Sorry for the late reply.

I am having deja vu about this, because it was something we dealt with several years ago in ILC studies.

> LcioPrimaryGenerator::generatePrimaryVertexFromMcpCollection()

Yes, that method should be where the event vertices are generated from the input MCParticle information (that is derived from the StdHep files).

From what I remember, decay positions are essentially ignored in the simulation, and a particle must have a valid decay time for the Geant4 predefined decay code to work properly.  For charged particles, the decay position would not be used anyways as the particle has to be tracked in the field.

Now, there might be some code in SLIC which attempts to compute a decay time based on the decay length from the StdHep file.  I would have to review more closely to remind myself of how/when this is activated as it was written several years ago.

I think the best way to do this is including valid decay times into your StdHep files and then it should work automatically, because this is what Geant4 expects. 

On Apr 18, 2015, at 8:49 AM, Sho Uemura <[log in to unmask]> wrote:

> Cool, thanks for tracking this down. It sounds like I need to make some changes in the LHE-StdHep and EGS5-StdHep converters.
> 
> It looks like this logic is in LcioPrimaryGenerator::generatePrimaryVertexFromMcpCollection()? I think I understand from your description but it's easier for me if I can read the actual code.
> 
> On Sat, 18 Apr 2015, Graf, Norman A. wrote:
> 
>> Hello All,
>> 
>> I think I might understand what is going on with the decay vertices
>> 
>> in slic.
>> "Pre-defined" decays in Geant4 are somewhat tricky.
>> The decay channel and kinematics can be set by pre-defining
>> the daughter particles. However, the lifetime of the
>> intermediate decay is set by the time of creation of its daughters.
>> If that time is zero (which is the default), two things can happen:
>> 
>> 1.) If Geant knows about the intermediate particle, then it will
>> throw a decay time from the known intermediate particle
>> properties. The intermediate particle is propagated for that length
>> 
>> of time and the (predefined) daughters are created at that time and
>> 
>> position, with modifications to account for change in direction and energy
>> loss. (Unless, of course, there was an interaction with material during
>> that time.)
>> 
>> 2.) If Geant does not know about the particle (such as a long-lived
>> A' particle), then the decay time defaults to zero and all the
>> daughters are created at the point and time of creation of the parent.
>> 
>> So, in order to get the behavior we want, you need to set the creation
>> time for all particles correctly. For daughters of intermediate particles
>> it is the lifetime of the parent.
>> I hope this is clear. If not, please let me know and I can try to answer
>> any specific questions.
>> 
>> 
>> 
>> And the reason why our lcsim code does not correctly read the stdhep
>> 
>> files is because the "version number" in the begin run record is empty.
>> 
>> This should be fixed before generating any more events.
>> 
>> 
>> 
>> Norman
>> 
>> ########################################################################
>> Use REPLY-ALL to reply to list
>> 
>> To unsubscribe from the HPS-SOFTWARE list, click the following link:
>> https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1
>> 
> 
> ########################################################################
> Use REPLY-ALL to reply to list
> 
> To unsubscribe from the HPS-SOFTWARE list, click the following link:
> https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1

########################################################################
Use REPLY-ALL to reply to list

To unsubscribe from the HPS-SOFTWARE list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1