How many calibration frames to light frames

[E_ri_k]

Just a quick question regarding the number of calibration frames to light frames. Just say I have 8x15 minute Light subs. As a minimum, I would imagine it would be acceptable to have 15 Darks, 15 Flats, and 15 Bias, which is what I have been averaging since I started imaging. But would I get a better result, and noticeable difference if to those 8 Light frames, I had say 50 of Dark, Flat and Bias each?

Thanks

Erik

[Shibby]

The number of light frames is actually irrelevant to the number of calibration frames!

The more calibration frames, the better, as the noise within them (and therefore noise added to your final image) will be reduced. This is probably most important with darks as they are long exposure. Until a certain point, where you eventually get diminishing returns for adding more subs.

[RobH]

Shibby's right.....too few calibration frames and you'll add noise.

As bias frames will be good for 6 months or so, and are quick to take, I make a master bias from 100 frames.

With darks and flats, 30 each is fine, and you'll be able to use your darks for a good while too. It's only the flats that need to be done again after each change to the optical train.

Rob

[steppenwolf]

I always take 20 of each. I seem to recall that it is a case of mathematical diminishing returns after the magic number 16?

[E_ri_k]

Wow, ok thanks guys. 100 frames is alot! I am as we "speak" processing 9 lights with 50 bias 30 Flats and 9 Darks. I'll see how it goes. I guess it makes sense with it being irrelevant of how many light frames you have, because the signal is always going to be mostly consistent, and not random like the noise.