Superbias

Roger Renberg

Hello Adam, 

New member here and love your courses! 

I was wondering if you could elaborate some more on why a superbias should be avoided if possible ? What are the pitfalls ? ? Everywhere else it says we should make a superbias. So would be great to get some details on that subject 

Kind regards

Roger

BlockHead

Hi Roger,

This is because for some sensors it causes problems in poor calibration of data. Thus the benefit, which is largely difficult to detect is not worth its use for any trouble that may ensue. It seems simpler to take 50 biases and call it good. You can always take 50 biases.... make a superbias and calibrate both ways. Can you tell a difference? Can you measure a difference. This is easy enough to test/compare.

Because I cannot possibly use all sensors that everyone has, I can only go on my experience and what other people smarter than me write. I have found a fixed pattern can be seen in calibrated data that matches the superbias. This is unacceptable to me. 

See Bernd's post:

https://pixinsight.com/forum/index.php?threads/superbias.12444/

and

https://pixinsight.com/forum/index.php?threads/calibration-integration-gave-birth-to-baby-stars.13012/

-the Blockhead