Ok I believe I found the issue and pushed a fix. It was due to a missing global settings function in dark library. Should be in 3.6.3 stable soon.

git pulled and tested, still as described earlier. The 1x1bin works, the 2x2bin doesn't find a suitable dark. Log attached. Maybe it's due to the fact that the number of pixels in x or y are not exactly 1/2 of the 1x1 frame? Logfile is OK now.

The second to last picture taken is the 1x1bin that is successful:
[2023-03-02T23:41:06.538 CET DEBG ][ org.kde.kstars.ekos] - Dark frame subtraction applied

The last picture taken is the 2x2bin that cannot find a suitable dark:
[2023-03-02T23:41:25.633 CET INFO ][ org.kde.kstars.ekos.capture] - "No suitable dark frames or defect maps found. Please run the Dark Library wizard in Capture module."

What I did is create two master darks: 5 exposures a 1s, ASI294MC, 1x1bin, gain200 and
5 exposures a 1s, ASI294MC, 2x2bin, gain200.
Both master darks are successfully acquired and listed in the "masters" tab.

I then took two preview shots:
1s, gain200, 1x1bin: successful.
1s, gain200, 2x2bin: failed to find suitable dark.

Funny, I tested again using the same master darks. Now taking 2x2bin shots is reported as successful and 1x1bin shots fail!

For me it seems to work, at least in dry run with telescope simulation.

But I found another bug in the darks wizard, sorry. I will create a new topic on this.


P.S.: Created a new library and first 1x1bin worked, 2x2bin not, despite setting temperature range to 30°C (which is maximum)... I had to restart and both worked. For the moment...

Ok I pushed another fix to master, can you please check now and report back?

Thanks Jasem for looking into this! Unfortunately I get the same result as before.

Well, just as Jurgen reported, once I started Ekos AGAIN, it worked for both bin modes!

Thanks Alfred, I pushed another fix. Can you please check now?

Jasem, the previous version did work fine from the second start of Ekos onward.

I just compiled the latest one and there is one thing that I noticed:
The first exposure I take looks like no dark was applied (judging from what I see in Fitsviewer). The 2nd, 3rd, etc. look very differently, clearly a dark has been applied to these.
When I change to a different bin mode, the same happens again: First exposure looks like there was no dark applied (although no "no suitable dark found" message is displayed) and all following ones clearly have darks applied.

Have to go now, will be back later tonight.

What currently happens is this: After starting Ekos, the first exposure is displayed without dark substraction, regardless of settings. No error message is displayed. All subsequent exposures in the same bin mode are shown correctly with dark substraction applied. Once you change the bin mode, the first exposure will be shown without dark substraction again with all subsequent exposures shown correctly. If you then go back to the first bin mode, the first picture is OK.

So it appears the check whether or not the "darks" box is ticked happens only after the first picture has been displayed. This happens in every bin mode separately.

I can't reproduce this behavior. Tested just now with fresh data and subtraction is working from the first time it is toggled.