Fuji RGB Scan merging weirdly

[bbatha]

I have tried two RGB scans starting from RAFs. They both have a huge bias towards the red channel. The images also exhibit square artifacts giving a screen door pattern. The process menu also chooses e6 or B&W rather than c41.

These were taken with a jackw scanlight

The 35mm scan was using ultramax 400 and the other scan was portra 400.

[bbatha]

I can share the raws with you if you can suggest a free service to host them.

[marcinz606]

it looks like it's not correctly merging the scans and only picking the red channel

[marcinz606]

https://wetransfer.com/ or google drive

[bbatha]

https://drive.google.com/file/d/1y_abLedto2ia7MYcxWsFAXXtpxKcptJQ/view?usp=drive_link i also included a successful inversion of a white light scan of the second image

[marcinz606]

merges on my end, important part is that you have to have the RGB Scan mode enabled before you load the files. I'll see if i can change it so that it reloads the files if toggled after files are already loaded.

I found small issue with fuji .raf files though when it comes to possible colorcasts on non-hq preview though, will try to fix it

[marcinz606]

should be fixed with 0.29.1

[bbatha]

I don't think the default setting was the issue. I had it turned on at the time. And I'm still seeing the issue with 0.29.1

[marcinz606]

you need to crop, having so much white space outside totally throws off the analysis

[cimelab]

I'm having a few similar issues on Mac.

1. I have to import one set at a time otherwise it randomly merges files from different images. Is there a specific naming or structure requirement? I thought it would just take groups of three in order
2. When getting a successful conversion, moving to another image and coming back may cause it to change.
3. previews don't update
4. When exporting I'm getting unmergerd jpegs. Workaround has been to hit the save before exporting
5. my preview images are never in order no matter what sorting option I choose