TheSkySearchers

My buddy Patrick regularly does 1800-second sub while shooting 3nm narrowband imaging. And he's not shooting with a really slow system. He just claims that he get's fainter details by going longer. He's always giving me a hard time because I've always limited myself to 600 second subs.
Last night, I decided to shoot my current subject at 600 sec and at 900 sec and compare the two in PixInsight. Other than a slightly brighter image, I didn't really see much difference in the fainter regions.
I know this is probably not the way to compare the two images but here's what I did. I opened both in PixInsight, Did a STF on both, copied the STF to HT and then exported each as PNG's. That's what you are looking at in this video. It starts with the full sensor capture and then zooms in on two different regions, one bright and one darker.
Tell me what you think. First, if this is the correct/best way to compare the two images, and second, do you think it's worth it to give up the total # of subs per evening by going longer? Figuring at this time of year I have about 6 hours of good imaging time between the end of Astronomical twilight and the beginning of astronomical dawn. So that's a difference between capturing 36 subs vs 24 subs.
BTW, it's a 4K clip if you have a decent 4K monitor to look at it.

I don't think there is an advantage to longer subs just for the sake of having longer subs. The difference between 24 subs and 36 subs is not significant. What matters most is the total integration time. In most cases, six hours will be six hours no matter how you divide it up.

The only reason to take longer subs is to get faint details out of the read noise threshold. If you are finding that you cannot get the faint details to show through the noise, then you might experiment with longer subs. If your faint details are fine, then longer subs just give more opportunities for losing integration time to problems like aircraft, cable snags, etc.

I take 300s subs for luminance, 600s for Ha and 900s for Oiii. My optics are fast (f/4), but it's an old camera and fairly noisy compared to more modern cameras.

I wouldn't use an automatic tool like STF when doing statistical comparisons on subs. STF, at least the way most people use it, uses different stretch parameters on every image. So you will be comparing apples to oranges.

Thanks for your input Kathy.
What would be a better way of comparing apples to apples?

AstroBee wrote: ↑Sat Jul 15, 2023 9:47 pm Re: Anyone doing really long exposure times?

Yeah, you!

Mine rarely go above 300. As long as there's a gap between the left side of the histogram and the zero of the X axis, I start capturing. Wasted shorter subs due to satellites etc are a smaller proportion of the total.

But I'm still interested in seeing how both your comparison experiment and your 900 second project goes.

Graeme

I used to do 20min subs years ago when I had a Skywatcher NEQ6, but I swapped that for an HEQ5 as lifting the NEQ6 regularly for astro camping was a bit heavy for me. The HEQ5 doesn't seem to be able to go longer than 600 secs without getting elongated stars, so I don't do that any more.

I think personally the only value in doing longer subs than 600secs is if the target is very very dim, such as perhaps the Squid and such like.

Carole

AstroBee wrote: ↑Sat Jul 15, 2023 11:08 pm Thanks for your input Kathy.
What would be a better way of comparing apples to apples?

One way (still not the best) is to use STF on one of the subs and then transfer that value to the other subs. That way the same stretch is applied to all (the stretch from the first sub that is). You can do this by applying the STF to HT and then use HT on the other subs.

I have found that 300sec seems like a good value for LRGB. Since I use 3nm NB filters, I go with 600sec with them. Haven't tried 900 on the Oiii as Kathy mentioned (I may try that on my Bubble Nebula). Since I'm using an AP1100 mount, long exposure tracking is not the problem. It's the planes and various satellites (and clouds).

Our team shoots with 15 minute subs, but we have a 24" CDK and a 16803 camera. One day we will switch to CMOS and our sub lengths will decrease accordingly.

Maybe your friend is using low-sensitivity sensors that benefit more from longer exposures than yours. Unlikely, I know - if they have a mount that can accurately track for 1/2 hour they probably use sensitive capture equipment.

SparWeb wrote: ↑Tue Aug 22, 2023 3:44 am Maybe your friend is using low-sensitivity sensors that benefit more from longer exposures than yours. Unlikely, I know - if they have a mount that can accurately track for 1/2 hour they probably use sensitive capture equipment.

Nope, ZWO 6200. Basically the same sensor as my 2600 except full frame instead of APS-C.

Interesting, not my area of work - the max I go for on my AVX is 300 seconds - I'm pushing the mount at that, also on an OSC the fainter stuff is harder to manage as you have not got the inherent sensitivity of mono cameras. To me, the 900-sec subs seem to cover up the detail more than show it as the brighter areas bloom a bit.

My two penneth

For me, it's the inherent low-level LP. Even though I live in a relatively dark area, I still have to contend with LP. I can't go any longer than around 600 seconds on my OSC because the LP noise floor rises above the faint detail and so it gets lost.

I tried 30min exposures of the Horsehead in Ha, only to see if a) I could pull it off with my Mach 1 mount, and b) do it in a ridiculously light-polluted neighborhood. There were nasty artifacts associated with overexposure of Alnitak, which could be managed with a starless workflow. Six hours total, Stellarvue SV110 and an SBIG STT8300M camera:--

DEnc wrote: ↑Tue Aug 22, 2023 8:25 pm I tried 30min exposures of the Horsehead in Ha, only to see if a) I could pull it off with my Mach 1 mount, and b) do it in a ridiculously light-polluted neighborhood. There were nasty artifacts associated with overexposure of Alnitak, which could be managed with a starless workflow. Six hours total, Stellarvue SV110 and an SBIG STT8300M camera:--

Horsehead.jpg

Yes, I would be afraid to try 30 min with a target like the Horse head, knowing Alnitak is going to blow out.

I instigated this question because of shooting the Flaming Skull Nebula which did not have any bright stars in the FOV.

AstroBee wrote: ↑Tue Aug 22, 2023 9:45 am

SparWeb wrote: ↑Tue Aug 22, 2023 3:44 am Maybe your friend is using low-sensitivity sensors that benefit more from longer exposures than yours. Unlikely, I know - if they have a mount that can accurately track for 1/2 hour they probably use sensitive capture equipment.

Nope, ZWO 6200. Basically the same sensor as my 2600 except full frame instead of APS-C.

I thought the idea of CMOS cameras was to do much shorter subs but lots of them.

I only do 600 on Ha or Lum, Oiii, Sii and RGB I only do 300secs RGB binned, or even 150secs binned as at home (Bortle 8) the LP is too great to do any longer and the sensor gets whited out). If I am at a Bortle 4 campsite I might do 300secs binned RGB.

Carole

It looks like you managed to pull it off. I like the black and white.