Log in

View Full Version here: : NGC 300 reprocessed


marc4darkskies
01-11-2022, 01:03 PM
Oh NOOOooooooo ... not another reprocess!! Well, what else am I supposed to do?? :shrug: We've had flooding rain and howling gales down here just yesterday! Besides, it's good practice! :bashcomp:

Anyway, this was an image from 2010. I understand there may be some who prefer the old version (:question:) or even a combination of the two :lol:, but I prefer the improved definition, depth and colour of the new. :D It's gratifying to see what new tools and more experience can do in 12 years!

Old left, new right ...

Full res (zoom & pan): https://www.flickr.com/photos/marc4darkskies/4662092145/in/dateposted-public/

AdamJL
01-11-2022, 01:13 PM
Very good, Marcus! Beautiful colours in both.
Must say, unless IIS has switched the versions around, I prefer the old one as there's less noise in the background.

marc4darkskies
01-11-2022, 01:58 PM
Cheers Adam!

Noise? What noise? :)

You won't be able to discern noise from the reduced res crop I posted (apart from maybe jpeg compression noise).

Attached is a full res crop of old (left) and new that includes some background. You can see the old one has more noise there. You may be thinking that the somewhat more visible stars in the galaxy & background are noise :question:

Please note though, that I did not NoiseXterminate the galaxy after I improved the contrast and I should probably do a little bit of that.

PS: And I don't think you'll be able to see background noise at full res in the new one - unless you're pixel peeping maybe? ;)
I'm happy for you to post a tight crop of an area where you perceive more noise so I can see exactly what you're talking about.

AdamJL
01-11-2022, 03:02 PM
Hi Marcus

I've always wondered if it's a monitor thing that allows us to see different aspects of an image. Plenty of time others have seen things I haven't and vice versa. Even room conditions and light sources change how we see detail.

And maybe noise is the wrong term. Mottling, maybe? I looked at your Flickr image before posting and I can still see it, and it's not anything related to background objects. There's a grey-black mottle in the background (again excluding background object, it's in the background itself) that I don't see (as much) in the old image. I can even see it in your right crop :question:

I can see it in this crop as well.

AdamJL
01-11-2022, 03:07 PM
Here's an extreme stretch of the above. It's the green/purple bits that I see around the image but I see them as light/dark mottling without the stretch.

Anyway, yes I pixel peep, that's why I'm always tinkering ;)

Plus I tend to pixel peep images from good photographers so I can learn as well.

marc4darkskies
01-11-2022, 03:22 PM
Hehe ... that's exactly what I was about to post :lol: Yes indeed, there is some colour noise if you look really closely and stretch the crap out of it :lol: I didn't do any colour noise reduction in the reprocess, nor did I go back to the original data - guilty as charged :ashamed:. Of course you realise I must now go back in and reduce colour noise just for you ;):D

Tough crowd! :)

AdamJL
01-11-2022, 03:26 PM
:D :D :D haha that gave me a good chuckle. Good onya, Marcus. It's a fantastic image, and considering I'm starting to shoot it myself, you've given me a high target to aim for.

marc4darkskies
01-11-2022, 03:58 PM
Appreciate that - I do the same thing :)

So I figured out what it is. I use Digimarc invisible watermarks in my posted images and it embeds colour coding that is visible as some colour noise when you hyper-stretch.

Below is with watermark (left) and without.

And it's official!!! You're the first person to EVER notice my INVISIBLE watermarking! :lol: You win the prize!

PS: And you're SPRUNG for illegally copying a portion of my image (notification below). The cops are on their way!

AdamJL
01-11-2022, 04:16 PM
Oh man, can I just win the lotto instead? :D :D

marc4darkskies
01-11-2022, 08:18 PM
Cheers Adam! Glad you liked it!

I've now posted the digimarc free image. I think they might be going out of business anyway. Not sure.

Dave882
01-11-2022, 10:04 PM
Gosh that’s just unbelievable res. I’m definitely going to pinch that and start selling prints :P

Ryderscope
02-11-2022, 07:56 AM
I definitely understand the need to reprocess old data at the moment Marcus given the lack of available clear sky. Comparing the two versions is good to see what can be done with the same data set. The earlier version has a softer appearance and the new version has more contrast revealing the structures in the galaxy and highlighting the Ha regions. Is see that some star reduction has been applied as well. I fall into the camp of a combination of the two images that retains some of the original look and feel whilst gaining a small amount of contrast enhancement from the new version. Purely subjective of course.

As for entertaining oneself during these times, on the weekend I resorted to testing my mount by emulating meridian flips and sitting there watching it move from one side of the mount to the other and back again. Alas, this is what it has descended to :rolleyes:

marc4darkskies
02-11-2022, 09:05 AM
Cheers Dave - please forward me a 60% cut of your profits! :lol: On a serious note, I did discover a couple of people in Germany stealing images a number of years ago. One was even displaying one in a gallery with the colours changed to look like NB!!! :mad2:



Fair comments Rodney! I prefer to draw out the maximum detail & depth possible without making it look forced. For comparison, see Steve Mohr's wonderfully deep & crisp rendition (https://www.flickr.com/photos/105968790@N03/30783299988/in/pool-3813124@N21) from 2018. Obviously his dataset was far superior to mine. Also see Rob Gendler's softer version from 2008 (http://www.robgendlerastropics.com/NGC300M.html). I think mine is somewhere in between.

Hehe ... watching the mount flip - that's a new one! :lol:

AdamJL
02-11-2022, 09:31 AM
Phwoar.. that shot from Steve is sublime.
I've just noticed you both did 10+ min subs as well. Yikes. I only did 5mins on my initial luminance run. I'll aim for 10mins at the next attempt.

marc4darkskies
02-11-2022, 09:19 PM
Yeah, give 10 a try. There are lots of variables that determine optimal sub length. With my setup 5 would be a little short, even for the brighter galaxies. Even with 10 or 15 subs the fainter elements can be a challenge. Lots of subs helps too of course.

Addos
07-11-2022, 03:11 PM
nice one marcus. good detail in the arms and the stars are pretty tight!

marc4darkskies
08-11-2022, 02:21 PM
Thanks Adam! In the original version from long ago I didn't have the tools and know-how to reduce/refine the stellar profiles. Relatively easy now though! :) I could have done a better job with the repro if I'd gone back to the original master frames - but I was too lazy. Maybe next week! :lol: