Dear Colleagues,

I have reconstructed skimmed samples of FEE, Møller and V0 candidates using the current HPS-PhysicsRun2016-Pass2 detector (aka v20) and the latest proposal from Alessandra (aka 31). The reconstruction was performed with today's snapshot of the git master differing only in the compact.xml description of the detector.

I have placed a few comparison plots on the slack #software channel and also attached them here for those of you not yet signed up to slack. Please take a look at them and post any feedback there.

For those of you wishing to independently analyze the recon output, the lcio files are available at SLAC at:

/nfs/slac/g/hps3/data/hpsrun2016/tpass2/localTests/v20v31Comparisons/

Please let me know if you have any questions or need any additional information.

Norman

-----Original Message-----
From: Nelson, Timothy Knight <[log in to unmask]> 
Sent: Thursday, December 6, 2018 9:28 AM
To: Alessandra Filippi <[log in to unmask]>
Cc: Graf, Norman A. <[log in to unmask]>; hps-software <[log in to unmask]>; [log in to unmask]
Subject: Re: Pass2 Readiness

Hi Alessandra,

Thanks!  It’s great that you did this.  I hope maybe you can present some usual results from this next Monday.  The motions in the xml all look reasonable: ones to tens of microns.

On your cautions:
- We don’t expect the elastic peak to fit at 2.300 GeV.  As Valery showed us for 1.06 GeV a couple of years ago, radiative effects in the target lower the apparent position of the elastic peak by something like 20 MeV.  We don’t know what this effect is at 2.3 GeV, but for sure chasing agreement with 2.300 is not motivated.  The peak SHOULD appear below the known beam energy.

- We don’t want global alignment to bring the beamspot to (0,0) unless it can be done in a way that is both physical and which brings it to (0.0) for all tracks, not just one particular set of V0, as we learned last fall.

I think we should test both this alignment and #20.  Unless there is a big difference in things like track chisquared, etc., this is the one we should use.

Cheers,
Tim

> On Dec 6, 2018, at 9:14 AM, Alessandra Filippi <[log in to unmask]> wrote:
> 
> Hi all,
> I can provide a new detector which is position-compliant (all offsets 
> are within physical limits, and only u translations and w rotations 
> -of central sensors- have been performed). It can be found at slac in 
> /nfs/slac/g/hps3/users/afilippi/hps/hps-java/detector-data/detectors
> and it is called HPS-PhysicsRun2016-Nominal-v2-31-NG_newField,
> compact.xml attached to this e-mail (take care of correctly addressing the fieldmap). It still needs to be tested (a couple of days more would be required).
> 
> Some words of caution:
> - as said, it wasn't tested extensively, neither on curved tracks nor 
> or straight tracks nor on Mollers nor fee's etc (no time to do it yet) 
> ==> this means that final resolutions weren't checked (but I don't 
> expect them to be different from what shown with other versions as 
> they never change), nor vertex positions (for which at least I need 
> Moller selected events)
> 
> - residuals are ok but those of version #20 are better. Layer 4 is, as usual, the worst.
> 
> - elastic peaks t/b calibrate nicely between themselves (3 MeV/c difference) but are both lower than the expected momentum, as it happens when the detector is not stretched in z (but we didn't want to do this). Lower means around 2.255 MeV/c (with a small statistics, though).
> 
> - global alignment to bring beamspot (x,y) to (0,0) is absent (no time to do it yet), but this is also true for detector #20.
> 
> If you want to start immediately with pass 2 and you are worried about being trapped in a relative millepede minimum using detector #20, you can use this version, and we'll all see what comes out after some runs have been processed.
> cheers
>    Alessandra
> 
> 
> 
> 
> 
> 
> 
> On Thu, 6 Dec 2018, Nelson, Timothy Knight wrote:
> 
>> Sounds good. The non-physical motions aren’t great, but we should solve that with better process next time if the data looks OK.
>> 
>> Tim
>> 
>>> On Dec 5, 2018, at 11:13 PM, Graf, Norman A. <[log in to unmask]> wrote:
>>> 
>>> Hello Tim,
>>> 
>>> I would like to start another test pass tomorrow using the current HPS-PhysicsRun2016-Pass2 detector (#20), generate our canonical physics distributions and make a decision after reviewing those results.
>>> 
>>> Norman
>>> 
>>> 
>>> From: Nelson, Timothy Knight
>>> Sent: Wednesday, December 5, 2018 11:02 PM
>>> To: Graf, Norman A.
>>> Cc: hps-software; [log in to unmask]
>>> Subject: Re: Pass2 Readiness
>>> 
>>> Hi Norman,
>>> 
>>> What alignment are you going to work with? Do we just go back to #20 now and see what we get?
>>> 
>>> Tim
>>> 
>>>> On Dec 5, 2018, at 10:56 PM, Graf, Norman A. <[log in to unmask]> wrote:
>>>> 
>>>> Dear Colleagues,
>>>> 
>>>> We have now resolved several code issues which were holding up the Pass2 reconstruction and the code has been merged into the git master branch. Thanks to everyone who helped identify and resolve the problems. I encourage everyone to checkout, build and test the latest code snapshot to make sure that what you expect to be in the output is there and correct. All tests should build, run and pass. If not, please post any issues immediately.
>>>> 
>>>> We still have a few remaining issues to iron out, but we would like to start the next test pass before the end of the week.
>>>> 
>>>> 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
> <compact_HPS-PhysicsRun2016-Nominal-v2-31.xml.gz>


########################################################################
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