Hi Eugene,
what we need is reconstructed momentum that has undergone radiation and
energy loss versus momentum that has not been radiated and that has not
lost energy.
Ideally, if you had the two ntuples of reconstructed MC with and
without MASCARAD/ELOSS based on the same seed, you could have this
momentum difference event-by-event. However I understand that the event
lists of reconstructed MC with and without radiation are different. That's
why I suggested to use the "tossed momentum" (or "generated momentum"
before radiation and eloss) instead. I think it is just pwl-pml or so.
Let me know if this is now clear to you.
Regards,
Michael
On Wed, 22 Mar 2006, Eugene J. Geis wrote:
> Just a quick clarification, the parameterization we need is of
> MASCARAD on compared to MASCARAD off for the electron ??
>
> I don't understand what is meant by "tossed momenta"
>
> -eugene
>
>
> Quoting Michael Kohl <kohlm@mit.edu>:
>
>> Hi,
>>
>> here are the minutes:
>>
>> -Assessment of v3_4_17 recrunch
>> v3_4_17 lrd recrunch done for hydrogen and deuterium 2005
>> Difference to v3_4_16 has been the extended magnetic field map.
>> -Adrian showed plots of kinematic offsets that indicate consistency
>> between small and large grid result.
>> -The observed offsets of th_p-th_p(th_e) and p_e-p_e(th_e) cannot be
>> simultaneously accounted for by an offset of th_e (would need
>> different signs for such an angle offset)
>> -Adrian's plots of z: blue=pleft, magenta=corresponding eright;
>> red=pright, green=corresponding eleft. Red/green pair consistent
>> with
>> each other but differs greatly from blue/magenta (which is
>> consistent
>> with each other). Reason was attributed to an inefficiency due to
>> losses in the cuts (those were not updated and were still valid for
>> the version before the latest geometry changes)
>> -Adam showed comparisons of d(e,e'p) data and MC for 2005 deuterium
>> of the old crunch v3_4_14 (lrn) from January and v3_4_17 (lrd)
>> from today. Plots show electron momentum in the left and right
>> sectors (columns) for 0.1<Q2<0.2 and 0.2<Q2<0.3 (plot containing
>> "1_3" in the filename) and 0.3<Q2<0.4 and 0.4<Q2<0.5 (plot
>> containing "3_5" in the filename). The agreement in the right sector
>> has improved considerably. The left sector has been affected only
>> little. The MC still shows broader distributions than the data,
>> indicating an issue with the resolution. Best agreement was achieved
>> for lowest Q2 bin, while the higher bins show discrepancies. In
>> general, the momenta of the data are smaller than those of the MC
>> which could be partially or potentially explained by radiative
>> effects.
>> -Chris generated code that calculates derivatives of p,th,ph,z with
>> respect to geometrical offsets of r,theta,phi,alpha,beta,gamma of
>> the
>> chamber locations. With the three chambers in each sector treated as
>> a rigid object, there's a total of
>> 2 sectors * (3 positions + 3 angles) = 12 parameters that can be
>> varied to find the most likely wire chamber geometry in a fit linear
>> in the offset parameters. There's a total of 12 constraints from
>> imposing elastic kinematics (2 sectors) * (vertex + coplanarity + 3
>> independent kinematic relations of momenta and angles +
>> time-of-flight relation), therefore the linear fit should have one
>> unique solution. (It's actually a linear regression).
>> Before imposing elastic kinematics constraints, the average effects
>> of energy loss and internal radiation on the momenta have to be
>> accounted for first.
>> Eugene had parameterized the momentum losses of protons and
>> electrons
>> by comparing the MoneCarlo momenta with energy loss turned on with
>> the tossed momentum, as a function of reconstructed momentum (email
>> from March 10). However, internal radiation (MASCARAD) was turned
>> off. This should be repeated as also the average momentum of the
>> electron due to internal radiation gets shifted as a result of the
>> finite resolution. Adrian had shown this in his comparisons of
>> radiated and unradiated Montecarlos in the meeting of 2006/03/01.
>> Needed are really parameterizations of delta p between (radiated +
>> energy loss)-MC and tossed momenta. Eugene, Adrian, this is input
>> that Chris needs while working on the procedure to find the optimum
>> wire chamber locations and orientations.
