Is MNI space voxel data good enough for fsaverage?

Hi all! I’m trying to redo some analysis from my lab from volume based analysis to surface based. I’m not super familiar, so please pardon my lack of knowledge here. Thank you.

I have some T1 anatomy images from different subjects, and corresponding beta maps in the functional space. I need these ultimately to be in the same space for group analysis on the beta values. For a quick first pass, I plan to just convert the beta maps to mni space using the registration provided by flirt and use nilearn’s vol_2_surf to sample it on fsaverage. Is this generally ok? I assume there would be loss in quality, otherwise people won’t be doing recon-all which takes forever, but how much worse is it in your experience?