Capture One and Fuji GFX 50S focus stacking

I received the following query from a reader:

I wonder whether you can give me a hint how to tackle an issue I ran into. Quite some people are using Capture One with their GFX but I have not heard yet of anyone using focus stacking with the TIFFs coming out of Capture One. 

I use Capture One as my main RAW converter. Importing DNGs with changing some metadata fields via ExifTools does the job. And the resulting TIFF files are marvelous. But when I focus stack, I get an error message in Helicon Focus stating that the file size is invalid. I tried Zerene Stacker and their error message is much clearer: there is a difference in file size between the files.

I checked this in the Finder of my Mac and the files indeed differ a bit in size. This is both with lossless and uncompressed RAF files used as input for Adobe DNG Converter. As a test, I then used Iridient Developer with the exact same DNGs, with the metadata adjusted just like I use with Capture One. And the resulting TIFFs are of the exact same size. So it is a Capture One issue. As an extra test, I converted my NEF files to DNG and made the exact same metadata changes as with the files for the GFX (so brand and model was changed to Phase One, IQ250). The rendered TIFFs are of the same size. So DNG in itself looks like not to be an issue for Capture One.

I suspected lens distortion correction and asked the reader to see if he could turn it off and report what happened. The next thing I heard was:

I just saw that Distortion Correction is set to the maximum. I will set that back to zero and see if that makes a change. Diffraction correction is in the lens tab, I left it activated when I tested it.

It worked!  With both V10 and V11.

I’m posting this in case other people have had the problem. I’m expecting that more folks will be stacking with the GFX now that progressive focus shift is in the firmware. I think that what is going on here is that the distortion correction is varying with subject distance, and thus the size of the output files. If that’s the case, I don’t think you want distortion correction on when you stack anyway, as any variation in the distortion correction within the stack will make things harder for the stacking program.