Every sample of Fujifilm GFX100 16-shot multi-shot mode I’ve seen so far (quite a few) shows that either (a) the results are out-of-register garbage, or (b) the results offer absolutely no increase in sharpness and usually have some problems too.
Why did Fujifilm implement a feature that appears to be no better than a science fair project when a simple 4-shot pixel shift feature (whole-pixels only) would have a far better chance of success, as with Sony and Pentax.
I’d not have objected to the 16-shot feature by itself, if only it worked at least as well as the multi-shot mode on the Olympus E-M1 Mark II, which can deliver stunningly good results. And if the 4-shot mode were offered, which by capturing a full RGGB value at each pixel could have extended the image quality into best in the world. Oh well...
Fujifilm even explicitly calls out that 4 shots are taken for “recording accurate RGB information”. This is precisely what Sony does, and it works great for static subjects. But then Fujifilm needlessly tacks on 3 more iterations of that process, in 1/2 pixel steps.