First attempt at M31

[Pitch Black Skies]

I made the schoolboy error of not adjusting the ISO back down after finding my target.

It's 52 x 120s, and my first ever 20 flats included with 20 bias, no darks.

ISO 25, 600

Edited November 5, 2021 by Pitch Black Skies

[Sunshine]

I'd be over the moon if it were mine, its great!

[Pitch Black Skies]

4 minutes ago, Sunshine said:

I'd be over the moon if it were mine, its great!

Thank you so much 🙂

[Richard_]

I echo Sunshine, a really good first! You've captured some nice detail in the lanes around the core. What software did you use to stack your subs and edit your final image?

The easiest improvement I'd recommend is to crop out the black bars at the top and bottom of your image as it's a landscape orientation posted in portrait orientation. 

[Pitch Black Skies]

7 hours ago, Richard_ said:

I echo Sunshine, a really good first! You've captured some nice detail in the lanes around the core. What software did you use to stack your subs and edit your final image?

The easiest improvement I'd recommend is to crop out the black bars at the top and bottom of your image as it's a landscape orientation posted in portrait orientation. 

Thanks Richard 🙂

I used DSS and then Startools.

I'm not sure what happened, I think I screenshot it.

[raadoo]

I had a stab at processing it (hope you don't mind) and I noticed a couple of things:

• there seems to be some backfocus issue with the sensor being to far back (stars in the corners take are circularly distorted around the center of the image)
• 120s @ ISO 25,600 is far too much exposure for Andromeda's core which, subsequently, got blown out. Now, there's not much you can do about blown highlights, so, when that happens to me, I just lean into it rather than try and fix it by going ham on the Highlights adjustment, as this leads to grey light instead of white and, to me, looks off.
• interestingly, even if the stars are not fully in focus, the arms have a ton of detail in them, which is a very cool surprise

My first attempt at M31 didn't even include a tracker of any kind, just a 300mm f/5.6 lens, a DSLR, and me manually cranking out 5s exposure like an absolute madman and the result is more akin to how you'd see Andromeda through a pair of binoculars. All that to say, really good job! 👏

[Pitch Black Skies]

11 hours ago, raadoo said:

I had a stab at processing it (hope you don't mind) and I noticed a couple of things:

• there seems to be some backfocus issue with the sensor being to far back (stars in the corners take are circularly distorted around the center of the image)
• 120s @ ISO 25,600 is far too much exposure for Andromeda's core which, subsequently, got blown out. Now, there's not much you can do about blown highlights, so, when that happens to me, I just lean into it rather than try and fix it by going ham on the Highlights adjustment, as this leads to grey light instead of white and, to me, looks off.
• interestingly, even if the stars are not fully in focus, the arms have a ton of detail in them, which is a very cool surprise

My first attempt at M31 didn't even include a tracker of any kind, just a 300mm f/5.6 lens, a DSLR, and me manually cranking out 5s exposure like an absolute madman and the result is more akin to how you'd see Andromeda through a pair of binoculars. All that to say, really good job! 👏

Wow, thanks so much! That is absolutely amazing.

I had a suspicion that my coma corrector might need to move a mm or 2 closer to the sensor. I think I had the focuser thumbscrew lock too tight which was tilting the focuser tube ever so slightly. I'm waiting for a clear night to do some more experimenting.

How did you get so much detail? Could I do this myself in Startools?

(Cranking out 5sec exposures like an absolute madman 😂 that got me).

Edited November 7, 2021 by Pitch Black Skies

[raadoo]

3 minutes ago, Pitch Black Skies said:

Wow, thanks so much! That is absolutely amazing.

I had a suspicion that my coma corrector might need to move a mm or 2 closer to the sensor.

How did you get so much detail? Could I do this myself in Startools?

(Cranking out 5sec exposures like an absolute madman 😂 that got me).

