Rosette Nebula in Ha

[Rodd]

3 hours ago, moise212 said:

Hi Rodd,

I much prefer the first version(s) where the dark areas/dust are/is not so black. Impressive image nevertheless, very clean, it's one of the ones of the best for my taste.

Alex

Thanks Alex.  I agree the bok globules should not be so dark--but I was trying to get the center hollow to be a bit darker and it is hard to do that without effecting the bok globules.  I know the number of versions is a turnoff (now), but I like the blend.  But it matters not--because this image will be a color image and I process the Ha channel completely differently when I use it as a lum channel.  I just like emphasizing the little details visible in an image from a 4" scope---I am that kid who won the golden ticket to Willy Wonka's chocolate factory.

[Rodd]

2 hours ago, Singlin said:

For my taste it seems a bit over cooked. I would have halved the subs to 15min.

What does it look like when you decrease the exposere in photoshop?

Thanks for looking and commenting Singlin.  But nah...30 min subs will definitely be better in an HaSHO or HaRGB image--which is the real goal.  I use Pixinsight, not Photoshop.  I agree the last version is too well done---but I think the blend is better than version #1

[Singlin]

I look forward to seeing The Ha rgb version where your Ha version will be better placed.

[Rodd]

47 minutes ago, Singlin said:

I look forward to seeing The Ha rgb version where your Ha version will be better placed.

It will probably be an HaSHO--at least to start. 

[ollypenrice]

Back on the original question about the effect of moonlight, I compared this carefully with mine. Unfortunately my shoot was compromized when I noticed that my PA had taken a nudge so I was stuck with 15 minute subs (24 of them) till I could fix it. I normally use 30 minutes in Ha and am convinced they go deeper. I don't see any reason to believe Rodd's data is saturated. Rodd's Ha signal goes deeper than mine but I think my background is darker - because it wasn't shot in moonlight. I think Rodd's data is excellent, though, and my guess is that the moon has done very little damage, just maybe lifting the darkest regions by a little.

The biggest difference is that I have a tad more contrast in the faint extension to the lower right in Rodd's orientation, some of which is down to my having less signal. I do think some of it is down to the moon, though.

As Barry says, you need to know when to stop. I feel an image should look as if the data has given 98% of what it has. 101% is 20% too much, not 3% too much!!

Olly

[Rodd]

7 minutes ago, ollypenrice said:

Back on the original question about the effect of moonlight, I compared this carefully with mine. Unfortunately my shoot was compromized when I noticed that my PA had taken a nudge so I was stuck with 15 minute subs (24 of them) till I could fix it. I normally use 30 minutes in Ha and am convinced they go deeper. I don't see any reason to believe Rodd's data is saturated. Rodd's Ha signal goes deeper than mine but I think my background is darker - because it wasn't shot in moonlight. I think Rodd's data is excellent, though, and my guess is that the moon has done very little damage, just maybe lifting the darkest regions by a little.

The biggest difference is that I have a tad more contrast in the faint extension to the lower right in Rodd's orientation, some of which is down to my having less signal. I do think some of it is down to the moon, though.

As Barry says, you need to know when to stop. I feel an image should look as if the data has given 98% of what it has. 101% is 20% too much, not 3% too much!!

Olly

Thanks Olly--That's a great way to put it regarding the data.  I think you are right about the Moon too.  When the Rosette had set I shot a few 30 min Ha subs of the Leo Triplets just to see if it would pay to add Ha to an RGB image.  I compared a sub that was shot with the Moon still up with one that was shot after the Moon had set and there was a noticeable difference in the darkness of the background in the raw sub taken after the Moon had set--and that was for a target more than 90 degrees removed from the Moon.  I might get a few more Ha subs of the Rosette without the Moon if nothing but to decrease the brightness in the center hollow a bit.  But adding the other filters might take care of that as well.

By the way--I was pleasantly surprised at the strength of the Ha signal in the Triplets.

