only caveat is that as Eugene pointed out long lived particles
will decay not at the main vertex but much later while tracked by
geant. And that is where geant has to do the decay. The lambda
is a good example.
Another point is in our case due to low energy it is probably not so 
important
that some particles could be generated down the line by a hadronic
interaction and then needs to be decayed within geant as well.

I hope my babbling makes some sense.

cheers,
Beni

> Hi Beni,
>
> Thanks for putting in the requested changes.
> I ran over 1M events newly generated, and got the
> following statistics from hd_dump:
>
> Number of EtaPrime   : 31719
> Number of omega      : 232782
> Number of phiMeson   : 11300
> Number of Lambda     : 50773
> Number of Sigma      : 16421
> Number of Xi         : 68
> Number of OmegaMinus : 0
> (For the above, I'm just grep-ing for particle names,
> so if there are events with multiple omegas or
> Lambda/AntiLambda pairs, they are counted multiple times)
>
> It does seem that the code is now able to tag
> eta', phi, omega events as advertised, and there are no rho
> meson events tagged as such.
>
> One thing I want to ask about is your statement:
> > At this point I would like to make a comment. Generally it is not a 
> good
> > idea to let
> > geant decay particles it is not designed for that. Phythia however is
> > exactly desinged
> > to do that.
>
> Could you clarify what the advantage is of having pythia
> doing the decays over geant? Is it angular correlations, polarizations,
> etc?  Do you know if there is any polarization information
> contained in the pythia output, and does pythia decay the particles
> based on these correlations?
>
> I'll have more updates at the offline meeting tomorrow.
>
> Thanks,
>     Kei
>
> On 9/17/13 9:06 AM, Beni Zihlmann wrote:
>> Hi All,
>>
>> Upon request:
>> I committed changes to bggen, hdgeant and particleType.h in order for 
>> the
>> use to be able to let omega, phi, eta', and the three rhos decay in 
>> geant.
>> the pid number are:
>>
>> Geant  Pythia
>> -------------------------------------
>>    33       223    ! omega(783)
>>    34       331    ! eta prime
>>    35       333    ! PHI
>>    42      113    ! rho0
>>    43      213    ! rho+
>>    44      -213   ! rho-
>>
>>
>> If you use bggen with the current pythia-geant.map translation table
>> omega, phi and eta' are decayed by geant and their pids will appear in
>> DMCThrown as well as their decay products with the appropriate 
>> association.
>>
>>
>> At this point I would like to make a comment. Generally it is not a good
>> idea to let
>> geant decay particles it is not designed for that. Phythia however is
>> exactly desinged
>> to do that. What we should do in the future, maybe when we move to
>> geant4 is to
>> change the code in DANA to take the full pythia output and create the
>> proper relation
>> between daughter and parent particles for the user such that the user
>> does not have
>> to do it him/her- self.
>>
>>
>> cheers,
>> Beni
>> _______________________________________________
>> Halld-offline mailing list
>> Halld-offline at jlab.org
>> https://mailman.jlab.org/mailman/listinfo/halld-offline
>>