Ticket #4855 (closed: fixed)
DiffractionFocussing - Incorrect answers for PEARL with parallelization enabled
Reported by: | Martyn Gigg | Owned by: | Martyn Gigg |
---|---|---|---|
Priority: | major | Milestone: | Release 2.0.2 |
Component: | Mantid | Keywords: | |
Cc: | Blocked By: | ||
Blocking: | Tester: | Roman Tolchenov |
Description (last modified by Martyn Gigg) (diff)
Matt Tucker on PEARL has found that the his reduction gives different answers between the September '11 release and Feb '12.
I have tracked this down to DiffractionFocussing and if I turn off the parallelization then it seems to give the correct answers.
LoadRaw(Filename=r'PEARL00073987.raw',OutputWorkspace='work',LoadLogFiles='0') ConvertUnits(InputWorkspace='work',OutputWorkspace='work',Target='Wavelength') LoadRaw(Filename=r'PEARL00073987.raw',OutputWorkspace='monitor73987',LoadLogFiles='0',SpectrumMax='1') ConvertUnits(InputWorkspace='monitor73987',OutputWorkspace='monitor73987',Target='Wavelength') CropWorkspace(InputWorkspace='monitor73987',OutputWorkspace='monitor73987',XMin='0.029999999999999999',XMax='6') MaskBins(InputWorkspace='monitor73987',OutputWorkspace='monitor73987',XMin='3.4500000000000002',XMax='3.7000000000000002') MaskBins(InputWorkspace='monitor73987',OutputWorkspace='monitor73987',XMin='2.96',XMax='3.2000000000000002') MaskBins(InputWorkspace='monitor73987',OutputWorkspace='monitor73987',XMin='2.1000000000000001',XMax='2.2599999999999998') MaskBins(InputWorkspace='monitor73987',OutputWorkspace='monitor73987',XMin='1.73',XMax='1.98') SplineBackground(InputWorkspace='monitor73987',OutputWorkspace='monitor73987',NCoeff='20') NormaliseToMonitor(InputWorkspace='work',OutputWorkspace='work',MonitorWorkspace='monitor73987',IntegrationRangeMin='0.59999999999999998',IntegrationRangeMax='5') ConvertUnits(InputWorkspace='work',OutputWorkspace='work',Target='TOF') Rebin(InputWorkspace='work',OutputWorkspace='work',Params='1500,-0.0006,19900') AlignDetectors(InputWorkspace='work',OutputWorkspace='work',CalibrationFile=r'pearl_offset_11_4.cal') DiffractionFocussing(InputWorkspace='work',OutputWorkspace='focus',GroupingFileName=r'pearl_group_11_2_TT88.cal')
The script above should give the answer in the attached NeXus file. It was produced with the release from September.
Attachments
Change History
comment:3 Changed 9 years ago by Martyn Gigg
Refs #4855. Fix DiffractionFocussing2 race condition.
This only affected workspaces that had no masked bins, like in the PEARL reduction.
Changeset: 013b8491ad40e3ce7b44bceca8871d7f2e3dd869
comment:4 Changed 9 years ago by Martyn Gigg
Refs #4855. Add back the protection for the Mac.
Changeset: f03f3075ca5dbfd9adc96aef436457f366ccc0be
comment:5 Changed 9 years ago by Martyn Gigg
- Status changed from accepted to verify
- Resolution set to fixed
comment:6 Changed 9 years ago by Martyn Gigg
Refs #4855. Add system test for PEARL reduction.
Changeset: 575579e4dbf1d61e064a9feb606acec3e6769c51
comment:8 Changed 9 years ago by Martyn Gigg
Backport the DiffractionFocussing race-condition fix.
Refs #4855. Fix DiffractionFocussing2 race condition.
This only affected workspaces that had no masked bins, like in the PEARL reduction.
Refs #4855. Add back the protection for the Mac. (cherry picked from 013b8491ad40e3ce7b44bceca8871d7f2e3dd869 and f03f3075ca5dbfd9adc96aef436457f366ccc0be)
Changeset: f1632bf3e671a5820c70175a2da380d0e8937a68
comment:9 Changed 9 years ago by Roman Tolchenov
- Status changed from verify to verifying
- Tester set to Roman Tolchenov
comment:10 Changed 9 years ago by Roman Tolchenov
- Status changed from verifying to closed
Ran the script and compared the result with the answer file. The differences are within 3e-20.
comment:11 Changed 5 years ago by Stuart Campbell
This ticket has been transferred to github issue 5702