Hi all,
there were a lot of discussions about noise in the past. One of the results everybody agrees with is, that you have to get the right amount of integration time to increase the SNR in any way to have good signal to work with in post processing. I remember somebody explaining how he is measuring the SNR in his (combined) images and decide, that taking more subs will barely increase the quality. That was the point he decided to stop taking subs of his target and processed his final image. The problem is, that I can't find the video anymore. But I liked the concept.
I use three setups so far. There is my Celestron EdgeHD 8" which has the worst f-ratio which is f/7 if I'm using the 0,7x reducer. It also has the longest focal length. With the amount of light I get with this scope, I need to expose longer to get the same SNR as lets say with my Explore Scientific 127 ED APO and its 0,7x reducer. I get an f-ratio of f/5,25 this way and may need approximately half of the total integration time (even less) to get the same SNR. The same is true for my TS 350 f/5 Quatruplet scope. The Newtons with an f-ratio of f/4 are even faster. You get the point…
The thing is, that it would be nice to measure the point when I got the "right" amount of images. The SNR graph converges against a value while taking more subs. It increases by the square root of the exposure time while the signal increases linearly. But how am I able to get these measurements (using lets say PixInsight)?
How do you deal with that question? Guessing the time based on images I see here in astrobin is surely not the right way. It works somehow, but I am sure there is a better solution to this.
Thanks for your help!
there were a lot of discussions about noise in the past. One of the results everybody agrees with is, that you have to get the right amount of integration time to increase the SNR in any way to have good signal to work with in post processing. I remember somebody explaining how he is measuring the SNR in his (combined) images and decide, that taking more subs will barely increase the quality. That was the point he decided to stop taking subs of his target and processed his final image. The problem is, that I can't find the video anymore. But I liked the concept.
I use three setups so far. There is my Celestron EdgeHD 8" which has the worst f-ratio which is f/7 if I'm using the 0,7x reducer. It also has the longest focal length. With the amount of light I get with this scope, I need to expose longer to get the same SNR as lets say with my Explore Scientific 127 ED APO and its 0,7x reducer. I get an f-ratio of f/5,25 this way and may need approximately half of the total integration time (even less) to get the same SNR. The same is true for my TS 350 f/5 Quatruplet scope. The Newtons with an f-ratio of f/4 are even faster. You get the point…
The thing is, that it would be nice to measure the point when I got the "right" amount of images. The SNR graph converges against a value while taking more subs. It increases by the square root of the exposure time while the signal increases linearly. But how am I able to get these measurements (using lets say PixInsight)?
How do you deal with that question? Guessing the time based on images I see here in astrobin is surely not the right way. It works somehow, but I am sure there is a better solution to this.
Thanks for your help!
but like i said, theres really no minimum or maximum time, it's when YOU are happy with it. ( im guessing theres technically an upper limit where more data wont improve the signal.. )