This is interesting, let's be sure I understand. We can implement new configurations with
GTEST in the name? Who can do this? How broad can the scope be?
All of us involved in SUSA implementation would like to have a TEST implementation
of it. Is that possible? I'm sure Steven Gardiner will be willing to do the work.
Of course, it's broader than just another FSI - maybe something like GTEST18_11a....?
As Robert says, we want these TEST configs to be related to what will later be implemented.
Originally, we were talking about a GTEST config that included all possible models and the
user can then choose as he/she wishes. (Of course, principal and FSI models must
be treated differently.) Is that still possible, desirable? Seems simple and solves a lot
of problems.
Steve
On 2/20/20 3:21 PM, Robert W Hatcher wrote:
Hi Robert,
yes, GTEST18_02c is exactly what I was thinking.
If you'd like I could commit GTEST18_02c and GTEST18_02d ...
-robert
Ok, for Steve giving the physics talk.Steve please make sure to point to the DocDb mention by Robert in your talk.
Cheers,Marco
On Thu, 2020-02-20 at 16:47 +0000, Robert W Hatcher wrote:
Nothing has been done specifically in terms of Test CMC. One was supposed to be with the Susav2. Hard to make it until the code is in not in.
I have test CMC's for both the INCL++ and Geant4 FSI's that just bootstrap from G18_02a ... it's basically the same diff as G18_02a -> G18_02b except yet a different FSI model:
diff -r G18_02a/ModelConfiguration.xml G18_02c/ModelConfiguration.xml54c54< - genie::HAIntranuke2018/Default---> - genie::HINCLCascadeIntranuke/Default59c59< <param type="alg" name="HadronTransp-Model"> genie::HAIntranuke2018/Default </param>---> <param type="alg" name="HadronTransp-Model"> genie::HINCLCascadeIntranuke/Default </param>
with G18_02c = INCL++and G18_02d = Geant4
Do we have a convention for naming experimental CMCs ?
I think GTEST18_02c (etc) would pass our regex if we want to go with that, leaving the clear association with G18_02*.
But this is definitely a place where it would be nice to have CMC allow the use of a different tune's x-sec splines (just like G18_02b fundamentally doesn't need different splines from G18_02a).
Do we need one with fancy decay distribution for the Delta?We can make some as well for FSI. I was expecting Robert or you to make the request.
As for a talk on INCL++/G4 FSI, I think Steve should give it as my contribution was more to the integration and technical bits than the physics, which he has studied and which is probably of more interest to Forum attendees then technical bits. Though, if desired I can give the same talk I gave earlier about integration (see docdb 174).
-robert
In case, let me know.CheersMarco
Marco Roda, PhD in Physics
University of Liverpool
department of physics
Oliver Lodge Laboratory
Liverpool L69 7ZE, UK
On 20 Feb 2020 15:25, Steven Dytman <[log in to unmask]> wrote:
Marco,
Not sure about v3.2 rush issue but I definitely worry about scope creep.
You make a good point. What do others think?
In January discussions, we decided to have a "test" CMC (I still prefer name
of model set) and you agreed to implement. Was that done?
regards,
Steve
On 2/20/20 9:27 AM, Dr. Marco Roda wrote:
> Hi Steve and Stephen,
>
> I don't think we are in a super rush.
>
> If we have to prioritize something, maybe prioritize the merging over
> the documentation. It's fair to have the code without documentation if
> we don't tag.
> And users can play wiht it anyway.
>
> But take your time, we don't want to release 3.2.2 in a month because
> we found bugs.
>
> Steve D it's ok if you want to give a talk at the next forum. I'll add
> it to my agenda. How much time do you need? Would it be FSI generic?
> Will it be instead of the talk by Robert or an addition?
>
> Cheers,
> Marco
>
>
>
> On Wed, 2020-02-19 at 23:17 +0000, Steven J. Gardiner wrote:
>> Hi Steve,
>>
>> Great news indeed about the INCL/Geant merge!
>>
>> I suspect I’m the last person with unmerged changes who still wants
>> to target v3.2. I’d still like to get both SuSAv2 and Noemi SF in.
>> I’ve been working hard this week to finish validations and get both
>> branches merge-ready. My proposed deadline is COB Tuesday (2/25) to
>> have code + documentation ready to go. As part of that effort, I can
>> produce a release candidate branch (with all anticipated changes
>> merged) and do some initial tests on the FNAL grid.
>>
>> I’m happy to negotiate a tighter deadline if others feel that Tuesday
>> is too far off.
