Inaccurate values near the end of the Q range

[Steve Williams]

Email from Richard.

Again Mantid has an extra bin at smallest Q ( 5ang next to beam stop) and though in the middle Colette (dots) agrees well with Mantid there are systematic differences at both the lowest and highest Q’s.  

Need to see whether there are pixels and/or wavelength bins at the stop & start of the ranges that are being over-intepolated rather than ignored. Perhaps we need to “round in” rather than “round out” in some places for safety in Mantid?

[Steve Williams]

A Mantid release in Apr 2010 (version 1.0.4241) showed points towards the end of the range away from the main curve but normally within the errorbar.

[Steve Williams]

(In [9950]) SANS masking at ISIS now always assumes the beam centre is at the centre of the detector re #2612

[Steve Williams]

(In [9974]) Run numbers entered into a RunFile widget ending in "-add" will cause a file whose name ends in -add to be loaded. Fix system tests re #2612 and #2337

[Steve Williams]

(In [10306]) Flatbackground now works for non-distribution data and SANS Display Mask is now clearer re #2612 fixes #2599

[Steve Williams]

(In [10307]) Better Display Mask in ISIS SANS re #2612

[Steve Williams]

(In [10309]) DoxyGen spelling mistake fix re #2612

[Steve Williams]

(In [10408]) Ensure it runs the ends of the loop re #2612

[Steve Williams]

(In [10482]) The GUI always overwrites settings from the user file re #2612

[Steve Williams]

The spurious counts at the low Q range of LOQ experiments have gone now because the mask is in the correct place

[Michael Reuter]

The spurious dropping points at low Q are no longer present, so the fix worked.

[Stuart Campbell]

This ticket has been transferred to github issue ​3459