Histogram, HFR, other stats are not standardized and are representing the PREVIEW and not the actual data.

[Ron Kramer]

This has bothered me for a while. Rather than piling more on your plates, I figured it would be addressed eventually. But today someone else mentioned it in the Facebook group. So I figured I should pass it on.
I suspect? this is also the case for the recommended exposure wizard… it’s likely using the stretch settings per user rather than actual image raw data.
HFR also is changed and history for HFR jumps when the user changes their 2 stretch settings.

[Simon Kapadia]

HFR explicitly uses the stretch. If it was to calculate on unstretched data there may well be very few visible stars to calculate it on, especially for narrowband images. For a colour camera you can choose whether to use the debayered image for HFR or not, but that’s really all you need. Remember that the purpose of this metric is just so you can get a relative (to each other) idea of how the imaging is going.

[Stefan B]

The change in statistics from loading a previous file is most likely caused by a bug in the FITS reader when loading in FITS files. I am already aware of it, but it won’t affect any actual image taking.

[Stefan B]

As this issue is a bit confusing so here is the summary:

• Histogram is based on the RAW data as is the rest like standard deviation mean etc
• HFR and Star Detection are using a STRETCHED image that is based in the image options as star detection on raw data will not work and not detect anything.
• Reading FITS files currently has a small bug, where values get shifted to the right and showing different statistics than they should have

‌

[Stefan B]

Fix for reading the FITS file will be available in 1.10 RC3

[Stefan B]

Closing the issue, as the fits issue will be adressed and the other one regarding hfr/star count is working as designed.