Hi Everyone,

One option could be to develop the new model by modifying the existing HPSEcal3
in lcsim (without checking it in).  This may make immediate testing easier since you
don't have to worry about modifying the various places in hps-java that have hardcoded
references to the existing HPSEcal3 detector model, nor bother with understanding what
needs to change when you move it from existing in lcsim to hps-java.  Then, when finished
testing, maybe Jeremy or another expert can help convert it over to existing in hps-java
with a new model name.

The proposed alternative, starting with a new ecal geometry model in hps-java
instead of lcsim, would be much easier if a working ecal example existed.  Currently
we're basing our understanding of how the software works on the existing
implementation of ecal geometry model in lcsim, HPSEcal3, while just the
compact.xml and lcdd reside in hps-java.  Ok, maybe this is really easy to move
to hps-java, but it's not obvious.

-Nathan



On Oct 1, 2015, at 6:09 PM, Holly Vance <[log in to unmask]> wrote:

> Dear Jeremy,
> 
> Annie and I modified the HPSEcal3 geometry so that it can be backwards compatible with everything that exists as well as accept new values of shifts in x and y and z if they are specified. 
> 
> I have already tested these changes (which of course required modifying the HPSEcal3 detector in detector/converter/compact/... as well as the one in geometry/compact/converter/lcdd/...
> 
> I am also aware that any new detector needs to access geometry/subdetector/....
> 
> So I see the directories in the hps-java/detector-model are the same. Are you suggesting we simply copy the HPSEcal3 necessary files to this file location? We could make a new HPSEcal4, but the 3 is backwards compatible by simply accepting new values. 
> 
> Please advise where we are to commit changes. Thanks!
> 
> -Holly and Annie
> 
> 
> On Thu, Oct 1, 2015 at 4:38 PM, McCormick, Jeremy I. <[log in to unmask]> wrote:
> We might as well just make a new model in hps-java as it was planned that any new development on these ECAL geometry classes would occur in hps-java and not lcsim, as we've now done with the tracker.  This means people can work on this geometry code without needing access to the lcsim SVN, which most people in the collaboration do not have (it is in a separate SVN repository from hps-java with different user accounts).
> 
> So I would suggest making a new "HPSEcalEngRun2015" class in trunk/detector-model based on HPSEcal3, rather than changing the existing class in lcsim, which is known to work well and should essentially be frozen, as there has been an enormous amount of data generated using it.  The slightest deviation there in the output geometry compared to the existing version could cause very nasty bugs in our software, so I believe it would be best to avoid this entirely by not changing it.
> 
> Also, one quirk of our framework is that you need two classes that implement the same geometry model, one for the recon "Java" geometry and the other which outputs the LCDD files for the simulation.  If it is not clear to you how this works, please let me know and I'll point you to examples of each.  I have generally worked on the simulation output model first, as you can easily visualize it in Geant4.
> 
> --Jeremy
> 
> -----Original Message-----
> From: [log in to unmask] [mailto:[log in to unmask]]
> Sent: Thursday, October 01, 2015 1:24 PM
> To: McCormick, Jeremy I.
> Cc: Holly Vance; hps-software
> Subject: RE: Ecal Geometry
> 
> Holly,
> 
> I think it is better to have separate parameters for the shifts and don't mix them with them beam gap parameters. I can do this changes too.
> 
> Annie.
> 
> > Don't change the existing HPSEcal3 detector model, as this could have
> > some unintended side effects like causing a mismatch between the
> > simulated and recon geometry models.
> >
> > You should instead make new classes in trunk/detector-model and we can
> > go from there.  You'll have a lot more freedom to change things this
> > way by making a new set of classes.
> >
> > If you want to copy and paste the existing HPSEcal3 detector model
> > into a new class in hps-java that's a reasonable place to start I think.
> >
> > -----Original Message-----
> > From: Holly Vance [mailto:[log in to unmask]]
> > Sent: Thursday, October 01, 2015 12:26 PM
> > To: McCormick, Jeremy I.
> > Cc: hps-software
> > Subject: Re: Ecal Geometry
> >
> > We are making the changes we discussed in the workshop last week.
> > These changes include (at the moment): x,y,z shifts to the top and
> > bottom crystals in the Ecal. Ani is working on alterations to how we
> > can space the crystals and angling.
> >
> > At the moment, the only place this happens is in HPSEcal3 in LCSIM. I
> > see that this can be done in a similar manner in
> > hps-java/detector-model. So I think I can copy the relevant code changes to there?
> >
> > On Thu, Oct 1, 2015 at 3:05 PM, McCormick, Jeremy I.
> > <[log in to unmask]> wrote:
> >
> >
> >       What changes are you trying to make to HPSEcal3?  Can you send me a diff?
> >  (output from 'svn diff' would be fine)
> >
> >       It would be preferable to move development over to the hps-java
> > trunk/detector-model package as we've done for the SVT detector
> > models, as you won't need to checkin anything to lcsim then.
> >
> >       The package names were left the same so it should be clear what needs
> > to go where.
> >
> >       We should probably make something new like an "HPSEcalEngRun2015"
> > model where you can make siginificant changes to how the model works
> > based on survey from the Eng Run etc., without any chance of affecting
> > the existing classes.
> >
> >
> >       -----Original Message-----
> >       From: [log in to unmask]
> > [mailto:[log in to unmask]] On Behalf Of Holly Vance
> >       Sent: Thursday, October 01, 2015 12:01 PM
> >       To: hps-software
> >       Subject: Ecal Geometry
> >
> >       Hello,
> >
> >       I've created a new HPSEcal3 detector in LCSIM. I do not currently
> > have credentials to commit to this repository-how can I proceed to
> > commit? Or can we branch the LCSIM trunk while we develop this detector further?
> >
> >       Thanks,
> >       Holly
> >
> >
> >       ________________________________
> >
> >
> >       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
> >
> 
> ########################################################################
> 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
> 
> 
> --
> BEGIN-ANTISPAM-VOTING-LINKS
> ------------------------------------------------------
> 
> NOTE: This message was trained as non-spam.  If this is wrong,
> please correct the training as soon as possible.
> 
> 
> Teach CanIt if this mail (ID 01PnUDOif) is spam:
> 
> Spam:        https://www.spamtrap.odu.edu/canit/b.php?i=01PnUDOif&m=4bfb8e90c713&t=20151001&c=s
> 
> Not spam:    https://www.spamtrap.odu.edu/canit/b.php?i=01PnUDOif&m=4bfb8e90c713&t=20151001&c=n
> 
> Forget vote: https://www.spamtrap.odu.edu/canit/b.php?i=01PnUDOif&m=4bfb8e90c713&t=20151001&c=f
> ------------------------------------------------------
> END-ANTISPAM-VOTING-LINKS
> 
> 
> 
> 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