[Rodd]

3 hours ago, ollypenrice said:

Back on the original question about the effect of moonlight, I compared this carefully with mine. Unfortunately my shoot was compromized when I noticed that my PA had taken a nudge so I was stuck with 15 minute subs (24 of them) till I could fix it. I normally use 30 minutes in Ha and am convinced they go deeper. I don't see any reason to believe Rodd's data is saturated. Rodd's Ha signal goes deeper than mine but I think my background is darker - because it wasn't shot in moonlight. I think Rodd's data is excellent, though, and my guess is that the moon has done very little damage, just maybe lifting the darkest regions by a little.

The biggest difference is that I have a tad more contrast in the faint extension to the lower right in Rodd's orientation, some of which is down to my having less signal. I do think some of it is down to the moon, though.

As Barry says, you need to know when to stop. I feel an image should look as if the data has given 98% of what it has. 101% is 20% too much, not 3% too much!!

Olly

How about this one Olly--darker background--deeper outer nebula.  Much less processing (not overcooked)

[ollypenrice]

No no no, that has flattened the bright stuff almost totally. Here's something you might find useful in comparing versions. Just paste one over the other in Photoshop and blink the top layer on and off. If you used that trick you'd chuck the last version in the bin in less than one second!

Olly

[Rodd]

7 minutes ago, ollypenrice said:

No no no, that has flattened the bright stuff almost totally. Here's something you might find useful in comparing versions. Just paste one over the other in Photoshop and blink the top layer on and off. If you used that trick you'd chuck the last version in the bin in less than one second!

Olly

Then how about this--I was trying to refrain from over processing  I can still go further if this is not far enough

[ollypenrice]

Rodd, honestly, too many posts! It's not that we don't want to look at them, it is that you need to become more judcious in looking at them. The picture is your picture. Take more time and become adept at deciding.

Olly

[Rodd]

2 minutes ago, ollypenrice said:

Rodd, honestly, too many posts! It's not that we don't want to look at them, it is that you need to become more judcious in looking at them. The picture is your picture. Take more time and become adept at deciding.

Olly

That's ok Olly--but I like to post because it lets me enlarge the image and look at it in a way that I cannot do just by zooming.  I will refrain from quoting people and will not get upset if no one comments.  At least one person (Maximidius) said he liked to see the process.  To end--I think this image is pretty decent--white stuff has lots of range now.  thank you for pointing that out.

[ollypenrice]

18 minutes ago, Rodd said:

That's ok Olly--but I like to post because it lets me enlarge the image and look at it in a way that I cannot do just by zooming.  I will refrain from quoting people and will not get upset if no one comments.  At least one person (Maximidius) said he liked to see the process.  To end--I think this image is pretty decent--white stuff has lots of range now.  thank you for pointing that out.

It's fantastically deep but has lost contrast in the bright parts. Walk away, come back in a month. You have to walk away. I have images which I loved when I did them and now I think, Whaaaat????

Olly

[Rodd]

1 hour ago, ollypenrice said:

It's fantastically deep but has lost contrast in the bright parts. Walk away, come back in a month. You have to walk away. I have images which I loved when I did them and now I think, Whaaaat????

Olly

I know the feeling--happens to me with every image.  I would like to end this train wreck by jumping off and tossing my last picture into the flames behind me.  This is a blend between the new and the old.  I think it has something to offer.  A month from now I will begin to tweak it if it needs it.  Uncannily deep though.

[Maximidius]

Hey Rod

Man, you have some image there. I took the liberty of downloading your first image and looked at it very closely, I am in awe at the quality of your data..I am no expert, but those 30mins subs really have picked up all there is to pick up. I'll jump straight in and say, I'm not keen on the last couple because they look like they have gone too far. I think the reason for the dilemma is because you have so much Ha signal present in the data it's easy to wash out the contrast while striving to show all of that faint gas, it's a difficult balance. And for me that's where the depth you're looking for lies. It's a gas and has pockets of density that change through distance and the only real way to duplicate that on a 2d surface is through tonal range when we look at it through NB and grayscale,,imho. Bok globules are opaque to light when they reach a certain density and thus will show transitions through tone from dark grey to near black, which yours do in the first 3 posts. I think as long as you bear in mind the physical characteristics of what you are imaging while seeking to use artistic license to portray this image in your own unique style, you'll get what you are looking for. 

