Originally posted by pschlute
Not my area of expertise, but the approach I would take is as you suggest...... take two separate exposures and combine in PP software. This will give greater flexibility, not only in changing the framing and sizing of the second exposure, but also in the opacity (selectively if needed)
Thanks , but I want to engage chance

in my process and therefore I want this to be done in camera - not in post.
---------- Post added 08-31-23 at 04:42 PM ----------
Originally posted by biz-engineer
IMO, which is first shot makes no difference in the final result, at least in raw file, because the sensor is linear.
Jpeg may be with the tone curve applied to underexposed file , then use "addition mode" instead of "average" the result could be different, this is something to try.
The question is not really which shot to create first, it is rather a way to be a bit more flexible when shooting these double exposures.
Let's say I want to make a double exposure of a head in silhouette and a shot of a forest (rather common subjects for double exposures). With Canon cameras e g it is possible to shoot a lot of heads in silhouette at home and to keep them on the camera SD-card. Once out in the woods you can choose any of the head shots as the first double exposure picture to show on the LCD as you (in live view) compose the second (forest) shot. Both will be shown superimposed on the LCD. Also if a particular head shot does not really work as you expected you could easily change that before exposing the second shot.
With the K-1 II I can only shoot ONE head shot in advance and then hike to the forest and hope that the actual forest scene will work out with the head shot. Or of course I could bring my head shot model out in the forest and experiment with both exposures consecutively. Fun - yes, but requires more planning