I'm using the following macro to output data for Rietveld plots from surface refinements.
xdd_out file load out_record out_fmt out_eqn
" %11.6f " = X;
" %11.6f " = Yobs;
" %11.6f " = Ycalc;
" %11.6f\n" = Yobs-Ycalc;
The data I'm working with are from a large area detector and are binned with a very small step size which I've matched in the calculation step. Plotting the raw data they look beautifully smooth, but when I plot Xobs from the output file (which I expected to be the same thing) it is full of noise. I've remedied this by rebinning the data at the start of each xdd in the inp file using "rebin_with_dx_of", but I don't see why this is necessary. Clearly X in the macro has been altered from the raw data, but why? Is this due to the "rebinning with equal step size" that happens at the start of some large refinements?