So heres what i did with your first image. 

Opened it up and checked the levels. I noticed that the white point may have been very very very slightly clipped. 

I opened it in camera raw and highlights showed overexposure in the stars but not in the gas. So i dropped the highlights bar to bring the stars in line and then a little more for the gas. What this does is compress the data a tad and move it back towards the midpoint increasing the white point

I then dropped the whites to reduce the washed out appearance and decreased the shadows bar towards the black end a touch.

Increased the clarity a touch

I then created a layer copy and with curves slightly reduced the curve value to reduce the Ha and make a darker center hole. and masked this so only the hole was blended at about 70%

I then went back and reduced the highlights a touch more for personal taste. 

Your original blackpoint was at 91% and this image is at 92% so the background space is just a touch darker. 

So i think what this did is increase the tonal range across the image thus portraying a sense of depth and reduced the washed out appearance from the exceptionally strong HA signal that you have. This is only my interpretation from what you have written and are striving for and of course is subject to personal taste, so it may not be to the tastes of everyone. But, i sincerely hope that it helps with what you are trying to achieve. 

[Rodd]

56 minutes ago, Maximidius said:

Hey Rod

Man, you have some image there. I took the liberty of downloading your first image and looked at it very closely, I am in awe at the quality of your data..I am no expert, but those 30mins subs really have picked up all there is to pick up. I'll jump straight in and say, I'm not keen on the last couple because they look like they have gone too far. I think the reason for the dilemma is because you have so much Ha signal present in the data it's easy to wash out the contrast while striving to show all of that faint gas, it's a difficult balance. And for me that's where the depth you're looking for lies. It's a gas and has pockets of density that change through distance and the only real way to duplicate that on a 2d surface is through tonal range when we look at it through NB and grayscale,,imho. Bok globules are opaque to light when they reach a certain density and thus will show transitions through tone from dark grey to near black, which yours do in the first 3 posts. I think as long as you bear in mind the physical characteristics of what you are imaging while seeking to use artistic license to portray this image in your own unique style, you'll get what you are looking for. 

So heres what i did with your first image. 

Opened it up and checked the levels. I noticed that the white point may have been very very very slightly clipped. 

I opened it in camera raw and highlights showed overexposure in the stars but not in the gas. So i dropped the highlights bar to bring the stars in line and then a little more for the gas. What this does is compress the data a tad and move it back towards the midpoint increasing the white point

I then dropped the whites to reduce the washed out appearance and decreased the shadows bar towards the black end a touch.

Increased the clarity a touch

I then created a layer copy and with curves slightly reduced the curve value to reduce the Ha and make a darker center hole. and masked this so only the hole was blended at about 70%

I then went back and reduced the highlights a touch more for personal taste. 

Your original blackpoint was at 91% and this image is at 92% so the background space is just a touch darker. 

So i think what this did is increase the tonal range across the image thus portraying a sense of depth and reduced the washed out appearance from the exceptionally strong HA signal that you have. This is only my interpretation from what you have written and are striving for and of course is subject to personal taste, so it may not be to the tastes of everyone. But, i sincerely hope that it helps with what you are trying to achieve. 

I like it--I just have to learn to do all that in Pixinsight, for to that I am committed.  Do it with the last one I posted--The last one is a blend of the first and another, so there is some of the one you like in it. 

[Maximidius]

Something like this ? 

 

[Rodd]

21 minutes ago, Maximidius said:

Something like this ? 

 

 

[Rodd]

23 minutes ago, Maximidius said:

Something like this ? 

 

Now you have the disease my friend--I think YOU went a bit too far.  How to do that in PI?  I think it needs color.  Unfortunately it will be some weeks before I can do it. 

[Maximidius]

Interesting that you think that. The adjustments I made, as requested, were less than the previous adjustments, which illustrates a point. The first image is probably the best one Imho and according to the fact that you liked it. 

PS with regard to PI i havent got a clue, i've never used it. It'll give it a go someday soon though so watch out, i'll be tapping you up for lots of advice.