DARKS, FLATS AND BIAS REVISITED - Forums

Topic starter

Dec 12, 2022

Hello everyone. I would like to seek feedback on best practices for darks and flats
Currently I do: 20 darks, 40 flats, 40 dark flats and I don't do bias
What is the right balance?. Does the number of light frames dictates the number of darks and flats?.
Does it make sense to do bias o top of flats and dark flats?.
How often should the library of darks be updated?.
Thank you for your inputs and clear skies.
Alicia

Jonny Bravo

Dec 12, 2022

More is better… to a point. I think James Lamb did a video on YouTube with some statistics…

Anyway, I typically don't take more than 30. I take darks, flats and flat darks. I do not take bias frames because of my camera (294MM Pro).

I keep my darks for months. No set rule for me.

andrea tasselli

Frequent contributor

Dec 12, 2022

Alicia Rossiter:
Hello everyone. I would like to seek feedback on best practices for darks and flats
Currently I do: 20 darks, 40 flats, 40 dark flats and I don't do bias
What is the right balance?. Does the number of light frames dictates the number of darks and flats?.
Does it make sense to do bias o top of flats and dark flats?.
How often should the library of darks be updated?.
Thank you for your inputs and clear skies.
Alicia

Question 1: There is no right balance because there isn't anything to be balanced
Question 2: No
Question 3: No
Question 4: When there is a significant change in the dark current or every 6/12 months or thereabout

Cheerio

Arun H

Frequent contributor

Dec 12, 2022

Alicia:

There was a long thead about this:

https://www.astrobin.com/forum/c/astrophotography/deep-sky/lets-discuss-about-dark-bias-dark-flats/?page=1

Ignore every other post in it, simply read John Hayes's post.

16 darks is more than enough for most people in most circumstances. Same with flats. There is a mathematical logic to this. Taking 40 wont hurt, but is a waste of time. And you should be taking dark flats (or bias if you trust your camera to give you a reliable bias).

Andy Wray

Dec 12, 2022

FWIW: I take 20x darks, 20 x flats and 20 x dark flats for each filter and that seems to do the job. I don't take bias frames anymore.

Well Written Concise

Alicia Rossiter

Topic starter

Dec 12, 2022

Thanks to all who shared their routine
Based on everything I have read, I think I will continue with the 20 darks + the 30 or sometimes 40 flats and dark flats. It has worked for me.
Clear skies

Bill Nichols

Dec 12, 2022

Alicia:

Attached is a link to a presentation, "Basics of Statistics for Astro Imaging" on the Astro Imagining Channel by John Hayes. It's long and detail, but he uses statistics to determine the number of calibration frames. Basically, he states that for every light frame you need 5 calibration frames, for example if you take 40 lights you need 200 calibration frames. There's a lot of good stuff in his presentation and it's very well done.

https://www.youtube.com/watch?v=T_k9B01AeFM

Clear skies,
Bill

Helpful

Karl Theberge

Dec 13, 2022

Well, I had a chat with DarkArchon and also the guy who created NINA and they plainly convinced to stop using darks altogether.

you should maybe give it a shot, your camera is not supposed to produce amp glow so.

Bias and Flat should be enough and the good thing is you dont need a lot of time to take them.

Last night I did 2 set of 25 flats and a set of 100 bias. Took me less than 5 min.

Alicia Rossiter

Topic starter

Dec 14, 2022

Good point about the amp glow. True, the 2600 MCPro has zero amp glow, however my 183MC Pro has awful amp glow, and sometimes it does not go away completely even with the darks. why is the added value of the bias ?. Thanks a lot for the info!.

Jonny Bravo

Dec 14, 2022

You use the biases to calibrate the flats. If you look at the typical (simplified) math done for calibration it looks like this:

(light - dark) / (flat - bias)

The advocates for the "don't take darks" method argue that the IMX571 sensor is so clean, calibrating with darks is not necessary and can actually be detrimental to your image. They will use other techniques like cosmetic correction and/or dithering during acquisition to get rid of the hot pixels that would have been removed by using a master dark. A benefit of not having to take/use darks is as @Karl Theberge wrote: calibration frames require minimal time. Granted, I have quite a few master darks I've taken at various gains/offsets/temperatures/bins. After all, there are far more hours I am not imaging than hours I am

.

My suggestion? Try calibrating your lights using darks, flats and biases... then calibrating your lights using only flats and biases. See the differences for yourself and determine what works best for you.

Helpful Insightful Respectful Engaging

Karl Theberge

Dec 14, 2022

Jonny Bravo:
You use the biases to calibrate the flats. If you look at the typical (simplified) math done for calibration it looks like this:

(light - dark) / (flat - bias)

The advocates for the "don't take darks" method argue that the IMX571 sensor is so clean, calibrating with darks is not necessary and can actually be detrimental to your image. They will use other techniques like cosmetic correction and/or dithering during acquisition to get rid of the hot pixels that would have been removed by using a master dark. A benefit of not having to take/use darks is as @Karl Theberge wrote: calibration frames require minimal time. Granted, I have quite a few master darks I've taken at various gains/offsets/temperatures/bins. After all, there are far more hours I am not imaging than hours I am .

My suggestion? Try calibrating your lights using darks, flats and biases... then calibrating your lights using only flats and biases. See the differences for yourself and determine what works best for you.

I tend to agree with the try and use what is best for you. At the end of the day, if the result is good, and you get same or better using different tool, why not.