search

scipp / esssans

SANS data reduction for the European Spallation Source
https://scipp.github.io/esssans/
BSD 3-Clause "New" or "Revised" License
0 stars 2 forks source link

Should the optional `WavelengthBands` be replaced by a mechanism like `IofQxy` vs `IofQ`? #144

Closed nvaytet closed 5 months ago

nvaytet commented 6 months ago

In the SANS workflow, we currently have an optional WavelengthBands parameter which if left as None will produce a one-dimensional $I(Q)$ result. However, if we define it, the workflow will produce $I(Q, {\rm band})$.

This strongly resembles the issue we had with $I(Q)$ vs $I(Qx, Qy)$.

See also in the powder diffraction workflow where we request a result binned in d-spacing and 2theta (as opposed to just d-spacing).

So instead of just setting the WavelengthBands and magically getting something different at the end, we could also require the user to compute e.g. NormalizedIofQinBands?

SimonHeybrock commented 5 months ago

Wouldn't that mean you need 4 different types? I feel that might be too confusing?

nvaytet commented 5 months ago

ok, I agree. Closing.