A question of time (integration)

[Coriorda]

Am just starting out in the AP jungle, and have yet to obtain a star tracker. As a consequence I need to take short exposures c1s and lots of them so I can stack them .

My question is. Will taking 400  x 1s subs give me to same result as say 2 x 200s subs. Is there a benefit over and above total acculumated  integration time of the longer exposures . ? Many thanks 

[PeterCPC]

Try reading this

[Coriorda]

Thank you, much appreciated and very enlightening 😃

[Aramcheck]

There's also a good talk on the subject by Dr. Robin Glover https://youtu.be/3RH93UvP358?t=56

 

Cheers
Ivor

[vlaiv]

Depends on read noise of your camera and it's relation to other noise sources.

Read noise is only type of noise that adds per exposure and is not time dependent. All other noise sources are time dependent and add like they grow in time.

If camera was with 0 noise had 0e read noise - it would be no difference between 200 x 1s and 1 x 200s. However, when you have read noise, 200 x 1s actually has 200 "doses" of read noise and 1 x 200s has only one "dose" of read noise.

Noises add in quadrature (like Pythagoras' theorem). If you have sides that are vastly different in size - resulting hypotenuse will be almost as long as longer side:

This means that if read noise is much smaller in magnitude than any other noise - resulting sum will be almost the same as that bigger component and read noise will not make much difference. In that case 200 x 1s will be almost the same as 1 x 200s. We can't distinguish by eye noise that is dozen or so percent higher - they look the same to us.

Edited November 8, 2020 by vlaiv
Need to think about what I'm writting :D

[Coriorda]

Thank you, your “Doses” analogy makes perfect sense the way you explain it. Thank you for taking the time to share your insight 😃