you’re right. I has managed to make it’s way into the release already then ;-)
So, that would be the way to do it. Make sure that you have duplicated all diffusion directions with opposing PE-directions and use —resamp=lsr.
> what is then the option --resamp lsr in eddy ?
> For the connectom data their acquired 288*2 raw data but the preprocess data have only 288 volume so I guess they did a pairwise restoration... no ? may be they have a special releas of eddy ?
When the pipeline for the HCP was finalised we didn’t yet have the lsr option, so the HCP data is actually resampled with the jac option and then simply averaged.
> Many thanks