>>
>>
>> -Strategy for coming steps / to do list
>> The strategy is two-fold: While we (Chris) continue to determine the
>> best
>> geometry we also have to prepare for the April 6/7 collaboration
>> meeting. Therefore:
>> +For the current recrunch version v3_4_17, the currently
>> corresponding residual kinematic offsets along with separate
>> parametrizations for electron radiation loss and proton
>> energy+radiation loss have to be determined and suitably
>> parameterized and be made available for everybody's use in
>> blast_anaware, (with energy/radiation loss separately accounted
>> for) (-> Eugene, Adrian). This has a very high priority.
>> +v3_4_17 in combination with the above corrections will be the data
>> basis for analysis updates that are going to be reviewed on
>> April 6/7.
>> +I will continue the v3_4_17 recrunch for deuterium 2004 asap.
>> By early next week (~03/27) the v3_4_17 recrunch of the entire data
>> should be available (hydrogen and 2005-deuterium is available
>> already), along with the corrections from above. From
>> then on, 10 days will be available to revisit cuts and macros to
>> update the analysis results. Cuts can be looked at already from now
>> on. In particular, it is desirable to have new estimates for th_d,
>> hPz and Pzz by the end of next week (Sunday April 2, 2006), with
>> three days left to apply them to the analyses.
>> +Chris will determine the optimum wire chamber geometry by early
>> next
>> week. Hopefully I will be able to crunch the hydrogen data with
>> the updated geometry before next Wednesday. If we're satisfied with
>> the
>> results, plan will then be to perform another recrunch of the
>> entire
>> dataset and to produce the "residual kinematic corrections" on top.
>> +This recrunch will be available by mid-April at the earliest.
>> "Final" (presentable and publishable) analysis results are then
>> envisioned for end of April.
>>
>> -Collaboration meeting is on Friday 2006/04/07 09:00 at Bates
>> -Extended analysis meeting on Thursday 2006/04/06 09:00 at Bates
>> -Next analysis meeting is on Wednesday 2006/03/29 13:30 at Bates
>>
>>
>>
>> Best regards,
>>
>> Michael
>>
>>
>>
>>
>> +-------------------------------------+--------------------------+
>> | Office: | Home: |
>> |-------------------------------------|--------------------------|
>> | Dr. Michael Kohl | Michael Kohl |
>> | Laboratory for Nuclear Science | 5 Ibbetson Street |
>> | MIT-Bates Linear Accelerator Center | Somerville, MA 02143 |
>> | Middleton, MA 01949 | U.S.A. |
>> | U.S.A. | |
>> | - - - - - - - - - - - - | - - - - - - - - -|
>> | Email: kohlm@mit.edu | K.Michael.Kohl@gmx.de |
>> | Work: +1-617-253-9207 | Home: +1-617-629-3147 |
>> | Fax: +1-617-253-9599 | Mobile: +1-978-580-4190 |
>> | http://blast.lns.mit.edu | |
>> +-------------------------------------+--------------------------+
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>
>
>
> --------------------------------------------------------------------------
> Eugene Geis
> PhD Student, Physics Department, ASU
> Research Affiliate, MIT-Bates Laboratory of Nuclear Science
> eugene.geis@asu.edu
> --------------------------------------------------------------------------
> http://quickreaction.blogspot.com
>
+-------------------------------------+--------------------------+
| Office: | Home: |
|-------------------------------------|--------------------------|
| Dr. Michael Kohl | Michael Kohl |
| Laboratory for Nuclear Science | 5 Ibbetson Street |
| MIT-Bates Linear Accelerator Center | Somerville, MA 02143 |
| Middleton, MA 01949 | U.S.A. |
| U.S.A. | |
| - - - - - - - - - - - - | - - - - - - - - -|
| Email: kohlm@mit.edu | K.Michael.Kohl@gmx.de |
| Work: +1-617-253-9207 | Home: +1-617-629-3147 |
| Fax: +1-617-253-9599 | Mobile: +1-978-580-4190 |
| http://blast.lns.mit.edu | |
+-------------------------------------+--------------------------+
This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:33 EST