Project

General

Profile

Bug #14764

Overlay reco needs correct digit collection

Added by Wesley Ketchum almost 3 years ago. Updated almost 3 years ago.

Status:
Feedback
Priority:
Normal
Assignee:
Wesley Ketchum
Target version:
MCC8
Start date:
12/06/2016
Due date:
12/08/2016
% Done:

80%

Estimated time:
Duration: 3

Description

Looks like in the overlay fcl file, the mixer and digitfilter are run in the same job with the right-ish configs. But, in reco, digitfilter is rerun looking at the daq collection, and thus the overlaid data is not picked up.

Suggest:
(1) Renaming the detsim output for the nonoise to be 'daqnonoise' (currently 'daq')
(2) Renaming the overlay module output to be daq (currently 'mixer')
(3) Removing the digitfiler from the overlay module chain

Then, downstream reco will run as usual, with the default configs.

History

#1 Updated by Wesley Ketchum almost 3 years ago

  • Status changed from Assigned to Feedback
  • % Done changed from 0 to 80

In feature branch:

feature/wketchum_mcc8_updates

In detsim I have now
--- standard_overlay_uboone.fcl doing the overlay process, running the mixer module with name 'daq'
--- standard_detsim_uboone_nonoise.fcl now names collections daqnonoise and pmtreadoutnonoise
--- there's a "microboone_overlay.fcl" file added to define the mixing module in DataOverlay

Remaining problem is that the optical information now has module name 'daq' and not 'pmtreadout', so that would imply a change of reco configs to accomodate. Need to think this through, but options are:
--- we copy the optical info from daq into pmtreadout in a seperate module
--- we just modify the reco fcl files, and we have special ones for data+MC overlays

Since we may have special ones for data and MC, maybe that latter is the right thing? I'd still like to avoid if possible...

Also available in: Atom PDF