Message: Re: Parameterisation and CopyNo Not Logged In (login)
 Next-in-Thread Next-in-Thread
 Next-in-Forum Next-in-Forum

Question Re: Parameterisation and CopyNo 

Forum: Geometry
Re: Question Parameterisation and CopyNo (Hanwook Bae)
Re: None Re: Parameterisation and CopyNo (John Apostolakis)
Date: 14 Sep, 2017
From: Hanwook Bae <Hanwook Bae>

Thank you very much for your reply.

There is another question about my problem and your reply.

In fact, there is something I forgot to mention: the logical volume what I want to parameterize is not simple volume. What I am saying is, the volume has many daughter volumes. So, one of the daughters is slightly different from a lower layer to an upper layer.

As I know about parameterization, I can't manipulate anything about the daughter in the target volume, but I can only transform the grand-mother volume in the parameterization situation.

In this case, I can't utilize the first solution because what I can in my parameterization class is to compute transformation of my detector cell.

Therefore, I think it is plausible to implement the second solution of your reply. However, unfortunately, I don't know anything to dispatch the parameterization to another parameterization class.

So, could you tell me the way to send a parameterization to another one or which is example source code about the second method?

I am really appreciated for your answer and looking for your kind understanding of my immatureness.

Thank you.

On Tue, 12 Sep 2017 13:03:22 GMT, John Apostolakis wrote:

> Dear Hanwook,
> 
> Your question and setup are clear.
> 
> The answer is simple: you can NOT decide what the first copy number in a 
> parameterisation is.  They must be numbered from the lowest permitted 
> value (0 or 1) to the copy number ( or max copy number -1 if 0 is first. )
> 
> But what you could do is use one parameterisation which describes both 
> the upper and lower layer.  I can see two ways of doing this
> 
> 1) By hand in code for a single parameterisation that handles both 
> cases.  It would be less elegant in some ways - it would need to have 
> something like "if" statements for all methods which depended on the layer.
> 
> 2) A different solution would be to create a 'hierarchy' of 
> parameterisations - one for the upper and one for the lower layer, and 
> then create a composite parameterisation that only knows the number of 
> copies in each one, and dispatches to the correct dependent 
> parameterisation.   ( The composite would own the other two, and simply 
> act as a switch between them.  )  This would make the code more elegant, 
> but I am not sure whether there would be a performance penalty.  Likely 
> some work would be needed (with templates) to avoid virtual function 
> calls, in order to ensure that persformance is not lost.
> 
> I think and hope that this clarifies how you should be able to create a 
> working program for your setup.
> 
> Best regards,
> John Apostolakis
> 
> On 11.09.17 17:37, Hanwook Bae wrote:
> > *** Discussion title: Geometry
> >
> > Hello, I am trying to use the parameterization class to make my detector
> > in the Geant4 framework.
> >
> > In detail, The detector is consisted of with a number of the same
> > geometry and has a multi-layered structure.
> >
> > However, the bottom layer has different structure slightly, so I divided
> > the detector into two parts is that the bottom part and upper one.
> >
> > The problem is that a sensitive detector that I am using can distinguish
> > the geometries only by their copy number and I can't modify the SD.
> >
> > So, I would like to modify a copy number of the upper parameterization
> > volume so that a copy number of the first volume in the upper part leads
> > to the last number of the lower one.
> >
> > Then I googled a way to modify a copy number of each volume in
> > parameterization but I can't find anything.
> >
> > Can I modify the copy number as I said? If it is possible, please tell
> > me the way to modify the copy number.
> >
> > Sorry for my poor English. If you can't understand the meaning of my
> > words, please ask me about it.
> >
> > Thank you.
> >
> > -------------------------------------------------------------
> > Visit this GEANT4 at hypernews.slac.stanford.edu message (to reply or unsubscribe) at:
> > http://hypernews.slac.stanford.edu/HyperNews/geant4/get/geometry/1547.html
> 

 Add Message Add Message
to: "Re: Parameterisation and CopyNo"

 Subscribe Subscribe

This site runs SLAC HyperNews version 1.11-slac-98, derived from the original HyperNews


[ Geant 4 Home | Geant 4 HyperNews | Search | Request New Forum | Feedback ]