Kevin_A I can test whether or not it would be better to go to 300s exposures from 180s in my bortle 4 sky using my 2600mc pro on my 115mm triplet.
I doubt it would make much difference today with low read-noise CMOS. The long exposures is more meaningful for CCD type read noise.
Basically, when you take an exposure, noise comes primarily from 4 sources. The target itself (in the form of shot noise), the sky background (also shot noise statistics), the dark current noise, and the read noise.
The first three are proportional to exposure time (i.e., longer exposures, more noise). The read noise is only once-per-exposure.
Longer exposures also improve signal of the target. Shot noise is the square root of the signal itself, so any longer exposure will only improve the SNR (of the signal part of the noise terms) by at most the square root of the exposure time.
I.e., the SNR of a 300 second exposure will at most be sqrt( 300/180 ) vs 180 second exposure. And we haven't added the other noise terms. Once we add them, the SNR improvement is very small.
Once the dark current noise integrates (long exposure) to 10 dB over the read noise, there is no reason to use any longer exposures. You can do the rest of the integration from stacking the subframes, and end up with the same resultant SNR.
The reason long exposures improves with CCD is that the read noise is huge. So it takes much longer exposures before dark current noise climbs over the read noise.
With CCD, you want as much exposure as possible per "read." But, for years now, the read noise from CMOS cameras from Sony and Canon has been tiny. And each time Sony comes out with a new sensor, it seems to keep improving.
Sensor improvements has nothing to do with ZWO, it is Sony who is improving it; ZWO is simply adding a microprocessor to control the exposure time and gain, and to download the data. That is also why the QHY cameras have much more flexibility in gain and mode control than ZWO cameras, even when they both use the same Sony sensor. QHY also uses industrial grade sensors for some cameras (one example is the ASI2600 equivalent) instead of the commercial grade sensors that ZWO uses for all of their cameras. It has to do with longetivity, temperature range, and amount of allowable hot pixels -- the chips are sorted, so the ones that meet industrial grade end up in the industrial grade bin, and the rest are sold as commercial grade. In the good old days, 10% carbon composition resistors never fall inside the 5% band, because the 5% ones have been sorted and sold as 5% resistors for more money :-).
If you haven't seen this presentation by Robin Glover, keep replaying it until the implications of the equations sinks in.
https://www.youtube.com/watch?v=3RH93UvP358
They are standard equations like you can see here, but presented so that hobbyists can understand.
https://www.photometrics.com/learn/imaging-topics/signal-to-noise-ratio
https://andor.oxinst.com/learning/view/article/ccd-signal-to-noise-ratio
If you are like me, you would prefer the latter than the YouTube stuff :-).
Now, Bortle 4 is a different story! Your enviable site will have much less shot noise from the sky background. But it does not mean that longer exposures will help, since longer exposure will also increase the sky noise, and thus not change the final SNR.
At my borderline Bortle 5 to Bortle 6 skies (I have Portland downtown to my SE, and Beaverton to my South, but the northern parts is nice and dark), I need narrowband filters to get emmision nebulas, but that does not help with galaxies and reflection nebulas. Even so, I usually never use any more than 300 sec exposures with narrowband.
Chen