>>
>> I’ve attached a new validation plot for Noemi SF for 560 MeV
>> electrons on 12C at 60°. The red curve comes straight out of Noemi’s
>> Fortran code. Blue is made from GENIE events with a scattering angle
>> of 60° ± 2°. The events were made with my feature branch as it
>> existed at the time of the last review meeting.
>>
>> I generated a larger sample (11M events) after merging in recent
>> changes from master, but something appears to have broken during the
>> merge. I am actively troubleshooting this today.
>>
>> Cheers,
>>
>> Steven
>>
>>
>>
>>
>>
>>
>>
>>> On Feb 19, 2020, at 12:36 PM, Steven Dytman <[log in to unmask]>
>>> wrote:
>>>
>>> No replies when I sent this to support, so maybe core is more
>>> appropriate?
>>>
>>> Marco reports INCL/GEANT is now merged into trunk, great news. We
>>> are now very close to v3.2 release. Does anyone have something
>>> that they strongly want included? If so, please give a solid
>>> estimate when code and documentation will be ready. Steven Gardiner
>>> is only developer who I know is in this situation.
>>>
>>> Steve
>>>
>>> ###################################################################
>>> #####
>>>
>>> To unsubscribe from the NEUTRINO-MC-CORE list, click the following
>>> link:
>>> https://nam05.safelinks.protection.outlook.com/?url=https%3A%2F%2Furldefense.proofpoint.com%2Fv2%2Furl%3Fu%3Dhttps-3A__www.jiscmail.ac.uk_cgi-2Dbin_webadmin-3FSUBED1-3DNEUTRINO-2DMC-2DCORE-26A-3D1%26d%3DDwIFaQ%26c%3DgRgGjJ3BkIsb5y6s49QqsA%26r%3DZ2n1_eNx066DXkZ8awih7ks9TFT77FjRGh9vHYSjeC4%26m%3D65fG5eFgA4QRU7KRg7mEKR0Km837kGyWaDt16nyWxec%26s%3DQblmBLhcbwkR217UljyK8ChW1Pwg82cUNoSA6rcJFZg%26e&data=02%7C01%7Cdytman%40PITT.EDU%7Cd96f5c2c2b03453fdcba08d7b6110b3c%7C9ef9f489e0a04eeb87cc3a526112fd0d%7C1%7C0%7C637178056636715237&sdata=K6Dy9FkCDUcP747H1AMYIrQV3MCFH%2Bfru7eP9FaJrlw%3D&reserved=0=
>>
>> To unsubscribe from the NEUTRINO-MC-CORE list, click the following
>> link:
>> https://nam05.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.jiscmail.ac.uk%2Fcgi-bin%2Fwebadmin%3FSUBED1%3DNEUTRINO-MC-CORE%26A%3D1&data=02%7C01%7Cdytman%40PITT.EDU%7Cd96f5c2c2b03453fdcba08d7b6110b3c%7C9ef9f489e0a04eeb87cc3a526112fd0d%7C1%7C0%7C637178056636715237&sdata=yZ%2BJ6E2ISAMkFt8QSQo12%2FEZpFRqXYrNYlURRxTyFmU%3D&reserved=0
> ########################################################################
>
> To unsubscribe from the NEUTRINO-MC-CORE list, click the following link:
> https://nam05.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.jiscmail.ac.uk%2Fcgi-bin%2Fwebadmin%3FSUBED1%3DNEUTRINO-MC-CORE%26A%3D1&data=02%7C01%7Cdytman%40PITT.EDU%7Cd96f5c2c2b03453fdcba08d7b6110b3c%7C9ef9f489e0a04eeb87cc3a526112fd0d%7C1%7C0%7C637178056636715237&sdata=yZ%2BJ6E2ISAMkFt8QSQo12%2FEZpFRqXYrNYlURRxTyFmU%3D&reserved=0
########################################################################
To unsubscribe from the NEUTRINO-MC-CORE list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=NEUTRINO-MC-CORE&A=1
To unsubscribe from the NEUTRINO-MC-CORE list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=NEUTRINO-MC-CORE&A=1
To unsubscribe from the NEUTRINO-MC-CORE list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=NEUTRINO-MC-CORE&A=1
To unsubscribe from the NEUTRINO-MC-CORE list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=NEUTRINO-MC-CORE&A=1
To unsubscribe from the NEUTRINO-MC-CORE list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=NEUTRINO-MC-CORE&A=1