Calibration preview in 1.0.75 not working?

I think the first image shows that your light frame did not have the calibration frames applied. The first column is missing MB, MD and MF that would indicate the bias, darks and flats calibration frames are applied. Why they weren't applied is another question.

Yes indeed, I still haven't figured it out. It works perfectly in 1.074.1 with the same files. I don't even need to explicitly run the Calibration step, just load the files and view them, it's awesome. But 1.075 behaves very differently. I assume the rules for matching calibration to light frames are a little different now.

I re-ran the test in both versions with just 1 light frame, 1 flat and 1 bias. I've attached the console output when previewing the calibrated frame.

What's interesting is that when I explicitly run Calibration in 1.075 I see this:

16:13:59 - 2) CALIBRATE: assigning masters to light frames...
16:13:59 - 2) CALIBRATE: found master calibration frames for frame: 2019-10-21_21-52-36_-0.30_180.00s_200.00g_2x2bin_1.fits

But when I preview the light frame then I see this:

16:15:58 - IMAGE VIEWER: loading into Linear Calibrated Image Loader: D:\AstroProjects\19.10.18 M27 [180s 200 bin2]\Lights\Group1\19.10.22 M27\2019-10-21_21-52-36_-0.30_180.00s_200.00g_2x2bin_1.fits
16:15:58 - 2) CALIBRATE: VERIFY CALIBRATION MASTERS: all frames are verified
16:15:58 - FRAME DETAILS UPDATER: starting...
16:15:58 - FRAME DETAILS UPDATER: no new frames to add
16:15:58 - FRAME DETAILS UPDATER: rebuilding all frame details...
16:15:58 - FRAME DETAILS UPDATER: checking if frames were identified earlier...
16:15:58 - FRAME DETAILS UPDATER: adding frame marks...
16:15:58 - FRAME DETAILS UPDATER: updated succesfully
16:15:58 - 2) CALIBRATE: trying to find calibration details of light frame: D:\AstroProjects\19.10.18 M27 [180s 200 bin2]\Lights\Group1\19.10.22 M27\2019-10-21_21-52-36_-0.30_180.00s_200.00g_2x2bin_1.fits
16:15:58 - 2) CALIBRATE: DID NOT FIND calibration masters for light frame: D:\AstroProjects\19.10.18 M27 [180s 200 bin2]\Lights\Group1\19.10.22 M27\2019-10-21_21-52-36_-0.30_180.00s_200.00g_2x2bin_1.fits

If anyone wants to try it out, here are the 3 files (light, flat and bias)

Interesting, I'll download them and have a look, thanks. Might it have to do with the light being binned x2? I'm not sure if the calibration files then match for the preview to work.

Only thing I see is that the calibration masters are 32-bit and your light is 16-bit. Tried that but also didn't work, so I'm now thinking a setting in the calibration data is wrong, I probably am overlooking something obvious so I'll add Mabula's eyes to this one. 😉

Hi @cshine, @astrogee, @vincent-mod,

The l-calibrated image viewer and calibration frames to light matching has not changed in 1.075 relative to 1.074.1.

Something else is causing this behaviour and it as the adjustment of APP's support for Bayer CFA sensors with uneven dimensions. Which is the case here.

What is happening is that for such a sensor, the masters created with APP 1.074.1 don't work in APP 1.075. Therefor the l-calibrated image viewer mode is not showing that the masters are applied to the light frames and so you will not see a calibrated image.

If you can create the masters again in 1.075 from the original calibration frames, please try that, then those masters should really work and you should see the l-calibrated preview 😉

If you no longer have the original calibration frames, please calibrate your frames in 1.074.1 and save them in 2) Calibrated. Then the calibrated frames are then fine for further processing in 1.075.

Apologies for this inconvenience, but as the changes are implemented now for better support for uneven Bayer CFA sensors, this incompatibility should no longer happen in future versions relative to APP 1.075.

Kind regards,

Mabula