I'll try and do a quick write-up of what I did, but keep in mind, this was all done on a non-linear, compressed JPG, so doing the below on the original data should yield much better results, both in terms of noise and star reduction, and detail enhancement. I've also attached a screenshot of the layers panel in Photoshop, so hopefully it's not completely dry reading 😀

• saved the JPG as a 16bit TIF (this was necessary for the next step and further edits being as non-destructive as they can be)
• used Starnet++ to separate the stars from the nebulosity
• opened both starless and original in Photoshop (from here on out, it's just Photoshop, by the way) and did the classic Difference blending of the two to create a Nebulosity layer and Stars layer
• Nebulosity:
  • applied frequency separation so I can work on the detail enhancement without affecting colour
  • colour layer was left mostly untouched until later steps (though I did do some White Balance and slight saturation adjustments in Camera Raw)
  • for the detail layer I mostly used High Pass and Camera Raw to sharpen and enhance texture and detail, while also throwing in the odd Noise Reduction along the way
  • in the detail layer (which is mostly 50% grey by this time, bar the actual detail in the galaxy), I also took a grey brush to the core of the galaxy, to remove the banding that was happening in the original
  • merged the frequency separated layers and did more cosmetic / colour work using Camera Raw together with some blended LUTs
• Stars:
  • slightly increased the saturation as the whole Starnet++ process usually leads to undersaturated stars
  • created a Highlights mask which was then used to reduce most of the stars' bloat
  • a few of the stars have diffraction spikes that the mask messed up, so I manually went in and brought back the original spikes for those
• merged the whole thing and uploaded it

My experience with StarTools has been very limited, but from what I can tell you could probably replicate the steps above without much hassle and with likely better results; especially as there seems to be a way to use Starnet++ in linear mode!

[alacant]

3 hours ago, Pitch Black Skies said:

myself in Startools?

Hi

Nice shot.

In StarTools, don't forget to leave room for the fainter bits of the galaxy. I reversed the development and included some of the outer areas for the second AutoDev.

Don't forget to correct the stars! The new SVDecon (available in the latest version) in the latest beta works very well. I'll attach a screen of it.

Agree with the cc closer comments. Then you'd have no need to correct the stars.

Cheers and HTH

 

Edited November 7, 2021 by alacant

[Pitch Black Skies]

9 hours ago, raadoo said:

I'll try and do a quick write-up of what I did, but keep in mind, this was all done on a non-linear, compressed JPG, so doing the below on the original data should yield much better results, both in terms of noise and star reduction, and detail enhancement. I've also attached a screenshot of the layers panel in Photoshop, so hopefully it's not completely dry reading 😀

• saved the JPG as a 16bit TIF (this was necessary for the next step and further edits being as non-destructive as they can be)
• used Starnet++ to separate the stars from the nebulosity
• opened both starless and original in Photoshop (from here on out, it's just Photoshop, by the way) and did the classic Difference blending of the two to create a Nebulosity layer and Stars layer
• Nebulosity:
  • applied frequency separation so I can work on the detail enhancement without affecting colour
  • colour layer was left mostly untouched until later steps (though I did do some White Balance and slight saturation adjustments in Camera Raw)
  • for the detail layer I mostly used High Pass and Camera Raw to sharpen and enhance texture and detail, while also throwing in the odd Noise Reduction along the way
  • in the detail layer (which is mostly 50% grey by this time, bar the actual detail in the galaxy), I also took a grey brush to the core of the galaxy, to remove the banding that was happening in the original
  • merged the frequency separated layers and did more cosmetic / colour work using Camera Raw together with some blended LUTs
• Stars:
  • slightly increased the saturation as the whole Starnet++ process usually leads to undersaturated stars
  • created a Highlights mask which was then used to reduce most of the stars' bloat
  • a few of the stars have diffraction spikes that the mask messed up, so I manually went in and brought back the original spikes for those
• merged the whole thing and uploaded it

My experience with StarTools has been very limited, but from what I can tell you could probably replicate the steps above without much hassle and with likely better results; especially as there seems to be a way to use Starnet++ in linear mode!

To be honest a lot of that has gone completely over my head but I will revisit in the future and can hopefully make some sense of it when I have accumulated more knowledge and experience.

For my image I just copied this gentleman's tutorial on YouTube. That is all I am really capable of at the moment.