Thanks! I didn't appreciate that a smaller box would have more weight, not less. My HDR assumption was that this image has a huge dynamic range but OTOH I suppose you could say that about any image with a specular highlight and a shadow. 

Hi Mike- thanks! I did the same with the magenta box, something similar with the yellow, but what's the thought process behind the white box? Also, what determines if an image is a candidate for HDR?

Just for entertainment, here's a RAW from my Z6 that has some nice highlights in the center that need to be recovered, and a tree with a shadow area that needs to be lightened up. I was shooting more or less into the sun and it just has to qualify as a good subject for HDR. It was the file I started having trouble with. My preferred RAW settings are no sharpening, no equalization, no noise filters, no smartcolor, no auto fill (but need that on so save works) and default everything else. I find the Z6 images need very little processing at that stage; everything else I like to mess with on an image-to-image basis.

http://www.conradhoffman.com/rawtest/DSC_1186.NEF

Enjoy,
CH

Getting closer to the bottom of this. Everything was working correctly until I went into preferences and put my RAW processing choices back to where I like them. Then the problem showed right back up. I'll investigate a bit further and determine which setting is culprit.

edit- I've been wrong before (just ask my wife!) but it appears that auto fill light must be turned on in RAW processing for fill to be saved in RAW refine and for HDR to work correctly. Is it a feature or a bug? 

The root of the problem is that the fill setting is never being saved. I've disabled pretty much everything on startup and the only virus program is Windows Defender. I've installed the latest version, also tried installation and launch as administrator. The rest of the raw refinements do seem to be saved. If it's a clue, the fill setting is never reflected in the thumbnail appearance. All suggestions gratefully accepted!

edit- if I hold shift while starting, everything works as it should, and continues to work correctly on subsequent launching, even if I don't hold shift. Perhaps I'm too old to be using computers?

Thanks,
Conrad

OK, I haven't used this much, but the controls aren't making a lot of sense to me. I start out setting exposure with one of the 9 squares. Then I use the fill and HDR sliders. Every time I touch the HDR slider, it reverts the fill back to zero. When I get what I think I want, and save it, the same image opened in the editor doesn't seem to have my adjustments. Nor are they preserved if I try to refine again. What am I doing wrong?

Thanks,
Conrad

My cardiologist says, don't even think stuff like that! 

I see this most with overly compressed video conference images, but I also see it in details of my photos that are overly processed. I'm just curious if there's any particular aspect of sharpening that makes it better or worse. Yes, I can just try things, but I want to know if anybody can give me more depth of understanding as to why it happens.

There's another way to do this. You could tell her about this fantastic printing program that doesn't cost an arm and a leg, that would improve everything she prints, handling color management and printer settings better than anything out there. IMO, it should be illegal to own a printer without it. Just saying.

I don't scan many color negs but I believe the effect of the orange mask with traditional wet process printing is quite complex. A simple "flat" correction doesn't accomplish the same thing. The color neg setting on my Epson scanner works well, better than I've been able to do by trying to  deal with the orange mask myself, though I don't know how sophisticated it is. Probably more than me!

edit- here's a better description of how the mask works: https://www.photo.net/discuss/threads/color-negative-mask-questions-and-confusions.254896/
Turns out it's two masks and not a simple filter.

Well, FWIW, I tried flipping the curve and doing all my editing in Qimage, modifying the curve slightly, applying DFS and cropping. The results were better than what I got using multiple programs. I'm mostly working with medium format black and white negatives I shot in the early '70s in high school. My scanner (Epson 4490) lets me select negative film and does the flip for me, but the results seem better if I stick with the positive film setting and work with the resulting negative image in Qimage. Tests with a step tablet confirm better range and noise. I start with a very large 16-bit tiff image and export the result as a down-sampled jpg. 

Ha! I'm an idiot- that's great, thanks!

Maybe too far out of the Qimage wheelhouse, but I've been scanning some black and white negatives and flipping them to positive in other software. Just curious- is there any way to do that in Qimage?

+1 on the divide and conquer approach. That's always worked for me when I've had the odd file with a problem.

Two words Fred, paper cutter! Then just print 5x7s. Maybe I'm too old school?