I see no diference in taking more data

[kindred swallow]

some time ago i got 17h on Caldwell 4 and this week I added more making it almost 30h but after procesing i couldn't get more out of it than erlier it's either my sky class or my procesing, and i hope it's the seckond one.

https://drive.google.com/drive/folders/1ILOfh9YlqJY72aECgDDmFFd8CwI6mnmr?usp=sharing

there is my data and my two processed images(guess witch is witch)

i used siril and graxpert on both

[marble flame]

which camera and scope did you use?

[pliant solstice]

it's a game of quickly diminishing returns.... 30h vs 17h only makes for an expected factor in SNR of 1.3

[kindred swallow]

Canon 450D
SW 150/750 but with a Barlow making it a 150/1740 so almost f/12

[kindred swallow]

Shouldn't it be almost cut in half?

[pliant solstice]

nope signal to noise (which is really all that matters in astrophotography) scales with the square root of exposure time

so sqrt(30/17)=1.3

I would argue that the slight improvement is visible in your images, but given that the processing likely isn't identical (different color choice) it can be pretty hard to tell

[kindred swallow]

Would you like to give it a go? I'm curious how much is there to pull out

[pliant solstice]

I can try, but I am better at statistics (part of the job) than AP processing 😉

I'm gonna try and remember this when I'm back home and have a GPU at hand.... 🙈

[kindred swallow]

Alright, no pressure

[marble flame]

i'm also processing it right now :)

@kindred swallow i did a thing

i linearfitted the images, and extracted their L* components, and combined them in a new color image, where the L* from the 30 hrs is put in red, and the L* from 17 hrs is put in green

and that image is part of the histogram, showing that the 30hrs is in fact less noisy, but not very much difference at all

[kindred swallow]

So I need more data?

[marble flame]

idk that's for you to decide

but i just finished to identical processings of both images

they're stretched to a point where the noise becomes much more apparent than you would make it for aesthetic purposes, but if you switch between them, you can see the difference

[pliant solstice]

with a barlow the optics probably suffer and the F-ratio is really painful, I would move on to another target, but that is a very personal decision

[marble flame]

i also think you may have had a problem with flats for the first part of the data

[pliant solstice]

but yeah @marble flame's side-by-side is convincing, no?

[marble flame]

even for this target, i would also prefer to not use the barlow

it would both give more FOV, and increase snr to the equivalent of 2x2 binning (witch i would do to this data anyways if i were to make an aesthetic process)

or maybe not, but i would definitely consider it

[white cipher]

the raw 30 hours also is less greenish

so in fact you should get better collors on it

[white cipher]

also i think in my process i can see a clear diference faight spots

[white cipher]

this one is def the 30 hours

look at the less greens

also 17hours one seens to have more dust spots

[marble flame]

I think that's mainly a flat issue but idk

Same with the dust spots

[white cipher]

i dont think so

[white cipher]

yeah this one defnetly

[marble flame]

I think so, because I used subtraction when I should've probably used division

[white cipher]

thats the 17 hours

no wait a sec

[kindred swallow]

My flats don't really work for some reason and the dust spots you can see there are mostly defects on my Barlow, the anti-reflection coating is crappy

[marble flame]

So you didn't use flats at all?

[white cipher]

thats the 17 hours

[kindred swallow]

I do but they don't seem to do much

[white cipher]

and thats the 30 hours

[marble flame]

There is also some vignetting that should've been removed my flats

[white cipher]

see how much i could get the less bright areas without exploding the core?

[kindred swallow]

I use masks in gimp to separately controll the core and outer stuff

[white cipher]

its a small diference for 15 hours of more data but plays a big role

[white cipher]

this was siriL only

no masks

siriL and graxpert

[kindred swallow]

It's so annoying, you can see hints of stuff being there but it's just so close to just being noise

[white cipher]

yeah

i had to play graxpert 2 times to get the better details

but seens way better than the 17 hours

[marble flame]

Get 30 hours more 👍

[kindred swallow]

But when 😭

[marble flame]

I'm wondering that myself too

[white cipher]

hahahaha i'll wait for this day

[kindred swallow]

My nights are 4h long and getting shorter

[marble flame]

Same

[white cipher]

my nights get 1 hours per night when i get a good night

i normaly get 1 good night per week

in a b9 zone

so thats simply bad

[kindred swallow]

In peak summer there is a month when the sun never goes below horizon far enough to get any astronomical night

But I'm planning on buying the minicam 8 so I should get muuuch more signal out of even those shor nights

[marble flame]

Once you go mono, you'll never want to go back

[kindred swallow]

i already want to go there 😆

[white cipher]

me too🤣

[thorn vault]

You just hit your rigs SNR limit. This happened to me too when I was first using my canon. I shot ngc 6888 for 18 hours and I didn't see a slight change. To see any improvement you have to change gear: modify your camera or buy a dedicated one; buy filters, narrowband or broadband. Shooting from less poluted skies is always gonna help.
Anyways, as someone mentioned to you earlier, SNR depends on the square of exposure time, so you get the idea, you would have to shoot for a gazillion years to see improvement.

[marble flame]

SNR limit isn't a thing if you're brave enough

[thorn vault]

That's your squid data?

[marble flame]

Yup, Ha still improving at 35 hrs

Just have to double every time

[thorn vault]

That's interesting

[kindred swallow]

I only see black