M63 revisit

[Rodd]

Another case of me coming to my senses--I realized the data just could not support what I wanted from it.  The data is not nearly as good as my other reprocesses--it was one of he first images I took with the TIA 130 and asi 1600.  It was shot at F7.7, and could use allot more data with longer exposures, which were 2min (R: 70, G: 97, B:71, L:190.  Again, the goal was realism and I think I succeeded.  New version then Old..  I took this image with the sensor rotated with the galaxy on its side, so rotating it this way kind of effects the size it is portrayed at by the forums.  I depicts it too large.But this is the best way to look at it.  The old one looks sharper, but it feels heavy handed to me--chuncky, not very gaseous.  The new one looks more real to me.

New

Old

 

 

 

[Laurin Dave]

Agree Rod, your new one looks better to me, more subtle and pleasing on the eye. 

 

Dave

[wimvb]

+1 for the rework. But this galaxy is well worth a revisit. Maybe shoot with longer exposures? It seems that it has a lot more to give.

Btw, since the galaxy is well separated from the edges, I would cut away the offending dust donut near the bottom.

[Rodd]

1 hour ago, Laurin Dave said:

Agree Rod, your new one looks better to me, more subtle and pleasing on the eye. 

 

Dave

Thank Dave

Rodd

[Rodd]

13 minutes ago, wimvb said:

+1 for the rework. But this galaxy is well worth a revisit. Maybe shoot with longer exposures? It seems that it has a lot more to give.

Btw, since the galaxy is well separated from the edges, I would cut away the offending dust donut near the bottom.

If I crop it too much it will change the scale and it will post too large--well beyond the zoom it can support.

Rodd

[wimvb]

There are two ways around this:

1. resample in PI before posting

2. when inserting the image in a post, ctrl + right-click and edit the image. Resize to about 400. This will show the image smaller (but still allow zooming in).

[Rodd]

1 hour ago, wimvb said:

There are two ways around this:

1. resample in PI before posting

2. when inserting the image in a post, ctrl + right-click and edit the image. Resize to about 400. This will show the image smaller (but still allow zooming in).

Thanks.. I will keep that in mind.  Sometimes full resolution is too much--though I try to process my images to look good under full resolution.  The ones I can't are the ones that I have the most problems with

Rodd

[MarsG76]

The new one is definitely better, much better.... looks very nice.

 

[Rodd]

45 minutes ago, MarsG76 said:

The new one is definitely better, much better.... looks very nice.

 

Thanks MG

Rodd

[gorann]

Yes Rood, the new version is lovely and a bit better, not the least the sky is better, but I also liked the old one. I have to say I did not notice the dust doughnut until Wim pointed it out and even then it is very hard to see on my screen in the morning light.

I noticed that since a week or so my images are posted in a larger format here on SGL (which is sometimes too revealing). I assume something in the system has been modified. I will try Wim's right click suggestion (if it works on a Mac).

[wimvb]

29 minutes ago, gorann said:

I will try Wim's right click suggestion (if it works on a Mac).

Maybe CMD and right click.

But resampling is the safest way. Or in GIMP/ PS, the scale function.

[Rodd]

4 hours ago, gorann said:

Yes Rood, the new version is lovely and a bit better, not the least the sky is better, but I also liked the old one. I have to say I did not notice the dust doughnut until Wim pointed it out and even then it is very hard to see on my screen in the morning light.

I noticed that since a week or so my images are posted in a larger format here on SGL (which is sometimes too revealing). I assume something in the system has been modified. I will try Wim's right click suggestion (if it works on a Mac).

Thanks Goran.  I too have trouble seeing the dust bunny (though I know its there because the calibrated subs have it--I had not hey perfected my flats).  The old image is ok....but I don't think it is supposed to be so red.  Then again, its not supposed to be purple either and the new one has that.  So, I guess the bottom line is they are both questionable.  The asi 1600 and the TOA 130 at F7.7 has a pixel scale of .78--the C11Edge reduced by .7x has a pixel scale of .57--so maybe I was asking too much of it.  The bad news is I don't think reprocessing this image again is likely to make much of a difference.  the good news is I will have to shoot this target again!.  BTW......you can add words to your spell checker...even names!?

Ro(dd)

[Rodd]

4 hours ago, wimvb said:

Maybe CMD and right click.

But resampling is the safest way. Or in GIMP/ PS, the scale function.

I have some trouble with this......Part of me says that the best images stand up perfectly to full resolution.  In the best images a sense of depth prevails at full resolution--more so than at normal viewing it seems.  As you get closer, the field seems to  gets deeper.  If my image can't stand up to full viewing, then it is flawed, and posting it at a reduced scale will not satisfy me-especially if I have 20-30 hours of exposure time and it still has a grainy, blotchy, pixelated (my usual complaints) background.  But part of me says  why not view at the best scale for the image.  Demon on my left--angle on my right.  Now--for drizzled images I think reducing a bit is necessary--even if they stand up perfectly to full resolution, because they are so big and when zoomed without rescale, you can't see much--so its useless for viewing (great for pixel peeping ghowever).

Rodd

[wimvb]

17 hours ago, Rodd said:

It was shot at F7.7, and could use allot more data with longer exposures, which were 2min (R: 70, G: 97, B:71, L:190. 

14.3 hours if I calculate correctly

2 hours ago, Rodd said:

.Part of me says that the best images stand up perfectly to full resolution.

I agree, but there is also "Lodriguss' devil": if you image under light polluted skies, you need to gather 2.5 times more data for every magnitude you lost, compared to a dark site. An additional problem is that you can't just take longer exposures. And sometimes you need longer exposures to get to the really faint stuff. Even if a cooled, low noise cmos allows short exposures, there may be a limit. As you remarked in your original post, the image could use longer exposures.

2 hours ago, Rodd said:

a grainy, blotchy, pixelated (my usual complaints) background.

My recipe (which can't replace the best noise reduction, more data):

(I view colour images as lightness (luminance) and colour (chrominance))

A grainy background is usually noise in luminance. My weapon of choice is TGVDenoise ON LUMINANCE ONLY in the linear scale, and once more after star reduction in the final stages of processing.

Blotchiness is ususally noise in chrominance. For this I use MMT ON CHROMA ONLY in the linear stage (after TGVDenoise) with noise reduction on 8 layers. This kills any blotchy colours in the background.

[peter shah]

big improvement over the first process...lovely image

[Rodd]

31 minutes ago, wimvb said:

Blotchiness is ususally noise in chrominance. For this I use MMT ON CHROMA ONLY in the linear stage (after TGVDenoise) with noise reduction on 8 layers. This kills any blotchy colours in the background.

Thanks-I'll have to give this a try

Rodd

[Rodd]

28 minutes ago, peter shah said:

big improvement over the first process...lovely image

Thanks Peter!

Rodd